Structure and optical function of amorphous photonic ... · distributions of scatterer size and...

doi: 10.1098/rsif.2012.0191 published online 9 May 2012J. R. Soc. Interface

Cao, Eric R. Dufresne and Richard O. PrumVinodkumar Saranathan, Jason D. Forster, Heeso Noh, Seng-Fatt Liew, Simon G. J. Mochrie, Hui angle X-ray scattering (SAXS) analysis of 230 bird speciesnanostructures from avian feather barbs: a comparative small Structure and optical function of amorphous photonic

Supplementary data

l http://rsif.royalsocietypublishing.org/content/suppl/2012/05/07/rsif.2012.0191.DC1.htm

"Data Supplement"

Referencesref-list-1http://rsif.royalsocietypublishing.org/content/early/2012/05/02/rsif.2012.0191.full.html#

This article cites 76 articles, 16 of which can be accessed free

P

J. R. Soc. Interface

on May 9, 2012rsif.royalsocietypublishing.orgDownloaded from

*Authors forichard.prum@Present addDepartmentOxford OX1OX1 3JA, UK

Electronic sup10.1098/rsif.2

doi:10.1098/rsif.2012.0191Published online

Received 9 MAccepted 17 A

Structure and optical function ofamorphous photonic nanostructures

from avian feather barbs: acomparative small angle X-ray

scattering (SAXS) analysis of 230bird species

Vinodkumar Saranathan1,5,*,, Jason D. Forster2,5, Heeso Noh3,5,Seng-Fatt Liew3,5, Simon G. J. Mochrie35, Hui Cao35,

Eric R. Dufresne2,4,5 and Richard O. Prum1,5,6,*1Department of Ecology and Evolutionary Biology and Peabody Museum of Natural History,

2Department of Mechanical Engineering and Materials Science,3Department of Applied Physics, 4Department of Physics, and 5Center for Research on

Interface Structures and Phenomena (CRISP), Yale University, New Haven, CT 06520, USA6Donostia International Physics Center (DIPC), 20018 Donostia-San Sebastian, Spain

Non-iridescent structural colours of feathers are a diverse and an important part of the phe-notype of many birds. These colours are generally produced by three-dimensional, amorphous(or quasi-ordered) spongy b-keratin and air nanostructures found in the medullary cells offeather barbs. Two main classes of three-dimensional barb nanostructures are known, charac-terized by a tortuous network of air channels or a close packing of spheroidal air cavities.Using synchrotron small angle X-ray scattering (SAXS) and optical spectrophotometry, wecharacterized the nanostructure and optical function of 297 distinctly coloured feathersfrom 230 species belonging to 163 genera in 51 avian families. The SAXS data provided quan-titative diagnoses of the channel- and sphere-type nanostructures, and confirmed the presenceof a predominant, isotropic length scale of variation in refractive index that produces strongreinforcement of a narrow band of scattered wavelengths. The SAXS structural data identifieda new class of rudimentary or weakly nanostructured feathers responsible for slate-grey, andblue-grey structural colours. SAXS structural data provided good predictions of the single-scattering peak of the optical reflectance of the feathers. The SAXS structural measurementsof channel- and sphere-type nanostructures are also similar to experimental scattering datafrom synthetic soft matter systems that self-assemble by phase separation. These results furthersupport the hypothesis that colour-producing protein and air nanostructures in feather barbsare probably self-assembled by arrested phase separation of polymerizing b-keratin from thecytoplasm of medullary cells. Such avian amorphous photonic nanostructures with isotropicoptical properties may provide biomimetic inspiration for photonic technology.

Keywords: biophotonics; organismal structural colours; amorphousnanostructures; non-iridescence; single scattering; self-assembly

1. INTRODUCTION

Structural colours are prevalent in nature, and gener-ally produced by the selective scattering and

r correspondence ([email protected];yale.edu).

ress: Edward Grey Institute of Field Ornithology,of Zoology, University of Oxford, South Parks Road,3PS, UK and Linacre College, St Cross Road, Oxford.

plementary material is available at http://dx.doi.org/012.0191 or via http://rsif.royalsocietypublishing.org.

arch 2012pril 2012 1

reinforcement of specific bands of wavelengths from bio-photonic nanostructures with variations in refractiveindex on the order of visible wavelengths of light[15]. Like photonic crystals [6], biophotonic nanostruc-tures vary in nanostructure in either one, two or threedimensions (figure 1ac). However, they may alsovary in whether they have long-range, crystallineperiodicity, or only short-range (nearest neighbour),structural correlations [15] (figure 1d f ). The latterreferred to as quasi-ordered or amorphous biophoto-nic nanostructures are characterized by unimodal

This journal is q 2012 The Royal Society
mailto:[email protected]:[email protected]://dx.doi.org/10.1098/rsif.2012.0191http://dx.doi.org/10.1098/rsif.2012.0191http://dx.doi.org/10.1098/rsif.2012.0191http://rsif.royalsocietypublishing.orghttp://rsif.royalsocietypublishing.orghttp://rsif.royalsocietypublishing.org/

crystal-like(biophotoniccrystals)long-range order

(a)one-dimension two-dimension

spatial variation in refractive indexthree-dimension

(b) (c)

(d ) (e) ( f )

quasi-ordered(amorphous photonicbiomaterials)short-range order

rang

e of

spa

tial p

erio

dici

ty

Figure 1. A classification of biophotonic nanostructural diversity based on the dimensionality of spatial variation in refractiveindex and its range of periodicity. (ac) One-, two- and three-dimensional biological photonic crystals with long-range periodicorder in refractive index modulations (after Joannopoulos [6]). (d) A chirped lamellar stack, a one-dimensional quasi-orderednanostructure with short-range spatial periodicity (currently unknown in birds; after Parker [2]). (e) TEM cross section of atwo-dimensional amorphous or quasi-ordered nanostructure with short-range order comprising of parallel collagen fibres in amucopolysaccharide matrix from the green tongue of magnificent bird-of-paradise (Cicinnurus magnificus, Paradisaeidae).( f ) TEM cross section of a three-dimensional amorphous or quasi-ordered nanostructure of b-keratin and spheroidal air vacuolesfrom the spongy medullary cells of the azure blue crown feather barbs of male Blue-crowned Manakin (Lepidothrix coronata,Pipridae) with short-range quasi-periodic order.

2 Avian amorphous photonic nanostructures V. Saranathan et al.


distributions of scatterer size and inter-scattererspacing, and a notable lack of any underlying period-icity beyond the span of a few nearest neighbours(figure 2e,f ) [3,79].

Vivid, non-iridescent structural colours in bird plu-mages (figure 2b,c) are taxonomically widespread andappear to have evolved numerous independent timesduring the evolutionary history of birds in many ecolo-gically diverse lineages [3,9]. They constitute animportant component of the phenotype of manybirds, and are frequently used in intraspecific communi-cation and camouflage (reviewed in [10,11]). Becausethe optical properties of biophotonic materials are inti-mately tied to the underlying nanostructures, a precisemechanistic characterization of organismal structuralcolour production is critical to study how such biologicalsignals evolve.

The non-iridescent structural colours of avian featherbarbs are generally produced by three-dimensional, quasi-ordered nanostructures composed of b-keratin (refractiveindex, n 1.58+0.01; [12]) and air in the medullarycells of feather barbs (reviewed in [3]). The colours so pro-duced do not appreciably change in hue with changes inangle of observation under natural lighting because back-scattered light dominates under such conditions [13].Spongy barb nanostructures have been directly examinedin only a relatively small number of avian taxa (approx.28 species from 16 families [3]), but typically known tooccur in one of two distinct morphologies. The channel-type nanostructures are characterized by a tortuous,interconnected bicontinuous network of air channels andb-keratin bars of similar widths and shapes (figure 2e)[3,79,14]. The sphere-type nanostructures consist of aquasi-ordered close packing of spheroidal air cavities thatare separated by b-keratin walls and frequently intercon-nected by tiny air passages (figure 2f ). However, DAlbaet al. [15] recently discovered a unique two-dimensional,medullary, feather barb nanostructure comprising bundles


of parallel, quasi-ordered,b-keratin nanofibres in air that isresponsible for the non-iridescent blue colour of BluePenguin (Eudyptula minor, Spheniscidae).

1.1. Small angle X-ray scattering (SAXS)

Previous research using Fourier analysis of two-dimensional transmission electron microscopy (TEM)images of amorphous barb nanostructures has documentedtheir isotropic quasi-order from their ring-shaped two-dimensional Fourier power spectra [3,79,16]. Fourieranalyses of TEM images were sufficiently accurate to falsifythe century-old, single particle (Tyndall or Rayleigh, andMie) scattering hypotheses, which assumed that thecolour comes from wavelength-dependent light scatteringproperties of isolated, spatially uncorrelated scatterers[3,79,16]. But two-dimensional Fourier power spectra ofTEMimages lack the resolution toaccount for thevariationin reflectance features of these complex three-dimensionalnanostructures [3,79,1720]. They also suffer from arte-facts owing to EM sample shrinkage and others related toanalysing a finite-thickness (approx. 90 nm), low-resol-ution, two-dimensional slice of a three-dimensionalnanostructure, such as aliasing, binning, etc. Nevertheless,most studies of avian structural colours have used two-dimensional electron microscopy to characterize theirunderlying three-dimensional biophotonic nanostructures[3,79,14, 1719,2125]. However, fundamental uncer-tainty remains about the exact organization of thesethree-dimensional amorphous feather barbnanostructures.Shawkey et al. [26] recently performed three-dimensionalelectron tomographic reconstruction of the channel-typebarb nanostructure in blue rump feathers of Eastern Blue-bird (Sialia sialis), and made a reasonable prediction of theoptical reflectance from the azimuthal average of the three-dimensional Fourier transform of the tomogram. However,sample shrinkage and tomographic distortion limited theaccuracy of structural and optical analyses [26,27].
http://rsif.royalsocietypublishing.org/

