Cumulative complexity meta-metrics as an efficiency measure and predictor of process mass intensity (PMI) during synthetic route design

Abstract

Functioning as a surrogate for step count, a cumulative complexity meta-metric (∑C_M*), calculated along the longest linear sequence of a synthetic route, is demonstrated to be a useful predictor of process mass intensity (PMI). In contrast, common theoretical measures of efficiency such as ideality and convergence, in this case, were found to be of limited use. A workflow and model are presented which allow prediction of PMI from ∑C_M* for small molecules (<600 Da) with good accuracy (R² > 0.9) when applied to a test dataset and a small number of literature examples. Requiring no empirical investigation, this method provides estimates of achievable, long-term PMI for synthetic routes and can be applied at the design phase. The overall procedure has been developed to be amenable to future automation, allowing rapid application across large numbers of synthetic routes.