Interview With Zhang Xiaoguang Of Beijing University Of Posts And Telecommunications: The Plight And Future Of Photonic Chips

In recent years, with the rapid growth of global traffic, the Internet, artificial intelligence and cloud computing application scenarios continue to increase, as one of the future trends in high-speed communication, photonic chips have also ushered in a golden period of development. Like its "predecessor" circuit, the optical circuit is also in the miniaturization, power reduction and low-cost requirements of the call for integration, integrated optical(PIC) came into being, and has become an important part of the modern communications industry can not be ignored. In order to further clarify the predicament and future of the integrated technology, Wikipedia had a face-to-face conversation with Prof. Zhang Xiaoguang from the School of Electronic Engineering, Beijing University of Posts and Telecommunications (BUPT), discussing the past and present life of photonic chips.
At the beginning of the conversation, Prof. Zhang Xiaoguang got right to the point, "Chip integration technology is now very mature, and is widely used in various electronic devices, such as cell phones, computers, smart home devices, and so on, all usingintegration technology. However, there are still shortcomings in China'sintegration technology, and integrated optical chips are one of the difficulties."

An integrated opticalis a functional circuit formed by a microchip containing two or more photonic components, hence the name photonic chip. The photonichas the advantages of high-speed parallelism and low power consumption, and its calculation speed and transmission rate are a thousand times that of electronic chips, while its power consumption is only one-ninety-thousandth of that of electronic chips, and its structural requirements are also lower, generally at the hundred-nanometer level, thus reducing the dependence on advanced processes, and it is considered to be the most suitable underlying technology for solving the bottleneck of the development of arithmetic power.
In terms of working principle, photonic chips utilize photons to transmit, sense, process and deliver information. Waveguides are utilized to control and direct the light, and through total internal reflection, photonic chips are comparable to wires used to transmit electrical signals. A laser source provides the light needed to drive the components, similar to a switch in a circuit. Through waveguide connections, multiple components can be integrated and fabricated on a single substrate, creating robust and miniaturized solutions. Components in photonic circuits can be passive and active, with passive components including switches and multiplexers and active components including detectors and lasers.
Compared to electronic chips, the difficulty of integrating photonic chips is mainly due to their discrete nature. From the manufacturing point of view, photonic integratedmanufacturing is not an easy thing. Photonic devices have a three-dimensional structure, much more complex than the two-dimensional structure of the semiconductor integration. To integrate lasers, detectors, modulators and other devices into the chip, it is necessary to repeatedly deposit and etch on multiple thin-film dielectric layers of different materials, such as indium phosphide, which is very demanding for the equipment and process requirements.
"However, I am still confident about overcoming these difficulties domestically." Prof. Zhang Xiaoguang said, "There is already a trend towards silicon-based integration in the industry, as this kind of system integration is already relatively mature, and domestic research on silicon optical chips has been basically synchronized with the world. But how to use silicon as the substrate of optical components and realize the technology mass production, there are still a lot of difficulties in the inside, and now some of the devices that perform better in our experimental process still need to rely on imports."
In Prof. Zhang Xiaoguang's opinion, the internationalindustry has been developed for decades. Europe, the United States, Japan and other countries in the opticalindustry started earlier, leading technology. The United States has long established the "National Institute for Photonic Integrated Manufacturing Innovation" to create a platform for the development and preparation of photonic integrated devices; the European Union to implement the "Horizon 2020" program, focusing on the deployment of optoelectronic integration research projects; Japan to implement the "Advanced Research and Development Program", the "Advanced Research and Development Program". Japan's implementation of the "Advanced Research and Development Program", the deployment of optoelectronic fusion system technology development projects. Overseas opticalcompanies have a first-mover advantage, through the accumulation of core technologies and production processes, and gradually realize the industry closed loop, to establish a high industry barriers.
Domestic opticalmanufacturers generally have the back-end processing capabilities in addition to wafer epitaxy, opticalcore epitaxial technology is not mature, high-end epitaxial wafer to international epitaxial plant procurement, limiting the development of high-end optical chip. Laser chip, for example, China can scale mass production of 10G and the following low and medium rate laser chip, but 25G laseronly a small number of vendors to realize the batch shipment, 25G or more rate lasermost of the vendors are still in the R & D or small-scale trial production stage.
At the time of the previous late start of domestic technology, the domesticintegration research is more to stand in the height of the ready-made research results of foreign countries to develop and innovate, with the changes in the international form of the domestic high-tech industry in the field of technology has received more and more restrictions, but also makes more people realize the importance of technological self-reliance. China launched the "large-scale photonic integrated chip" as early as 2016, the Chinese Academy of Sciences strategic pilot scientific and technological special projects, promoting the vigorous development of domestic independent innovation.
"However, the most important thing for the country to achieve real industrial breakthroughs lies in the construction of the system," Prof. Zhang Xiaoguang emphasized at the end of the talk, "from the point of view of the actual production, the most important step to achieve self-reliance is to establish the overall industrial system, in particular, the formation of a complete set of process flow, if the principle can be understood in the first place, then it is necessary to establish the system. If in the principle of the case of all can get through, because of the immature process can not guarantee the product yield, independent innovation is still only on paper."