(a) (b) (c)

(d) (e) ( f )

(g) (h) (i)

Figure 2. Diversity of non-iridescent feather barb structural colours in birds and morphology of their underlying three-dimensionalamorphous photonic nanostructures with short-range quasi-periodic order. (a) Female Silver-breasted Broadbill (Serilophus lunatus,Eurylaimidae). (b) Male Eastern Bluebird (S. sialis, Turdidae). (c) Male Plum-throated Cotinga (Cotinga maynana, Cotingidae).(d) SEM image of a rudimentary nanostructure with a very thin layer (1 mm or less) of a disordered network of spongy b-keratinbars present at the periphery of the medullary barb cells from the pale blue-grey primary coverts of S. lunatus, (e) TEM image of achannel-type b-keratin and air nanostructure from royal blue back contour feather barbs of S. sialis. ( f ) TEM image of a sphere-type b-keratin and air nanostructure from the dark turquoise blue back contour feather barbs of C. maynana. (g i) Representativetwo-dimensional small-angle X-ray scattering (SAXS) diffraction patterns for the rudimentary, channel- and sphere-type featherbarb nanostructures in (d f ), respectively. The SAXS patterns for both channel- and sphere-type nanostructures exhibit ring-like features that demonstrate the isotropy and short-range spatial periodicity of these nanostructures, whereas the rudimentarybarb nanostructure shows a diffuse, disc-like pattern. The false colour encoding corresponds to the logarithm of the X-ray scatteringintensity. Scale bars: (d) 250 nm; (e,f) 500 nm; (g i) 0.05 nm21. Photo credits: (a) Yiwen Yiwen (image in the public domain);(b) Ken Thomas (image in the public domain); and (c) Thomas Valqui (reproduced with permission).

Avian amorphous photonic nanostructures V. Saranathan et al. 3


Here, we use synchrotron small angle X-ray scattering(SAXS) to quantitatively characterize the nanostructureand optical function of a large sample of structurallycoloured feathers with spongy barb nanostructures,from across the phylogeny of birds. We examine thenanostructure and optical properties of 297 distinc-tly coloured feathers from 230 species belonging to163 genera in 51 avian families (see electronic sup-plementary material, table S1). SAXS is a precisionstructural tool routinely used in material science todirectly measure bulk structural correlations in com-plex nanostructured morphologies [2831]. SAXSenables a direct experimental measurement of the two-dimensional projection of the three-dimensional Fouriertransform of the scattering structure (figure 3) withessentially no sample preparation, allowing for rapidthroughput inconceivable with electron microscopymethods [29,31,32]. The azimuthal average of theSAXS pattern gives the X-ray scattering intensity as a


function of q, the scattering wavevector, or spatialfrequency of variation in electron density (which is aproxy for variation in refractive index). The SAXSpatterns resolve spatial correlations of dimensions 2p/qthat range from a few tens to several hundred nano-metres (figures 35). X-rays also interact only weaklywith soft biological tissues because of the relatively lowelectron density of biological media [2831,35]. Hence,SAXS provides single scattering data that are highlysuited to quantitatively predict the interactions ofvisible light with the nanostructure without artefactsresulting from multiple scattering. Recently, we appliedSAXS to a few species with non-iridescent featherbarb structural coloursEastern Bluebird (S. sialis),Purple-throated Cotinga (Cotinga maynana), BlueCotinga (Cotinga cotinga), Asian Fairy Bluebird(Irena puella), Indian Roller (Coracias benghalensis)and Blue Penguin (Eudyptula minor)and successfullymodelled the directional light scattering properties of

101

q (nm1)

Porod asymptote (q4)

M. caerulescensA. clypeataG. victoria (unstructured)

H. mustelina (control)

101

A. laminirostris

scal

ed s

catte

ring

inte

nsity

visible q

(a) (b)

visible q

Figure 4. SAXS structural diagnosis ofweaklystructured, controland unstructured feather barbs. (a) Representative azimuthalSAXS profiles for the rudimentary sphere-type nanostructure(structured*, electronic supplementary material, table S2) inA. laminirostris (Ramphastidae), and the rudimentary channel-type nanostructures (structured, see electronic supplementarymaterial, table S2) in Melanotis caerulescens (Mimidae) andAnas clypeata (Anatidae) as well as unstructured feather barbsfrom Goura victoria (Columbidae) and Hylocichla mustelina(Turdidae). The azimuthal profiles are normalized to one alongthe intensity axis for ease of comparison. (b) The azimuthal

inte

nsity

I(q

)

kks

ki

q

q

synchrotron X-rays

Figure 3. Experimental schematic for SAXS experiments onfeather barb nanostructures. A small (approx. 50 mm2)sample of the distal pennaceous portion of the feather vane isshown affixed to cover a 3 mm diameter hole on an aluminiumblock, which is then mounted in a plane perpendicular to theincident X-ray beam. The two-dimensional SAXS diffractionpatterns for both channel- and sphere-type nanostructures exhi-bit ring-like features. Exploiting the circular symmetry of theSAXS diffraction patterns, the scattering intensity (I) is azi-muthally averaged as a function of q to obtain scatteringprofiles, where the peaks correspond to the rings observed inthe respective two-dimensional diffraction patterns. The scatter-ing wavevector q measures the momentum transfer or themagnitude and direction of the scattering of incident photons(ki into ks) as a result of constructive interference from structuralcorrelations of size 2p/q within the nanostructure.



their underlying amorphous photonic nanostructures[13,15,3638].

SAXS profiles for 18 weakly structured (blue lines), five control(grey lines) and 16 unstructured feather barbs (black lines) on asemi-log scale. The azimuthal profiles are vertically displacedalong the intensity axis for clarity. The azimuthal scatteringprofiles of the control feathers, many purple, magenta andbright white feathers as well as several marginally blue-grey(black lines) feathers did not deviate from Porods Law even atlow q (,0.04 nm21). Thus, these feathers do not possess anyunderlying barb nanostructure, ruling out any contribution ofconstructive interference to their observed colours.The azimuthalSAXS profiles from feathers with mainly slaty blue-black to palegreyish-blue colours show slight to moderate deviations fromPorods Law at low q, with these features resembling a shoulderrather than a peak. Nevertheless, the spatial correlations thatthese feather barbs do possess appear to be at the appropriatelength scales to be able to produce visible structural coloursthrough interference. (a,b) The thick horizontal line indicatesthe range of spatial frequencies relevant for avian visiblestructural colour production.

1.2. Self-assembly by phase separation

Macro-molecular self-assembly through phase separationis a fundamental property of soft condensed matter sys-tems [28]. The stability of a molecular mixture isdetermined by its temperature, the strength of intermole-cular interactions (x) and the relative volume fractions ofthe component materials. At lower temperatures andintermediate volume fractions, a mixture may becomeunstable and can proceed to unmix by one of two funda-mental physical processes [28]. Phase separation of acompletely unstable mixture can proceed via spinodaldecomposition (SD), which usually produces a character-istic morphology of interconnected bicontinuous channels[39,40]. The observed fractal-like patterns or motifs beginat small length scales and spontaneously coarsen orthicken over time roughly maintaining the same averageshape in a scale-independent fashion (self-similarity). Bycontrast, a meta-stable mixture can unmix throughnucleation-and-growth, which proceeds via the develop-ment and subsequent coarsening of spherical droplets ofthe minority component [41,42]. Unlike SD, however,nucleation requires the crossing of an activation barrier.If nucleation is fast and growth, relatively slow, nearlyidentical (or monodisperse) spheres can form [43,44].For simplicity, we have summarized here the


morphologies observed during the classical phase separ-ation of a simple binary fluid mixture. However, phaseseparation phenomenology should be modified to includenonlinear viscoelastic mechanisms when the two nascentphases have distinct rheological (i.e. mechanical) pro-perties, such as in mixtures of a network-forming orpolymerizing component and a fluid [4547]. In this scen-ario, polymerizing proteins may form networks that resist

(a) (b) (c) (d )

1 2 3 1 2 3103

102

1

10

1 2 3 4

37

Porod asymptote (q4)amorphous PS spheres(C. maynana)(Ta. larvata)

1 2 3 4

scal

ed in

tens

ity I

(q)/

I(q p

k)

101

103

102

1

10

101

scaled q (q/qpk) scaled q (q/qpk)

spinodal polymer blendsI. puellaS. sialisP. iris

Porod asymptote (q4)

Figure 5. SAXS structural diagnosis of amorphous photonic nanostructures in feather barbs. (a) and (c) depict, respectively,representative normalized azimuthal SAXS profiles for channel-type nanostructures from I. puella (Irenidae), S. sialis (Turdidae),and Pitta iris (Pittidae) and sphere-type nanostructures from C. maynana (Cotingidae) and Tangara larvata (Thraupidae) on aloglog scale exhibiting clearly distinguishable structural differences. The azimuthal profiles are scaled to compare across differ-ent colours and nanostructural sizes. (b) and (d) show, respectively, scaled azimuthal SAXS profiles for 159 channel- and 96sphere-type nanostructures on a semi-log scale. The colour of each profile is coded to the approximate colour of the correspondingfeathers based on its primary optical peak hue (pure UV colours shown in black). These azimuthal profiles are vertically displacedalong the y-axis for clarity. In addition to the primary peak, the channel-type nanostructures (a,b) either have a weak to a pro-nounced shoulder at approximately twice the dominant spatial frequency, 2*qpk or lack any other significant feature, while thesphere-type nanostructures (c,d) exhibit one or more pronounced higher-order scattering peaks in addition to the primary peakat ratios of approximately

p3 and

p7 times qpk. The grey dashed lines in all figures plot the scaled experimental scattering profiles

from two polymer mixtures undergoing spinodal decomposition [33,34] (a,b) and an amorphous film of self-assembled colloidalpolymer spheres (c,d). The vertical lines at 1,2,3 (a,b) and at 1,

p3 and

p7 (c,d) are visual guides for the expected positional

ratios for the SAXS peaks based on experimental observations of classical spinodal and nucleated, close-packed spheremorphologies, respectively.



coarsening while the solvent lacks such dynamic elas-ticity. Such viscoelastic processes lack self-similarity inthe coarsening of domains, and are also hypothesized todynamically self-arrest either owing to the cross-linkingof networks or from the evaporation of solvent duringphase separation [4547].

