What is the next step in quantum computing?

Famed science fiction author William Gibson once said, "The future is here; it's just not very evenly distributed." These were the opening words of Rich Uhlig, Intel Senior Fellow, Vice President and Director of Intel Labs, at the recent Intel Labs Day. During the event, Intel highlighted research initiatives across several areas. One, in particular, was quantum computing, which introduced a second-generation malingering cryogenic quantum control chip.

Speaking about the uneven distribution of potential benefits of technology, Rich Uhlig emphasized that genotyping technologies and home DNA testing are available today. Still, they are not available to the world's population on a large scale, proving that the future is not "evenly distributed". At the same time, from an IT perspective, large amounts of data must be moved, stored, analyzed, protected and computed at scale. The planet is increasingly covered by sensors that generate massive amounts of new data daily. New challenges are on the horizon.

We must make huge improvements across technology areas of computing and memory interconnection. Intel believes that these gains require a new way of thinking, and they will emerge when experts can collaborate by crossing multidisciplinary areas of knowledge, science and technology. Intel highlighted one of these areas at the event: quantum computing.

Quantum computing has been a very active area of research in recent years, and many companies are working on it. Quantum computers have some unique characteristics that make them very powerful.

In an interview with EE Times, Dr Anne Matsuura, Director of Quantum Applications and Architecture at Intel Labs, and Dr James Clarke, Director of Quantum Hardware at Intel, spoke at the event on the topic, highlighting applications of disruptive technologies and taking a deeper look at Intel's quantum efforts.

The discovery and design of new materials are one of the early applications where quantum technology could have a meaningful impact. "We believe that commercial-scale quantum computers will allow the simulation of these materials so that in the future, we can also design materials and chemicals with the desired properties," Matsuura said. "Today's 100 quantum bits or thousands of quantum bits will not lead us to this result. We will need a commercial-scale quantum computer system with millions of quantum bits to achieve the quantum utility of such ambitious problem solving."

What are quantum bits? Quantum and Classical Computing

Quantum bits are "bits" of information in quantum systems and certain elements of quantum mechanics. But how are quantum bits physically created? How can electronic devices effectively control quantum bits in a quantum system?

Unlike conventional bits, quantum bits do not exist simultaneously in zero or one state but in a "superposition" of two states. Moreover, in a quantum processor, more quantum bits in superposition can be connected so that they express a group behaviour called entanglement. This entangled state is the basis for the incredible computing power of quantum computers and the source of their potential to solve complex tasks beyond the capabilities of traditional supercomputers.

As part of Rich Uhlig's Lab Day keynote, Matsuura gave an introductory talk for the quantum section, describing what Intel is doing and the latest market developments.

"An easy way to get an intuition about quantum computing is to think of a computer bit as a coin," she said. "It can be in a heads state or a tails state - it's either in one state or the other. Now imagine the coin is spinning. It's spinning in the sense that it's in both a heads state and a tails state; it's a superposition of two states."

This is similar to a quantum bit or quantum bits. Matsuura adds, "Now imagine that we put two quantum bits together and entangle them. Now we have four states at the same time. So two entangled quantum bits represent a mixture of four states simultaneously. More generally, n quantum bits represent 2n states."

The computational power of a quantum computer grows exponentially with the number of quantum bits. Theoretically, if we have 50 such entangled quantum bits, we can access more states than any possible supercomputer. If we had 300 entangled quantum bits, we could simultaneously represent more states than the atoms in the universe.

Like a classical computer, a quantum computer consists of a quantum circuit composed of elementary quantum logic gates. Quantum computers can potentially solve problems that traditional computing solutions cannot solve. The underlying technology is quantum physics; because a quantum bit (or quantum bits) can exist in multiple states simultaneously, it can be used to compute all possible states simultaneously, significantly speeding up the solution of complex problems.

"Quantum bits do not have long lifetimes, and noise or observation can lead to information loss," Matsuura said. "Therefore, in reality, we need hundreds of thousands, or more likely millions, of high-quality quantum bits for commercial-scale quantum computers. In other words, we need to scale quantum to be useful for real-world applications. Intel's quantum research program focuses on several key areas: spin quantum bit technology, cryogenic control technology, and full-stack innovation. Each of these areas addresses key challenges associated with scaling quanta, and Intel is systematically addressing each of them to enable scaling."

Quantum Bit Technology

Quantum bits are available in the small quantum computing systems we see today. They are not of sufficient quality or quantity to get them into commercial-scale systems. We will need many stable or noise-resistant quantum bits and efficient connections between quantum bits to scale to commercial-scale quantum computers with millions capable of running quantum algorithms. After Matsuura's talk, James Clark gave an interesting talk.

