Network Security Internet Technology Development Database Servers Mobile Phone Android Software Apple Software Computer Software News IT Information

In addition to Weibo, there is also WeChat

Please pay attention

WeChat public account

Shulou

The dawn of superconducting chips! Superconductors can also conduct electricity in one direction, which was once considered impossible.

2025-02-14 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > IT Information >

Share

Shulou(Shulou.com)11/24 Report--

The artistic imagination of superconducting chips. Photo Source: TU Delft We are in the information age, computing power is indispensable, but computing energy consumption is becoming more and more important. Since the current in the computing chip is constantly changing, can we use superconductors to make the chip?

As early as 1911, Dutch physicist Heike Kamerlingh Onnes found that when the temperature dropped to 4.2 K (about-268.95 ℃), the resistance of metallic mercury immersed in liquid helium would disappear. But it was not until 1957 that there was the first theory that could explain the phenomenon of superconductivity, the BCS theory. The theory was established by American scientists John Bardeen, Leon Cooper and John Schrieffer based on "wave-particle duality". They believe that when there is a voltage, the free electrons in the outer layer of the metal will flow through the lattice to form an electric current, but usually the lattice is defective and the current is hindered by thermal vibration. In superconductors, electrons are bound to form a "Cooper pair", resulting in waves of collective condensation, which, unlike free electrons, can pass through the lattice without hindrance.

Since there is no hindrance in the flow of electricity through the superconductor, this means that it is almost impossible to suppress or even block the current-let alone let it flow in one direction. The invention of unidirectional superconductors is like the invention of a special kind of ice that does not cause friction when it slips from one direction to another, but creates insurmountable friction when it slides in another direction. In fact, scientists at IBM had the idea of superconducting computing as early as the 1970s, but they soon stopped trying. In a later paper, IBM mentioned that it was impossible to build a superconductor computer without non-reciprocal superconductivity (non-reciprocal superconductivity).

Symmetry breaking so how do today's semiconductors achieve unidirectional conduction?

Semiconductors can achieve unidirectional conduction through PN junctions. The PN junction consists of P-type semiconductors (Positive) and N-type semiconductors (Negative) respectively. P-type semiconductors lack free electrons and conduct electricity by holes, while N-type semiconductors have more free electrons and can conduct electricity. When P-type semiconductors and N-type semiconductors are combined together, the free electrons and holes near the contact surface of the two semiconductors attract each other, are bound to the boundary and cannot move freely. This area is called the depletion layer and no longer conducts electricity. When the voltage from N to P is applied, the electron moves from P to N direction, the depletion region expands, and the PN junction does not turn on; when the voltage from P to N is applied, the electron moves from N to P direction, the depletion region decreases, and the PN junction conducts. As a result, the PN junction realizes the characteristic of unidirectional conduction.

PN shows the intention. Image source: the wikipedia superconducting principle itself does not have anything to do with directionality. The anisotropy can be forcibly applied to the superconductor by the external magnetic field, so as to realize unidirectional superconductivity. However, without the application of external magnetic field, the strong directionality of unidirectional conduction seems to have nothing to do with superconductivity.

What's more, the Cooper pair in BCS theory exists in the lattice of metal crystals, while the PN junction is based on the special properties of semiconductors, and the difference between metals and semiconductors is an insurmountable gap. Today, computing consumes more and more energy, and the training process of neural network is even more energy-consuming. Some people even suspect that human beings will be dragged down by the high energy cost before the real artificial intelligence is realized. Superconducting technology has gradually entered our lives. In December last year, Shanghai Xuhui Commercial Core Zone put into production the superconducting cable transmission project with a total length of 1.2km, rated current of 2200 amperes and rated capacity of 133MVA. Can't we really use superconducting technology to reduce energy consumption?

Some time ago, Mazhar Ali, an associate professor at Delft University (TU Delft), and his team published a paper in the journal Nature, which achieved unidirectional superconductivity for the first time without an external magnetic field. They used two-dimensional quantum materials to make "quantum material Josephson junctions" (QMJJs), which rekindled the hope of superconducting computing.

Josephson junction,JJ is a phenomenon discovered by British physicist Brian David Brian David Josephson in 1962. This structure consists of superconductors and insulators (or other materials that can form barriers). The structure is usually superconductor-insulator-superconductor (S-I-S). In this structure, quantum tunneling can occur in the Cooper pair in the superconductor. For this discovery, he was awarded the 1973 Nobel Prize in Physics.

The schematic diagram of the Josephson junction shows that An and B are superconducting materials and C is the material that provides the barrier. Image source: wikipedia but Josephson junction technology is also symmetrical, there is no special symmetry breaking mechanism, resulting in the current "forward" and "backward" difference. In Professor Ali's quantum material Josephson junction, the classical barrier material in Josephson junction was replaced by the quantum barrier material Nb "Br". Nb "Br" is a two-dimensional material similar to graphene on which the team can peel off several layers of atoms to create a very thin "sandwich" structure. The inherent characteristics of this quantum material can adjust the coupling between the two superconductors in a very fine way, thus realizing unidirectional superconductivity. In the end, Professor Ali's team realized the unidirectional conduction of superconductors using the structure of NbSe "/ Nb" Br / NbSegment. In the environment of 20mK without magnetic field, they proved that the material is superconducting at positive current and resistive at negative current.

The achievement of a bright future is enough to make people think about it. Professor Ali said that such technology may greatly increase the computing speed of chips to the THz level, while the current frequency of advanced chips is generally between 1~5GHz, equivalent to hundreds of times faster. But before the chip can actually be put into use, they still need to solve two problems. The first is to raise the working temperature of this structure to above 77K, and liquid nitrogen refrigeration can be cooled to this temperature, which can greatly reduce the refrigeration cost. The second is to expand the production scale, this experiment only created a few quantum Josephson junctions in nanodevices, and then the problem of large-scale mass production of this material needs to be solved.

In 2020, the achievement of room temperature superconductivity of carbon-containing hydrogen sulfide superconductor C-S-H at 2.6 million times atmospheric pressure brought great hope. But the study received a series of questions last year, including suspicions that the team's data were suspected of being falsified. This undoubtedly makes people more cautious about the achievements in superconductivity. Mazar Ali's team kept their experiments as rigorous as possible and kept them as repeatable as possible. They have used different batches of materials and equipment from different countries to manufacture and measure this structure, and all of them have achieved the expected results.

However, it may be difficult for you to use superconducting chips on your computer or even your mobile phone in a short period of time. Even if the research team succeeds in raising the operating temperature of the chip above 77K, the cost of liquid nitrogen cooling is unbearable for most people. However, in today's computing energy consumption is more and more can not be ignored, even if only in the data center using such technology, the entire human society to save energy, can not be ignored.

Reference link:

Https://www.nature.com/articles/s41586-022-04504-8

Https://www.eurekalert.org/news-releases/950503

Http://stdaily.com/index/kejixinwen/2021-12/22/content_1241108.shtml

Https://en.wikipedia.org/wiki/Josephson_effect

Https://news.sciencenet.cn/htmlnews/2021/10/467672.shtm

This article comes from the official account of Wechat: global Science (ID:huanqiukexue), author: Wang Yu

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.

Views: 0

*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.

Share To

IT Information

Wechat

© 2024 shulou.com SLNews company. All rights reserved.

12
Report