Prum et al. [25] used TEM of serial sections to observethe development of channel-type nanostructure in thegrowing feather germ of parrot feathers. They foundthat these amorphous intracellular nanostructuresdevelop spontaneously without any underlying biologicaltemplate or prepattern of cytoskeletal fibres or mem-branes and thus evidently self-assembled. Furthermore,they pointed out that these nanostructures bear a quali-tative similarity to morphologies self-assembled duringSD. Later, based on the SAXS data from two species,Dufresne et al. [38] hypothesized that both channel andsphere nanostructures in birds could be self-assembledby arrested phase separation of filamentous b-keratinprotein from the cellular cytoplasm. Further, they [38]proposed that the two classes of nanostructures, channel-and sphere-types, could possibly be self-assembled by SDand nucleation-and-growth mechanisms, respectively.

Here, we further test the hypothesis that constructiveinterference of light scattered by three-dimensional


quasi-ordered photonic nanostructures is responsible forthe non-iridescent plumage structural colours found indiverse avian lineages [3,8,9]. We also quantitativelycompare the SAXS data from hundreds of feathernanostructures with experimental scattering data fromself-assembled, synthetic soft condensed matter systems.

Using single scattering theory [8,13,16,48], we pre-dict the optical reflectance of each nanostructure fromthe SAXS structural data and compare it with normalincidence optical measurements, and in addition per-form angle-resolved spectrophotometry on a subset ofthe feathers. Further, we quantitatively explore thedifferences in the nanostructure and optical functionof channel- and sphere-type barb nanostructures.

2. MATERIAL AND METHODS

2.1. Taxon sampling

We surveyed the birds of the world to identify all avianfamilies and genera with probable non-iridescent structu-rally coloured barb colours (usually blues, violets, greens,etc.) from museum specimens and published illustrations(see electronic supplementary material, table S1).We cross-checked target species for the presence of



non-iridescent structural colours (peaked reflectance pro-files) by visual inspection and optical spectrophotometryof museum skins. We sampled across the gamut of knownstructural hues including the near ultraviolet (visible tobirds but not to humans), violet, blue, cyan and green,as well as saturated (peaked) yellowred hues whenthey co-occur on birds with obvious barb structuralcoloration (blues and greens). We included feathersamples of multiple plumage patches with different col-ours from the same species as well as differentlycoloured, but homologous patches from both sexes ofsexually dichromatic species (see electronic supplemen-tary material, table S1). We included in the samplesome species with ambiguous bluegrey (e.g. Polioptilacaerulea, Pachyptila vittata) and dull slate-coloured(e.g. Brachypteryx montana, Rhyacornis fuliginosus)feathers (see electronic supplementary material,table S1). We also sampled five control feathers witheu- and phaeo-melanin, carotenoid pigments, and unpig-mented matte white colour, chosen from the avian generaCorvus, Hylocichla, Carduelis, Saltator and Larus thatshould in theory lack any structural colour-producingbarb nanostructure (see electronic supplementarymaterial, table S2). In order to assess the variability ofbarb structural colour within and among individuals ofa single species, we assayed feathers from multiplestudy skins of S. sialis and C. maynana (see theelectronic supplementary material).

In total, including controls, we examined the nano-structure of 297 distinctly coloured feathers from 230species belonging to 163 genera in 51 avian families(see electronic supplementary material, table S1).Feathers were obtained from study skins of the taxaof interest from Yale Peabody Museum of Natural His-tory (New Haven, CT, USA), University of KansasNatural History Museum and Biodiversity ResearchCenter (Lawrence, KS, USA), American Museum ofNatural History (New York, NY, USA), Natural His-tory Museum at the Academy of Natural Sciences(Philadelphia, PA, USA), Harvard UniversityMuseum of Comparative Zoology (Cambridge, MA,USA) and University of Oxford Natural HistoryMuseum (Oxford, UK) (see electronic supplementarymaterial, table S1). Immature individuals or specimenswith obvious fading or degradation were avoided.

2.2. Small angle X-ray scattering

For SAXS data collection, small (approx. 50 mm2)samples of the distal pennaceous portion of the feathervanes were affixed to an aluminium block using SuperGlue (DuPont, Wilmington, DE, USA) over a 3 mmdiameter hole. The block was mounted in a plane per-pendicular to the incident X-ray beam as shown inthe experimental schematic (figure 3). Pinhole SAXSdata on two to three individual barbs per feathersample were collected in transmission geometry, atbeamline 8-ID-I of the Advanced Photon Source,Argonne National Laboratories (Chicago, IL, USA).We used a 15 mm (Horiz. Vert.) beam (1.68 A,7.35 keV, 50 0.2 s exposures, sample-detector dis-tance 3.56 m, flux 2.7 109 photons s21). Beam sizewas minimized to sample as few spongy medullary


cells as possible (they are typically approx. 1015 mm3 but vary with taxon [49]). The azimuthallyaveraged scattering profiles were calculated from theCCD-collected two-dimensional SAXS speckle diffrac-tion patterns using the freely available Matlab-implemented software, XPCSGUI, developed by beamline8-ID (http://8id.xor.aps.anl.gov/UserInfo/Analysis/)at 200 equal q-partitions, and with customized masksto filter out the beam stop [15,38,50]. SAXS data fromthe feathers of Cittura cyanotis and the yellow throatof Psarisomus dalhousiae (see electronic supplementarymaterial, tables S1 and S2) were collected using a7.35 keV beam (1.68 A, 50 mm horizontal 50 mm ver-tical, 9.24 m camera length, 50 0.1 s exposures) on aPilatus2M detector at beamline I22 of the DiamondLight Source, Didcot, UK.

Biomimetic amorphous samples that closely mimicquasi-ordered arrays of a nucleation-and-growth struc-ture were prepared by drop casting a 50 : 50 bidispersemixture of 258 and 286 nm diameter polystyrene (PS)spheres [51]. SAXS measurements of biomimeticsamples were carried out by sandwiching the samplein an aluminium sample holder between two pieces of0.0025-inch (approx. 63.5 mm) thick adhesive Kaptontape, purchased from McMaster-Carr (Catalogue no.7648A33). Light scattering data for phase-separating spi-nodal morphologies were obtained from Takenaka &Hashimoto [33] and Hayashi et al. [34].

2.3. Normal incidence and angle-resolvedspectrophotometry

Normal incidence reflectance measurements were madefrom two or three different locations within eachsampled plumage patch and averaged. Whenever poss-ible, the same museum study skins that were sampledfor the SAXS structural assays were used for the corre-sponding reflectance measurements. For 77 plumagepatches, the original study skins were locally unavail-able for spectrophotometry, and reflectance wasmeasured from the individual feathers collected forSAXS assays (see electronic supplementary material,table S1). These feather-based reflectance measure-ments were essentially identical to those measuredfrom other museum specimens of the same species. Allmeasurements were made in relative darkness using anOcean Optics S2000 (Dunedin, FL, USA) fibre opticspectrophotometer and an Ocean Optics DH-2000-BAL deuteriumhalogen light source, following stan-dard procedure [79]. The S2000 provides 2048 datapoints between 178 and 879 nm. In order to shieldany ambient light and control the irradiance, the bifur-cated fibre-optic cable was inserted into a probe holder.Reflectance was measured using normally incident lightat a distance of approximately 6 mm from approxi-mately a 3 mm2 illuminated patch of the integumentwith a 500 ms integration time and calibrated usingan Ocean Optics Spectralon matte white reflectancestandard and with a matte black velvet cloth asdark reference.

We also conducted angle-resolved spectrophotometryin diffuse scattering geometry [13,36,37] on a smallerset of 22 individual feather samples with both channel-
http://8id.xor.aps.anl.gov/UserInfo/Analysis/http://8id.xor.aps.anl.gov/UserInfo/Analysis/http://rsif.royalsocietypublishing.org/



and sphere-type nanostructures, to differentiatebetween double scattering versus pigmentary origin ofthe short-wavelength secondary reflectance featuresusing the effects of angular dispersion, as well as to esti-mate the nanostructural parameters independent of theSAXS data. The bird feathers were mounted horizon-tally so that their axes were perpendicular to therotation axis of a goniometer. Collimated white lightfrom an Optics DH-2000-BAL deuteriumhalogenlight source was incident on the sample at normal inci-dence and with a spot size of approximately 1 mm.The scattered light was collected by a lens and focusedonto an optical fibre connected to a spectrometer(Ocean Optics S2000). The spectral resolution was1.5 nm and the angular resolution, determined mainlyby the collection angle of the lens, was about 58.To measure the scattered light, we fixed the feathersample and rotated only the detection arm, in whichcase, the illumination angle remained constant whilethe angle of observation changed. The measured spectraof scattered light were normalized by the incident sourcelight spectrum after dark subtraction.

2.4. Electron microscopy

We followed standard specimen embedding proceduresfor TEM [79]. For scanning electron microscopy(SEM), longitudinally and cross-sectionally fracturedfeather barb samples were gold-coated and studied on aHitachi SU-70 and a Philips XL 30 environmental SEMat a range of tilt angles.

