Quantum Breakthrough: New Algorithm Cuts Chemical Modeling Costs Dramatically

A team of scientists has developed a new method to model complex chemical systems more efficiently using quantum computing. Jeremy Canfield, Dominika Zgid, and J K Freericks introduced a refined version of the unitary coupled cluster (UCC) algorithm, called quadratic UCC (qUCC), which cuts down the computational workload by balancing tasks between quantum and classical processors.

The qUCC approach tackles a key challenge in quantum chemistry: reducing the circuit depth in variational quantum eigensolver calculations. By using a Taylor series expansion of the energy around small angles, the method approximates the effects of many UCC factors without heavy processing. This allows the quantum computer to store the quantum state while shifting the most demanding nonlinear optimisation steps to classical systems.

Tests on hydrogen chains and the BeH₂ molecule confirmed the method's reliability. As more UCC factors were treated exactly, the calculations showed systematic convergence. The team expanded the number of exactly handled factors from around 30 to roughly 300, enabling stricter checks against full configuration interaction (FCI) benchmarks.

The study also identified the toughest phase for convergence: the shift from weak to strong electronic coupling. Here, the method's sensitivity to correlation changes became most apparent. Despite this, qUCC required only a fraction—between one-third and one-half—of the total UCC factors to be processed directly on the quantum computer for accurate results.

The hierarchical design of qUCC reduces strain on quantum hardware by offloading intensive tasks. Classical processors handle the nonlinear optimisation for exact angles, while the quantum system manages the state generated by the ansatz. This division of labour makes the approach far more scalable for large, strongly correlated systems.

The new qUCC method significantly lowers the computational cost of modelling complex chemical systems. By combining quantum and classical processing, it allows for more efficient simulations of larger molecular structures. The findings open the door to broader applications in quantum chemistry, particularly for systems with strong electronic correlations.