Practical Quantum Computing on the Horizon
IBM, Google, Intel, and Silicon Quantum Computing - University of South Wales (SQC-USW) report on practical quantum computing advances and ongoing challenges in the race towards quantum supremacy at the APS March Meeting 2022. While system proposals diverge in their approach to specific shortfalls, private and public organizations agree that applications may begin appearing ahead of initial forecasting.
Quantum computing, which exploits quantum mechanics to vastly improve computation, currently generates too many errors for pragmatic applications but sustained development has shrunk estimates. IBM, an early entrant to the quantum computing arena, predicts this technology will be a reality as soon as 2026. Meanwhile, Google and SQC-USW expressed more modest expectations ranging between 2029 and 2033. This is a significant revelation for a field considered to be in its early infancy–and far from practical applications–just four years ago, according to a study conducted by the University of California, Santa Barbara.
IBM: Circuit Knitting
Early quantum pioneer IBM showcased steady advances in hardware, software, tools and online operability of 24 quantum systems of different sizes. Its development roadmap anticipates it will deliver a 433-qubit system by the end of 2022 and a 1,000+ qubit system in the cloud by 2023. Looking towards the future, IBM’s quantum researcher Hanhee Paik emphasized the need for material research to push advances in working cold temperatures and creating user-friendly environments–such as its serverless architecture.
Google: Error Correction & Scaling
The company has bet its quantum advantage on superconducting qubits, a system plagued with random noise that causes unacceptable error rates. In redress, Google has opted for surface coding, which refers to the distribution of quantum information to make the system immune to logical errors. Presently, this solution exhibits significant overhead and extra operation costs that requires additional innovation. Immediate goals pertain to building logical qubit prototypes at a run distance of three and five distance codes to compare their performance, an important analysis for future ambitions.
“In contrast, the scaling behavior is critical. If we cannot get bigger to be better, than the whole enterprise will not work,” said Satzinger.
Intel: Cryogenic Wafer
Leading CMOS manufacturing company, Intel entered the quantum computing arena in anticipation of leveraging its expertise in transistors to scale up faster. The company’s explanation for its quantum dot qubits was lean, which the company creates with silicon-germanium on silicon technology. Intel has progressed in evaluating wafers, which as it turns out is a very similar process to what it already does in its advanced complementary metal-oxide-semiconductor (CMOS) line.
Like IBM, Intel has expressed a need to invest in generating data at operability at below zero temperatures. “As we start to move into the spin qubit space, the information that we are interested in is really more relevant at cryogenic temperatures.” There is not “any information about that today at room temperature, and you need to go to lower temperatures to really characterize your quantum dot characteristics,” said Otto Zeitz, Quantum Hardware Engineer, Intel.
SQC-USW: Single Atom Qubits in Silicon
In partnership with the University of South Wales, Founder of Silicon Quantum Computing Michelle Simmons, presented steady progress in their work to understand and develop techniques to create single atom qubits, or phosphorous, in silicon. Using scanning tunneling microscopy (STM), she and her team are able to fab and align single atoms with precision, allowing them to “deterministically put dopants in silicon crystal surface.” By connecting them with lead states and measuring their position, the research group can create sensors to later “do the spin states.”
Figuring how to benchmark the readout between qubit gates will inform future ambitions related to cavity coupling and coherent control.