TechCapsules

AI-summarised brief · reviewed before publication

OTI Lumionics and Samsung Advanced Institute of Technology published research in the Journal of the American Chemical Society detailing their Iterative Qubit Coupled Cluster algorithm. The team successfully emulated over 200 logical qubits on a single commercial 32-core AMD processor with 800GB of RAM, avoiding supercomputing clusters. This emulation targeted 14 phosphorescent transition-metal organometallic complexes for OLED display materials. The algorithm achieved a Mean Absolute Error of 0.05 eV and an R² correlation of 0.94 against experimental data, outperforming standard classical methods like DFT. Implementing the code on NVIDIA Blackwell architectures delivered a 90x speedup, reducing complex calculations to one hour. The study demonstrates variational stability in multireference states where single-reference methods failed. By optimizing circuits with over 1.5 million parameters, the collaboration established a rigorous performance baseline for future fault-tolerant quantum processors in molecular electronic-structure calculations, proving accessible hardware can handle significant quantum algorithm emulation for materials discovery.