Fully integrated Quantum processing units

For several sectors looking to drastically increase their procedures and capacities, quantum computing opens the next frontier of computing power. To achieve this, massively scalable qubit technologies must be created. Additionally, rising quantities of qubits must be controlled, and error levels must be kept as low as possible to prevent measurement from being impacted.

Oxford Ionics and Infineon Technologies established a partnership to develop fully integrated quantum computing units (QPUs). Within the next five years, the commercial production of QPUs with hundreds of qubits will be made possible, thanks to the electronic qubit control (EQC) technology developed by Oxford Ionics combined with Infineon’s Ion Trap technology as well as engineering, manufacturing, and assembly capabilities.

The objective is to translate quantum computing technologies from research laboratories into useful industry applications.


One of the big challenges of building quantum computers is finding ways to build quantum processors that can be fully integrated and that can be fabricated scalably. Typically, trapped–ion qubits are controlled by individual laser beams supplied by a painstakingly aligned optical assembly. As the number of qubits increases, this approach rapidly becomes untenable. The introduction of future chips will improve the scalability of quantum computers by reaching thousands, or even millions, of qubits, reducing the complexity of the quantum processor — one of the critical barriers to achieving the viability of quantum computers.

Trapped–ion quantum computers implement qubits using single atoms. These atoms of a given material are ionized so that they have a net–positive charge and, therefore, can interact via Coulomb interaction. This simplifies the realization of 2–qubit gates that can facilitate qubit entanglement. The atoms are arranged in the hardware in a lattice structure by means of electromagnetic fields that confine the atoms to a specific location. Quantum gates are made using laser radiation that, by interacting with the electrons, can change their state.

The main challenge is finding a way of controlling the qubits that can be fully integrated at the chip scale,” said Chris Balance, co–founder of Oxford Ionics. “Trapped–ion quantum computers work by manipulating the quantum state of atomic ions [the qubits] trapped some tens of microns above the surface of a chip. Conventionally, these qubits are controlled using lasers, which are challenging to integrate at the chip scale and can lead to intrinsic errors in the quantum computations. Oxford Ionics’ patented electronic qubit control technology is a way to control the qubits using electronic currents flowing in integrated structures in the chip, instead of incorporating the lasers.”

From the Infineon perspective, we will work on integrating certain aspects of the control electronics and the optics whilst managing the complexity and maintaining our improved trap properties,” said Stephan Schaecher, director of new application and quantum computing for the Power System & Solutions Division of Infineon Technologies. “Other identified challenges that Infineon is already working on in various projects include enabling faster gates, increased connectivity on– and off–chip, and generally better processor architectures.”


The big challenge in quantum computing is scalability and performance improvement. According to Oxford Ionics, the company’s technology can offer both, and collaboration with Infineon and its mature and flexible semiconductor process will accelerate the accessibility of a commercial QPU.

Ballance said the devices that Infineon and Oxford Ionics have produced so far are optimized to develop new capabilities and control only a handful of qubits.

What’s really important about them is that they control the qubits using electronics built into the chips, rather than lasers, giving pathway to very large–scale integration. In the EQC architecture, Oxford Ionics has shown single–qubit gate error rates below 0.0001% [1 ppm], and 2–qubit error rates at the 0.1% level [99.9% fidelity]. The goal of Infineon and Oxford Ionics is to offer within five years individual, fully integrated QPUs offering hundreds of qubits networked together into a networked quantum supercomputing cluster.”

By the end of 2022, the first Oxford Ionics devices will be cloud–accessible, giving users access to these quantum computers, with the target to expand to hundreds of qubits in less than two years. With Oxford Ionics’ quantum networking technology, Infineon and Oxford Ionics hope to provide standalone, fully integrated QPUs with hundreds of qubits within five years. It will be Infineon’s responsibility to provide the technological, manufacturing, and assembly foundations to enable a considerable amount of qubits with low error rates.

Schaecher pointed out one of the most challenging aspects of moving quantum into industry: “Making quantum computers practical requires experts as well as managers who combine an understanding of physics with application know–how and business insight. At the moment, these are scarce. That’s why we co–founded QUTAC, in order to jointly build up the necessary knowledge base. As a technology supplier, there are still quite some technical challenges to be solved, and above all, a lot of staying power is needed. Something comparable was perhaps the development of EUV lithography for semiconductor manufacturing — very long and intensive development but now impossible to imagine chip fabrication without it.”

According to Schaecher, there will be adjustments and improvements done to the process and the materials side to further improve the performance of the trap properties. “Infineon is leveraging its unique know–how in the development and manufacturing of specialized technologies, such as 3D MEMS structures or the integration of exceptional materials into a semiconductor fabrication,” he concluded.