Quantum Science: The Quest for Quantum Information Technology Expands

本文出现在激光重点世界and has been published here with permission.

Exploratory research inquantum sciencehas yielded promising advances towards new technologies:量子传感器, quantum communications, and量子计算. The United States joined the European Union, United Kingdom, China, Japan, Canada, and Australia in launching quantum initiatives to nurture those new technologies. A push by the National Photonics Institute, a joint effort of the Optical Society (OSA) and SPIE, led to bipartisan passage of a law establishing the National Quantum Initiative (NQI), signed into law on December 21, 2018. Plans call for the government to invest some $1.2 billion in quantum science over five years, with industry to invest additional money.

It’s an important and ambitious project, with government agencies and an industrial consortium hoping to transform the weirdness of quantum science into a practical powerhouse ofquantum information technology. The first results are expected in a matter of years, but maturity may be decades away.

政府和行业团队

On the government side, the National Quantum Coordination Office, part of the White House Office of Science and Technology Policy, coordinates quantum efforts by the National Institute of Standards and Technology (NIST), the National Science Foundation (NSF), and the Department of Energy (DoE). Research grants flow to scientists through the three agencies. Military and intelligence agencies will continue their own separately funded quantum programs and coordinate with the NQI.

NIST为量子经济发展财团（QED-C）提供了启动资金，以实现和发展美国量子业。超过100家美国公司已签署了加入的意向书，并有正式的支持协议即将出版。Qed-C的副总监Celia Merzbacher说，Qed-C的重点是使工业和政府与彼此的计划和需求紧密接触是“世界上独特的”。详细信息可在QED-C的网站上找到：https://quantumconsortium.org.

财团成员将支付会费，并选择支持什么工作。竞争前的研究将很多，以满足行业需求。冷冻技术是主要需求，因为量子处理器现在接近绝对零。Qed-C已经为其开发了路线图。另一方面，梅尔兹巴赫（Merzbacher）预计，该财团将避免仅在学术兴趣或竞争激烈的领域资助项目。

Science First, Technology Follows

The big-picture goal is “science first,” says Michael Raymer, a quantum optics researcher at the University of Oregon (Eugene, Ore.), who co-authored a white paper proposing the NQI and co-led a lobbying effort to see it through. Scientific direction on the government side will come from the bottom up, with researchers writing proposals and peer reviewers selecting which to fund. The goal is to go beyond the use of quantum transitions in lasers and atomic clocks to “Quantum 2.0” applications that usequantum entanglement，包括计算，通信和传感。

Conventional digital computing stores and processes information as digital bits that each have two states, 0 or 1, so its capacity to describe digital states increases linearly with the number of bits. Quantum computing stores and processes quantum objects called “qubits” (quantum bits) which can occupy superpositions of two states, 0 or 1.

That subtle-seeming difference means that digital bits can have only one of two possible values, but qubits can have many more values because quantum processes can superpose their two quantum states in many ways as long as they maintain coherence, which as shown inFigure 1means that the superposition can be anywhere on the surface of the sphere at right. Thus their capacity increases exponentially, so a line of 50 individual ion qubits trapped in a vacuum could represent up to 2⁵⁰(about 1.125 × 10¹⁵)可能的量子态。

Quantum computers are not going to replace digital computers for balancing your checkbook or doing arithmetic. But quantum computers could excel at other types of processing that don’t come easily on digital computers because their quantum processing power increases exponentially with the number of qubits. That promises major advances in applications, including “simulating nuclear and high-energy physics; designing new chemicals, materials, and drugs; breaking common cryptographic codes; and performing more speculative tasks such as modeling, machine learning, pattern recognition, and optimizing hard logistical problems such as controlling the electric energy grid or traffic control systems,” Raymer and two colleagues wrote in the journalScience.¹

Computing Challenges

Quantum computers are much more vulnerable than digital ones to errors arising from noise and failure to maintain coherence. The “quality” of quantum bits depends on their isolation from the rest of the world needed to avoid overwhelming the signal with noise. So far, many groups have demonstrated qubits and simple types of quantum computing. In October 2019, Google claimed it had achieved “quantum supremacy” (solving a problem no classical digital computer could solve) with an array of 53 tunable qubits.²