2.5. Parametrization of small angle X-rayscattering structural data and opticalreflectance spectra

The two-dimensional SAXS diffraction patterns forboth channel- and sphere-type nanostructures exhibitring-like features (figure 2h,i). Exploiting the circularsymmetry of the SAXS diffraction patterns, we azi-muthally integrated them using the XPCSGUIpackage after masking out the beam stop pixels, toobtain profiles of the scattered intensity as a functionof scattering wavevector, I(q), at 200 equal q-partitionsor spatial frequency bins (figures 35). The peaks in theazimuthal profiles correspond to the rings observed inthe respective two-dimensional diffraction patterns(figure 3). The azimuthally averaged profiles weredeconvolved, or peak-fitted, to estimate the peak qvalue, intensity, and the full-width at half-maximum(FWHM) of the scattering peaks, using the freely avail-able peak-fitting software, Fityk (v. 0.8.2; [52]) on aWindows platform. We used a Porod background (q24

dependence; see 3 and figures 45) and the split-Pearson VII function with a LevenbergMarquardtleast square method to fit all the observed scatter-ing features (peaks and shoulders) present in theazimuthal profiles. The Pearson VII function is a com-bination of Gaussian and Lorentzian (Cauchy) typepeak profiles that is generally used to closely approxi-mate X-ray scattering peaks [53,54]. The split-PearsonVII accommodates any asymmetry in peak shapes.


The unprocessed optical reflectance measurementswere also similarly deconvolved to estimate the relevantoptical peak parameters such as wavelength of peak reflec-tance, intensity and FWHM of the reflectance peak usingFityk. The FWHM characterizes the saturation of optical(reflectance) signals [55]. We used a constant backgroundand a Gaussian or a split Gaussian function with a Leven-bergMarquardt least square method to fit all theobserved spectral features (peaks and shoulders) presentin the reflectance profiles. The split Gaussian functionwas used for asymmetrical peak profiles.

2.6. Small angle X-ray scattering single-scattering reflectance predictions

We used the azimuthally averaged SAXS structuralspectra to directly predict the optical reflectance spec-tra of the respective amorphous barb nanostructuresusing single-scattering theory by mapping the SAXSintensity from wavevector or spatial frequency (q) towavelength (l) space [8,13,16,48,50]. This result followsfrom Braggs Law, for under normal incidence of lightand back-scattering geometry (the angle between inci-dence and observation, u 08), the scatteringwavevector (q) and the wavelength of light (l) aresimply related as

l 2 2pq

navg; 2:1

where 2p/q is the average inter-scatterer or nearest-neighbour spacing D, and navg is the average or effectiverefractive index of the nanostructure [13].

2.7. Statistical analyses

Regression analyses were performed using the statisticstoolbox of Matlab 2008a (The MathWorks Inc., Natick,MA, USA) and MINITAB statistical software, release 16(Minitab Inc., State College, PA, USA) running on aWindows platform. One-way ANOVA and generallinear model tests for the statistical difference of theslopes and intercepts of two regression lines [56] wereall performed in MINITAB. The p-value for statistical sig-nificance was set at 0.05.

3. RESULTS

3.1. Comparative structural diagnoses of featherbarb photonic nanostructures

Of the 297 feathers assayed in this study, the azimuthalscattering profiles of the control feathers (n 5)did not deviate from Porods Law even at low q(,0.04 nm21) (figure 4, and electronic supplementarymaterial, table S2). The scattering profiles of some feath-ers with longer-wavelength reflectance, including deeppurple and magenta (n 9), bright whites (n 3) andsome blue-grey feathers (n 4) did not deviate fromPorods Law as well (figure 4 and electronic supplemen-tary material, table S2). Porods Law (I(q)/ q24) is thenull expectation for the scattering from an unstructuredmaterial characterized by sharp interfaces or edges separ-ating two media [31]. Thus, these 21 feathers do not

slope = 1.87, R2 = 0.94 (channels)slope = 1.69, R2 = 0.97 (spheres)

slope = 2 (spinodal)slope = 3 (close-packed spheres)

0.025 0.030 0.035 0.040

0.040

0.050

0.060

0.070

0.080

0.090

first SAXS peak position, qpk1 (nm1)

seco

nd S

AX

S pe

ak p

ositi

on, q

pk2

(nm

1 )

Figure 6. Regression plot of the first and second SAXS peaksof channel- (open triangles) and sphere-type (shaded circles)amorphous barb nanostructures. The colour of each triangleor circle is coded to the approximate colour of the correspond-ing feather (UV colours in black). The thin vertical andhorizontal lines at each data point indicate the standarderror of the mean (s.e.m). The solid blue and green lineswith corresponding slopes of 2 and

p3 indicate the expected

positional ratios for the second SAXS peak based on exper-imental observations of spinodal and nucleated, close-packedsphere morphologies, respectively. The solid and dashed greylines, respectively, indicate the 95% confidence interval ofthe regressions.



possess any underlying barb structure at optically relevantlength scales ruling out any contribution of constructiveinterference to their observed colours.

The SAXS data from a further 18 out of 297 feathers(see electronic supplementary material, table S2) withslaty blue-black to pale greyish-blue colours, such asthe primary coverts of the Silver-breasted Broadbill(Serilophus lunatus, Eurylaimidae; figures 2a and 4)exhibited diffuse, disc-like two-dimensional SAXS dif-fraction patterns (figure 2g). Their correspondingazimuthal averages showed slight to moderate devi-ations from Porods Law at low q; these featuresresembled a shoulder rather than a peak (figure 4).Nevertheless, the spatial correlations of these featherbarbs appear to be at the appropriate length scales toproduce visible structural colours through constructiveinterference (figure 4). For 16 of these 18 feathers, theprimary scattering feature could not be estimatedwith peak-fitting procedures (called structured, seeelectronic supplementary material, table S2).

We examined some of these slaty, blue-grey featherswith distinctive disc-like SAXS patterns using SEM.SEM images from two structured species, Serilophuslunatus (Eurylaimidae) and Melanotis caerulescens(Mimidae), revealed a very restricted and thin layer(1 mm or less) of a disordered channel-like network ofspongy b-keratin bars present at the periphery of themedullary barb cells (figure 2d and electronic supplemen-tary material, S1c). All of these 16 structured speciesare closely related to other species known to possess thechannel-type medullary barb nanostructures (figure 5band electronic supplementary material, table S2). Thetwo species for which the peaks could be estimated(called structured*, see electronic supplementarymaterial, table S2) were the toucans Andigena laminiros-tris and Pteroglossus viridis (Ramphastidae). Othertoucans have sphere-type nanostructures (figure 5d andsee electronic supplementary material, table S2). SEMimages of A. laminirostris revealed a very thin layer(1 mm or less) of hollow spheroidal concavities of highlyvariable sizes and shapes (highly polydisperse) in thespongy b-keratin at the periphery of the medullary cells(see electronic supplementary material, figure S1d).Thus, these blue-grey and slate-grey feathers characterizedby diffuse, disc-like SAXS patterns possessed rudimentaryand highly disordered versions of channel- (structured)and sphere-type (structured*) nanostructures, found intheir close relatives.

The two-dimensional SAXS diffraction patterns ofrest of the barb nanostructures assayed exhibit ring-likefeatures that demonstrate strong nanostructural isotropyand short-range spatial periodicity (figure 2h,i). Theazimuthally averaged scattering profiles display peaksthat correspond to the rings observed in the respectivetwo-dimensional diffraction patterns (figure 3). For com-parison of feathers across different structural colours,nanostructural dimensions and scattering intensities,we normalized all azimuthal SAXS profiles by the pri-mary peak spatial frequency (qpk) and intensity (I(qpk))(figure 5). In the high q region (q . 0.1 nm21), theSAXS profiles follow Porods Law. However, at low andintermediate q values (q , 0.1 nm21), the azimuthalSAXS profiles of most barbs exhibit clearly


distinguishable features that can be used to identifythe barb nanostructures.

Of the remaining 258 feathers, 218 were readily clas-sifiable into the two known classes of three-dimensionalbarb nanostructures based on the features (observedversus expected number and relative positions ofpeaks) of their SAXS patterns. Many feathers thatlacked any higher-order scattering features besides theprimary structural correlation peak (n 48) or exhib-ited a low intensity second-order shoulder atapproximately twice the dominant spatial frequency(1.879+ 0.005; n 95) were identified as the channel-type (figures 5a,b and 6). Both the structure factorscalculated from TEM images of channel-type nano-structures [25], as well as experimental scattering datafrom other interconnected bicontinuous network nano-structures [33,34] support these conclusions (see 3.2).By contrast, many other feathers (n 75) diagnosti-cally exhibited two or more pronounced, higher-orderpeaks indicative of spherical form-factor scatteringfringes [25,57] at ratios of approximately

p3 (1.693+

0.004; n 75), p7 (2.498+ 0.010; n 75) and p11(3.475+ 0.063; n 11) and were recognized as thesphere-type (figures 5c,d and 6). The secondaryshoulder at approximately 2*qpk in channel-type nano-structures is comparatively much broader in width andweaker in intensity relative to the primary peak than isthe second-order peak (approx.

p3*qpk) from sphere-

type nanostructures (see electronic supplementarymaterial, figure S2). We further corroborated many ofthese SAXS structural assignments of barb nanostruc-tures based on previously published (see [3] andreferences therein) and our own TEM and SEMimages for 39 channel- and 27 sphere-type barb



nanostructures (see electronic supplementary material,table S2).

