PIC & Mix: How Quantum Technologies are Shaking up the Photonic Integrated Circuit Market

Estimated reading time: 3 minutes

Photonic integrated circuits (PICs) are optical systems fabricated on semiconductor wafers, allowing complex optical processes to be performed on a chip-scale device. PICs have enabled a range of applications, notably as optical transceivers for high-speed communication within AI data centers, where they are widely used as the backbone of communication across servers to train the most complex machine learning models.

Most of the PICs today are based on silicon or silica, as the fabrication techniques for these materials are the most mature. However, silicon/silica have properties which make them suboptimal or even unusable for some emerging applications within quantum technologies. As a result, quantum tech is one of the frontier applications that is driving interest in new material platforms for PICs, and in this article we will dive deeper into insights from the IDTechEx Materials for Quantum Technologies report that forecasts PICs for quantum to be a US$12.6 Billion market opportunity by 2046.

Why are Photonics so Intertwined with Quantum Tech?

Photonics is the realm of technology concerned with generating and manipulating light, and the study and use of photonics has long been intertwined with the cutting-edge of experimental physics: think lasers, microscopes, and optical systems that can take up entire rooms in a lab.

Quantum technologies, which covers quantum computing, quantum sensing, and quantum communications, is a branch of commercial technology that has largely been spun out of experimental physics research centers and universities. However, as quantum technologies move from the lab to market, real-world products can no longer be based on bulky optical tables and delicate systems of many individual lasers and lenses. This is where the core value proposition of PICs comes in – they offer a route for quantum technologies to shrink their complex optical systems down to robust and mass-manufacturable chips.

Computing with Light

For example, many of the most advanced hardware approaches to quantum computing are heavily reliant on photonic systems, including those based on neutral atoms, trapped ions, or photonic qubits. In neutral atom quantum computers, such as those developed by Infleqtion or Pasqal, and in trapped ion computers, which are built by IonQ and Quantinuum, complex systems of lasers, waveguides, and cameras are used to manipulate and measure individual atoms and ions – the ‘qubits’ in these machines. The photonic quantum computers developed by PsiQuantum, ORCA Computing, and Quandela take this a step further and use photons (single particles of light) as the qubits themselves.

Developing PICs suited for these applications is therefore crucial for the scale-up of quantum computing for these players. As a result, the last 24 months have seen a flurry of acquisitions by major quantum computing players of photonics companies. These deals are typically focused on bringing expertise and fabrication capabilities to these quantum computing companies, padding out the skillset they need to turn theory and experiments into a commercial product.

Going Beyond Silicon

Silicon and silica (silicon dioxide) are the most mature wafer materials for semiconductor manufacturing, but some of their properties are a poor match for the requirements of quantum technologies. One fundamental barrier is that silicon is not transparent in the visible range of light, where many important frequencies for quantum technologies lie, such as atomic or solid-state spin transitions. Also, unlike datacoms applications which typically prioritize speed and high-power capacity above all, quantum technologies are generally more concerned with achieving very low noise and high stability to protect fragile quantum states.

These different requirements have pushed quantum tech players to experiment with materials beyond the usual silicon-based platforms. Promising candidates with suitable properties include silicon nitride (SiN), which is largely compatible with existing silicon-based processes, or thin-film lithium niobate (TFLN) and barium titanate (BTO), which have a high electro-optic coefficient suited for extremely fast modulation of light. However, TFLN and BTO are far less commercially mature than silicon photonics, limited to smaller wafer sizes, high price tags, and only a handful of foundries that are able to work with these materials.