这一主张受到质疑，仍然存在重大挑战，但大多数研究人员认为，随着量子位的数量扩大并证明了合适的有用算法，将达到所谓的至高无上。

Many types of qubits have been demonstrated, and their performance differs widely. Many consider the most versatile to be semiconductor devices made by photolithography, but they must be cooled very close to absolute zero to maintain coherence. Google’s Sycamore system operated at 20 mK; other semiconductor qubits operated at up to 100 mK. In April 2020, two separate groups reported using spin resonance in silicon to make qubits operable above 1 K,³which may significantly simplify cooling requirements. One at the University of New South Wales (Sydney, Australia) reported a record high temperature of 1.5 K. A second, at QuTech and the Technical University of Delft (Delft, Netherlands), reached 1.1 K.⁴

被困在静电陷阱中并由激光耦合的原子离子也可以用作量子。许多团体在该领域工作，主张是多种多样的。IONQ（马里兰州大学公园）在其网站上声称拥有79个QUAT的最大阵列记录，但没有列出详细信息。⁵2018年，量子光学和量子信息研究所（奥地利维也纳）和同事的本·兰尼（Ben Lanyon）报告了20 Quitbits，这是最大的完全可控的，纠缠的系统。⁶In March 2020, Honeywell (Charlotte, N.C.) claimed a record “quantum volume” benchmark of 64 using four ion-trapped qubits cooled to 12.6 K.⁷Figure 2shows the ion trap apparatus. Honeywell says its advantage was “having the highest-quality fully connected qubits with the lowest error rates.” The company plans a commercial version in later this year.⁸

D-Wave Systems (Burnaby, BC, Canada) already sells a quantum computer with 2000 supercomputing qubits, but outside observers say the qubits have limited connectivity and very short coherent times, and have yet to demonstrate entanglement to everyone’s satisfaction.⁹Xanadu（加拿大多伦多，安大略省多伦多）使用集成的光子学来制作产生挤压光的光子量子量，并将其用于量子计算和量子机学习。¹⁰PsiQuantum (Palo Alto,Calif.) is fabricating silicon photonic qubits in a conventional semiconductor lab, and says on its website “Error correction is at the center of everything we do.”¹¹It seems likely that each of these different platforms may find their own niches where their hardware is most suitable.

Quantum Sensing

Development barriers are lowest for quantum sensing, so it is likely to be the first quantum technology to be widely deployed, says Chris Haimberger, technology manager of Toptica-USA (Farmington, N.Y.). Indeed, NIST already has reported an amazingly precise systematic uncertainty of 9.4 × 10^-19基于铝27离子过渡的量子逻辑光谱的原子钟。¹²

Quantum atomic interferometers can sense gravity and acceleration for navigating without GPS or for geological exploration. Quantum sensing can functionally image molecules on a nanometer scale, assess strain within materials, and measure magnetic fields in, for example, living cells.

Lasers can excite atoms with the high precision needed to create excited Rydberg states, which are just below the ionization threshold, and very sensitive to electric or magnetic fields. “Most of the sensors are much more about very precise control of the system than entanglement,” says Haimberger. Rydberg Technologies (Ann Arbor, MI) is offering radio-frequency field probes serving as optical frequency trackers.

Quantum Communications and the Quantum Internet

Quantum communications will be needed to transport qubits for quantum computing and other applications. Photons will carry the quantum information through space, air or optical fibers, like they carry digital bits. Depending on the wavelength used, quantum transmitters and receivers might need to be kept very cold to preserve quantum states and entanglement.

存储在量子存储器中的Qubits是固定物理对象，但是它们的量子状态可以通过将其编码为光子特性（例如极化）来传输到遥远的内存中。可以通过首先纠缠另外两个光子并将其发送到不同位置来传输该编码的信息。一旦这些光子到达目的地，就可以将基于内存的值的量子状态“传送”到从一个位点到另一个位置的遥远存储器。

The process is not teleportation in the science-fiction sense of moving an object or a person through space. (It also is not instant, because it requires sending a subsidiary classical bit, which takes time.) Instead, it transports information on the quantum states, which can be used for remote quantum computing or other purposes.