For the remaining 40 feathers, we provisionally ident-ified the nanostructures as the structural interpretationof their SAXS data was less straightforward and/or insome feathers, the azimuthal profiles were comparativelynoisier, i.e. jagged (see figure 5 and electronic supple-mentary material, table S2). The azimuthal SAXSprofiles of 21 such feathers (sphere*, see electronic sup-plementary material, table S2) exhibited only a broadsecond-order peak between 1.604 and 1.737 times qpk,but consistent with the distribution for unambiguoussphere-type morphologies (see electronic supplementarymaterial, figure S2). The scattering intensities of thesecond-order peaks from most of these feathers werealso considerably higher than the mean intensity of thechannel-type second-order shoulders (see electronicsupplementary material, figure S2). By contrast, the azi-muthal profiles of eight feathers were noisy with onlythe primary scattering peak present, consistent with thechannel morphology, while a further 11 feathers exhibitedsecond-order shoulders but at smaller positional ratios(1.7471.796) than expected (channel*, see electronicsupplementary material, table S2). These tentativenanostructural assignments were validated based on EMimages (nine channel* and five sphere*) and/or theunambiguously identified nanostructure present in otherstructurally coloured plumage patches on the samespecies or in a few cases, indirectly assumed from thatin closely related taxa within the same genus (see elec-tronic supplementary material, figure S2 and table S2).We also evaluated the relatively noisy barb morphologiespresent in Myiomela leucura (Turdidae), Chiroxiphia cau-data (Pipridae) and Euneornis campestris (Thraupidae),using SEM. SEM images of M. leucura revealed a spindlychannel morphology with anastomosing networks ofb-keratin bars of variable thickness (see electronic sup-plementary material, figure S1a). SEM images ofC. caudata and E. campestris revealed sphere-type nanos-tructures with a greater degree of polydispersity in thesize and the shape of the air spheres than in typicalsphere nanostructures found in their close relatives (seeelectronic supplementary material, figure S1b,f ). Thesenoisier nanostructures appear to be more variable (poly-disperse) versions of the types of nanostructures foundin their closest, structurally coloured relatives.

3.2. Structural comparisons of amorphous barbnanostructures and synthetic soft mattersystems

We compared the normalized azimuthal scattering pro-files of the unambiguously diagnosed instances ofchannel (n 143) and sphere (n 75) barb nanostruc-ture to experimental light-scattering data from polymermixtures in early and late stages of SD [33,34], and aself-assembled, amorphous film of colloidal polymer(PS) spheres, mimicking a quasi-ordered nucleation-and-growth nanostructure (figure 5ad).

The shape of the scattering profile of a classical spi-nodal mixture is scale independent; the overall structureof spinodal morphologies are universal even though thespecific structure in a phase-separating sample may


differ locally [39,40], but experimentally, there is sub-stantial variation between early and late stages of SDat intermediate and high q [33,34,58,59]. The exper-imental spinodal polymer profiles [39,40] provide areasonable fit to the normalized channel-type scatteringprofiles from feather barbs at low q (figure 5a), eventhough the width (FWHM) of the primary SAXSpeak of channel-type nanostructures narrows as thedominant length scale of the nanostructure increases,i.e. for those producing longer wavelength colours(figure 5b and electronic supplementary material,figure S3a). At high q, the polymer spinodal mor-phologies lack any higher-order feature at the earlystage or exhibit a second-order shoulder at approxi-mately 2 or approximately 3qpk at the late stage[33,34]. Similarly, the channel-type barb nanostructureseither lack or exhibit a shoulder or second-order maxi-mum at approximately twice the peak spatialperiodicity (mean 1.87) (figures 5a,b and 6). However,the positions of the secondary shoulder from the channelnanostructures are probably underestimated becausesuch shallow (broad and low intensity, see electronic sup-plementary material, figure S2) features are difficult toprecisely estimate through curve-fitting procedures.

The SAXS scattering profile from a film of self-assembled, quasi-ordered, colloidal polymer spheresreveals a series of higher-order form factor diffraction(fringes) peaks at similar relative positions (

p3 andp

7) to those seen in azimuthal profiles of sphere-typenanostructures (figure 5c,d). The width (FWHM) ofthe primary SAXS peak of sphere-type nanostructuresis also in good agreement with that of the self-assembled,amorphous PS spheres. (The X-ray scattering from anarray of solid spheres is indistinguishable to that fromits inverse structureair spheres in solidand thereforethis direct comparison is valid according to Babinetstheorem [31].) Although the PS spheres are self-assembled into an amorphous structure, the spheresthemselves were not synthesized in situ by a nuclea-tion-and-growth process; however, the scattering profileof a three-dimensional amorphous array of spheresgrown by a nucleation-and-growth mechanism shouldbe similar [57]. The number and strength (intensityand width) of the higher-order peaks in the sphere-typenanostructures are sample-specific, and reflect thedegree of quasi-periodic or nearest-neighbour order andsphere size monodispersity.

3.3. Comparative structural properties ofamorphous barb nanostructures

We examined the width (FWHM) of the primary SAXSpeak, Dq, to quantitatively characterize the extent ofspatial periodicity in the channel and sphere classes ofamorphous nanostructures (n 255; structured andstructured* were excluded; see electronic supplementarymaterial, table S2).

The spatial coherence length, j, is given by 2p/Dq.For ordered systems, j describes the crystallinedomain size, whereas in quasi-ordered or amorphoussystems, the coherence length (after scaling by the cor-responding peak spatial periodicity, j/D) can provide ameasure of the extent of short-range nearest-neighbour



order [13,60]. For both classes of amorphous barbnanostructures, j is only a few times the dominantlength-scale of spatial correlations, D (2p/qpk), reflect-ing the very local nature of spatial order in thesesystems [13]. However, the mean structural FWHM(Dq) of sphere nanostructures is significantly smallercompared with channels (one-way ANOVA, F 26.82, p , 0.001, n 255), even after scaling by thecorresponding peak spatial frequency of structural cor-relations, qpk (one-way ANOVA, F 28.51, p , 0.001,n 255), suggesting that sphere-type nanostructuresare more ordered, or have a larger coherence lengththan channel-type nanostructures (see electronicsupplementary material, figure S3).

In addition, the FWHM of the primary SAXS peakincreases significantly with the dominant spatial fre-quency of structural correlations, qpk, for both channel-(r2 0.22, p , 0.001) and sphere-type (r2 0.29, p ,0.001) nanostructures (see electronic supplementarymaterial, figure S3a,b). However, the statistical signifi-cance of this relationship persists only for channel-typenanostructure (r2 0.067, p 0.001), after scaling theFWHM by the corresponding qpk (see electronic sup-plementary material, figure S3c,d). In other words,channel-type nanostructures with larger size scale ofspatial periodicity (i.e. D) producing longer wavelengthcolours have a smaller structural (SAXS) peak width,and consequently a larger coherence length or greatershort-range order. In contrast, the sphere-type nanostruc-tures appear to be nearly scale invariant as revealed bythe same relative widths of the structural peaks acrossall length scales (see electronic supplementary material,figure S3d). These structural differences between channeland sphere morphologies are perhaps a result of their dis-similar processes of phase separation and arrest duringintracellular self-assembly, which are also probablyaffected by the subsequent desiccation of medullarybarb cells in different ways.

3.4. Comparative optical function of amorphousbarb nanostructures

The slaty blue-black to pale greyish-blue feathers withrudimentary (structured and structured*) barbnanostructures generally exhibited a broad, low inten-sity, sperm-whale-shaped reflectance profile with agradually decreasing reflectance at longer wavelengthsand a more rapid decline at shorter wavelengths. Thepeak parameters from the optical reflectance of thesefeathers could not all be consistently estimated (seeelectronic supplementary material, table S2) and aretherefore excluded from further optical analyses. Never-theless, the SAXS results demonstrate that thesefeathers are sufficiently nanostructured at the appropri-ate length scales to produce the observed colours viaconstructive interference (figure 4).

The spectral peaks in the optical reflectance measure-ments of the structurally coloured feathers with channel-and sphere-type nanostructures characterized in thisstudy varied from 343.83 to 639.37 nm (n 255; seeelectronic supplementary material, table S2). Manyfeathers, particularly royal (medium) blue to turquoise(light) blue ones, with either class of barb nanostructure


(n 86), exhibited a characteristic bimodal reflectanceprofile with an additional peak in the ultraviolet (UV)/violet distinct from the primary reflectance peak in thevisible (400700 nm) region (figure 7df,mo, andelectronic supplementary material, table S2). These sec-ondary (short wavelength) peaks are qualitatively quitedifferent from the relatively low intensity UV pig-mentary (carotenoid) transmittance peaks seen innanostructured feather barbs with cortical pigmentation,for instance, in structural greens (figure 7gi,q). The dipbetween the two peaks in the former are relatively shal-lower and the secondary peak is nearly of the sameamplitude or higher than the primary single-scatteringpeak (figure 7). Unlike the UV peaks in spongy barbnanostructures with cortical pigments, these secondarypeaks are of structural origin owing to the double scatter-ing of light and not explained by the higher-orderstructural correlations in the X-ray scattering data[36,37]. Diagnostically, the double scattering peaksoccur at nearly constant relative spectral ratio to oneanother (1/

p2) as expected from optical theory [36,37]

(criterion 1). Moreover, double scattering peaks arealso depolarized (criterion 2) and exhibit reverse angulardispersion (see electronic supplementary material, figureS4bf ) that is specifically predicted by optical theory(criterion 3) [36,37], whereas the static spectral featuresthat are produced by pigmentary absorption are not(see electronic supplementary material, figure S4g,h).We have described the complete mechanistic basis ofthe double scattering phenomenon in detail elsewhere[36,37]. We tentatively identified 58 of these 86 feathersas double scattering candidates based on criterion 1alone (see electronic supplementary material, table S2),while we were able to unambiguously document doublescattering in the remaining 28 feathers based on criterion2 and/or 3 (see electronic supplementary material, figureS4bf ). Here, we mainly consider the primary opticalreflectance peak, which originates from the single scatter-ing of incident light whereby each incident photon isscattered only once before it exits the nanostructure [13].

