Researchers have developed a novel energy-only approach called tReax_expedition for identifying competing reaction pathways in chemical processes, particularly when potential-energy gradients are unavailable. This method combines Monte Carlo exploration with stochastic quench labeling and restrained continuation to resolve candidate channels and sample chain maxima. Benchmarks on various potentials and gas-phase reactions demonstrated the approach's ability to accurately recover intermediate states and competing routes, with sampled chain maxima differing from refined saddle energies by less than 0.76 kJ/mol in most cases. While neural committor learning aids in connecting paths, molecular shooting diagnostics suggest limited committor calibration. AI
RANK_REASON The cluster contains a scientific paper detailing a new computational method for chemical reaction pathway search. [lever_c_demoted from research: ic=1 ai=0.7]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
