Quantum computers could give scientists a new way to explore the hidden behaviour of matter

Spectroscopy has long been one of science's most trusted windows into the hidden workings of matter. By shining light or other forms of energy onto molecules and materials and analysing what comes back, researchers can deduce structures, identify substances, and probe behaviour that would otherwise remain invisible. The technique underpins an enormous range of modern science: chemists use it to determine the shapes of newly synthesised molecules, astronomers use it to work out the composition of distant stars, and condensed matter physicists use it to map the electronic properties of materials that may one day form the basis of new technologies. Now, a team involving researchers at Queen Mary University of London has shown that a quantum computer itself can act as the spectroscopic instrument, offering a new route into quantum systems that conventional computers struggle to model.

The new study, published in Nature Communications, describes a generalised approach to quantum computational spectroscopy, an idea that extends the reach of a technique traditionally performed in laboratories with lasers, detectors and samples. Instead of measuring a physical material directly, the researchers used a quantum processor to carry out simulations of spectroscopic measurements, reconstructing key indicators of quantum behaviour from within the computation itself. The paper, titled "Generalised quantum computational spectroscopy on a quantum chip," suggests that quantum hardware could serve not merely as a calculator but as an exploratory tool for understanding complex quantum matter.

The motivation behind the work lies in a fundamental limitation of classical computing. Quantum systems, such as interacting molecules or engineered materials with exotic electronic properties, can be exceptionally difficult to model using conventional machines. The difficulty arises because the number of quantum states that must be tracked grows exponentially with system size: adding just a handful of interacting quantum particles can multiply the computational burden many times over. Even the most powerful supercomputers in the world eventually run out of capacity, forcing scientists to rely on approximations that may miss important physics. Quantum computers, which operate on quantum-mechanical principles themselves, hold the promise of representing such systems far more naturally, because their fundamental units of information — quantum bits, or qubits — can exist in superpositions and become entangled with one another in the same way that the particles of a real quantum system do.

Computational approaches have long complemented experimental spectroscopy by allowing scientists to predict properties using theoretical models and simulations before any measurement is attempted. In fields ranging from quantum chemistry to condensed matter physics, researchers routinely calculate the spectra of candidate molecules and materials on classical computers, comparing the results with laboratory measurements to refine their understanding. What the Queen Mary-led team has done is generalise the computational version of spectroscopy so that it applies to a much broader range of quantum systems. In particular, their method is not restricted to relatively simple, static systems in isolation. It can also handle systems that are affected by their surrounding environment, a scenario known as open-system dynamics, as well as systems whose properties change over time. These are precisely the situations where real-world materials and molecules live, bathed in thermal noise and undergoing constant evolution, and they are among the hardest cases for existing computational techniques.

The distinction between closed and open quantum systems is central to modern quantum science. In an idealised textbook description, a quantum system evolves in isolation, its behaviour governed purely by its internal structure. In reality, no system is perfectly isolated: molecules collide with their surroundings, electrons in a material interact with vibrations of the crystal lattice, and qubits in a quantum computer are disturbed by stray electromagnetic fields. This environmental coupling, often described as noise or dissipation, is usually treated as an obstacle to be minimised. But in many contexts it is an essential part of the physics, shaping chemical reaction rates, energy transport in photosynthetic complexes, and the behaviour of quantum devices. A spectroscopic method that can incorporate open-system dynamics, rather than treating them as an afterthought, therefore brings the computational tool closer to the conditions of real experiments and real materials.

At the heart of the method is a quantum computing technique known as an ancilla-assisted Hadamard test. In this procedure, an additional quantum bit, or ancilla, is used alongside the quantum system being simulated to extract measurable quantities from the computation. The ancilla acts as a probe: by preparing it in a particular state and allowing it to interact with the simulated system through controlled quantum operations, the researchers can read out information encoded in quantities that would be prohibitively expensive to estimate using classical algorithms. The Hadamard test, a well-established primitive in quantum computation, is the mechanism that makes this readout possible, allowing the team to evaluate complex quantities that encode the spectral characteristics of the system under study. By running this test on a quantum chip, the team was able to reconstruct a key measure of quantum behaviour, effectively performing the analogue of a spectroscopic measurement entirely within a quantum processor. This is the sense in which the spectroscopy is "on a quantum chip": rather than probing matter with light, the quantum computer probes a simulated quantum system and returns the kind of information that a spectroscopic experiment would yield.