Many feathers (n 77) producing structural greensand longer wavelength hues with reflectance peaksabove 500 nm have distinct spectral indications of thepresence of carotenoid or psittacofulvin pigments in theouter b-keratin cortex of the barb, the mechanisticbasis of which are well established for many species[3,14,61]. Unlike a typical sigmoidal reflectance profileof carotenoid pigmented barbs that plateau at higherwavelengths [10], the reflectance from these featherswere distinctly peaked or saturated, but with a relativelysharp cessation of the reflectance intensity on theshort-wavelength side of the peak (figure 7gi,pr andelectronic supplementary material, figure S4g,h). Inaddition, the reflectance spectra of these feathersshowed a minor UV transmittance peak at approximately350 nm and/or several low intensity sub-peaks at inter-mediate wavelengths [3,14,61]. In contrast to doublescattering feathers, these pigmentary spectral featureswere angle-independent (see electronic supplementarymaterial, figure S4g,h). Therefore, we conservativelyexcluded these nanostructures for analyses involving thesaturation or widths (FWHM) of the primary reflectancepeaks, as the short and middle wavelength pigmentary

0

10

20

30

0

10

20

0

20

40

60

0

50

100

0

50

100

0

50

100

150

0

5

10

0

10

20

0

20

40

0

50

100

0

20

40

60

0

50

100

0

20

40

0

10

20

30

0

50

100

400 600 8000

50

100

400 600 8000

20

40

60

400 600 8000

20

40

60

mea

sure

d (

colo

ur)

per

cent

opt

ical

ref

lect

ance

wavelength (nm) wavelength (nm)wavelength (nm)

pred

icte

d (b

lack

) si

ngle

-sca

tteri

ng (

SAX

S) r

efle

ctan

ce (

arb.

uni

ts)

(a) (b) (c)

(d)

(g) (h) (i)

( j) (k) (l)

(m)

(p) (q) (r)

(n) (o)

(e) ( f )

Figure 7. Single-scattering SAXS reflectance predictions for the primary optical peaks of channel (a i) and sphere-type ( jr)amorphous barb nanostructures. SAXS single-scattering reflectance predictions (black lines) and measured normal incidencereflectance curves (coloured lines) for (a) UV (black) belly feather barbs of Charmosyna papou (Psittacidae), (b) violet primaryfeather barbs of Acryllium vulturinum (Psittacidae), (c) royal blue rump feather barbs of S. sialis (Turdidae), (d) sky blue rumpfeather barbs of Alcedo atthis (Alcedinidae), (e) deep azure blue back feather barbs of Irena puella (Irenidae), ( f ) electric bluewing covert feather barbs of Pitta maxima (Pittidae), (g) emerald green back feather barbs of Ailuroedus buccoides (Ptilonor-hynchidae), (h) emerald green back feather barbs of Charmosyna papou (Psittacidae), (i) emerald green back feather barbs ofCalyptomena whitehadi (Eurylaimidae), ( j) deep blue throat feather barbs of Tangara chilensis (Thraupidae), (k) royal bluewing covert feather barbs of Wetmorethraupis sterrhopteron (Thraupidae), (l ) violet scapular feather barbs of Conirostrum albi-frons (Thraupidae), (m) dark turquoise blue back feather barbs of C. maynana (Cotingidae), (n) sky blue back feather barbs ofmale Tersina viridis (Thraupidae), (o) azure blue rump feather barbs of Lepidothrix serena (Pipridae), ( p) golden yellow crownfeather barbs of Lepidothrix vilasboasi (Pipridae), (q) electric green back feather barbs of Chloronis riefferii (Thraupidae), (r)golden crown feather barbs of Tangara larvata (Thraupidae). The colour of the measured reflectance curves is approximatelycoded to the colour of the feather barbs based on the spectral position of the primary reflectance peak.




0.025 0.030 0.035 0.040

0.010

0.012

0.014

0.016

0.018

0.020(a) (b)

SAXS peak spatial frequency, qpk (nm

1)SAXS peak spatial frequency,

qpk (nm1)

0.020 0.030 0.0400.008

0.010

0.012

0.014

0.016

0.018

0.020

navg = 1.201; R2

= 0.77

f = 34%navg = 1.265; R

2 = 0.84

f = 46%

optic

al p

eak

spat

ial f

requ

ency

,k p

k (n

m

1 )

Figure 8. Regression plots of the primary optical peak hue from normal incidence reflectance measurements expressed as peakspatial frequency (kpk 2p/lpk) against the dominant spatial frequency of structural correlations (qpk) measured using SAXSfor (a) channel- (shaded triangles) and (b) sphere-type (shaded circles) nanostructures. For both nanostructural classes, thesize of the nanostructural periodicity measured by SAXS strongly predicts, i.e. scales with the measured primary peak hue,demonstrating that the underlying barb nanostructures are tuned to produce the observed structural colours. The inverse oftwice the slope of the regression yields navg, the average or effective refractive index (and hence f, the keratin volume fraction)for each class of nanostructure. The estimated navg and f for sphere nanostructures on the whole (1.265, 46%) is significantlyhigher than that for channel morphologies (1.201, 34%) and congruent with predictions of the phase separation hypothesis.The colour of each triangle or circle is coded to the approximate colour of the corresponding feather (UV colours in black).The vertical and horizontal lines at each data point indicate the standard error of the mean (s.e.m).



absorption could lead to an underestimation of theiractual widths (see figure 7gi,pr and electronic sup-plementary material, figure S4g,h). However, thespectral position of the reflectance peak or hue (lpk) isrelatively unaffected by the pigmentary absorption andthe inclusion or exclusion of these data here did not sig-nificantly alter the results.

3.4.1. Single-scattering optical predictions of amorphousbarb nanostructuresBased on single scattering theory, we applied BraggsLaw (equation (2.1)) to the azimuthal SAXS profiles toobtain the single scattering optical reflectance predic-tions of amorphous barb nanostructures [38,50]. Fromequation (2.1), for a given nanostructure, the predictedpeak hue depends on the size of the nanostructuremeasured by SAXS (D) and the average or effectiverefractive index of the nanostructure (navg), while thepredicted optical saturation or peak width depends onthe spatial coherence length, j (2p/Dq) alone. Althoughwe predicted the optical reflectance curves for all 255barb nanostructures, here we present these results onlyfor a small subset of feathers, owing to space limitations(figure 7). We summarize below the goodness of fit ofSAXS single scattering optical predictions to normal-incidence reflectance measurements, based on pairwiseregressions of peak hue and saturation of the reflectancemeasurements and reflectance predictions for bothchannel- and sphere-type nanostructures.

There is a strong positive correlation between themeasured primary optical peak hue (expressed in spatialfrequency, k 2p/lpk) and the dominant spatialfrequency of structural correlations, i.e. the primarySAXS peak wavevector (qpk) for both channel and


sphere-type nanostructures (figure 8a,b). For bothclasses of barb nanostructure, the size of the nanostruc-tural periodicity scales strongly with primary peak hue,demonstrating that the underlying barb nanostructuresare tuned to produce the observed structural colours.The correlation is stronger for sphere-type (r2 0.84)than for channel-type nanostructures (r2 0.77). Thisrelationship persists even for those barb nanostructuresproducing longer wavelength colours (peaking atapprox. 550 nm and higher) that probably involvecortical pigments.

Although the average inter-scatterer spacing D or thedominant length scale of nanostructural periodicity canbe directly measured using SAXS, there is no directmethod to measure the average or effective refractiveindex of the amorphous barb nanostructure, navg. Weused two independent methods to estimate navg: (i) bycorrelating normal incidence optical measurements withSAXS structural data using equation (2.1) and (ii)from angle-resolved optical reflectance measurements.

First, we use the regression relationship in figure 8 toestimate the average or effective refractive index (navg),and hence the filling or volume fraction (f) of b-keratinfor each class of nanostructure as a whole, using the Max-wellGarnett effective medium approximation [51]. Theinverse of twice the slope of the regression yields navg (seeequation (2.1)). The estimated global navg for spherenanostructures (1.265; 46% f) is significantly largerthan that for channel morphologies (1.201; 34% f; one-way ANOVA, F 31.45, p , 0.001, n 255). Thisresult is congruent with predictions of the phase separ-ation hypothesis, since, under similar thermodynamicconditions (kBT/x), nucleation-and-growth shouldoccur at higher volume fractions of keratin (hence navg)



compared with SD [25,38]. Further, for each of the 255barb nanostructures, we calculated its navg by substitut-ing the values of qpk from SAXS profiles and lpk fromnormal-incidence optical reflectance data in equation(2.1) (see electronic supplementary material, table S2).The navg estimated thus is significantly positively corre-lated with the peak optical hue, lpk for both channels(r2 0.44, p , 0.001) and spheres (r2 0.17, p ,0.001) (see electronic supplementary material, figureS5a,b). The increase in navg with lpk suggests that pro-duction of a longer wavelength hue by either class ofnanostructure involves increases in both the lengthscale of spatial periodicity (D) (figure 8) as well as thekeratin volume fraction (navg), instead of independentlyvarying one parameter or the other. The scatter in theplot probably reflects variation in the length scale ofspatial periodicity and keratin volume fraction (hencenavg) among different species with similar structurallycoloured plumages.

Since amorphous feather barb nanostructures havethe same angular dispersion for specular reflection anddiffuse scattering peaks, they both share a commonphysical origin, and under directional lighting con-ditions, the reflectance peak depends only on theangle between incidence and observation [13,36,37].Exploiting this, we measured the angle-resolved diffusescattering spectra of barb nanostructures for a smallsubset of channel- (n 11) and sphere-type (n 11)nanostructures. We estimated navg independent of theSAXS data by analysing the angular dispersion of theprimary optical peak using a modified form of BraggsLaw (equation (2.1)), taking into account the reducedangle with respect to the normal at which light travelsinside the barb nanostructure (i.e. Snells Law):

lpku 2Dn2avg sin2u1=2; 3:2

where the primary optical peak position, lpk, varieswith u, the angle between incidence and observation.A plot of l2pk against sin

2u yields navg (p

y-intercept)but also D (0.5/

pslope) [62,63] (see electronic sup-

plementary material, figure S6a). The correspondingvalues of D (see electronic supplementary material,figure S6b) and navg (see electronic supplementarymaterial, figure S6c) obtained from these two indepen-dent methods are consistent and agree to be within 7percent of each other. Unlike photonic crystals, however, theoptical diffuse scattering and specular reflection intensitiesof amorphous barb nanostructures falls off rapidly atshallower angles [13] and so the SAXS and normalincidence optical characterization (method 1) of theamorphous barb photonic nanostructures is probablymore accurate.

