Major Quantum Computing Advance: Scientists Break 25-Year Barrier in Chip Fabrication

Researchers at UCL have created a novel manufacturing technique for constructing quantum computers that achieves an almost negligible failure rate and exhibits significant potential for scalability, as indicated by recent findings.

A paper published in Advanced Materials details the first dependable method for accurately arranging single atoms in a grid, an accomplishment 25 years in development. This technique provides near-perfect precision and is amenable to scaling up, representing a crucial advancement towards fabricating practical quantum computers. However, significant engineering challenges still exist before this ambition can be realized.

In theory, quantum computers might solve problems that traditional, transistor-based computers cannot tackle. One promising strategy involves using individual atoms in silicon as quantum bits, or qubits. These atoms are cooled to extremely low temperatures to maintain their delicate quantum states and can be controlled using electrical and magnetic fields. This enables them to process information similarly to how classical transistors alternate between binary states of 0 and 1, albeit with far more intricate possibilities.

Harnessing the Strength of Quantum Mechanics

This capability allows the computer to utilize the power of quantum mechanics, the foundational laws of physics that explain how the universe operates. This encompasses phenomena such as superposition, the capacity of qubits to exist in multiple arrangements simultaneously, and quantum entanglement, the phenomenon where qubits become fundamentally connected.

These attributes allow for complex issues to be represented in innovative ways. For a problem with an extraordinarily large number of potential outcomes, a quantum computer can evaluate possibilities concurrently, unlike a conventional computer, which would process each option sequentially—an endeavor that would take today’s top supercomputer millions of years to complete.

Various methods for constructing a quantum computer are in progress, but none have yet achieved the necessary scale and minimal error rates.

One method for creating a quantum computer involves accurately placing individual ‘impurity’ atoms in a silicon crystal, facilitating the manipulation of their quantum characteristics to establish qubits. A significant advantage of this technique is its inherently low qubit error rates, supported by scalable silicon microelectronics technologies. The common method utilizes phosphorus as the impurity atom, but due to the fact that single phosphorus atoms can only be positioned with a 70% success rate, this approach remains distant from the near-zero failure rate essential for building a quantum computer.

In this research, UCL scientists speculated that arsenic could be a superior material compared to phosphorus for achieving the low failure rate necessary for building a quantum computer.

They employed a microscope capable of detecting and manipulating individual atoms, akin to the needle on a vinyl record player, to accurately insert arsenic atoms into a silicon crystal. They subsequently repeated this process to construct a 2×2 array of single arsenic atoms, prepared to transform into qubits.

A Move Towards Scalable Quantum Devices

Dr. Taylor Stock, lead author of the study from UCL Electronic & Electrical Engineering, remarked: “The most sophisticated quantum computing systems currently under development are still dealing with the dual challenges of mitigating qubit error rates and scaling up the qubit count.

Reliable, atomically precise manufacturing could facilitate the construction of a scalable silicon-based quantum computer. The prevailing belief was that single-atom fabrication using arsenic would encounter the same issues as phosphorus. However, based on our calculations, we discovered that single arsenic atoms might be positioned more dependably than phosphorus, and we have successfully accomplished this. We’ve been conservative in estimating that we can place atoms with 97% accuracy, but we are confident that this can be improved to 100% in the near future.”

Currently, the technique proposed in this study requires each atom to be individually positioned manually, a process that takes several minutes. Theoretically, this operation can be repeated indefinitely, but practically, it will be essential to automate and industrialize this method to create a universal quantum computer—necessitating the development of arrays containing millions, tens of millions, or even billions of qubits.

Collaboration with Industry and Future Prospects

The authors assert that the silicon semiconductor industry, valued at approximately $550 billion, should be able to aid in advancing this field, as both arsenic and silicon are widely utilized in semiconductor fabrication for classical computing. The strategy established in this study is anticipated to be highly compatible with existing semiconductor processing and could potentially be integrated once the engineering challenges have been resolved.