Optimization of a reaction solvent is typically performed when a chemistry if progressed from discovery to scale up. Typically, a large number of solvents are screened to determine which solvent gives the highest rate and yield. Samples are drawn out during the reaction and are analyzed by HPLC. This screening method suffers from a long idle time as the HPLC methods are long and is limited to a small number of samples due to the large number of HPLC samples generated. What is described in this work is an in-situ UV/vis method to perform an on-line analysis of multiple reactions to quickly determine which solvents give the fastest rate. A fiber optic probe is placed directly into the reaction vessel and UV/vis spectra are collected simultaneously from each reaction. Composition profiles and pure component spectra of reactants, intermediates, and products are estimated, using iterative target transformation factor analysis (ITTFA), a type of self- modeling curve resolution (SMCR), without the aid of referee measurements or standards. The results indicate that the method can successfully predict which solvent and can be used as a broad screening tool for the optimization of the reaction. Pairwise analysis of consecutive batches can be used to perform standardless comparisons between the two batches to determine if the reaction proceeded faster or slower, and made more or less product.
|