I'd read yesterday that Grignard reagents can't give ketones on reaction with carboxylic acids. It is because of the fact that Grignard reagent is a "super-base", and it will deprotonate the acid to form a carboxylate salt, rather than doing a $\ce{Nu-}$ attack.

However, after watching a video on Youtube, on the reaction of organolithium with acids (example reaction below),


I wondered,

  • After the acid-base reaction, why couldn't Grignard reagent (when taken in excess) do a $\ce{Nu-}$ attack on the carboxylate salt, just like organolithium does and then form ketone? Just like it does with an ester, acid chloride, etc.
  • $\begingroup$ Does this answer your question? Why are organolithium and organomagnesium compounds nucleophilic in nature? $\endgroup$ Commented Aug 29, 2020 at 7:44
  • $\begingroup$ @Safdar, No. I couldn't think how a difference in Nu nature can prevent the grignard reagent from attacking the carboxylate salt? $\endgroup$ Commented Aug 29, 2020 at 7:53
  • $\begingroup$ After formation of salt, the $\ce{O-}$ makes carbon way less electrophilic due to the distribution of electron density, so we need a much better nucleophile. $\endgroup$ Commented Aug 29, 2020 at 7:56

1 Answer 1


Two ideas come to mind for how a Grignard reagent could differ from an organolithium compound. Both are related to the bulkier nature of the Grignard molecule.

  1. Magnesium forms a more ionic bond than lithium.

In this hypothesis, the reason the organolithium compound attacks the carboxylate is that we have significant covalent bonding between one oxygen and the lithium, so we don't really have an ion: the carbonyl function is preserved with the non-lithiated oxygen. If the $\ce{MgX^+}$ moiety from the Grignard reagent were bonded more ionically, then the Grignard substrate would behave more like a true carboxylate ion with poor electrophilicity. The larger size of the Grignard fragment (including the halogen atom) combined with only one net positive charge (rather than two as on a bare magnesium ion) could promote such greater ionic character.

The diagram below shows how the extra atom could promote more ionic character in the Grignard product (using bromine as the halogen). Assuming magnesium bonds covalently primarily with its $3s$ orbital, that orbital forms a three-center four-electron bond with the more electronegative neighbors, which gives an ionic contribution to each individual linkage. With the corresponding lithium compound a similar delocalization does not occur because the lithium is bonded only to a single atom.

enter image description here

Note that a similar delocalized bonding model can be rendered for the Grignard reagent itself, thus enhancing the ionic character of the Grignard reagent's carbon-magnesium bond.

  1. The ketone precursor with a Grignard reagent is sterically hindered.

When an organolithium compound reacts, two $\ce{OLi}$ groups become attached to the original carboxylate carbon and these are later hydrolyzed to a carbonyl oxygen. With a Grignard reagent, two bulkier $\ce{OMgX}$ groups would have to be attached to form the ketone precursor, creating more possibility for unfavorable steric conditions.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.