Xinwu (ID: ) Wen | Dong Wenshu
Xinshixun reported on August 26 that just two days ago, the research team of carbon nano-semiconductor materials at Peking University published a paper in the world's top academic journal "Nature Electronics". The research team has developed "radiation-resistant" carbon nanotube transistors and integrated circuits, which can be used in special application scenarios with strong radiation such as aerospace and nuclear industries. This research result means that my country's carbon-based semiconductor research has successfully broken through the worldwide problem of radiation resistance, laying a solid foundation for the development of radiation-resistant carbon-based chips.
"Carbon nanotubes (CNTs) have excellent electrical properties, quasi-one-dimensional lattice structure, good chemical stability, and high mechanical strength. They are ideal semiconductor channel materials for building new CMOS transistors and integrated circuits and are expected to promote future electronics. Moreover, because carbon nanotubes have the characteristics of strong carbon-carbon covalent bonds, nanoscale cross-sectional area, and low atomic number, they can be used to develop a new generation of ultra-radiation-resistant integrated circuit technology. However, how to use carbon nanotubes is a challenge faced by scientists all over the world. One of the main problems." Professor Zhang Zhiyong, a member of the joint research team and deputy director of the Carbon-based Electronics Research Center of Peking University, said.
This is the second time the team has published a paper in a top academic journal this year. Back in May, the team topped the list in science with an innovative method for making carbon nanotube transistors. So, what is this "carbon nanotube transistor" that is frequently favored by top journals? What is the point of research?
To clarify this issue, we have to start with the golden rule "Moore's Law" in the semiconductor industry.
Moore's Law predicts that the number of transistors that can fit on an integrated circuit doubles approximately every two years. As long as the number of transistors on a chip increases as expected, the chip industry will reap dividends.
Unfortunately, Moore's Law pointed out the direction the semiconductor industry should pursue but did not point out a path.
Since 1995, there have been predictions that transistor size will soon reach its limit. To prevent the hanging sword of "Moore's Law will eventually end", some researchers in academia and industry have turned the tide several times and "continued" Moore's Law by improving transistor architecture and manufacturing equipment.
On the other hand, a group of "maverick" researchers focused on semiconductor materials. They proposed simply switching to other materials to make transistors and abandoning the "dead end" of silicon-based materials. Among many alternative materials, carbon nanotube materials are the most promising. In May of this year, the research team of Peking University in my country achieved world-leading results in the preparation of carbon nanotubes, and is expected to advance the chip manufacturing process below 3nm! Just two days ago, the latest progress of the Peking University research team in carbon nano-semiconductor materials was included in the academic journal "Nature Electronics".
As early as 2008, the ITRS research report pointed out that the focus of future research should be on carbon-based electronic products.
In addition to being the "son of tomorrow" of the chip industry, carbon nanotube transistors have different meanings for my country's chip industry. Since the beginning of this year, the United States has further tightened its import and export controls on my country's chip technology. Huawei is the first to bear the brunt, and behind Huawei is the impact of the entire domestic industrial chain.
my country's research and development of carbon nanomaterials started earlier, and the recent research results of the Peking University team are all "firsts" in the world. Today, when the importance of "independent research and development" is infinitely magnified by external forces, the advantages of carbon nanotube technology cannot be ignored.
What exactly are carbon nanotube semiconductors? Is it feasible to use carbon nanotubes as an alternative? Is this a brilliant move? This is the question Chise wants to answer today.
01 The accidentally discovered "son of tomorrow" in the semiconductor industry
In the early stages of semiconductor development, transistors were made of germanium. After discovering that chips made of germanium materials could not withstand high-temperature working conditions, the researchers opened the periodic table of elements and chose silicon, which is the same family as germanium, has more reserves, and better heat resistance, as a substitute.
Carbon-based semiconductor materials have been discovered more by accident than silicon materials. Carbon nanotubes are arranged in a tubular shape by carbon molecules, which can be regarded as a "cylinder" wound by a single layer of graphite. They need to be prepared by special methods using carbon materials such as graphite rods as raw materials.
In 1991, Japanese physicist Sumio Iijima worked at the NEC (Nippon Electric) Basic Research Institute in Tsukuba, Japan, specializing in nanoscience, electron microscopy, and other fields.
