Quantum Computers Now Simulate Heat Behaviour Without Exponential Data Demands

authorsScientists at the Institute of Physics, Ecole Polytechnique Fédérale de Lausanne (EPFL), in collaboration with IBM Quantum and the IBM T.J.

Watson Research Centre, present the Quantum Finite Temperature Lanczos Method (QFTLM). The method uses quantum computers to compute thermal expectation values, enabling calculations beyond the reach of classical approaches. Demonstrations using the transverse-field Ising model confirm QFTLM’s ability to accurately reproduce thermal observables across varying temperatures, marking a key step towards practical finite-temperature quantum simulation. Quantum computation extends materials modelling beyond classical limits Quantum simulations of previously intractable problems in materials science and quantum chemistry are now possible by circumventing the exponential scaling limitations of classical computation.

The Quantum Finite Temperature Lanczos Method (QFTLM) extends established techniques to utilise quantum computers for this purpose. Understanding the behaviour of quantum many-body systems at finite temperatures is crucial for predicting material properties and designing novel quantum technologies, but classical methods struggle due to the exponential growth of the Hilbert space with system size. This exponential scaling arises from the need to account for all possible quantum states, making accurate simulations of even moderately sized systems computationally prohibitive. QFTLM addresses this challenge by leveraging the principles of quantum mechanics to represent and manipulate these complex states more efficiently. Real-time quantum Krylov methods, combined with efficient state preparation, enable the accurate computation of thermal expectation values, essential for understanding the behaviour of complex quantum systems at varying temperatures. Krylov subspace methods are a class of algorithms used to approximate the solution to linear equations or, in this case, to estimate the expectation value of an operator. The ‘real-time’ aspect refers to the evolution of the quantum state in time, allowing for the simulation of dynamic processes. Efficient state preparation is vital for initialising the quantum computer in a state representative of the thermal ensemble, which is a statistical distribution of states weighted by their Boltzmann factors. This ensures that the simulation accurately reflects the system’s behaviour at a given temperature. The Boltzmann factor, proportional to exp(-E/kT), where E is the energy, k is Boltzmann’s constant, and T is the temperature, determines the probability of occupying a particular quantum state at thermal equilibrium. The transverse-field Ising model confirms the algorithm’s ability to model complex quantum behaviour through accurate reproduction of thermal observables, such as magnetization and specific heat, across a broad temperature range. The transverse-field Ising model is a fundamental model in condensed matter physics, exhibiting a quantum phase transition between a ferromagnetic and a paramagnetic state. Magnetization, a measure of the net magnetic moment, and specific heat, a measure of the energy required to raise the temperature, are key indicators of this transition. Demonstrating accurate reproduction of these observables validates the QFTLM’s ability to capture the essential physics of the system. Performance is impacted by the Krylov dimension, which defines the size of the computational space explored, and the number of trace estimator states used to approximate thermal averages. Increasing the Krylov dimension allows for a more accurate representation of the quantum state, but also increases the computational cost. The number of trace estimator states affects the precision of the thermal average; more states generally lead to a more accurate estimate, but also require more quantum resources. Regularization techniques proved important for maintaining stability during simulations on noisy quantum hardware, mitigating errors inherent in current quantum computers. These techniques help to suppress the effects of noise, such as qubit decoherence and gate errors, which can distort the simulation results. While these successful simulations were performed on limited systems, future work will focus on scaling to realistically complex materials, containing hundreds or thousands of interacting quantum particles, and refining error mitigation strategies. Advancing materials modelling through quantum simulation despite hardware limitations Fundamental to designing new technologies, from superconductors to more efficient solar cells, is calculating the thermal behaviour of materials. Understanding how materials respond to temperature changes is critical for optimising their performance and discovering new functionalities. For example, in superconductors, thermal fluctuations can destroy the superconducting state, limiting their operating temperature. In solar cells, thermal energy can reduce efficiency by generating electron-hole pairs that do not contribute to electricity generation. This method offers a pathway to more accurately model complex materials, vital for advances in these fields. Current, imperfect quantum hardware requires careful consideration, introducing a critical tension as sophisticated error mitigation is demanded to achieve accurate results. The limitations of current quantum computers, such as limited qubit count, short coherence times, and gate errors, pose significant challenges to implementing QFTLM on realistically complex systems. Error mitigation techniques are therefore essential for extracting meaningful results from noisy quantum computations. SGian Gentinetta and colleagues at the Institute of Physics, EPFL, in collaboration with IBM Quantum, have created a more efficient framework for understanding quantum many-body systems by adapting a technique for averaging system properties at a given temperature to utilise quantum computers. The finite-temperature Lanczos method, originally developed for classical computers, provides a way to calculate thermal expectation values by iteratively constructing a basis of states that approximate the thermal ensemble. Adapting this method to quantum computers allows for a more efficient representation of these states and a reduction in the computational cost. Demonstrating accurate reproduction of thermal observables confirms the potential of this framework, though extending its application to larger, more realistic materials remains a key focus. The ability to accurately simulate the thermal properties of materials with QFTLM could accelerate the discovery of new materials with tailored properties, leading to breakthroughs in various technological fields. Further research will focus on improving the scalability and robustness of the method, as well as exploring its application to a wider range of materials and quantum systems. The development of more powerful and reliable quantum computers will be crucial for realising the full potential of QFTLM and other quantum simulation techniques. The researchers successfully adapted the finite-temperature Lanczos method for use on quantum computers, creating the Quantum Finite Temperature Lanczos Method (QFTLM). This new approach allows for more efficient calculation of thermal properties in quantum materials, bypassing limitations of classical computation. Accurate reproduction of thermal observables in the transverse-field Ising model demonstrates its potential, and this improved modelling capability promises to accelerate materials discovery. Future work will concentrate on enhancing QFTLM’s scalability and robustness, paving the way for simulating increasingly complex systems with greater accuracy. 👉 More information 🗞 Quantum Finite Temperature Lanczos Method 🧠 ArXiv: https://arxiv.org/abs/2603.25394 Tags: