Why can't the reaction of grignard reagent with carboxylic acid give a ketone?

Question

I understand that a Grignard reagent can act both as a base and a nucleophile. It acts as a base in the presence of a proton source (eg: alcohols, amine, water, etc). But when reacting with a ketone or an aldehyde, it acts as a nucleophile and uses a specific reaction mechanism. Why isn't that same reaction mechanism not viable for reaction with carboxylic acid?

Why is this reaction mechanism wrong? Is the formation of a diol not possible due to repulsions or is there some other reason?

Answer 1

Huy Ngo says it all. A Grignard reageant will react first with just about any hydrogen that has even a modicum of proton donation capability, before "resorting" to the slower nucleophilic attack.

To get a ketone (in most cases), avoid the dissociable proton by using a nitrile instead, assisted by a copper salt. This forms an intermediate imine which is hydrolysed to a ketone after water/acid workup:

$\ce{CH_3MgI + CH_3COOH -> just a salt}$

$\ce{CH_3MgI + CH_3CN ->[Cu^+] (CH3)2C=N-M ->[H2O] (CH3)2C=O}$

Answer 2

It is known fact that organolithium compounds react with carboxylic acid to give ketones (e.g., Ref.1). Until recently, however, no report has been published on Grignard reagents on this manner. Now, it is evident that Grignard reagents can also be used to prepare ketones under specific conditions (Ref.2):

References:

1. Robert. Levine, Marvin J. Karten, “Reactions of carboxylic acids with organolithium compounds,” J. Org. Chem. 1976, 41(7), 1176–1178 (https://doi.org/10.1021/jo00869a020).
2. Kilian Colas, A. Catarina V. D. dos Santos, Abraham Mendoza, “$\ce{i-Pr2NMgCl•LiCl}$ Enables the Synthesis of Ketones by Direct Addition of Grignard Reagents to Carboxylate Anions,” Org. Lett. 2019, 21(19), 7908–7913 (https://doi.org/10.1021/acs.orglett.9b02899).