At that time, Iijima used a high-resolution transmission electron microscope to observe carbon fiber products produced by the arc method, and accidentally discovered carbon nanotubes. Iijima Sumio has guided a large number of Chinese students in the United States and Japan. In 2011, was an the of.
▲ Juno Iijima
Through the study of carbon nanotube materials, it is found that it has more excellent semiconductor characteristics than silicon-based materials, especially in terms of high mobility, nanometer size, flexibility, permeability, and biocompatibility. These excellent properties mean that carbon-based integrated circuits will have the advantages of high speed and high energy efficiency.
Based on the above performance advantages, compared with silicon-based transistors, carbon nanotube transistors have 5 to 10 times the speed and energy consumption advantages, and are suitable for high-end electronic applications, high-frequency device applications, optical communication circuit applications, and flexible thin-film electronic applications.
02 More than 20 years of growth history: IBM/Stanford entered the game
Juno Iijima's discovery opened up the research of carbon nanotubes as materials and also paved the way for the semiconductor application of carbon nanotubes. As far as practical applications are concerned, IBM is "the first warrior to eat crabs".
In 1998, IBM researchers made the first working carbon nanotube transistor. For a long time thereafter, IBM showed interest in carbon nanotube transistors.
In 2012, IBM researchers fabricated a carbon nanotube transistor with a channel length of 9nm. This is the world's first transistor that can work below the 10nm node. That same year, IBM developed a technology that could integrate more than 10,000 carbon nanotube transistors into a single chip based on standard semiconductor manufacturing processes.
▲Schematic diagram of a 9nm channel length carbon nanotube transistor prepared by IBM
In 2014, IBM made bold claims, saying that by 2020, it will use carbon nanotubes to prepare semiconductor chips that are five times faster than existing chips. In 2017, IBM research advanced the size of carbon nanotube transistors to 40nm. IBM has also set up a team dedicated to the research and development of carbon nanotube semiconductor technology, led by IBM's T·J·Washington Research Center.
▲IBM carbon nanotube project leader
In addition to IBM, Stanford, MIT, Intel, and other institutions have also started research on carbon nanotube technology.
In 2013, a research team at Stanford University created a carbon-based chip with 178 carbon nanotube transistors. At that time, researchers at Stanford University commented: Maybe one day Silicon Valley will become Carbon Valley.
▲ Carbon-based chips manufactured by Stanford University in 2013
In 2015, Intel technology analyst Rob revealed that Intel was considering using carbon nanotube transistors in future chip technology.
In 2019, researchers at the Massachusetts Institute of Technology collaborated with chip manufacturers to create the world's first fully functional, programmable 16-bit RISC-V architecture carbon-based processor. A processor is capable of fully executing the entire instruction set. It also performs a modified version of the classic ",!" program that prints ",! I am, made from CNTs (, ! I am, made from )".
▲The 16-bit RISC-V architecture carbon-based processor manufactured by the MIT research team
In July 2019, the "Electronic Revival Project" was held. The Stanford-MIT Carbon Nanotube Project receives funding.
Semiconductor manufacturer giants and academic research leaders have bet on carbon-based semiconductors, once again proving the great potential of carbon nanotube materials in the semiconductor field.
However, this does not mean that the research and development of carbon nanotube semiconductor technology will be smooth sailing. Since the first carbon nanotube transistor came out in 1998, carbon nanotube semiconductor technology has been encountering material bottlenecks. Gate lengths for the smallest carbon nanotube CMOS devices have long been stagnant at 20nm (IBM 2014).
Specifically, to meet the requirements of large-scale high-performance integrated circuits, carbon nanotube transistors are required to meet two requirements at the same time:
1. In terms of arrangement and density, it is required to adopt a high orientation arrangement, and it is required to place 100~200 carbon nanotubes within 1 micron to ensure the number of transistors;
2. In terms of purity, the purity of the semiconductor is required to be greater than 99.9999%, or the content of metal-carbon tubes is less than 0.0001%, to ensure the performance of the semiconductor.
At present, the foreign companies that have developed well in the field of carbon nanotube semiconductors are memory chip manufacturers in the United States. According to official sources, the company plans to launch NRAM products based on carbon nanotube transistors later this year.
▲ NRAM products
03 My country's carbon-based semiconductor research "unique"
The research and development of foreign carbon nanotube semiconductor materials has stagnated for six or seven years. According to media reports, IBM's carbon nanotube research and development team has been disbanded, and most of the relevant personnel have entered universities for academic research.
