Two-Dimensional Materials toward CMOS-Compatible Scalable Quantum Hardware

The quest to build a quantum computer is less about inventing a new kind of machine and more about solving a problem of extreme fragility. At the heart of these machines are qubits, the tiny units of information that can exist in multiple states at once, unlike the simple on-off switches of a standard computer. To work, these qubits must be kept in a state of perfect order, isolated from the slightest vibration, heat, or electrical noise. In the real world, however, the materials used to build them are rarely perfect. They contain microscopic imperfections, rough edges, and chemical leftovers from the manufacturing process that act like static on a radio line, scrambling the delicate information before it can be used. For decades, scientists have tried to smooth out these surfaces and purify these materials, but the very act of building a complex chip often introduces new flaws that kill the quantum state.

A new perspective from researchers at Eindhoven University of Technology suggests that the solution might lie in a different kind of material: two-dimensional crystals. These are substances that are only a single layer of atoms thick, like a sheet of paper that is one atom deep. Because they are so thin and their edges are naturally clean, they offer a way to build quantum devices without the messy, rough interfaces that plague traditional materials. The researchers are asking a practical question: can the industrial methods already being developed to put these ultra-thin sheets into next-generation computer chips also be used to build the quantum processors of the future? The answer is not yet a simple yes, but the path forward is becoming clearer. The team has mapped out where these materials are currently working, where they are failing, and exactly what experiments need to happen next to turn laboratory curiosities into mass-producible technology.

The journey from a single, hand-assembled device to a factory-made quantum processor is filled with hurdles. In a laboratory, a scientist might carefully peel a tiny flake of a two-dimensional material off a larger crystal and stack it with another flake under a microscope to create a single qubit. This works for one or two devices, but it cannot be scaled up. To build a computer with thousands of qubits, the process must be reproducible, uniform, and compatible with the massive factories that currently make silicon chips. The researchers point out that while we have seen impressive results with individual two-dimensional qubits, we have not yet seen them working together in large, reliable arrays. The gap between a single successful experiment and a manufacturable product is wide, and it is filled with issues like chemical residues left behind by the manufacturing process, bubbles trapped between layers, and variations in the material that change from one spot on a wafer to another.

To understand where we stand, the authors sorted the current research into four distinct categories. First, there are the fully demonstrated qubits, where scientists have successfully prepared the quantum state, controlled it, and read the result. Second, there are demonstrated components, which are parts of a circuit that work as intended but have not yet been combined into a full qubit. Third, there are physical phenomena, where the underlying science is proven but no device has been built to harness it. Finally, there are proposed concepts, which are ideas on paper that have not yet been tested. This sorting reveals that while we have some working qubits made from these materials, many of the most exciting ideas remain in the "proposed" or "component" stages. For example, scientists have built a superconducting circuit using a capacitor made of a two-dimensional material that can hold a quantum state for a significant amount of time, but they have not yet shown that this can be done consistently across a whole factory wafer.

One of the most promising areas involves using these atomically thin sheets to create better interfaces for superconducting circuits. In traditional superconducting qubits, the barrier between two metal layers is often a rough, disordered oxide that causes the quantum information to decay. By replacing this with a perfectly flat, two-dimensional crystal, researchers have created junctions that are much cleaner. They have shown that these new junctions can transport electricity without resistance and can be tuned by electrical gates, offering a level of control that was difficult to achieve before. However, the paper emphasizes that while these individual devices work, the statistical proof is missing. We do not yet know if every junction on a large sheet will perform the same way, or if the process of making them introduces hidden flaws that only show up when thousands of them are connected.

Another major avenue is the use of two-dimensional materials to trap and control single electrons, which act as spin qubits. In these devices, the electron's spin, a property similar to a tiny magnet, holds the information. The researchers highlight that materials like bilayer graphene can confine these electrons so effectively that their spin can last for a very long time. Specifically, a single-hole device in bilayer graphene exhibited a spin–valley relaxation time of 38 seconds at extremely low temperatures. This is a remarkable feat, suggesting that the material itself is very quiet and does not disturb the electron. Yet, the challenge remains to take this single success and replicate it across a large chip. The current experiments rely on small, carefully selected pieces of material, and it is unclear if the same performance can be achieved when the material is grown on a large scale and processed by machines rather than by hand.

There is also a unique opportunity with two-dimensional materials to create qubits that can be controlled by light. Certain defects, or missing atoms, in a crystal called hexagonal boron nitride can act as a quantum bit that can be read and written using lasers, even at room temperature. This is a rare and valuable property, as most quantum systems must be kept near absolute zero to function. The researchers have identified specific defects that can be controlled this way, but the work is still in the early stages. The goal is to find a way to create these defects exactly where they are needed, in large numbers, and to ensure that they all behave the same way. Currently, the ability to place these defects with precision and to get a high yield of usable ones is the main bottleneck.

The path forward requires a shift from looking at single, perfect devices to looking at the statistics of many devices. The paper argues that the next critical step is to measure how these materials perform across an entire factory wafer. This means checking not just if one qubit works, but if a hundred of them work, and if they all work the same way. It involves developing new ways to measure the quality of the material and the manufacturing process without destroying the device. The researchers propose a cycle of testing where data from the factory floor is linked directly to the performance of the quantum device. If a specific step in the manufacturing process causes a drop in performance, that step must be identified and fixed. This kind of rigorous, data-driven approach is what separates a scientific curiosity from a technology that can be built at scale.

The authors also look ahead to three specific concepts that could only be realized with these two-dimensional materials. The first involves inserting rare-earth ions into the gaps between the layers of the crystal to create new types of qubits. The second idea uses the unique twisting of the layers to create a special kind of electrical junction that could act as a qubit without needing external magnetic fields. The third concept uses the pattern created by twisting two layers of material to create a multi-level quantum system, which could potentially store more information than a standard qubit. While these ideas are theoretically sound and supported by some experimental evidence, they remain unproven. The paper is clear that these are not finished products but rather directions for future research that need to be tested with the same rigor as the current technologies.

Ultimately, the value of these materials lies not in replacing the existing silicon-based technology, but in adding specific functions that are difficult or impossible to achieve with traditional methods. They offer a way to reduce the noise that kills quantum information, a way to integrate control electronics directly next to the qubits, and a way to create interfaces that are cleaner and more controllable. The researchers conclude that the potential is real, but it is conditional. The benefits will only be realized if the manufacturing processes can be refined to produce consistent, high-quality devices. Until then, the focus must remain on proving that these materials can be made to work reliably, not just in a single experiment, but in the vast, complex environment of a quantum processor. The story of two-dimensional quantum hardware is still being written, and the next chapter depends on the ability to turn a few successful lab demonstrations into a reliable, scalable industry.