3D Polarization-Independent Invisibility Cloak at Visible Wavelengths
Authors (from left to right):
Tolga Ergin, Joachim Fischer, and Martin Wegener
Institute of Applied Physics, Karlsruhe Institute of Technology, Germany
Link to Martin Wegener's Group >>
Invisibility cloaks are a fascinating subject for laymen, poets, movie directors, and scientists alike. In recent years, remarkable progress was made in realizing such devices. One design, the “carpet cloak” , turned out to be especially promising. Here, the cloak sits on top of a bump in a mirror. An object can be hidden under the mirror (the “carpet”), yet without cloak, the bump would still be seen as a distortion in the mirror reflection. The cloak redirects and bends the light such that this distortion is countered – the mirror appears to be flat and the object is invisible.
Multiple exciting realizations of the carpet cloak in effectively two-dimensional geometries and at different wavelengths were demonstrated, both for microscopic [2,3,4,5,6] and macroscopic devices [7,8]. However, due to the two-dimensional nature and the consequential polarization dependence of these structures, the cloaking effect disappears once the structure is inspected from the third dimension or with a different polarization. Following our 2010 work , which demonstrated 3D polarization-independent cloaking at optical frequencies for the first time, we were able to shrink our device such that it now operates at wavelengths that are visible to the human eye . To accomplish this, we used a lithographic technique called stimulated-emission-depletion-inspired direct laser writing .
The demanding part in the fabrication of these structures is to create a locally and gradually varying index of refraction. Our design accomplishes that by using a woodpile photonic crystal in the long-wavelength limit, meaning that the feature size of the woodpile structure has to be smaller than the operating wavelength. Here, the woodpile acts as an effective material – the light does not “see” the substructure of polymer rods and air, it rather “sees” a homogeneous material with a certain index of refraction at every point in space. The value of the refractive index is controlled via the local polymer-air ratio (see Fig. 1b).
Figure 1: (a) Electron micrograph of the reference (top) and the cloak (bottom) structure. (b) To show the interior, the structures are cut by focused-ion-beam milling. The constant filling fraction of the reference is clearly visible, whereas the cloak shows a locally changing polymer-air ratio. The photos in (c) and (d) are taken with a usual digital camera through a standard optical microscope at 700 nm wavelength. (c) is taken from the air side, where we expect to see the bump’s distortion (two black stripes) both in the reference and in the cloak. (d) When the sample is flipped around and inspected from the glass side, the distortions in the cloaking structure disappear. [Image reproduced from the paper published in Optics Letters. Link]
To document the distortion due to the bump that we are trying to hide, we fabricate reference samples right next to the cloak. These references have the same bump in the carpet, but a constant woodpile filling fraction (and hence, constant refractive index) – no cloaking is expected here. As a control experiment, we image the structures from the air side with a usual microscope (Fig. 1c). The depicted photographs give a very good impression of the image that one sees with the human eye and a color filter. As expected, both structures show an identical distortion (dark stripes). In sharp contrast, the distortion in the cloaking structure is almost completely gone when the sample is flipped around and inspected from the glass side (Fig. 1d) – the carpet mirror looks flat.
Figure 2: Dark-field mode. The sample is tilted by 30°. Here, the bump in the reference structure (top) lights up due to reflections off of the bump’s side. For the cloak (bottom), the bright stripe disappears – just like a flat mirror would look like. [Image reproduced from the paper published in Optics Letters. Link]
The exciting fact about this realization of the carpet invisibility cloak is that it works in three dimensions and with unpolarized light – both facts that are natural to the human sight and experience. To show the 3D performance of our cloak, we tilted the sample by 30° with respect to the optical axis (“dark-field mode”, Fig. 2). Here, in contrast to normal incidence imaging (“bright-field mode”, Fig. 1) , an ideal flat mirror looks dark, since most of the illuminating light is reflected to the side and is not collected by the microscope objective. The bump in the carpet, on the other hand, lights up as a bright stripe, since light is scattered off of it. With the cloak in place, the bright stripe disappears. Again, this corresponds to the reflection at a flat mirror.
Figure 3: Measurements of the cloak similar to Fig. 1c, but for different illumination wavelengths. Cloaking persists down to 650 nm. [Image reproduced from the paper published in Optics Letters. Link]
We also measured the wavelength dependence of the cloaking effect (Fig. 3). We found that for all wavelengths, for which the woodpile photonic crystal can be regarded as a good effective medium, the cloak works very well. This is true for wavelengths larger than 650 nm. In fact, we expect the cloak to be extremely broadband and to keep its functionality up to about 3 µm, where absorption of the polymer sets in. For wavelengths shorter than 650 nm, the light starts to “see” the polymer rods and air holes, which leads to scattering and refraction.
The carpet cloak is a fascinating example of newly developed devices based on the theory of transformation optics and demonstrates the fabrication finesse that is possible with the 3D laser lithography of present days.
 Jensen Li and J. B. Pendry, "Hiding under the Carpet: A New Strategy for Cloaking", Phys. Rev. Lett. 101, 203901 (2008). Abstract.
 R. Liu, C. Ji, J. J. Mock, J. Y. Chin, T. J. Cui, D. R. Smith, "Broadband Ground-Plane Cloak", Science 323, 366 (2009). Abstract.
 Jason Valentine, Jensen Li, Thomas Zentgraf, Guy Bartal & Xiang Zhang, "An optical cloak made of dielectrics", Nature Materials, 8, 568 (2009). Abstract. 2Physics Article.
 Lucas H. Gabrielli, Jaime Cardenas, Carl B. Poitras & Michal Lipson, Nature Photonics, 3, 461 (2009). Abstract.
 J. H. Lee, J. Blair, V. A. Tamma, Q. Wu, S. J. Rhee, C. J. Summers, and W. Park, "Direct visualization of optical frequency invisibility cloak based on silicon nanorod array", Optics Express, 17, 12922 (2009). Abstract.
 Majid Gharghi, Christopher Gladden, Thomas Zentgraf, Yongmin Liu, Xiaobo Yin, Jason Valentine, Xiang Zhang, "A Carpet Cloak for Visible Light", Nano Letters. To be published. DOI:10.1021/nl201189z
 Baile Zhang, Yuan Luo, Xiaogang Liu, George Barbastathis, “Macroscopic invisibility cloak for visible light”, Phys. Rev. Lett. 106, 033901 (2011). Abstract.
 Xianzhong Chen, Yu Luo, Jingjing Zhang, Kyle Jiang, John B. Pendry & Shuang Zhang, "Macroscopic invisibility cloaking of visible light", Nature Communications, 2, 176 (2011). Abstract. 2Physics Article.
 Tolga Ergin, Nicolas Stenger, Patrice Brenner, John B. Pendry and Martin Wegener, "Three-Dimensional Invisibility Cloak at Optical Wavelengths", Science 328, 337 (2010). Abstract. 2Physics Article.
 Joachim Fischer, Tolga Ergin, Martin Wegener, "Three-dimensional polarization-independent visible-frequency carpet invisibility cloak", Optics Letters, 36, 2059 (2011). Abstract.
 J. Fischer et al., Optics Materials Express, submitted (2011).