Compared with my country, the research on carbon nanomaterials in my country started earlier. In 1997, Peking University established the first nanotechnology research institution in China: Peking University Nanoscale Science and Technology Research Center.
In 2002, the team of Professor Wu Dehai of Tsinghua University cooperated with Professor PM of Rensselaer Polytechnic Institute in the United States and used the floating chemical vapor deposition method to prepare carbon nanotube bundles with a diameter of about 300-500 microns for the first time.
In the same year, the team of Professor Fan Shoushan from Tsinghua University reported a method for preparing carbon nanotube fibers using carbon nanotube arrays. In addition to these achievements, according to the data in 2014, more than 1,000 universities and scientific research institutes in my country are engaged in the research activities of carbon nanomaterials, and they are constantly innovating in the research of carbon nanomaterials.
In terms of research and development of carbon nanotube semiconductor technology, compared with the situation abroad in 1967, we have a feeling that "the scenery here is unique".
Representative research institutions of carbon nanoelectronic devices in China include the Chinese Academy of Sciences, Peking University, and Tsinghua University. In May of this year, Beijing Yuanxin Carbon-Based Integrated Circuit Research Institute, initiated by Peking University, published a new carbon nanotube preparation method in the authoritative academic journal "Carbon Nanotubes", and at the same time realized the high density of carbon nanotube transistors, high purity. first. Require.
The purity of carbon nanotubes prepared by this method can reach 99.9999%, and the array density can reach 120/micron. With this technology, researchers are expected to advance integrated circuit technology below the 3nm node!
▲Wafer-level high-quality carbon tube array film developed by the Peking University team
As soon as the news came out, the stocks of many companies engaged in the research and development and production of carbon nanomaterials in my country rose one after another. Representatives include Chujiang New Materials, Yinlong Carbon-based Research Institute, Zhongke Electric, Danbang Technology, etc.
Among these companies, Danbang Technology can be called a strong force in the field of carbon-based semiconductors in my country. Danbang Technology was established in 2001, specializing in the R&D and production of flexible circuits and materials. It is a listed company on the small and medium-sized board of the Shenzhen Stock Exchange.
In 2019, the TPI quantum carbon-based film independently developed by Danbang Technology was successfully trial-produced. As the only company in the world that can mass-produce TPI quantum carbon-based films, it is said that its technology has been favored by Apple and Huawei. It is reported that Huawei has entered the pilot testing phase.
TPI quantum carbon-based film has a multi-layer graphene structure, which is mainly used in 5G mobile phones, chips, notebooks, flexible screen heat dissipation, and other usage scenarios.
On June 30, Danbang Technology disclosed its 2019 annual report. The data shows that in 2019, Danang Technology's operating income was about 347 million yuan, of which the PI film business accounted for about 1.67 million yuan, accounting for 0.48%.
04 Conclusion: Carbon-based materials are expected to provoke the "beam" of the future semiconductor industry
According to the forecast of the International Semiconductor Technology Roadmap Committee (ITRS), the performance of silicon-based semiconductors will reach the physical limit around 2020.
In 2020, the most advanced chip process node will advance to 5nm. After 5nm, where will the global semiconductor industry go? Even if it is not 2020, the size limit of silicon-based transistors will eventually come. What should the chip industry do then?
Facing the size limitation of silicon-based materials, switching to carbon-based materials is a good way. The achievements made by the research team of Peking University in my country in the preparation of carbon nanotube transistors are only a step forward for the industrial application of carbon-based semiconductors, and will also help my country prepare for the prediction of "the end of Moore's Law".
On June 1, after the Peking University team published the " paper, the research on carbon nanotubes at the Massachusetts Institute of Technology also made progress. According to a paper published in the journal Nature Electronics, researchers at the Massachusetts Institute of Technology have demonstrated that carbon nanotubes can be produced in large quantities in factories.
Since their discovery in 1991, carbon nanotube technology has been developing steadily. Perhaps in the future, carbon-based materials will become the "leader" in the semiconductor field, let us wait and see.
Data recommendation
Past aggregated data:
download
(5)
Order information:
:
Community recruitment
"Chip Semiconductor Technology Community" is recruiting! There are:
Industry analysis and interpretation, report data sharing, industry big V communication