Micromanipulation - as an assembly tool for three-dimensional photonic crystals

One of the desired applications of photonic crystals is an optical integrated circuit which is equivalent to a highly integrated circuit of an electronic device. Electronic devices have adopted a three-dimensional (3D) arrangement of elements and wiring in order to formulate highly integrated components. To bring a similar success of the present electronic devices to the field of optoelectronics, the development of fabrication technologies of 3D photonic crystals must be undertaken. If all the waveguides and elements are supposed to be arranged in one 2D plane, high integration cannot be desired; waveguides will become unnecessarily long and complicated. Furthermore, realization of integrated optical circuits means that we need to be able to lay out elements made of various materials suitable to each purpose at exact positions as we designed, and connect them with intricate waveguides. The materials of the crystals are supposed to be semiconductors if we want to add active elements such as lasers and LEDs. There have been no technologies which fulfill all these requirements. Recently we have introduced a novel fabrication technology for a semiconductor 3D photonic crystal by uniting integrated circuit (IC) processing technology with micromanipulation. Four- to twenty-layered (five periods) crystals for infrared wavelength (3-4.5 micrometer) were integrated at predetermined positions on a chip with a structural error of within 50 nm. Observation of the PBG confirmed the precision of our technique. A crystal with a controlled defect was also arranged on the same chip. Numerical calculations revealed that a transmission peak observed at the upper frequency edge of the bandgap originated from the excitation of a resonant guided mode in the defective layers. This technology offers immense potential in becoming a breakthrough in the production of optical wavelength photonic crystal device. In the field of photonic crystals, our method is considered as a very eccentric approach. However, when we widely overlook the industrial production lines, it turns out that manipulation technique has already deeply taken in the field. For example, mass production of the high-density mounting circuits for mobile computing products are enabled by a manipulation robot's ultra high-speed assembly based on image recognition. In the case of photonic crystal devices, things are not easy as the case of electronic devices, because assembly in the micro world is quite different from what we experience in the macroscopic scale; surface effects and electrostatic force rule over more strongly than inertia. Moreover, part size and accuracy required for photonic crystals are about 1000 times smaller in scale of that required in electronic products. Because of these difficulties, an operator operates micromanipulation system to assemble elements one by one for now, thus not a few people claim that our technique is just a toy for research. However, if the systems for image recognition and feedback in micro and nano scales are established, mass production of photonic crystal devices by automatic manipulation is not a dream. Also the technology which can perform stable assembly in these scales is indispensable not only to the field of photonic crystal device but to the field of the electron device and micromachining in which miniaturization is progressing at an increasing speed.