[w4rlock]

Hi, I'm new here.

Anybody know of anything safer and easier to use than NaH or potassium tert-butoxide for Claisen/Dieckmann condensations? Those two superbases are the only things I found in literature.

thanks,
w4rlock

[Arctic-Nation]

NaH and KtBuO are hardly superbases. In the Claisen and Dieckmann condensations, one simply needs a strong, non-nucleophilic base (except when working with simple methyl and ethyl esters), and these come with limitations. You won't find anything safer and easier to handle than KtBuO, and if you're worried about atom economy and waste reduction, you can't beat NaH.

[w4rlock]

Thanks for the info. I am in fact using methyl esters.

Well, it was suggested to me that benzene should be used as solvent to NaH, which makes it very non-user/environmentally friendly. Will try out other dry solvents such as diethyl ether to see if it has an effect on reaction too.

KtBuO is very attractive and I think I will probably settle for that. LDA is another option i'm looking at.

[Dan]

You can also use the alkoxide that matches your ester - so a methoxide in your case, NaOMe is the commonest. By doing so, although the base is nucleophilic there is no net reaction from transesterification.

[w4rlock]

Essentially that's what you're doing with sodium hydride too, right? Preparing the methoxide in situ using catalytic amounts of alcohol to start the reaction off.


First off I'm going to go with KtBuO and see where that gets me.

thanks for the responses so far.

[Dan]

Essentially that's what you're doing with sodium hydride too, right? Preparing the methoxide in situ using catalytic amounts of alcohol to start the reaction off.

If you're running the reaction with methanol then yes - but you didn't mention that until now so I hadn't assumed it was present.

You will essentially get the same thing if you use ^tBuOK in methanol:

^tBuOK + MeOH MeOK + ^tBuOH

If methanol is the solvent you will have mostly methoxide in solution.

[w4rlock]

Decided to try out NaH in Benzene since I found a paper which details the method for Dieckmann cyclisation of a molecule similar to mine. The procedure I'm following states to prepare an oil-free suspension of sodium hydride in Benzene, dissolve the diester and to initiate by adding methanol. I then go on to reflux under nitrogen for 4-6 hours. The paper says to quench with acetic acid and then water.

I keep getting 5 peaks by GC (unfortunately don't have an MS so I can't get more info then that) and none of these are the starting compound. Separating these products is proving to be very difficult since they seem to elute from the column together.

My strongest feeling is that the highly basic conditions (or possibly the acidic conditions on adding acetic acid) are hydrolysing my ester bonds. Another possibility is that there are some intermediates formed (ex. enolates) which are not being acidified/hydrogenated properly during the work up http://www.orgmech.co.uk/mechbyyear.php?Year=Second&Num=120)

My starting compound is a linear tertiary amine with two ester groups attached RN((C₂H₄)COOMe)₂

Any suggestions?

[w4rlock]

OK I think I was over-analysing the situation. I just realised for a SM like mine (with a hexyl-aine chain) the yield that was obtained in the paper was just 43%. So it could be that my reaction went  on as expected. My only issue now is to carry out a successful separation and figure out which one of the products is the cyclic ketone.

From first principles I can assume it will show up with a lower Rf relative to the SM (since it is less polar) on TLC and probably a longer RT by GC. Unfortunately without GCMS I will have to do some guess work and then check out the fractions using IR and NMR.

Still, if you have any insight on to this reaction and my issues they would be appreciated.

Has anyone used TMG or other amidine-type bases to catalyse such reactions?