[Nescafe]

Hi,

Can someone guide me through how this deprotection works.


R-P(=O)- (OEt)₂  to R-P(=O)- (OH)₂

via N,O-Bis(trimethylsilyl)acetamide and iodotrimethyl silane.

Does the oxygen attack the silyl groups so you get

R-P(=O)- (OSiMe₃)₂  then acid work up gives R-P(=O)- (OH)₂?

I also do not understand why we need two sources of silane, is it just to speed it up? or do they have different roles.

Nescafe.

[discodermolide]

The oxygen of the OEt groups is silylated by TMSI. I^- then attacks the one of the hydrogens of the terminal carbon atom of the ethyl group (i.e. CH₃) to give HI and ethylene and the bis-OTMS ester. The acetamide mops up the HI formed.

[Nescafe]

The oxygen of the OEt groups is silylated by TMSI. I^- then attacks the one of the hydrogens of the terminal carbon atom of the ethyl group (i.e. CH₃) to give HI and ethylene and the bis-OTMS ester. The acetamide mops up the HI formed.

That makes sense. So the acetamide acts as a base, didnt know that. Why use such a weird base and not DIPEA? Also when you make the bis-OTMS, this is liberated to give the phosphonic acid via work up? Can't the HI do this for us?

[discodermolide]

No you need an aqueous or alcoholic system to hydrolyse the OTMS esters. I usually used methanol with a trace of acetic acid.
HI is far to strong an acid and you will start to get side products if you used aqueous HI.

[Nescafe]

No you need an aqueous or alcoholic system to hydrolyse the OTMS esters. I usually used methanol with a trace of acetic acid.
HI is far to strong an acid and you will start to get side products if you used aqueous HI.

I did this reaction following a procedure in literature which if I remember correctly (did this four months ago) was just BSA/Iodotrimethyl silane in DCM. I didn't even do a work up. I had a mixture of protected and unprotected material. Perhaps water in the solvent or during work up does the hydrolysis? Wouldn't methanol just give you a methyl ester?

[discodermolide]

No you can't make phosphonate methyl esters this way.
Actually the best way is to hydrolyse then with lithium hydroxide to give the di-lithium salt. Then you acidify carefully with phosphoric acid. Lithium phosphate is totally insoluble and precipitates out quantitatively, leaving the bis phosphonic acid in solution. Evaporate solvent and you have it.

[Nescafe]

No you can't make phosphonate methyl esters this way.
Actually the best way is to hydrolyse then with lithium hydroxide to give the di-lithium salt. Then you acidify carefully with phosphoric acid. Lithium phosphate is totally insoluble and precipitates out quantitatively, leaving the bis phosphonic acid in solution. Evaporate solvent and you have it.

Alright thank you, any idea what would be the pKa of this acetamide used in this reaction would be?

[discodermolide]

No, sorry.

[Babcock_Hall]

Was the TMSI in catalytic or stoichiometric amounts?  Are you quite sure that the "TMSI" you were using was iodotrimethylsilane, and not trimethylsilyl imidazole?  Unfortunately, both are sometimes abbreviated as TMSI.  Iodotrimethylsilane and its equivalents are often used to remove alkyl groups in phosphate and phosphonate chemistry; therefore, it probably was iodotrimethylsilane.

The technical bulletin on BSA and TMCS says:  "TMCS increases the reactivity of BSA (or other silylation reagents)."  Apparently it is used at the 1-20% level.  http://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Aldrich/General_Information/bsa_tmcs.pdf

There is also a commercial preparation which is BSA, TMCS, and trimethylsilyl imidazole:  http://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Aldrich/General_Information/bsa_tmcs_tmsi.pdf  Its existence is why it occurred to me to ask about it.  However, I get the impression that this reagent is mostly used to silylate hydroxyl groups, even hindered ones.

Here is a little bit of background information on BSA:  "However, displacement of protons from other amides is bound to be incomplete when monosilylamides are used as silyl donors; it is here where the thermodynamically greater silylating strength of disilylamides is of special significance."
http://pubs.acs.org/doi/pdf/10.1021/ja00966a038 J. F. Klebe, H. Finkbeiner, and D. M. White JACS 1966 88 3290.

[Nescafe]

The oxygen of the OEt groups is silylated by TMSI. I^- then attacks the one of the hydrogens of the terminal carbon atom of the ethyl group (i.e. CH₃) to give HI and ethylene and the bis-OTMS ester. The acetamide mops up the HI formed.

How does the I^- attack one of the hydrogens? Is it that strong a nucleophile to carry out this attack?

Nescafe.

[discodermolide]

In this case, yes. The oxygen is protonated so the driving force is the formation of ethylene (gas) and the removal of the positive charge on the oxygen.

[Nescafe]

In this case, yes. The oxygen is protonated so the driving force is the formation of ethylene (gas) and the removal of the positive charge on the oxygen.

Thank you. I was also wondering why we need to first react with TMS-I, then hydrolyze, why cant we just hydrolyze the esters the usual way of just LiOH, why turn them to the (-O-Si-Me3) first?

Nescafe.

[discodermolide]

You may have other base labile groups in your molecule. This method was developed to assist in phospholipid chemistry where you have labile acyl esters.
You can use LiOH, but then you must acidify. This is best done with phosphoric acid in a suitable solvent (usually water). You get lithium phosphate, which is totally insoluble in water and filtered off and the phosphonic acid in the solvent layer.

[Nescafe]

You may have other base labile groups in your molecule. This method was developed to assist in phospholipid chemistry where you have labile acyl esters.
You can use LiOH, but then you must acidify. This is best done with phosphoric acid in a suitable solvent (usually water). You get lithium phosphate, which is totally insoluble in water and filtered off and the phosphonic acid in the solvent layer.

So the O-SiMe3 is a better leaving group than O-Et and therefore more labile? If so, silicon is more electropositive than a carbon, wouldnt this mean ^-O-SiMe3 would be more reactive than ^-O-Et?

[orgopete]

I have some skepticism about the mechanism. If ethylene is produced, the reaction can be catalytic in TMSI as the HI released could regenerate TMSI, the phosphoric acid, and ethylene. As Babcock has pointed out, the bissilylated acetamide is a catalyst for the silylation step, just as dimethylaminopyridine is a catalyst for an acylation transfer. I had believed iodide dealkylated the silyl-oxonium phosphate to give a silyl ester and ethyl iodide and thus consuming the iodide. As a base, chloride or imidazole would be better, especially if ethylene were being produced.