Clark noted that Intel is interested in building quantum computers with millions of quantum bits. "We are using the same technology as transistors to build our quantum bit chips. We're using CMOS control electronics to control our quantum bits that are built with our transistor technology."

Short-term applications will be in chemistry, materials, biology and medicine. In the long term, applications such as optimization algorithms, cryptography and machine learning are being sought.

Clark notes several ways to make quantum bits for quantum computers. The first is to capture ions in which you can use lasers to study the excited states of metal ions. These excited states can represent zero and a quantum bit. This technique is similar to the atomic clock that won the Nobel Prize in physics in 2012.

The second technique is the superconducting quantum bit technique. Small rings of superconducting metal are used to create an artificial atom whose states represent zero and one of the systems.

The third technology is silicon quantum dots or spins quantum bits, which essentially control an electron's spin, representing the zero-sum one-state of a quantum bit. "We think this is a very interesting technology for Intel," Clark said.

The analogy with the transistor allows us to explain this latest technology. A transistor is essentially a switch. When you apply a voltage or potential, a current flows through the device.

"The transistor is the most ubiquitous man-made object on the planet," said Clark. "At Intel, we believe we ship 800 trillion transistors a year; that's an incredible number. The truth is that every person on the planet has a few transistors every minute of every day. Some predict that by mid-century, there will be more transistors on the planet than there are human cells. Transistors are everywhere."

Instead of allowing a current of many electrons to flow through the device, individual electrons are captured. In this device, individual electron transistors are created. By combining many of these individual transistors, we can create a network of electrons. We can control the interaction between two neighbouring electrons by controlling the potential between individual transistors.

The engagement of a single electron transistor in a magnetic field will result in that single electron having two states of energy used for quantum bits. "There are two states where we are essentially controlling an electron," Clark said. "We create spinning quantum bits in the same way we create transistors. There are two states where we are essentially controlling an electron. We create spin quantum bits the same way we create transistors. Our current technology is based on fin-FET geometry; we're making our quantum bit structure in the same way that we're making fin-based ones."

At this point, improving quantum bit technology involves solving the same challenges as transistors: size variability, gate oxide defects and voltage variability. Fast characterization of quantum bits is important in this process.

Quantum bit control

One of the challenges of quantum computing is quantum bit control. As Matsuura points out, this quantum information is very fragile.

Many electronic racks control today's quantum bits with complex wiring connected to the quantum bits located in cryogenic refrigerators to protect the fragile quantum bits from the thermal and electrical noise of commercial-scale quantum computing. "This is an area that Intel is addressing with cryogenic quantum bit control chip technology with scalable interconnects," Matsuura said.

Making more quantum bits creates millions of lines, which makes the hardware too complex. One approach Intel has taken to improve wiring efficiency is using CMOS technology to bring control very close to the quantum chip. In the long run, it will allow fewer wires and a more elegant interconnect system. The technology is Intel's Horse Ridge, produced using Intel's 22FFL CMOS process. Its functionality has been proven in 4 kelvins.

During the Intel Labs event, the company unveiled its second-generation cryogenic quantum control chip, Horse Ridge II. Horse Ridge II builds on the capabilities of the first-generation SoC and generates RF pulses to manipulate the state of quantum bits, called quantum bit driving. It introduces two additional control functions: quantum bit readout, a function that reads the current quantum bit state, and multi-gate pulses, which can control multiple quantum bit gates simultaneously. Adding a programmable microcontroller running in the integrated circuit allows Horse Ridge II to provide a higher level of flexibility and sophistication in performing these control functions.

With Horse Ridge, Intel hopes to increase the scalability of quantum computers to thousands or even millions of quantum bits by reducing the complexity of interconnecting quantum systems, one of the key obstacles to achieving quantum utility and solving real-world problems through quantum computers.

Programming Quantum Computers

Quantum computers do not work like classical computers. Instead of running on binary arithmetic, quantum computers manipulate the probability magnitude of a quantum wave function and then sample the resulting probability distribution. "Programming a quantum computer is very different from programming a classical computer," Matsuura said. "Quantum bits are fragile; the ability to correct quantum bit errors that occur will be very important. But because today's quantum computers do not implement error correction systems, we are developing noise-resistant quantum algorithms and error mitigation techniques to help us run these algorithms on today's small quantum bit systems.

"The quantum bit control processor sends the microcode to the control electronics," she added. "And it converts all the logic operations in the algorithm needed to run that quantum algorithm into microcode. This tells the control electronics what pulses to send and when to send them to the quantum bits. Runtime software running on a classical processor loads and executes the quantum program and your algorithm and provides the sequence of these quantum operation instructions to the quantum bit control processor for execution. The program code consists of classical and quantum instructions and is generated by a quantum compiler. The compiler takes the algorithm, compiles it, and calculates how to map and schedule your quantum upstream quantum bits based on the connectivity between the quantum bits and the specific properties of the quantum bits.

