Finnish quantum computing startup IQM Finland Oy today said it has raised $13 million in a seed funding round to help it build hardware that it says can fix quantum errors and drive greater adoption of the emerging technology architecture.

The investors are Matadero QED, Maki.vc, MIG Fonds, OpenOcean, Tesi (Finnish Industry Investment), and Vito Ventures.

IQM began life as a joint research project at Aalto University in Helsinki and the VTT Technical Research Center of Finland before establishing itself as a company to commercialize the high-speed quantum processors it has developed.

Quantum computing is a fundamentally different and vastly more powerful computer architecture that, although still nascent, has the potential to solve extremely complex problems that are impossible, or would take years for today’s computers to tackle.

The main difference is quantum processing can take place in multiple states simultaneously. Whereas traditional computers use binary digits or “bits” that can be represented as 1 or 0, quantum computing uses “qubits” that can be “superpositioned,” allowing them to be represented as 1s, 0s or both states at the same time.

In addition, qubits can use a method called superdense coding that allows them to hold two bits simultaneously. So two superpositioned bits held in one qubit means they can process four times the data of ordinary computers.

The other important distinction of quantum computers is “entanglement,” or the ability of qubits to correlate with each other so that each is aware of the state of all the others. That means quantum computers grow in power exponentially as qubits are added.

Fixing quantum errors

One of the problems IQM is trying to solve with its hardware is something called quantum error correction, which it says will be essential if universal fault-tolerant quantum computation is ever to be achieved.

In an interview with SiliconANGLE, IQM Chief Executive Jan Goetz explained that quantum computing is plagued by unwanted changes in the state of the system. That’s because of the fundamental differences in quantum architecture, where information is stored in single “excitations,” for example a single photon or a single electron, as opposed to classical computers where large numbers of electrons store information collectively.

With classical computers, individual errors aren’t a problem because the impact is too small to be measured, Goetz said. But in quantum systems, these errors are a big deal because only a single excitation is used to store that data, which is consequently lost forever if the excitation is lost.

“If you assign a task to a computer, you want to be sure that the answer is correct,” Goetz said. “This is not the case anymore if there are errors in the system.”

Quantum errors occur from the unstable nature of the excitations that store their data, which are altered whenever they interact with the environment, Goetz said. Errors can occur both before computation and afterwards, during the readout process once a result has been found.

“If you want to start computing, you have to start from a well-known state,” Goetz said. “Basically, you want that the quantum memory has only zeroes stored in it before you start. If one of the memory entries changes from the zero state by interacting with the environment, you already have caused an error. The same goes for the readout.”

The general strategy to deal with these errors is to apply what Goetz calls “quantum error correction,” wherein the errors are detected on the fly and corrected afterwards. And that’s where IQM’s quantum processors can help, he said.

Noting that stringent temperature control is necessary to avoid overheating quantum systems and therefore changing the state of the excitations, the company has built what it calls a “quantum-circuit refrigerator” into its chips that comes with a unique “on/off capability” that can cool each qubit directly.

“When activated, it takes the qubit state to zero,” the company explained. “By enabling unprecedented temperature control, it addresses a key quantum computing bottleneck: slow and imperfect reset of the quantum memory.”

The company has also developed a new type of multi-channel readout process for qubits that speeds up the extraction of quantum information.

“With our implementation, the reset and readout tasks will not only be more accurate but also faster,” Goetz said. “If you perform these tasks faster, the quantum systems have less time to interact with the environment and therefore, they have less time to create errors.”

Goetz said IQM is now building on the early breakthroughs with an aggressive roadmap to develop new hardware systems. The systems will be designed for easy integration into existing quantum frameworks, he added.

“Quantum computing is redefining computing for good, and the potential upsides seem massive,” said Holger Mueller, principal analyst and vice president of Constellation Research Inc. “Many of the newly funded startups in the space are spinoffs from university research teams, and IQM is no exception. But research is one thing and commercial readiness quite another, so we’ll have to wait and see how IQM progresses.”

Image: IQM