Intel and QuTech: Quantum Processors

2021: Progress in achieving quantum scalability by programming silicon qubits with Horse Ridge

Intel and QuTech, a joint project of the University of Technology Delft and the Netherlands Organization for Applied Scientific Research, announced on May 14, 2021 the publication of key research results in the field of quantum computing aimed at eliminating the "bottleneck of interconnections" between quantum chips that are in cryogenic dissolution refrigerators and complex electronics for controlling qubits that operate at room temperature. Innovations presented in the industry scientific journal Nature mark an important stage in solving one of the serious problems of scaling quantum computing using the Intel Horse Ridge cryogenic controller chip.

Intel and QuTech Have Made a Breakthrough in Quantum Interconnection

Operating control electronics with high accuracy at cryogenic temperatures would overcome the so-called "interconnect/wiring bottleneck." Intel took the first step to the solution of this problem with the announcement of Horse Ridge - the cryogenic managing director of the chip for qubits constructed with use of 22-nanometer Intel FinFET LowPower technology. The second generation of chips was introduced in 2020. Horse Ridge transferred the main control functions of a quantum computer to a cryogenic refrigerator - as close as possible to the qubits themselves - to simplify the complex control circuits of quantum systems.

The study presented in the publication demonstrates successful results of randomized testing, which show that a mass-produced CMOS-based crypto controller provides coherent control of a two-qubit processor with the same level of accuracy (99.99%) that provides electronics at room temperature. This marks a key milestone in crypto electronics research for quantum computing, Intel and QuTech emphasized.

In particular, companies have successfully demonstrated frequency multiplexing using a single cable to control two qubits. This is an important confirmation of the operability of the concept, since each qubit is individually controlled by a separate cable - an approach that does not scale as the number of qubits increases. Horse Ridge is designed to eliminate this limitation by multiplexing to reduce the number of radio frequency cables needed to control qubits, the researchers explained.

The study demonstrated the programmability of a controller on a two-qubit Deutsch-Jose algorithm, which is more efficient on quantum computers than on traditional computing systems.

The results of the study, tested by randomized testing, confirm the original potential of Horse Ridge as a highly integrated and scalable solution to simplify quantum control electronics, and prove that this technology can be applied directly to multi-qubit algorithms and noisy quantum devices of intermediate scale, emphasized in Intel and QuTech.

According to the researchers, further research in this area can lead to the full integration of the controller chip and qubits on one chip or in one package - especially since they are all made of silicon, thus paving the way for quantum scalability.

2020: Horse Ridge Cryogenic Quantum Processor

On February 19, 2020, Intel Labs, together with QuTech, revealed the key specifications of their cryogenic quantum processor, code-named Horse Ridge. Details are presented in scientific work, the article considers the key technical capabilities of Horse Ridge, which solve fundamental problems when building a quantum system that is powerful enough to demonstrate the practicality and feasibility of quantum computing, as well as the advantages that they carry (quantum practicality): scalability, flexibility and accuracy.

As reported, as of February 2020, researchers in the field of quantum computing are working with only a small number of qubits in small, specially developed systems surrounded by complex control and interconnect mechanisms. The Intel-introduced Horse Ridge chip greatly simplifies all these complex tasks.

As of February 2020, the quantum computing research community is only at the first stages of its work to demonstrate the feasibility of this technology. The ability to use quantum computing to solve practical problems depends on the ability to scale the system to thousands of qubits and simultaneously control it with high accuracy. The Horse Ridge chip greatly simplifies the modern complex control electronics necessary for the operation of such a quantum system, thanks to the use of a highly integrated system on a chip (SoC). It allows you to speed up the configuration of the system, increase the performance of qubits and ensure its effective scaling to a larger number of qubits, necessary for the application of quantum calculations in solving practical problems.