To demonstrate the power of the approach, the researchers applied it to unusual quantum phenomena that sit at the frontier of modern physics. The first is parity-time symmetry breaking. In conventional quantum mechanics, physical systems are expected to obey certain fundamental symmetries, and when those symmetries break, the behaviour of the system can change dramatically. Parity-time symmetry and its breaking describe transitions that have been studied in specially engineered systems and have connections to novel optical and electronic behaviour. Capturing such transitions computationally is demanding, because they involve subtle features of a system's spectrum that approximate methods can easily miss.

The second phenomenon, topological holonomy, belongs to the family of topological effects that have become central to condensed matter physics. Topological properties are global features of a quantum system that are remarkably robust against local disturbances, which is why they are of interest for potential technologies such as fault-tolerant quantum computing and protected electronic states. Holonomy describes how a quantum system changes when it is transported around a loop in parameter space, a geometric effect that leaves a fingerprint in the system's behaviour. Such geometric phases have deep roots in physics, appearing in contexts from molecular dynamics to the design of quantum gates, and they often reveal structure that is invisible to more conventional measurements. By successfully probing both parity-time symmetry breaking and topological holonomy with their quantum chip method, the researchers showed that quantum computational spectroscopy can reach insights into quantum behaviour that are difficult to access either through conventional spectroscopy or through existing quantum computational approaches.

The significance of these demonstration cases lies in what they represent. Both phenomena involve aspects of quantum dynamics and spectral structure that standard experimental spectroscopy may struggle to isolate, and that earlier quantum algorithms were not designed to capture in full generality. A generalised framework that can address static systems, open systems interacting with their environment, and time-dependent dynamics in a single coherent method gives researchers a more flexible toolkit. It means that phenomena which are difficult to reproduce in a laboratory, or difficult to calculate on classical machines, could potentially be studied through quantum simulation instead. The choice of test cases is also telling: rather than demonstrating the method on trivial model systems chosen for convenience, the team targeted examples with genuine physical content, strengthening the case that the approach can capture meaningful science rather than merely executing a formal procedure.

The broader implications extend across several scientific disciplines. According to the research team, the work could ultimately be relevant to physics, chemistry and materials science. Computational spectroscopy allows researchers to investigate the properties of real or hypothetical materials before they are ever produced experimentally. In molecular engineering, this could mean screening candidate molecules for desired optical or electronic characteristics without synthesising them first, dramatically accelerating the search for new compounds. In drug design, it could aid in understanding how molecular structures respond to energy and interact with their surroundings, questions that lie at the heart of how pharmaceutical molecules bind to their biological targets. In advanced materials research, it could help scientists predict the behaviour of novel materials, including those whose exotic quantum properties are of technological interest, before committing resources to fabricating them. In each of these areas, the ability to simulate spectroscopic signatures on a quantum device could complement, and in some cases substitute for, costly and time-consuming laboratory work.

Dr Jinzhao Sun, from the School of Physical and Chemical Sciences at Queen Mary University of London, led the theoretical aspect of the study. The research, the team notes, represents a step towards using quantum computers not simply to perform calculations, but as tools for exploring and understanding the behaviour of complex quantum systems. That distinction matters. Much of the current excitement around quantum computing focuses on speed: the hope that quantum machines will eventually solve certain problems faster than any classical computer. This work points toward a complementary vision in which quantum processors function as scientific instruments, generating knowledge about quantum phenomena in ways that neither classical computation nor laboratory experiment can easily replicate. In this view, the quantum computer occupies a conceptual position closer to that of the telescope or the particle accelerator — an instrument through which new observations become possible — rather than simply a faster successor to the desktop machine.

As with any emerging approach, the work comes with caveats. The demonstrations described in the paper were carried out within the constraints of current quantum computing technology, and today's quantum processors remain limited in size and susceptible to errors. Qubits can lose their quantum state through interaction with their environment, and the operations performed on them are imperfect, which places practical limits on the depth and complexity of the computations that can be reliably executed. The study is best understood as a proof of principle: a demonstration that the generalised quantum computational spectroscopy framework works and can capture phenomena of genuine physical interest. Scaling the approach to systems of practical complexity will depend on continued progress in quantum hardware, including larger numbers of high-quality quantum bits and improved error correction. The researchers themselves frame the method as something whose usefulness will grow as quantum computing technology develops.

Even so, the study contributes to a growing body of work aimed at finding practical applications for near-term quantum devices. Around the world, research groups are exploring how modest-sized quantum processors, imperfect as they are, might already deliver value in areas such as simulation of quantum materials, optimisation and machine learning.