A Computational Study on Destabilizing Effects on Grignard Reagents

Abstract

The diarylmethane derivative 4MDM shows beneficial immunomodulatory effects in models of pulmonary inflammation. We hypothesized that saturated carbocycles would improve solubility and stability of the compound. A Grignard reaction was proposed to make saturated analogues of 4MDM. However, formation of the Grignard reagent proved to be unreliable using traditional methods. Therefore, we worked to better understand formation of the Grignard reagent by a Mg/Cl exchange reaction. Traditionally, only the Mg-C bond was expected to be important for predicting stability. Ethynylmagnesium chloride, ethenylmagnesium chloride, and ethylmagnesium chloride were modeled as controls because of their known decreasing order of relative stability, which correlated with increasing bond lengths. All compounds were constructed through Avogadro and minimized via GAMESS with a STO 3G basis set using the B3LYP level of theory on the Argo Cluster supercomputer. Unsurprisingly, the shorter Mg-C bonds correlated with the increased stability of the compounds. When LiCl was added to the models, the Mg-Cl bonds became longer but the Mg-C bond remained constant. We concluded that increased Mg-Cl and Mg-C bond lengths indicate lower stability. We hypothesized that ethylmagnesium chloride derivatives with electron releasing groups (i.e., alpha Si, beta Si, or alpha O) would destabilize the Grignard reagent. The designs confirmed the predictions as hypothesized. Further, our results suggest LiCl is necessary to afford consistent behavior of the Grignard Reagent. Future directions include experimental validation of our results.