For decades, chemistry textbooks have taught that electron-donating groups weaken carbon–carbon bonds because they stabilize the organic radicals that can form when these bonds dissociate. New research by a team of VU theoretical chemists reveals a fundamentally different bonding mechanism.
The team quantum-chemically analyzed the bonding mechanism of a series of archetypal C–C bonds, namely, those between two substituted methyl radicals in RCH2–CH2R. By changing the substituent R from H to CH3, NH2, OH, and F, they tracked how and why the central carbon–carbon bond strength changes. The pattern was clear: the more electron-donating the group, the weaker the bond.
However, the analyses also revealed that the traditional explanation does not hold. The long-standing idea has been that electron-donating groups stabilize the carbon radical formed when the bond breaks, reducing the driving force for bond formation. Instead, the study found that nearly all substituents actually destabilize the radical rather than stabilizing it. Fluorine was the only exception.
So, what is actually going on?
When an electron-donating group is attached to a carbon radical, it pushes some of its own electron density onto the carbon center. This generates a lobe of an occupied orbital on the carbon radical center pointing in the direction in which a C–C electron-pair bond is formed between two such carbon radicals. This radical electron, surrounded by a lobe of an occupied orbital, is what the researchers describe as a "lone-pair shielded radical".
The study also shows that lone-pair-shielded radicals are more common in organic chemistry than previously appreciated and can be exploited as a tool to deliberately tune bond strength by choosing a substituent that makes the lone-pair-like lobe at the carbon center larger or smaller. Because carbon–carbon bonds form the backbone of organic molecules, from plastics and fuels to pharmaceuticals and biomolecules, understanding what determines their strength is essential. These findings offer chemists a more accurate framework for predicting and designing molecular behavior.