How do single-scattering structure-factor and form-factor models compare with multiple-scattering treatments in predicting angle-independent color in disordered colloidal assemblies?

Why single-scattering models get the peak right but the color wrong

In a disordered colloidal assembly, the main color peak comes from short-range correlations between particles—essentially, the typical spacing between neighbors sets the wavelength that is most strongly reflected. A single-scattering model that uses an effective refractive index for the material predicts the position of this peak accurately [1]. That success shows the peak's origin is indeed short-range order, not long-range periodicity.

But the same model does not quantitatively reproduce the observed color. The measured main peak is much broader, and there is much more reflectance at low wavelengths than the model predicts [1]. These discrepancies are not minor: they change how saturated the color looks. A broader peak and extra short-wavelength reflectance both tend to wash out the color, reducing saturation [1].

What multiple scattering adds to the picture

Multiple scattering means light bounces around inside the material before it exits. One significant contributor to peak broadening is light that scatters, is totally internally reflected at the boundary of the sample, and then scatters again [1]. This path effectively spreads the reflected light over a wider range of wavelengths, broadening the main peak.

The high reflectance at low wavelengths also comes from multiple scattering, but its origin is different: the scattering cross-section of individual particles increases as wavelength decreases [1]. In other words, shorter wavelengths are scattered more strongly per particle, so they are more likely to undergo multiple scattering events and eventually exit the sample. Both effects—boundary-reflected light and the wavelength-dependent cross-section—reduce color saturation [1].

How Monte Carlo models turn multiple scattering into a color prediction

Monte Carlo simulations track many possible paths of light through a disordered structure, including repeated scattering events. When these simulations account for surface roughness on the film, particle polydispersity (a mix of particle sizes), and wavelength-dependent absorption in the components, they produce quantitative agreement with measured reflectance spectra [2]. This means the model can predict not just the peak position but the full spectrum that determines the perceived color.

The approach is parameterized in terms of experimental variables, which simplifies the connection between simulation and fabrication [2]. It has been used to reproduce the color of the male mountain bluebird in an experimental system and to find the limits of angle-independent structural colors for a given set of materials [2]. A related multiscale model extends this to films of disordered photonic balls—nanostructured spheres tens of micrometers across—and accurately predicts their reflectance spectra as well [3][5].

When multiple scattering matters most—and when it can be suppressed

Multiple scattering becomes especially important when the material is strongly scattering, such as when the particles have a high refractive index contrast with their surroundings or when the sample is thick. In the Rayleigh scattering regime—where particles are much smaller than the wavelength of light—Monte Carlo simulations show that multiple scattering and absorption can interact counterintuitively: adding black absorbents to a white, opaque material can actually produce a blue color in diffuse reflection [4]. This is a case where multiple scattering is not just a nuisance but part of the color-generating mechanism.

For applications that require saturated structural color, the practical lesson is that single-scattering models can still be useful for design—but in a different way. They can be used to design materials in which multiple scattering is suppressed, even without absorbing components, so that the color becomes saturated [1]. In other words, the single-scattering model tells you where the peak will be and how to minimize the multiple-scattering effects that dilute it; the multiple-scattering model tells you what color you will actually see when those effects are present.

Sources used in this answer

Related research questions