The reporter learned from China Information and Communication Technology Group that my country's self-developed "100G silicon optical transceiver chip" was put into production and used in Wuhan a few days ago, and passed the user's existing network test, with stable and reliable performance. This marks the successful development of my country's commercial 100G silicon optical chip, and will promote my country's independent silicon optical chip technology to a new level.
This silicon optical chip was jointly developed by the National Information Optoelectronics Innovation Center, the State Key Laboratory of Optical Fiber Communication Technology and Network, Wuhan Guangxun Technology Co., Ltd. and China Information and Communication Technology Group. It is integrated on a silicon chip less than 30 square millimeters. It includes nearly 60 active and passive optical components including optical transmission, modulation, and reception. The chip has the advantages of ultra-small, high-performance, low-cost, generalization, etc., and can be widely used in transmission networks and data center optical transmission equipment.
The core concept of silicon photonics technology is "replace electricity with light", that is, using laser beams instead of electronic signals to transmit data, and integrating optical devices and electronic components into a single microchip. The use of light instead of traditional copper wire as the information transmission medium on the silicon chip greatly improves the connection speed between the chips.
In recent years, silicon photonics technology has continued to develop, and companies represented by Luxtera, Intel and IBM have continuously launched commercial-grade silicon photonics integrated products. In 2018, the global silicon optical chip and its packaging device market will be close to 200 million US dollars, and the overall market is expected to maintain rapid growth. According to Yole's prediction, the silicon photonics market will exceed US$1.3 billion by 2025, of which more than 90% will come from data center applications.
Experts said that the commercialization of 100G silicon optical chips indicates that my country has the conditions and foundation for the commercial design of silicon optical products. We believe that with the continuous explosion of traffic, the "light advancement and retreat" at the chip level will become the general trend, and silicon-optical integration is expected to be applied on a large scale in the future.
Yu Shaohua, director of the Expert Committee of the National Information Optoelectronics Innovation Center and academician of the Chinese Academy of Engineering, said that the commercialization of 100G silicon optical chips indicates that my country has the conditions and foundation for the commercial design of silicon optical products. In the future, silicon optical technology will be deployed and applied on a large scale in optical communication systems, pushing my country's independent silicon optical chip technology to the direction of ultra-high speed, ultra-large capacity, ultra-long distance, high integration, high performance, low power consumption, and high reliability. development of.
Problems faced by silicon optical chips
Silicon Photonics is attractive because of its low cost, low power consumption, mature CMOS technology, high integration, compact structure, and ultimately replace copper interconnects to realize optical interconnection. It is undoubtedly the general trend of future development, but the current problems limit its application.
One is the problem of light sources. Because the silicon material itself has an indirect band gap and it is difficult to emit light, silicon-based lasers were once recognized as a world problem. In other words, in addition to lasers, almost all other optical device functions can be realized by silicon optics, and foreign manufacturers have also made finished products ready for mass shipment. However, silicon-based lasers have not yet made major technological breakthroughs. At present, the relatively hot research on indium-phosphorus gain media and silicon-based waveguide hybrid integrated (non-monolithic) semiconductor lasers can barely be regarded as silicon-based, but there are also problems such as coupling and tuning. , To be truly commercial, a certain technological improvement is needed.
The other is cost. Although the cost of silicon-based raw materials is low, traditional CMOS production lines cannot directly produce silicon optical devices, and certain optimizations are needed. Moreover, from the laboratory to the commercial use, it is necessary to consider the packaging scheme and the module yield and other issues, so the cost of silicon optical devices is not necessarily much lower than that of the three and five groups. If the cost remains high, the operator is unwilling to buy it, so how to promote it
For now, it is difficult to break the barriers, and it will take some time for silicon photonics technology to really get out of the laboratory. At that time, Moore's Law or something will probably not be used.
Source: The content comes from "Economic Information Daily", thank you.