An Overview of Intel’s Quantum Computing Efforts

Everyone is well aware that Intel is the leader in producing classical microprocessors, so there has been a lot of curiosity about what they are doing in the quantum realm. They have a research group in Hillsboro, Oregon that has been working on this for several years and we had the opportunity to speak with Intel Lab Quantum Hardware Manager Jim Clarke about their activity and also attended the recent March APS meeting in Chicago where Intel presented 14 papers. Intel has partnered with QuTech in the Netherlands on quantum computing and recently announced that it will install a quantum computing test bed at Argonne National Laboratory later this year. They have also worked with NIST, as evidenced by a few joint papers written by the companies at the APS meeting.

Here is a summary of some of the key concepts behind Intel’s quantum strategy:

Intel thinks useful quantum computing is still a long way off. Maybe ten years from now. They realize the technical complexities associated with the development of quantum technology and have a long-term roadmap to deliver powerful error-corrected quantum processor chips in the future. They do not plan to offer targeted offers for NISQ applications.

Intel realizes that its biggest advantage in the QC race is leveraging the high-performance capabilities they’ve developed over the past 50 years in building silicon transistors. Although Intel experimented with superconducting technology a few years ago and built a 49-qubit chip called Tangle Lake, they are now focusing on building qubits with spin qubit (aka quantum dot) technology. Spin qubits have a significant advantage over superconducting qubits because the matrix area per qubit is several orders of magnitude smaller. This allows them to fit over 10,000 quantum dot arrays on a single 300mm wafer. Using spin qubits also allows them to build the quantum dots in the same high-volume manufacturing facility where they build their microprocessors. Unlike some other groups that also build spin qubits using electron beam lithography, atomic layer deposition, and lift-off silicon processing, Intel uses its standard EUV (Extreme Ultra Violet) optical lithography, etching and its CMP (Chemical Mechanical Polishing) and building their chips using a high-volume 193 nm lithography process. Because they have optimized these process steps for their transistor technology for many years in their manufacturing facilities, this approach will provide them with high yields, high precision, low contamination, high uniformity, and high reproducibility. Intel told us that no new manufacturing equipment will need to be installed to build these wafers.

Intel has a lot of experience working with suppliers to get high quality materials and chemistries and they will leverage that expertise to get a more purified silicon starting material called 28Si. It is an isotope of the silicon atom that has no nuclear spin and is useful for improving coherence times. For conventional transistor processing, they use natural silicon which contains a mixture of 28Yes, 29If, and 30Yes. These last two isotopes have an inherent nuclear spin, which is why they want to avoid them in a quantum qubit. But to build a classic transistor, this is not a problem. Although some of Intel’s early spin qubit results used natural silicon, they will be converted to purified silicon 28If in the future.

The devices that Intel mentioned at the March APS meeting were linear configuration devices with what Intel called gate sizes of 55, 23, 17, and 7. This nomenclature is different from what we normally called a quantum gate because it represents the number of accumulation, plunger and barrier structures in the design that control where an electron can be trapped. (See the diagram below.) These configurations only support nearest-neighbor coupling, with the smallest configuration providing 3 qubits and the largest supporting a few tens. Interestingly, the technical results they reported were based on qubits built with natural silicon, but they still reported good coherence and fidelity numbers with T1 at around 14-65 milliseconds and T2* at around 1 microsecond. Single-qubit randomized benchmarking fidelities approached 99.9%. These measurements will no doubt improve once they build new devices with the 28Yes.

Conceptual diagram showing the structures in a quantum dot design. Credit: Intel

While Intel doesn’t need any new equipment to build the chips, they worked with BlueFors and Afore to create new test equipment to probe the wafers. They worked with these companies to create a first cryoprobe capable of probing a wafer at a cryogenic temperature of 1.6 kelvin in a very short time. One particular challenge for teams developing superconducting or spin qubit chips is that the chips must be cooled to kelvin or millikelvin temperatures before they can be tested. Normal transistors do not have this problem because their wafers can be tested at room temperature to provide engineers with quick feedback on the results of a newly fabricated wafer. The cryoprober allows Intel to take a fresh stick of wafer from the chamber and start getting results within hours. This allows them to quickly assess how well the processing went for the process engineers, determine which dice on the wafer are the best (called the Hero devices), then package those best chips and install them. in a dilution refrigerator for further testing. . Without a cryoprobe, testing a chip requires sticking the device in a dilution refrigerator and waiting a few days for the refrigerator to cool to millikelvin temperature so an engineer can start testing the chip. Intel claims that this cryoprobe capability provides a 1000X improvement and is a major advantage for Intel engineers to allow them to iterate and improve their designs much faster.

Another key technology Intel is working on is developing a cryoCMOS control chip that can be placed close to the qubits and eliminate much of the wiring that would otherwise be required to run through the dilution refrigerator from room-temperature electronics to the qubit chip. This can be a major concern in mechanical engineering, especially as the number of qubits continues to grow towards 1000 or more per system. Intel has developed such a chip called Horseridge II with QuTech which they are currently testing.

An interesting point about Intel’s quantum research developments is the fact that they don’t stop at hardware development. They pursue a comprehensive approach including software, as shown in the diagram below:

Diagram of Intel’s Full Stack QC approach. Credit: Intel

They developed their own software development kit (SDK) with an LLVM-based C++ compiler and system software workflow designed for efficient execution of classical/quantum variational algorithms. It will include its own optimization compiler which will take a user’s program and compile it to use the processors native gate set in the most efficient way and control all interactions between the classical processor and the quantum processor so that they can work effectively together. The SDK will support a few different simulators they have developed as backends as well as a quantum dot chip. Intel also has a software team researching algorithms to figure out how these might be run on a spin qubit-based quantum processor.

As we mentioned at the beginning of this article, Intel considers the development of quantum technology to be in its infancy and does not plan to offer end-user access via the cloud anytime soon. However, they have announced that they will partner with Argonne National Laboratory and provide Argonne with a quantum testbed later this year. We expect this testbed to have a small qubit count and its primary purpose will be to provide additional testing and feedback on Intel’s spin qubit technology. It is not clear if the configuration of this machine is yet finalized, but it was not specified in the press release.

Longer team, it doesn’t appear that Intel’s ultimate business model for delivering commercial quantum products has yet to be fully determined. They have a number of options, including selling individual chips, providing complete quantum computers to OEMs, partnering with a cloud provider, or even becoming a cloud provider themselves. But there is still a lot of technical development work to be done and the quantum ecosystem will surely change dramatically over the next few years. It may therefore be better for them to keep their options open in the short term. In any case, we wish them the best of success.

For more information on Intel’s recent quantum activities, you can view a recent article published in Nature magazine that outlines their latest advancements in spin qubit technology here and you can register and access Intel’s recent presentations at the March APS meeting here.

April 23, 2022

Sherry J. Basler