We obtained measured and predicted optical band-widths for both channel- and sphere-type nanostructuresby scaling the saturation (FWHM) of the measured reflec-tance and the width (FWHM) of the corresponding singlescattering azimuthal SAXS profiles by the respective peakhue (lpk) and peak spatial frequency (qpk), in order tocompare across feathers of different colours and nanos-tructural length scales (feathers with cortical pigmentswere excluded because pigmentary absorption can leadto an underestimation of the true peak widths). The


bandwidth of the primary optical peak is positively corre-lated with the scaled widths of the primary SAXS peakfor both channel (r2 0.47) and sphere-type (r2 0.71)nanostructures (see electronic supplementary material,figure S7a,b). This result indicates that the width of theprimary single scattering SAXS peak reasonably predictsthe optical saturation of the nanostructure. Concordantwith the structural results (see 3.3), nanostructureswith larger size scales of spatial periodicity (i.e. D) gener-ally make more saturated colours (smaller FWHM).Variations in the inter-scatterer spacing, D (whichincrease the SAXS peak width thereby decreasing thecoherence length, j), result in broader, less saturatedstructural colours. The measured optical bandwidth isconsistently larger than the single-scattering structuralprediction for both nanostructural classes (see electronicsupplementary material, figure S7a,b), probably becauseof multiple scattering [36].

4. DISCUSSION

We have characterized the nanostructure and optical prop-erties of hundreds of structurally coloured feathersencompassing the gamut of non-iridescent structuralhues from diverse taxa across the phylogeny of birdsusing a combination of SAXS, electron microscopy,normal incidence and angle resolved spectrophotometry.The SAXS structural information enabled quantitativediagnoses of the channel and sphere-type nanostructures,and documented the presence of a predominant, isotropic,short-ranged order. The nanostructural variation in refrac-tive index is of the appropriate length scales to producestrong reinforcement of a narrow band of scatteredwavelengths.Noisyand ambiguous cases of structural diag-noses from SAXS data were corroborated by EM data.Additionally, we have identified a previously unknownclass of slaty blue-black to blue-grey structural coloursthat are produced by rudimentary or highly variable ver-sions of channel- and sphere-type nanostructures.Overall, the SAXS results represent a substantial improve-ment over Fourier analyses of EM [3,79,1719] andthree-dimensional electron tomography data [26].

SAXS structural data also provided good predictionsof the primary, single-scattering peaks in optical reflec-tance measurements. Both the spectral position andshape (FWHM) of the peaks in the azimuthal SAXSprofiles were highly correlated with those of thecorresponding primary peak of optical reflectancemeasurements (figure 8 and electronic supplementarymaterial, figure S7). The discrepancies between theoptical measurements and structural predictionsespecially for short wavelength peaks (lpk , 450 nm;electronic supplementary material, figure S7) barbscan be explained in part by the multiple scattering oflight, since scattering (and multiple scattering) is stron-ger at shorter wavelengths of light, which could result insignificant broadening of the optical reflectance peaks.

We have also documented quantitative differences inthe nature of structural colour production by sphere-and channel-type nanostructures. On average, the coloursgenerated by sphere-type nanostructures are significantlymore saturated (smaller FWHM) than those produced by



channel-type, and this has a strong nanostructuralbasis (see electronic supplementary material, figure S3).However, the FWHM of the primary optical peaks pro-duced by both types of nanostructures decreases(i.e. saturation increases) with increasing peak hue (i.e.longer wavelength colours), reflecting the underlyingincrease in short-range quasi-periodic order within thenanostructure with increasing nanostructural size (seeelectronic supplementary material, figures S3 and S7).

4.1. Development of amorphous feather barbnanostructures

The development of the channel-type amorphous featherbarb nanostructure in Blue-and-Yellow Macaw proceedsin the spongy medullary cells in the telling absence ofany precursor biological template or pre-pattern createdby cytoskeleton, organelles or membranes [25]. Duringfeather cell maturation, capillary forces owing to thedrying of the spongy cells apparently drive higher mol-ecular weight materials to the cells periphery, resultingin dense peripheral aggregations of granular materialsand a large, electron-lucid cytoplasmic volume in thecentre of the cell. The channel-type nanostructurearises spontaneously from within the peripheral regionsof dense, granular cytoplasmic material, coarsens overtime and grows to fill the volume of the cell [25]. Wehave hypothesized that this self-assembly processoccurs by phase separation, possibly regulated by therates of b-keratin expression and polymerization[25,38]. How could phase separation stop at the correctsize to be able to produce the appropriate colour?Phase separation could be arrested by either mechanicaljamming or a glass-transition in the b-keratin proteinphase [25,38], thus determining the characteristiclength scale of the nanostructure. Phase separation hasbeen studied in detail in other protein solutions such aslysozyme, etc. [45,64]. These hypotheses are ultimatelytestable with experimental analyses of the self-assemblyof b-keratin polymers. Upon barb cell death, the cyto-plasm dries out completely, and is replaced by airresulting in the final keratin-and-air amorphous photonicnanostructure.

The morphological similarities in the previouslypublished EM images [3,25] and experimental X-rayscattering data reported here (figure 5ad) forchannel- and sphere-type amorphous barb nanostruc-tures and synthetic self-assembled soft matter systemsis congruent with the hypothesis that avian barb nanos-tructures probably self-assemble via arrested phaseseparation of polymerizing b-keratin from the cellularcytoplasm, as suggested earlier for a few avian species[25,38]. Although the channel- and sphere-type barbnanostructures, respectively, appear to be similar tomorphologies observed during classical phase separ-ation via SD and nucleation-and-growth, to concludethat they indeed develop via phase separation,let alone assign a particular mode of phase separa-tion using just morphology is not straightforward[4547]. The lack of a perfect agreement between chan-nel-type barb nanostructures and classical spinodalmorphologies perhaps suggests that there may beimportant differences between a biological soft matter


system and a simple binary fluid de-mixing, perhapsinvolving some viscoelastic phase separation processes[4547]. Nevertheless, careful observations of b-keratinself-assembly in developing feathers, together within vitro investigations are necessary to pinpoint theprecise mechanisms of their self-assembly.

4.2. Double scattering of light by amorphousfeather barb nanostructures

The phenomena of double scattering and cortically pig-mentation of spongy nanostructures are distinguishableby their starkly differing angular dispersion (see elec-tronic supplementary material, figure S4bh),predictable spectral position of the relatively strongdouble-scattering peak in relation to the primarypeak, polarization dependence [36,37], and by the lim-ited classes of available pigments in birds [10]. Wehave documented the widespread occurrence of doublescattering in the optical function of amorphous featherbarb nanostructures (figure 7d f,mo and electronicsupplementary material, table S2). Many such huessuch as turquoise (light) blue (e.g. in male Cotingaspp., figure 7m and male Tersina viridis) include twostrong and distinct spectral peaks, one in the UV(visible to birds but not to humans) and the other ingreen. To birds, these hues are distinct colours stimulat-ing non-adjacent cone types (i.e. UV and medium-wavelength spectral sensitivities) in the avian retinathat will be perceived by birds as distinct colours[65,66]. Thus, the double scattering spectral featuresprobably contribute significantly to the colours ofnon-iridescent structural plumages perceived by birds,given that most birds can see in the UV/deep violet[11,67,68]. This suggests that the double scatteringfrom amorphous barb nanostructures constitutes asource of rich UV signals in birds [69].

The occurrence of double scattering in the opticalreflectances of amorphous barb nanostructures is prob-ably underestimated here, since at shorter wavelengths,it is much harder to separately estimate the double scat-tering peak from the primary peak and the sensitivity ofthe spectrophotometer steeply decreases. In fact, manyreflectance profiles with a short-wavelength primarypeak but without an obvious double-scattering peakhave a distinct shoulder on the short-wavelength sideof the reflectance spectrum (e.g. many Malurus spp.and tanagers, see figure 7j,k).

4.3. Structurepigment interactions

The combination of spongy medullary barb photonicnanostructure and carotenoid or psittacofulvin pig-ments is well known to produce longer wavelengthcolours that cannot be produced by pigments alone,such as structural greens [3,14,61]. Structural analysesof these feather barbs demonstrate that the underlyingbarb nanostructures are larger in spatial periodicitythan those producing purely structural hues (such asUV, violet and blue). Indeed, the underlying spongymedullary keratin nanostructure in each case wastuned to the appropriate length scale to produce theobserved reflectance peak by constructive interference



alone (figure 8 and electronic supplementary material,table S2). Therefore, contrary to prevalent simplisticnotions of colour mixing (structural blue pigmentaryyellow green) [70], the peaked or saturated green col-ours in feather barbs are produced by a combination ofmedullary barb nanostructures tuned to produce thoselonger wavelength colours and the absorption of someportions of the shorter wavelength double-scatteringpeak and the intermediate wavelengths between thetwo peaks. Similar structurepigment interactionshave been proposed in saturated yellow (Ramphastostoco, Ramphastidae) and orange (Tragopan caboti,Phasianidae) colours in avian skin [3,71].