The program code consists of classical and quantum instructions generated by the quantum compiler. It calculates how to map and program quantum operations on quantum bits, considering their connectivity and properties. Quantum compilers have challenging tasks. They have to choreograph a dance of quantum bits that locate and move them to the right position at the right time. The algorithm works on deadline because quantum bits have a very short lifetime, typically a fraction of a second, and the operations require important and often changing time scales.

Error correction is another interesting topic. Much work is needed for commercial-scale systems with millions of quantum bits to select the correct logical quantum bits without error. In the meantime, Intel is developing noise-resistant quantum algorithms and error mitigation techniques to help these algorithms run on all of today's quantum bit systems.

Correcting quantum errors is the foundation of most quantum computer projects because it helps maintain the fragile quantum states on which quantum computing depends. The operations required for error correction are not only very complex but must also keep quantum information unchanged.

One way to improve fault tolerance is to delegate some of the computation to the CPU. And indeed, the entire stack requires this classical-quantum hybrid approach. "We expect many of our algorithms for commercial full-scale quantum computing systems to be hybrids consisting of classical and quantum parts, taking advantage of the unique strengths of each of the two computational models," Matsuura said.

Validation of quantum computers

Research and development begin with a deep understanding of the workload the system is about to perform. The nature of the workload guides the design of the complete computer system.

Following Clark's lecture, Matsuura's talk highlighted that even with a large number of quantum bits, we would never achieve our goal of running useful applications on a quantum computer without building all the elements of the computational stack. "There is still a long way to go before we have a practical, commercial-scale, useful and impactful quantum computer," she said. "We won't have a quantum computing system without a complete stack, hardware and software.

She added: "Intel has introduced equipment that helps test our quantum bits quickly on CMOS wafers in our factories." "We're taking hours rather than days to get information; we're essentially mimicking the cycle of information in standard transistor development. Without the new cryogenic detector, a custom-designed device that we developed with our partners Bluefors and Afore, we can get test data and learn from our research devices at a rate of 1,000 times faster, thus significantly accelerating the development of quantum bits."

At Intel, we are using the same technology to make our quantum bits as our advanced transistor technology," Clark explained in his presentation. This is being done at our facility in Oregon. We're using 300 mm wafers to make these devices; for every wafer we get, we produce thousands of quantum devices to test these quantum bits. We're using what's called a dilution refrigerator to cool our quantum chips at very low temperatures to preserve the quantum effect. We're thinking of temperatures a fraction of a degree higher than absolute zero."

The first cryogenic wafer detector, built by Intel, Bluefors and Afore, is a low-temperature detector tool designed to test and verify the quantum bits needed for quantum computing. The cryogenic wafer probe allows researchers to test quantum bits down to a few Kelvin on a 300 mm wafer, making it the first test tool for quantum computing.

In the process, several probes are available to characterize these transistors. With Intel's cryogenic probes, 300 mm wafers can be scanned to characterize quantum bits quickly. One way to analyze this situation is to apply small microwave pulses near the quantum bit to observe the up-state and down-state changes.

Future applications of quantum computing have captured the world's imagination, and, as a result, it has been the subject of much media hype. Quantum computers may one day impact areas such as transportation and routing logistics, designing new drugs and protein folding, and perhaps even modelling climate risk analysis and the price of financial options. Of course, one of the initial applications piqued interest in quantum computing: cryptography. But in reality, these uses depend on quantum computing, hardware and software discoveries and will take many years to materialize.

Because quantum computing is a new type of computing that runs programs differently, we need hardware, software, and applications developed specifically for quantum computing.

"This means that quantum computing requires new components at all levels of the computing stack, from control electronics to compilers to quantum bit control processors and quantum bit chip devices," Matsuura said. "Intel is developing all components of the complete quantum computing stack."

Matsuura points out that getting these quantum components to work together is some quantum orchestration. She added that collaboration with outside entities could provide additional impetus for progress in the field. "We are collaborating with several universities, including the University of Chicago. We recently committed to participate in the National Research Center for Quantum Information Science, called Q-NEXT, announced by the Department of Energy, where Intel will provide a complete quantum stack for research partners."

As Matsuura pointed out in our interview and at the Intel Labs Day event, there are still many challenges with quantum computing technology. One is scalability. Today's quantum computing systems are scalable versions of brute force that avoid the problems associated with scaling to millions of quantum bits. "We're trying to figure out how to scale to a large number of quantum bits," she said.