QuiX Quantum: Shaping the Future of Measurement-Based Photonic Quantum Computing
Most qubits require ultra-low temperatures to preserve their quantum characteristics. Photons stand out as they can serve as qubits even at room temperature and maintain their quantum states indefinitely—unless they are absorbed.
One of the main challenges in photonic quantum computing is the loss of photons. The other is entangling a large number of photons, which is required to perform meaningfully large quantum computations and solve problems in domains like materials and drug development, finance, or logistics.
Quix Quantum pioneers measurement-based quantum computing, diverging from the mainstream focus on gate-based quantum computing and harnessing the power of photonics to develop large and powerful quantum computers. Founded by Hans van den Vlekkert and Jelmer Renema in 2019, Quix Quantum has sold more than a dozen quantum systems to date, and in the summer of 2022, it secured a €5.5M Seed investment from PhotonDelta, FORWARD.one, and Oost NL.
Learn more about the future of measurement-based photonic quantum computing from our interview with Caterina Taballione, former quantum system engineer and now commercial & partnership lead at Quix Quantum:
Why Did You Start QuiX Quantum?
I have been fascinated by quantum computing since my Master’s studies in physics when I first encountered quantum information theory. I had studied information theory, and then, during my Master’s, I looked into quantum optics, leveraging photonics and quantum effects for information processing.
The more I learned about photonics, the more passionate I became, so I pursued a Ph.D. in integrated photonics at the University of Twente. I was in the midst of my PhD when Jelmer, a postdoc at the University of Oxford at the time, came to visit our lab. We talked about using integrated photonic chips for quantum computing, and I was hooked immediately, so we started collaborating on a joint research project.
Eventually, Jelmer proposed founding a company and turning our research in photonic quantum computing into a tangible product. This opportunity resonated deeply with me, as it felt like Quix Quantum was a direct extension of my PhD thesis. I was excited to see how my research was used for practical applications and saw the potential for Quix Quantum to pioneer large-scale quantum computers and tackle significant problems, which led me to take the leap and join the startup.
How Does Your Photonic Quantum Computer Work?
In general, quantum computers are a new type of computer based on the principles of quantum mechanics. It’s a fascinating theory describing the world on small length scales, typically smaller than a nanometer. At such length scales, you observe quantum effects such as entanglement and superposition that don’t occur for macroscopic systems and are at the core of why quantum computers can be powerful.
Our goal is to build such a quantum computer using integrated photonics. This means we miniaturize optical components and waveguides and integrate them onto a microchip so that light is confined within these tiny structures, and we can leverage its quantum properties for computing. Working on a chip level and with a high component density enhances performance, reduces costs, and increases efficiency.
One of the challenges is photons getting absorbed, which is the only source of loss in photonic quantum computing. We use a three-layer (“triplex”) structure of one silicon dioxide layer sandwiched between two silicon nitride layers, which allows the waveguides to support a wide range of optical wavelengths with the lowest losses available on the market. Still, some photons get absorbed, and their information is lost, so we need to produce many photons spread over time for error correction to ensure we have enough to continue the computation.
The other big challenge is entangling the photon qubits. Most quantum startups pursue gate-based quantum computing, where qubits are static, and two-qubit gates—operations in a quantum algorithm involving two qubits—are used to entangle and exchange information between the qubits, but they often also lead to errors.
Performing two-qubit gates with photons is extremely hard and not viable from our point of view. That’s why we opted instead for a different paradigm called measurement-based quantum computing, where the complexity of creating two-qubit gates is shifted to creating a large, entangled state of many photon qubits—a so-called cluster state—before the quantum computation starts.
Entanglement is a quantum effect that links the properties of the entangled particles. For example, suppose you entangle two photons and measure the polarization of one of them. In that case, you automatically know the polarization of the other one since they’re linked, even when the photons are separated.
Measuring one photon’s state impacts the state of all the other photons entangled with this specific one. Thus, once we create a cluster state of many entangled photons, we do the simplest thing possible and just use single qubit gates, i.e., rotating single qubits, which reduces the complexity of a quantum computation a lot.
In the simplest case, you measure the first set of qubits, which, thanks to entanglement, impacts all the other entangled qubits. The outcome of this first set of measurements determines what measurements to perform on the second set of qubits, a process called feedforward. Analogously, this determines the measurements on the third set of qubits and so forth. Any quantum algorithm can be implemented using an appropriate series of measurements, and you can change the quantum algorithm at any time by changing the measurements.
In a gate-based quantum computer, you must have all the qubits available to perform your computation from the beginning. In a measurement-based quantum computer, you only need to generate the required entangled qubits per computational step, an additional advantage of measurement-based quantum computing.
Let me give you an analogy: imagine passengers going via train from city A to city B. In gate-based quantum computing, the entire track (the qubits) gets built upfront, and it’s pre-determined when trains will meet at which station to let passengers swap trains (the algorithm, i.e., where and when the quantum gates have to be applied) to get to their final destination. In measurement-based quantum computing, you don’t have the entire track laid out from the start. Instead, you create the tracks (the entanglement between the qubits) you need for each step of the quantum computation, i.e., you measure where you are on the journey after each step and then build the next piece of track (feedforward).
You can build a quantum computer in many ways, and there’s no definite winning technology platform yet. However, photonics has a few distinct advantages, and we believe it will be the winning technology. For example, it works at room temperature and thus consumes less energy and requires less infrastructure as you don’t need cooling equipment. Also, photons intrinsically don’t suffer from decoherence, i.e., if you encode a quantum state in a photon, it will preserve it forever unless the photon gets absorbed or gets lost by escaping the quantum circuit.
Finally, photonics can leverage established telecommunications infrastructure, which is great for building quantum processors, integrating them with optical networks, and connecting different quantum computers. This has the potential to make distributed quantum computing and the quantum internet a reality.
Our motto at Quix Quantum is that we don’t make promises; we deliver. We are one of the few companies that have sold quantum hardware systems, 15 to date, mostly to quantum research labs, national computing centers, and other government bodies. We have actual customers and researchers using our machines.
How Did You Evaluate Your Startup Idea?
Once fully developed, quantum computing can have many applications and open up large markets. So, the question is more about unlocking this opportunity.
We start by building special-purpose quantum computers, which are not necessarily application-specific but specialized in the sense that you can’t program any arbitrary quantum algorithm. The search space of that computer will be somewhat restricted. We are investigating its potential applications in different fields and researchers will be able to use it through the cloud. That way, we will improve the system, aiming to scale it to a point where it becomes a general-purpose quantum computer that can run any quantum algorithm.
We are actively seeking collaboration with end users to make ends meet: knowing better about end users’ problems will allow us to engineer quantum algorithms that can solve these problems and meet requirements.
Some problems will be solved sooner than others as the size and quality of quantum computers increase. As the technology matures and reaches a new level, new problems will be unlocked. The first problems that quantum computers will be able to solve could be in chemistry, but there will also be applications in finance, logistics and optimization in general. When we’ll reach quantum advantage for a given problem will be largely determined by the size of the problem and the amount of data required to solve it.
What Advice Would You Give Fellow Deep Tech Founders?
When you found a startup, initially, you might have just a few other people as cofounders or employees working on the startup, but finding the right people is crucial. Make sure you have a team with different profiles and skills to tackle the challenges that will inevitably come your way. Think about your strategy. Of course, you can’t plan every detail in advance. But think about a plan for how your startup can succeed.