[owk9688]

I'm trying to understand the role of DMAP or Imidazole in the TBS silylation of alcohols (alcohol and TBSCl with imidazole and TEA in DMF to yield the TBS ether.) I was reading a few reviews of the mechanism of this reaction and I can gather that the weaker base (DMAP or even imidazole can be used) acts as an initial catalyst in binding to the silyl halide and generates the reactive silylinium species. The alcohol then attacks that, freed imidazole/DMAP deprotonates the alcohol, and then the auxiliary base like TEA deprotonates the imidazole to reform the catalyst. All that makes sense to me but I have a few dilemas with that understanding:

Why not just use TEA as the catalyst and the auxiliary base? I think I know this one...I'm guessing its because the TEA would attack the TBSCl and form an irreversible adduct. So youd want a base strong enough to activate the TBS group but weak enough to then allow the alcohol to nucleophilically displace it.

But if thats the case...how is that realistic when the imidizole loading is less than 30mol% and the TEA is more than 1 equivalent. There is far more TEA in these reactions than imidazole so wouldnt the TEA, being more basic and I'm assuming more nucleophilic, attack the TBSCl way faster than the imidazole and just shut down the reaction? I'm just very confused as to how the imidazole activates the TBS here rather than the TEA

Thanks for any insights

[Babcock_Hall]

Other people here probably have more practical experience; however, I would like to make a general point.  For a chemical species to be an effective nucleophilic catalyst, it must be both a good nucleophile and a good leaving group.  You seem to be saying something similar to this, but your emphasis is on basicity.

[pgk]

The mechanism is not fully clarified yet.
Even Prof. Corey himself (nobel prize awarded) who invented this reaction, “assumed” that the reaction proceeds via N-tert-butyldimethylsilylimidazole formation (but without putting his hand in the fire).
J. Am. Chem. Soc., (1972), 94(17), 6190–6191
http://pubs.acs.org/doi/abs/10.1021/ja00772a043

[wildfyr]

I always thought of it in this fashion: imidazole or DMAP are somewhat nucleophilic and can attack the silicon center much better than TEA. They are also amine bases, so they can act as proton scavangers as well. I personally run such reactions with 2+ equivalents of imidazole instead of using imidazole and TEA. Both are cheap and easy to handle. I also, frankly, always felt that the amine attacks and displaces the chloride, rather than making a 5 bonded silicon anionic species (silylinium as you say?). I have heard of such anionic species being proposed as a SuFEx intermediate, but the idea is really the same for this.

Corey does comment on this as pgk said:

7: This process seems likely to proceed via N-dimethyl-tert-butylsilylimidazole, the conjugate acid of which can be expected to be a very reactive silylating agent.

I do have one point of contention with Corey's paper though. He states:

...the use of imidazole as catalyst and dimethylformamide as solvent proved to be exceedingly effective...

In my hands the use of DMF vs another polar aprotic solvents such as MeCN or THF to make TBDMS ethers was not important. I only used DMF where solubility of the starting material was an issue.

[owk9688]

Thanks for the responses. I actually was thinking in terms of basicity a bit too much. Now that Babcock_Hall pointed that out, in addition to the other comments I realize that TEA is fairly non nucleophilic (not as much so as DIPEA but still moreso than the rigid DMAP or Imidazole) and so that likely hinders its ability to attack the TBSCl as opposed to the catalystic bases.

Thanks so much!