A Simple Topology-based Model for Predicting Activation Barriers of Chemical Reactions at 0 K


This works exploits two topological concepts for developing a model capable of predicting activation energies at 0 K. Experimental barriers of 17 exergonic reactions were fit via simple linear regression analysis, leading to the model which accurately predicted activation energy of both organic and organometallic reaction systems under different conditions, for instance, gas-phase/solvent media and temperature. This linear function was further recalibrated to enhance its predicting capabilities, generating such a refining process the equation characterized by a squared Pearson correlation coefficient (r2 = 0.9774) 1.1 times higher. Surprisingly, the performance of the corrected fit equation was only slightly better. Moreover, both linear fits failed to provide reasonable energy values for the epoxidation of isobutene and -methylstyrene by dimethyldioxirane. This issue might be explained considering our results suggest that it is paramount to use a level of theory that guarantee an adequate description of the physical and chemical properties of the reacting system in order for the model to predict reasonable activation barriers. Since the fitted data involves reactions experimentally studied within the range [293, 748] K, a high degree of inaccuracy should be expected when applying the model to reactions investigated under laboratory conditions involving temperatures outside the indicated interval.