Quantum state teleportation has stringent requirements, but it helps realize the potential of quantum computing and sensing. Encoding qubits as entangled photons would allow sharing their states to distribute quantum-computing tasks between locations. Quantum entanglement also offers a physical guarantee of security because it allows detection of eavesdropping. That enables secure distribution of a quantum key, which recipients could use to securely encrypt and decrypt a message.

另一个有趣的可能性是“盲”或私人量子计算，其中量子链接的一端的人可以指示另一端的量子计算机执行量子操作，然后将结果传输回原始源。得益于纠缠的安全性，可以将结果传输回原始发件人，而另一端任何人都知道结果。

纠缠的光子可以通过光纤在信号变得太弱而无法可靠地检测到可靠地检测到一百公里。他们可以在自由空间中走得更远。In 2017, a group led by Jian-Wei Pan, a physicist at the University of Science and Technology of China (Shanghai) stretched transmission of entangled photons to 1200 km by sending them from China’s Micius satellite through free space to a pair of ground stations, as shown inFigure 3.¹³Too few photons got through for practical use, but the test helped stimulate U.S. support for the NQI.

去年夏天,美国能源部测试纠缠的传播photons through 50 km of fiber linking its Argonne and Fermi National Laboratories in Illinois as a first step toward a “quantum Internet” linking its labs. DoE is working with NASA on plans for satellite-based tests using a technique called “entanglement swapping” using very precisely timed pulses of entangled photon states.¹⁴One or two satellites would be based in low- or medium-Earth orbit, so signals could sweep across continents to connect with many locations.

Quantum optical signals decay with distance because of decoherence as well as attenuation, so maximum transmission distance is limited without quantum repeaters. Conventional optical amplifiers will not suffice because stimulated emission cannot reliably clone qubits, as spontaneous noise photons are always added at random. What’s needed to make a quantum repeater is a way to teleport qubits from one end to the other in order to retain entanglement. It’s a big challenge, but many concepts are under investigation.

One example is using a cluster of about 10 photons to encode a qubit, and when one photon is lost, intercept the remaining ones, detect what was lost, and restore a copy of it. The idea is to be able to restore the missing photon without knowing the original quantum state it represented, but Raymer says it is still far from full implementation.

Plans and Progress

As it emerges from startup mode, the QED Consortium is focusing on identifying quantum-enabling technologies and what developers need from them. They already have looked at cryogenics, a crucial field because most quantum devices operate near absolute zero, and the special cooling equipment is large and costly. Merzbacher says they are planning a workshop on quantum-enabling laser technologies.

一个Qed-C研究小组正在研究如何以有意义的方式报告进度，以便用户社区可以理解系统功能。梅尔茨巴赫说：“在量子计算中，人们喜欢报告量子位的数量，但这并不是一个很好的衡量性能，因为存在质量和数量问题。”在阅读最近的新闻稿时，这个问题非常明显。其他问题包括开发量量子技术的劳动力，以及协调行业，政府和学术界之间的信息流。光学行业发展协会（OIDA）正在研究开发量子技术所需的光子学路线图。这一切加起来。

Progress is encouraging, but quantum technology is still in its early stage. “In two to five years we will have quantum computers that do something useful,” says Mark Tolbert, CEO of TOPTICA-USA, who is actively involved with QED-C and the NQI. Honeywell, IonQ, Google, and others have demonstrated devices, but he says they are “still trying to figure out what they can do with them.”

The mainstream view, Tolbert says, is that more general use of quantum computers is 10 to 20 years away. He says quantum sensing is likely to come first, followed by quantum communications, where he has seen been major advances in the past six months. Raymer thinks the full suite of technology may take 20 to 30 years to mature.

Look back 20 to 30 years, and you can see the birth of the Internet and a time when 56 kbit/s modems seemed fast and we thought a megabit to the home would be blindingly fast. Look forward that far, and you can expect another transformation of information technology.