How to achieve regioselective (base-catalyzed) enolization of a carbonyl group in a diketone?

One can convert a ketone to an enolate, either base- or acid-catalyzed.

Let's say you now have a diketone. You choose to treat this compound with base (e.g. hydroxide anion, alkoxide anion, LDA/THF, NaH).

How can one control the regioselectivity of this enolization reaction such that only one specific carbonyl group is converted to an enolate group?

• You should read up on kinetic vs. thermodynamic control. But you're probably going to end up with a mixture.
– Zhe
May 3 '20 at 15:13
• Selective protection, as shown here might help... May 3 '20 at 15:14

As @Zhe and @Zenix have mentioned, you have to hope that there is a difference in the rate of enolate formation to differentiate the two ketones.

To do this cool a rigourously dried solution of the diketone and a trapping reagent ($$\ce{Me3SiCl}$$ if you're doing an anhydrous workup otherwise TIPS-Cl or MeI) to $$\mathrm{-75 \, ^\circ C}$$ and slowly add 1 eq. of LDA keeping the reaction mixture cold. This will trap the first formed enolate (kinetic enolate). Allow to warm, dilute with hexanes, filter (or wash with aq if you used TIPS or MeI), concentrate and analyse.

If this gives a mixture then you try making the more stable (thermodynamic enolate) and trap that. This means forming the enolate without the trapping compound ($$\ce{Me3SiCl}$$, MeI or TIPS-Cl), giving it time to equilibrate then adding your trapping compound. This might be done with NaH in dry DMF, left to stir at $$\mathrm{-30 \, ^\circ C}$$ for an hour then the trap is added, worked up and analysed.

Even then with this example I think you'll get a mixture, but you don't know until you try. If you do get a mixture then you have to rearrange your synthetic pathway so that you don't have this problem - you'll probably have a different one but that's the way it goes!

• Better to add ketone to LDA because it minimizes the generated enolate from exchanging with ketone that has not been enolized. May 3 '20 at 19:51
• I agree that for trapping the kinetic enolate of a single ketone that is a good way to go, but in this case as it is a diketone I think it risks forming the substituted di-enolate. May 3 '20 at 20:38