A new approach to manipulating light propagation

A new, facile method can change the light propagation properties of photonic crystals made of chromogenic materials. Precious opals are formed in nature by regularly ordering silica spheres on a light wavelength scale, causing them to display bright colors due to the interference and diffraction of white light. Photonic crystals are essentially artificial opals achieved by the spatial periodic arrangement of thin films (Bragg mirrors), rods, or particles, in one, two, or three dimensions, respectively, on the light wavelength scale.¹ A light beam traveling through these crystals experiences a periodic variation of dielectric constants (refractive index) that creates special optical effects. Some wavelengths (colors) are forbidden from passing through because a photonic band gap (PBG) develops. This leads to many possible features such as wavelength selective flow, localization, and control of spontaneous emission of light for photonic components such as switches, tunable filters, and reconfigurable networks.

Much work has been done to design and fabricate passive photonic crystals with a fixed PBG. The PBG depends mainly on the medium's refractive-index contrast and the spatial-periodicity scale. However, tunable PBGs are desirable so that a single photonic crystal can be used for multiple-wavelength ranges. PBG tunability with liquid crystals or ferroelectric materials with a large variation in optical and dielectric properties under an applied electric field or temperature has been tried.^2,3 However, a fine tunability is yet to be realized. Variations of lattice constants by external mechanical,⁴ electrical, or magnetic fields⁵ or light⁶ also have severe limitations due to structural damage and/or device conception.

Thus, we used transition metal oxide-based photonic crystals as a facile approach to PBG tunability. Transition metal oxides (TMOs) such as tungsten trioxide (WO₃) are known for their chromogenic properties, such as a reversible optical property change under the influence of an external electric field (electrochromic), or heat (thermochromic), or light (photochromic).⁷ The multiple valence states in the TMO allow for an easy transfer of electrons between these states through a small externally applied activation accompanied by a reversible optical property change. The idea is to build photonic crystals out of transition metal oxides and to externally influence their refractive index change. We recently demonstrated the feasibility of this approach through an electrochromic process.⁸

Applying a small voltage triggers a double insertion/extraction of electrons and ions as shown below:

The inserted electrons can be localized on tungsten ion sites or can be free in the WO₃ matrix, depending on whether the film is amorphous or crystalline. The modulation in optical properties occurs either through absorption (amorphous WO₃) or through reflection (crystalline WO₃). Both of these optical changes are accompanied by a drastic change in the optical constants.

We have built photonic crystals from WO₃ to exploit their electrochromic property. A four-component device (see Figure 1) was fabricated where, upon applying an electric field between the two transparent conducting ITO layers, lithium ions from the LiClO₄electrolyte and electrons from the cathode are inserted into the periodically nanostructured WO₃ inverse opal. The systematic movement of the PBG that can be expected in WO₃ is a function of the quantity of lithium inserted (see Figure 2).

Figure 1. An electrochromic photonic-crystal device based on tungsten trioxide (WO₃)inverse opal (dark region, WO₃; white. air gaps). ITO: Indium tin oxide. LiClO₄/PC: Lithium perchlorate propylene carbonate.

Figure 2. Reflection spectra of a 300nm inverse-photonic crystal resulting from complex changes in optical constants with lithium insertion: (a) movement of band gaps (light wavelengths that are not transmitted across the sample, but are reflected back). Expanded views of the movements of (b) the gap at 532nm to lower wavelengths, and (c) the gap at 344nm to higher wavelengths.

While our earlier work shows the feasibility of this approach, significant improvements in the optical efficiency and wavelength range need to be achieved. We are currently working to reduce the absorption by using predominantly crystalline films where mainly reflection changes are expected. We also are working on photonic crystals based on other transition metal oxides.