S17 - Material Sciences

Quantum computing (QC) holds tremendous potential for accelerating the simulation and design of energy materials, where classical computing methods are limited by the exponential divergence of complexity in high-dimensional materials configuration spaces. We develop and adopt tailored approaches for the integration of quantum-computing methods into modeling workflows for electrochemical materials [1,2]. Electrochemical reactions involve the transfer of both electrons and ions within the bulk, or at the surface, of active materials. Reliable theoretical predictions therefore require accurate descriptions of both the electronic and ionic structures of active phases, both of which pose tremendous challenges for classical simulation methods. A characteristic feature of active battery materials is the presence of occupational disorder within the ionic lattice. During charging and discharging, lithium ions, or other mobile ion species, are extracted, or inserted, into the lattice. The exponentially scaling number of possible arrangements of ions on the partially occupied sub-lattice makes the creation of representative atomistic models particularly challenging. Classically, this problem is addressed by the sampling of low- or lowest-energy configurations, e.g., by Monte Carlo methods or other classical optimization heuristics [3]. Quantum optimization techniques, such as adiabatic quantum annealing (QA), offer new avenues for tackling configurational combinatorics in battery materials modeling. QC has the potential to bypass these classical bottlenecks, in particular the use of quantum algorithms can allow for more efficient ground-state solutions of complex many-body electronic Hamiltonians with annealing-inspired methods, extending beyond simple approximations. Additionally, QC algorithms enable probing of larger materials spaces, which are needed for direct comparisons between simulation and experimental results.

In materials science research, combinatorial optimization plays a fundamental role in many tasks, such as experimental design and candidate material selection. For example, experimental design can be formulated as the problem of selecting which combinations of conditions to evaluate, making it a typical combinatorial optimization problem. For such problems, the application of quantum and quantum-inspired optimization methods, including quantum annealing, is highly anticipated.However, in practical materials research, there is still a lack of systematically organized knowledge regarding how individual problems can be formulated as quadratic unconstrained binary optimization (QUBO) problems, as well as which types of problems are well-suited for quantum annealing-based approaches. Furthermore, while materials informatics is being actively introduced, the application of quantum-related technologies is also being considered, potentially imposing an additional burden on researchers due to the introduction of new computational paradigms.In this study, we investigate methodologies for formulating concrete problems in materials research as QUBO instances and evaluate their practical applicability through case studies in physical simulation and data analysis. In addition, we explore implementation strategies that embed quantum-related optimization techniques into user-accessible tools in a way that minimizes the need for users to be aware of the underlying computational paradigm.