Photonic crystals (PCs) represent an interesting class of materials with a variety of potential uses in optical telecommunication systems.1 They are created when a material with a high refractive index (e.g., silicon) and a substance with a low refractive index (e.g., air) form a structure such as a crystal lattice. The periodic variation leads to properties that may facilitate improvements in optical telecommunication systems by increasing transmission speed, simplifying integration, and reducing size and costs.
PCs do not allow the propagation of light at certain wavelengths inasmuch as they possess a ‘photonic bandgap’ comparable to the electronic bandgap in semiconductors. Rather surprisingly, however, they can serve as new materials for photonic devices operating at precisely these forbidden wavelengths. A perfect PC can be transformed into a photonic component such as waveguide, splitter, or filter by disturbing the periodicity of the crystal lattice in a controlled fashion. This is done through the introduction of so-called defects by adding or removing material from the PC. Defects locally close the bandgap, allowing propagation of light within this wavelength range. A line of defects, for example, acts as a simple straight waveguide. Introducing more complex defect structures into 3D PCs enables the production of photonic devices with complex functions.
Figure 1. Scanning electron microscope picture of 1000nm polystyrene beads arranged in an opal structure.
We have fabricated PCs of 1000nm polystyrene beads arranged in an opal structure on cover glass substrates using the self-assembly technique.2 Figure 1 shows such a crystal. The bead diameter of 1000nm was chosen to provide a photonic band gap around the 1.5μm telecom wavelength in the inverted crystal. To create defect structures within the crystal we first fill the interstitial space between the balls with a liquid polymer (ORMOCER). The next step involves local solidification of this liquid through a polymerization process that forms the desired defect structure. Usually UV light is employed for this purpose, but to write 3D structures deep within the PC, we use visible or infrared light via a two-photon process. Two-photon polymerization requires high photon densities that can only be provided by tightly focused femtosecond laser pulses.
In recent experiments we used a mode-locked frequency-doubled ytterbium-doped glass laser operating at a wavelength of 515nm with a pulse width of 240fs. Because single-mode waveguides require diameters as small as 1μm, the laser light must be tightly focused by using, for example, a 100× oil immersion microscope objective. For proof of principle we used a 20× objective to write an array of vertical waveguides with diameters ranging from 5 to 50μm.
Figure 2. Two optical microscope images of a photonic crystal show an array of vertical waveguides with varying diameter (5–50μm) and constant (top) and varying laser power (bottom).
After laser exposure the remaining nonpolymerized ORMOCER is removed in an isopropanol bath. Figure 2 shows two microscope pictures of a PC with an array of vertical waveguides. ORMOCER and the polystyrene beads have similar refractive indices. Due to this index matching, any part of the PC with polymerized ORMOCER transmits light, whereas the remaining area appears opaque. Hence, the vertical waveguides appear as bright spots. In this experiment the minimal waveguide diameter was 5μm. This can easily be scaled down to smaller diameters using a focusing objective with higher magnification and higher numerical aperture.
We have also written simple horizontal waveguides3 and other more complex structures such as Mach–Zehnder interferometers.4 Combining the self-assembly technique with simple two-photon polymerization offers several advantages, including precise control of location, shape, and alignment of embedded defects in photonic crystals.
The remaining nontrivial step is the ‘inversion’ of the polystyrene opal template. This step is necessary because the index difference between polystyrene and air is not large enough to form a complete photonic bandgap. In this process the interstitial space between the polystyrene beads is first filled with a high-index material. Subsequently, the beads and polymerized ORMOCER are removed by simply heating the crystal in an oven, leaving behind an inverted photonic crystal with the desired embedded defect waveguides. Successful inversions of opal templates with titanium dioxide and silicon have been achieved. Encouraged by these results, we will continue to improve the quality of the polystyrene opals and the two-photon polymerization process to produce high-quality templates for new optical devices in telecommunication.
This work has been supported in part by the European Commission within the Sixth Framework Program (STREP) NMP-4-CT-2005- 017160 (NewTon).5 The authors would also like to thank Cefe L´opez for the silicon inversion.
Thorsten Schweizer, André Neumeister, Rainer Kling
Laser Zentrum Hannover (LZH) e.V.
Thorsten Schweizer received his PhD in physics from the University of Hamburg and spent six years at the Optoelectronics Research Centre in Southampton, UK. He recently joined the Microtechnology Group at LZH to work on defect structures in photonic crystals.
André Neumeister received his degree in engineering physics from the Münster University of Applied Sciences. He is currently working toward a PhD at LZH in the area of laser-assisted microengineering with an emphasis on additive micro-manufacturing using polymers.
Rainer Kling heads the department of production and systems at LZH and has been working to advance laser micromaterial processing. He earned his PhD from the University of Hannover in 1997 and began work with lasers at the National Institute of Standards and Technology, Washington, DC, in 1998.
Wendel Wohlleben, Reinhold J. Leyrer
Wendel Wohlleben heads a laboratory in the department of polymer physics. His main interest is in nanotechnology, including self-assembly, innovative analytical techniques of nanosuspensions, and the risk assessment of nanomaterials.
Reinhold J. Leyrer is a group leader in the department of polymer colloids. His research is focused on synthesis by emulsion polymerization with applications for functional polymers, including elastic polymer opals and templates for inverse opal photonic crystals.
3. M. Boyle, A. Neumeister, R. Kiyan, C. Reinhardt, U. Stute, B. Chichkov, W. Wohlleben, R. J. Leyrer, Production of 3D photonic components with ultrafast micromachining, Proc. SPIE 6462, pp. 646212, 2007.doi:10.1117/12.698564