In July 2007, after working for quite a while on Silicon integrated data pipes (called WaveGuides), Intel’s photonic lab disclosed an optical modulator capable of encoding data onto an optical beam at a rate of 40 gigabits per second. At the time of announcement, Intel had already developed the Continuous Wave Silicon Raman Laser, so all that was missing in order to create an all-in-one communication chip was an optical detector capable of receiving data at the same high speeds.
In order to connect between the Silicon Raman Laser and the optical modulators, Intel used what they call waveguide devices, which are Silicon pipes designed to smoothly transmit light of a certain wavelength for optical communication purposes. In order to convert the waveguides into optical detectors, the company’s engineers needed to find a material that can absorb light. Germanium was found to be a material that can not only detect light, but is also much cheaper and more efficient than the exotic materials traditionally used for light detection purposes.
Based on these findings, Intel researchers added a very thin layer of Germanium onto the Silicon layer of the waveguide device. However, Intel’s engineers encountered a problem caused by the crystal lattices of Germanium and Silicon. Put simply, the Germanium layer did not perfectly match the Silicon layer. This mismatch caused an additional “noise” that disturbed the performance of the detector.
At a press briefing during the Intel Developer Forum last month, Intel Photonic Technology Lab Director Mario Paniccia announced the development of a photodetector that can reach a maximum speed of 31 Gigabits per second, a significant achievement in Silicon Germanium photodetectors. The breakthrough was achieved by improving the process of coating the Silicon with Germanium. The optimized thermal growth process minimized the impact of defects and allowed for a better match between the layers of the Germanium and the Silicon.
Intel’s researchers are currently working on developing a 40Gbps photodetector. However, according to Intel Research Lab engineer Tao Yin, there is still a lot of work to be done before this goal can be reached. Most of the work has to do with combining and packaging all three components – the hybrid silicon laser, the modulator, and the photodetector- onto a single Silicon chip. Once mass production of the optical communications chips commences, it will become possible to use them in order to integrate computers and optical devices and transmit terabits of aggregate data per second. Intel hopes to enable tera-scale computing in the not too distant future.
Further discussion of Intel’s Photonic Labs research can be found on the TFOT forums.