Say you need to silylate something, like an alcohol or a ketone. Boy, have I got the recipe!

  1. Dissolve 1.25 eq dry sodium iodide in dry acetonitrile.
  2. Dissolve your starting material in hexanes (pentane or heptane, all good). Don’t worry if everything doesn’t dissolve properly; solubility will increase during the course of the reaction.
  3. Mix the two in a suitable container. I like beakers. And stir bars.
  4. With efficient stirring of the bi-phasic mixture (no inert gas necessary), add 1.25 eq Et3N followed by 1.25 eq TMSCl of decent quality in one squirt.
  5. Stir for 15 minutes at rt.
  6. Separate the layers. That’s right, acetonitrile and petroleum ethers do not mix at all. Phase separation is very rapid.
  7. The top hexane layer has your desired product, salt free and, curiously, base free. Evaporate. Done. Quantitative yield is the general rule.


  • I guess you create TMSI in situ, a notoriously unstable and hence very reactive species. It would silylate Chuck Norris.
  • Perfect separation without water in any of the steps is nice.
  • All the crap goes into the bottom acetonitrile layer, which you can flush down with water.
  • 1 gram/10 mL total solvent volume is a good rule of thumb.
  • This reaction is fast! It is typically complete before your first GCMS or LCMS is done.
  • If there ever was a lit ref for this, I haven’t seen it in many years. Sorry. Anyone?


6 Responses to My favorite reaction Redux

  1. balaji says:

    wow wonderful, i am blown away

  2. Maksim says:

    If everything were soluble in hexanes…

    • drfreddy says:

      That’s the coolest feature with this method! A slurry or even an oiled out starting material works equally well. The silylated product is almost always soluble in hexanes, in my humble experience.

  3. TMSCl says:

    Is it even working with other silylating reagents e.g. TBDMSCl, TBDPSCl?

    Does anyone know?

Leave a Reply

Your email address will not be published. Required fields are marked *