The main technical details covered in the scientific article are:

• Scalability: An integrated system on a chip (SoC), implemented using 22nm CMOS Intel FFL technology (FinFET Low Power), combines four RF channels in one device at once. Each channel is able to monitor up to 32 qubits using frequency multiplexing, a method that divides the total available frequency band into a series of disjoint frequency bands, each of which is used to transmit a separate signal. Using these four channels, Horse Ridge can potentially control up to 128 qubits with a single device, which allows you to significantly reduce the number of cables and infrastructure equipment, compared to what was previously required.
• Accuracy: The increase in the number of qubits causes other problems that make it difficult to increase the power of a quantum system and questions the possibility of its operation. One of these potential negative consequences is a decrease in the accuracy and productivity of the qubit. In developing the Horse Ridge chip, Intel optimized multiplexing technology that scales the system and reduces errors from "phase shift," a phenomenon that can occur when controlling many qubits at different frequencies, which leads to crosstalk between qubits. The various frequencies used in Horse Ridge can be "adjusted" with high accuracy, which allows the quantum system to adapt and automatically correct the phase shift when controlling several qubits through the same radio frequency channel, thereby improving the accuracy of the qubit gates.
• Flexibility: The Horse Ridge chip can work with a wide range of frequencies, allowing you to control both the operation of superconducting qubits (so-called transmons) and spin qubits. Transmons for February 2020 usually operate at a frequency of about 6 GHz-7 GHz, while spin qubits - at a frequency of 13 GHz to 20 GHz.

Intel is studying silicon spin qubits, which can work at temperatures high enough for qubits up to 1 degree Kelvin. This study opens up opportunities for integrating silicon spin-qubit devices and the cryogenic control system implemented in Horse Ridge to create a solution that allows you to combine qubits and controls in one convenient package.

2017: Experimental quantum 17-qubit processor

In October 2017, it announced the supply of an experimental 17-qubit processor Intel based on superconductivity technologies to the Dutch research center, QuTech which is engaged in physics research with Intel quantum. Manufactured Processor at Intel's production facilities, it features a unique crystal structure that allows you to increase the yield of usable crystals on the plate and achieve a significant increase in productivity.

In essence, quantum computing systems represent the pinnacle of the development of parallel computing. These systems are able to solve the most complex computing problems that are inaccessible to traditional computers. In particular, quantum computers allow you to simulate natural processes in the interests of specialists in chemistry, materials science and molecular physics. With the advent of quantum computers, scientists will finally be able to create a catalyst for absorbing carbon dioxide from the atmosphere, superconductors capable of operating at room temperature, and new drugs for incurable diseases.

Intel has created a superconducting quantum chip of 17 qubits

However, despite significant progress in research and active discussions about the success of scientists, the problem of overcoming natural obstacles to the creation of viable large-scale quantum systems capable of demonstrating the required accuracy of calculations remains urgent. One such obstacle is the problem of producing homogeneous and stable qubits (the basic elements of quantum computing systems).

Qubits require extremely gentle treatment. Random noise and even random observation of the qubit can lead to data loss. For the stable operation of qubits, an extremely low ambient temperature is needed - at the level of 20 milliKelvin, which is 250 times colder than the temperature of open space. Such a temperature regime poses the strictest requirements for the design of the enclosures of quantum systems, which include qubits. In an effort to realize the full potential of quantum computing, Intel experts from the Component Research Group (CR) in Oregon and the Experimental Manufacturing Complex (ATTD) in Arizona are working hard to create innovative architectures and enclosures to fulfill the unique requirements and tasks of quantum computing systems.

With a size not larger than a ten-ruble coin, the new 17-qubit experimental processor has the following capabilities:

• The chip architecture made it possible to achieve increased reliability, improve temperature characteristics and reduce the level of radio interference that occurs during the collaboration of qubits.
• Scalable connectivity provides 10x to 100x greater I/O throughput than traditional wire-pin chips.
• Technology, materials, and design solutions have enabled Intel engineers to house quantum integrated circuits in the chip package that are significantly larger than those of traditional silicon processors.

Progress in quantum computing:

• Intel- QuTech cooperation in quantum computing is far from limited to the development and testing of qubit-based superconducting devices. As part of the collaboration, partners are engaged in research on all aspects of quantum computing systems, from qubits themselves to software and hardware architectures for controlling qubits and quantum applications. All these elements are necessary to overcome the path from the research laboratory to real products.
• Unlike competitors, Intel studies several types of qubits at the same time. In addition to the superconducting qubits used as elements of this latest experimental processor, Intel experts are developing another type of qubit - spin qubits in silicon. These spin qubits resemble a single-electron transistor and in many aspects behave similarly to traditional transistors, which allows them to be manufactured using similar technological processes.
• Thanks to outstanding performance and efficiency, quantum computers of the future will be able to solve a number of critical problems, but they will not be able to replace traditional computing systems and other innovative technologies, including, in particular, neuromorphic computing (see Loihi (neuromorphic processor)). Building on the principles of Moore's Law, we must continuously move forward by inventing new and developing existing innovative computing technologies.