In feathers with structurepigment interactions, theincident light first passes through the pigmented outercortex layer of the barbs, a portion of which is absorbedby the pigments, and the rest is transmitted to theunderlying medullary layer of nanostructure. The pig-ment acts like a band-pass filter, i.e. only the range ofwavelengths that is selectively transmitted by thepigments in the cortex can be scattered by the nano-structure underneath. However, the nanostructure istuned to constructively reinforce and scatter only a por-tion of the transmitted pigmentary reflectance, with thelonger wavelengths getting destructively interfered andthereby reducing the typical broadband longer wave-length reflectance of the pigments, just like in avianskin [3,71]. The light scattered by the nanostructureonce again passes through the pigmented cortex on itsway out of the feather barb. The result is a uniquelysaturated, brighter longer wavelength colour that isspectrally quite distinct from both the superficial pig-ments and the nanostructure below (figure 7).Furthermore, just as in avian skin [71], there appears tobe no intrinsic constraints to the production of longer-wavelength structural colours in feather barbs by con-structive interference from amorphous nanostructures(figure 8). However, given the propensity for doublescattering owing to the amorphous nature of spongybarb nanostructures [36,37] (figure 7df,mo and elec-tronic supplementary material, table S2), pure (highlysaturated) non-iridescent long-wavelength structuralhues are unlikely to occur in feather barbs [72].

The deposition of spectrally absorbing pigments super-ficially in the barb cortex can result in the attenuation ofthe double scattering short-wavelength structural peak inthe optical reflectance (see above), resulting in more satu-rated long-wavelength hues (e.g. purer structural greensand yellows as in many tanager feathers). However, a por-tion of the short-wavelength double-scattering peak can bereinforced, if the transmission spectra of the depositedpigments have an overlapping but relatively less intenseUV peak (characteristic of carotenoids like lutein, forinstance; figure 7gi,p,q) or not (as in the case of manytanagers; figure 7r). In bird species with both non-irides-cent blue and structural green plumage patches, therespective absence or presence of cortical pigments inthe barb appears to be the general mechanism by whichplumage reflectance differences arise between blue andgreen patches of sexually monomorphic species (e.g.Corythaeola cristata, Musophagidae; Merops viridis,Meropidae; Eumomota superciliaris, Momotidae;Aulacorynchus prasinus, Ramphastidae; Forpus


xanthopterygius, Psittacidae; Calyptomena hosii,Eurylaimidae; Vireolanius pulchellus, Vireonidae;Erythrura trichroa, Estrildidae; Chlorophonia occipita-lis, Fringillidae; Tangara chilensis, Thraupidae; seeelectronic supplementary material, table S2), andbetween homologous blue (usually in males) and green(usually in females) patches in the opposite sexes ofsexually dichromatic species (e.g. Lepidothrix spp.,Pipridae; Cyanerpes cyaneus, Te. viridis, Thraupidae;see electronic supplementary material, table S2).

In the optical reflectance of a further 21 (out of 255)feather barbs examined (see electronic supplementarymaterial, table S2), a longer wavelength pigmentaryreflectance plateau from apparently modified keto-carotenoids [73] or phaeo-melanins complements ashorter wavelength (violet to blue) single or doublestructural peaks, producing distinct extra-spectral col-ours including vivid shades of violets and purples toblue hues with hints of lilac (e.g. Acryllium vulturinum,Numididae, figure 7b; C. maynana, figure 7m; Maluruscoronatus, Maluridae; Sitta oenochlamys, Sittidae).Thus, complex interactions between structural and pig-mentary mechanisms produces hues that areunavailable to either modes of colour production, butadditional research is required to understand thenature of these interactions more precisely.

4.4. Evolution of non-iridescent barb structuralcolours

Non-iridescent, structural colour-producing three-dimen-sional amorphous barb nanostructures appear to haveindependently evolved at least 44 times within 41 familiesacross Aves, conservatively assuming a single evol-utionary origin within each family examined, with anadditional five families possessing species with marginalbarb structural coloration (see electronic supplementarymaterial, table S2). However, there appears to be mul-tiple origins of non-iridescent barb structural colourwithin some families (see electronic supplementarymaterial, table S2). An accurate estimate of the numberof evolutionary gains and losses of barb structural coloursin extant birds requires a well-resolved phylogeny of allbirds. Therefore, a formal, detailed macro-evolutionaryanalysis of barb structural colours will have to await sig-nificant progress in unravelling the evolutionary tree ofbirds. Nevertheless, the phylogenetic distribution ofphotonic barb nanostructures at the inter-ordinal levelis distinctly non-random. The channel nanostructures(n 36) occur more frequently than the spheres (n 8)among both passerine (perching) and non-passerinebirds (electronic supplementary material, table S2). Foreach avian family with barb structural colours, we haveexamined multiple species in different genera and multipleplumage patches within at least one species. In nearlyevery instance, all specimens sampled within a familyshared the same class of nanostructure. However, weidentified a few instances in which the nanostructuresvaried within a single avian family. Within the tanagers(Thraupidae), Cyanicterus cyanicterus, Diglossa cyanus(see electronic supplementary material, figure S1e) andXenodacnis petersi possess channel-type nanostructurewhile the rest of the tanagers examined have spheres.



But these three species are closely related members of adistinct branch of the family [74], and this couldrepresent a genuinely independent evolutionary origin ofcolour-producing nanostructure within the family. Simi-larly within the cardinals and grosbeaks (Cardinalidae),all taxa examined appear to possess a channel-type nano-structure except Cyanoloxia glaucocaerulea, which hasthe sphere-type morphology, based on SAXS data andSEM images (see electronic supplementary material,figure S1g,h).

4.5. Biomimicry lessons for amorphousphotonics

Recently, there has been a flurry of interest in the opticalproperties, design and synthesis of amorphous nanos-tructures at optical length scales, as they possess bothisotropic optical properties (non-iridescence) and omni-directional photonic bandgaps at high refractive indexcontrast [7580], unlike conventional, angle-dependentphotonic crystals [6]. In contrast to amorphous opalnanostructures of synthesized dielectric spheres featuredin contemporary engineering approaches [75,7880], thespongy feather barbs possess an amorphous inverse-opal nanostructure, which is likely to possess better opti-cal properties than their opal counterparts, by analogyto photonic crystals [6]. Indeed, the optical properties offeather barbs currently surpass those of engineeredamorphous colloidal photonic materials [51]. Amorphousphotonic nanostructures that are probably self-assembled in bird feather barbs with pronounced isotro-pic short-range order could therefore provide a useful,tunable biotemplate for positive cast replication ordielectric infiltration. A thorough understanding of thephysics and development of organismal structuralcolour that have evolved over millions of years of selec-tion for a consistent optical function may thus guidebio-inspired technological innovations [5,51,8183].

This work was supported with seed funding from the YaleNSF-MRSEC (DMR 1119826) and NSF grants to R.O.P.(DBI-DBI-0078376), H.C. (PHY-0957680) and E.R.D.(CAREER CBET-0547294) as well as Yale University fundsto V.S. and R.O.P. R.O.P. would like to acknowledgesupport of the Ikerbasque Science Fellowship and theDonostia International Physics Center. We thank twoanonymous reviewers for their helpful comments, ThomasValqui for his kind permission to use his photograph ofC. maynana and Kristof Zyskowski for help with birdtaxonomy. We are grateful to the Yale Peabody Museum ofNatural History, the University of Kansas Natural HistoryMuseum and Biodiversity Research Center, the AmericanMuseum of Natural History (Paul Sweet), the NaturalHistory Museum at the Academy of Natural Sciences (NateRice), the Harvard University Museum of ComparativeZoology (Jeremiah Trimble and Scott Edwards), and theUniversity of Oxford Natural History Museum (MalgosiaNowak-Kemp) for the feather samples. Tim Quinn obtainedTEM images of some bird feather barbs. SAXS data on birdfeathers were collected with the help of Alec Sandy andSuresh Narayanan at beam line 8-ID-I of the AdvancedPhoton Source at Argonne National Labs, and supported bythe US Department of Energy, Office of Science, Office ofBasic Energy Sciences, under Contract No. DE-AC02-06CH11357. We thank Nick Terrill and Tobias Richter forhelp with SAXS data collection at beamline I22 of the


Diamond Light Source (sm6905-1) that contributed to someof the results presented here. V.S. initiated, designed andperformed the research with R.O.P.; J.D.F. prepared thebiomimetic bidisperse PS sphere films; H.N. and S.-F.L.collected angle-resolved data on a few feathers; V.S.analysed and discussed the data with all authors; and V.S.wrote the manuscript with E.R.D and R.O.P.

REFERENCES

1 Kinoshita, S., Yoshioka, S. & Miyazaki, J. 2008 Physics ofstructural colors. Rep. Prog. Phys. 71, 076401. (doi:10.1088/0034-4885/71/7/076401)

2 Parker, A. R. 2000 515 million years of structural colour.J. Opt. A Pure Appl. Opt. 2, R15R28. (doi:10.1088/1464-4258/2/6/201)

3 Prum, R. O. 2006 Anatomy, physics, and evolution ofavian structural colors. In Bird coloration, volume 1mechanisms and measurements (eds G. E. Hill & K. J.McGraw), pp. 295353. Cambridge, MA: HarvardUniversity Press.

4 Srinivasarao, M. 1999 Nano-optics in the biological world:beetles, butterflies, birds, and moths. Chem. Rev. 99,19351961. (doi:10.1021/cr970080y)

5 Vukusic, P. & Sambles, J. R. 2003 Photonic structures inbiology. Nature 424, 852855. (doi:10.1038/nature01941)

6 Joannopoulos, J. D., Johnson, S. G., Winn, J. N. & Meade,R. D. 2008 Photonic crystals: molding the flow of light, 2ndedn. Princeton, NJ: Princeton University Press.

7 Prum,

Structure and optical function of amorphous photonic ... · distributions of scatterer size and...

Documents

Transcript of Structure and optical function of amorphous photonic ... · distributions of scatterer size and...