High-end optoelectronic chips are the biggest short board
In the process of optical communication transmission, the transmitting end converts the electrical signal into an optical signal, and then modulates it to the laser to emit a laser beam, which is transmitted through the optical fiber, and then receives the optical signal at the receiving end and then converts it into an electrical signal, after being modulated and demodulated. It becomes information, and the role played by the optoelectronic chip is to realize the mutual conversion between the electric signal and the optical signal. It is the core of the optoelectronic technology product and is in the pyramid tip of the field of optical communication.
At present, there are not many enterprises that can produce optoelectronic chips in China, and there are about 30 companies, most of which can produce low-end chips in large quantities.
Only a few manufacturers such as Lightspeed Technology, Hisense, Huawei, and Bonfire can produce medium and high-end chips, but the overall supply is limited, and the market share is less than 1%. High-end chips rely heavily on Broadcom, Mitsubishi and other US and Japan companies. China Telecom Science and Technology Commission Director Wei Leping said that in the core construction cost of optical networks such as routers, base stations, transmission systems, and access networks, the cost of optical devices accounts for as much as 60-80%, and the core reason for the high cost of optical devices is that high-end chips cannot be completely domestically produced. It needs to rely on imports, so high-end optoelectronic chips should become the key point for China's optical communications industry to overcome.
What's the problem
China's optoelectronic chip industry is relatively backward, both related to internal R&D strength and external environment.
In terms of internal research and development, optoelectronic chips are highly integrated components, including integrated components such as lasers, modulators, couplers, beam splitters, wavelength division multiplexers, detectors, etc. There are two major categories in the industry. Chip packaging solutions, one is III-V family, the other is silicon light, the former technology is relatively mature, there is a mature monolithic integrated solution, the latter's laser integration and packaging solutions are still perfect.
In general, the development process of optoelectronic chips can be decomposed into three parts, namely epitaxial material design, epitaxial material growth and back-end process preparation. Epitaxial material design refers to the design of chip epitaxial structure that meets application requirements by means of simulation software. The preparation of epitaxial materials refers to the use of related methods to grow epitaxial materials that meet the design requirements. The quality of the epitaxial materials is often an important factor affecting the performance of the optical chip. The back-end process is prepared by using semiconductor-related processes to make the epitaxial materials have a certain surface. Structure of the optoelectronic chip.
China's optoelectronic chip research and development, design, film processing, packaging, etc., compared with foreign countries, are somewhat lacking. According to the China Electronic Components Industry Association issued "China optoelectronic device industry technology development road map (2018-2022) "The domestic enterprises have only mastered the laser, detector, modulator chip, and PLC/AWG chip manufacturing process and supporting IC design, packaging and testing capabilities, overall level and international benchmarking companies with 10Gb/s rate and below. There is a big gap, especially the high-end chip capability is more than 1-2 generations behind the developed countries of the United States and Japan. Moreover, China's optoelectronic chip chip processing is also heavily dependent on the United States, Singapore, Canada and other countries.
In terms of the external environment, in the global electronic information industry, China is still a newcomer to the industry. According to the law of industrial development, newcomers must first start from the relatively easy industrial links such as the whole machine and system. Dr. Zhu Shaozhen, Director of the Super Moore Research Office of Di Chiku Integrated Circuit Research Institute, told the "Communication Industry News" (network) reporter that the current stage of the industry in China determines that the whole machine and system enterprises will preferentially purchase optical chips from global leading suppliers. Thus, the determination to independently develop chips is relaxed.
Obviously, the research and development process of optoelectronic chips is extremely complicated. It requires not only a certain amount of technology accumulation, but also a large investment, and the R&D and production cycle are also long. This is especially true for high-end chips. This means that SMEs are difficult to use in high-end optoelectronic chips. R&D has made a difference, and even large enterprises have not received enough user feedback during the R&D process, and timely corrections have been made in the commercial process.
How to break through
To overcome high-end chips, we must first break the loop of domestic communication equipment manufacturers that are unwilling to use them, and R&D vendors are not willing to invest heavily. This requires increasing investment in information infrastructure construction, coordinating industrial layout, improving corporate survival environment, and creating a good industry. Ecology.
In addition, Mao Qian, director of the Optical Communications Committee of the China Communications Society and director of the Asia-Pacific Optical Communications Committee, believes that the chip technology is very complicated, and the success of the process is closely related to people. It is also a top priority to train and attract high-tech talents.
Zhu Shaozheng suggested that domestic communication equipment companies should judge the current international situation, increase the research and development of independent products, and use the advantages of the system to increase the proportion of trial and error. It is also necessary to use optical chips to start the application of consumer electronics in the east (such as Face recognition VCSEL optical chip), to take advantage of technology reserves, increase product and market layout, and enhance the competitiveness of optical chip products by market means.
Fortunately, many companies and research institutes in China have taken the lead in the field of optoelectronic chips, and have achieved certain results. For example, Wuhan Institute of Posts and Telecommunications and other scientific research institutions have been deeply involved in the field of optical devices for many years, with deep technical reserves. In 2013, Sis joined the chip battlefield through the acquisition of the Belgian silicon photonics company Caliopa, and later acquired the British photonics integration company CIP. At present, the company has mastered the 100G optical module technology, ready to further overcome the difficulties and achieve mass production.
The company's optical Xun technology recently launched the 120GCXP module and the 100GQSFP28SR4 module, which is the first domestic chip to achieve 100G rate optical module in China. In addition, Hisense, Huagong Technology are cultivating high-end optoelectronic chip market.
However, it takes a long time for all of the above-mentioned companies' chip products to achieve internal procurement. Zhu Shaozhen emphasized that for China's optical chip development, some necessary 'credits' still have to be repaired.