Quantum Algorithms Unlock Simulations of Neutrino Behaviour in Supernovae

Scientists at University of Toronto, led by Katarina Bleau, have developed a novel quantum simulation approach to model the intricate behaviour of neutrinos in extreme astrophysical environments, notably supernovae. The research team presents algorithms leveraging Dicke states and the $su(2)$ spin algebra to accurately model collective neutrino oscillations. These algorithms address limitations inherent in previous quantum simulations by more fully exploiting the symmetries present within these complex systems, thereby enabling more efficient and accurate modelling of dense neutrino gases. Performance evaluations on both classical and quantum hardware represent a significant step towards simulating phenomena previously considered computationally intractable.

Reduced qubit counts unlock simulation of dense neutrino systems

Neutrino simulations now require four times fewer qubits, decreasing from eight to just three for systems comprising seven electron neutrinos and seven electron antineutrinos. Previous simulation methods struggled to model systems beyond a handful of neutrinos due to the exponentially increasing computational demand associated with increasing particle numbers. This new approach surpasses those limitations by employing a more efficient encoding scheme. Based on Dicke states and $su(2)$ spin algebra, the newly developed algorithms enable the simulation of substantially larger and more realistic neutrino densities, reaching approximately 10³⁰cm⁻³, a density characteristic of the core-collapse phase of supernovae. Understanding neutrino behaviour at such densities is crucial, as they influence the dynamics of the supernova explosion and the synthesis of heavy elements.

This density was previously inaccessible to quantum simulation, opening new avenues for astrophysical research and potentially resolving long-standing questions about the mechanisms driving supernovae. Performance on the IBM Boston quantum computer is comparable to conventional encoding methods in terms of accuracy, yet substantially reduces the hardware demands, making larger simulations feasible with current technology. The qubit-efficient approach has been successfully extended to bipolar systems, simulating seven electron neutrinos and seven electron antineutrinos using only three qubits, a significant compression from the fourteen qubits needed by standard methods. This reduction in qubit count is particularly important given the limitations in the number of qubits available on current and near-term quantum computers.

A “diagonal Dicke” encoding exploits the symmetry inherent in neutrino-antineutrino interactions, effectively reducing the computational space required for the simulation. Specifically, the encoding reduces the Hilbert space dimension from (N+1)² to N+1, where N represents the number of neutrinos. This compression is achieved by mapping the many-body neutrino state onto a smaller Hilbert space spanned by Dicke states, which are symmetric states with well-defined total spin. Simulations conducted on the IBM Boston device demonstrate that this encoding maintains coherence for a longer duration than conventional methods, effectively delaying the onset of noise-induced errors in the calculated oscillation probability of the neutrinos. This improved coherence is vital for obtaining reliable simulation results. Successfully modelling a system with a mixing angle of π/2 and a J/|∆| ratio of 2.5 showcased the algorithm’s flexible nature and ability to handle different parameter regimes, although the current implementation relies on a first-order Trotter approximation. While computationally efficient, this approximation introduces some error not present in the more complex, second-order approach typically used in conventional simulations, representing a potential area for future refinement.

Reducing quantum resource requirements for modelling stellar collapse neutrinos

Simulating supernovae, the spectacular deaths of massive stars, demands increasingly sophisticated computational tools to unravel the behaviour of neutrinos within these extreme environments. These elusive particles play a vital role in the explosion’s energy transport, mediating the transfer of energy from the collapsing core to the outer layers of the star. However, modelling their collective oscillations, a phenomenon where neutrinos interact and change flavour in a correlated manner, presents a formidable challenge for even the most powerful supercomputers. The computational complexity arises from the large number of neutrinos involved and the intricate interactions between them. A sharp reduction in the qubits needed for these simulations is now achieved, and acknowledging concerns about circuit complexity is important. A reduction in the number of qubits needed for a simulation is valuable even if it necessitates more computational steps, as it expands the scope of simulations possible with existing hardware.

Dense neutrino gases offer a refined approach to simulating these complex astrophysical systems by exploiting inherent symmetries within the neutrino interactions. Algorithms with improved qubit efficiency have been developed by utilising a specific arrangement of quantum particles, Dicke states, and a mathematical framework describing particle interactions, namely the $su(2)$ spin algebra. This advancement enables the modelling of larger, more realistic scenarios, such as those occurring in supernovae, with reduced computational demands, which is particularly advantageous considering the constraints of present quantum hardware. The $su(2)$ spin algebra provides a powerful tool for describing the collective behaviour of neutrinos, simplifying the simulation and reducing the required computational resources. Demonstrated performance on both conventional and quantum computers indicates a promising new method for examining neutrino behaviour, optimising the quantum computing process, potentially at the cost of increased circuit depth. Further research will focus on mitigating the impact of the Trotter approximation and exploring the potential for implementing higher-order approximations to improve the accuracy of the simulations. The ability to accurately model neutrino transport in supernovae has implications for our understanding of nucleosynthesis, the formation of neutron stars, and the overall evolution of the universe.

This research successfully developed new algorithms for simulating dense neutrino gases using fewer qubits. Reducing the number of qubits needed is important because it allows scientists to model more complex systems, such as those found in supernovae, with current quantum computing technology. The algorithms utilise Dicke states and the $su(2)$ spin algebra to improve efficiency, and have been demonstrated on both classical and quantum computers. The authors intend to refine these simulations by addressing approximations and improving accuracy in future work.