Skip to Content

Safety Warning: Togni’s Reagents

togni'sSome of you may have used the second Togni reagent (shown) as a trifluoromethylating agent. Well, there’s a new paper in Organic Process R&D that brings word that it’s an explosive hazard. A group at Novasep, in Leverkusen, Germany finds that it has a powerful exothermic decomposition, and (in addition) it’s about as combustible as black powder. Sensitivity to impact and friction varied from sample to sample as well, which isn’t what you want to hear, either. Their conclusion is that the reagent is “dangerously explosive and may only be transported by approval of the national competent authority.” The first Togni reagent (with a dimethyl in place of the carbonyl) wasn’t fully evaluated, but it may be just as bad. That one has been available from Aldrich; I imagine that this may change shortly. Too bad – these are useful reagents, but not if you have to suit up to use them safely.

11 comments on “Safety Warning: Togni’s Reagents”

  1. Dave says:

    Not too surprising. That trivalent Iodine can’t be too happy, and that five membered ring should be trying to come apart. It might be interesting to see the decomposition products, just to determine how it wants to rearrange itself. It’s hard for me to believe that it would be quite as explosive as black powder, but it may be close in terms of sensitivity (Black powder is a pain to work with!).

  2. Robert says:

    There’s also a S-trifluoromethyl dibenzothiophenium (tetrafluoroborate?) salt that seems to be used in electrophilic trifluoromethylation papers as well. The dibenzothiophene byproduct might be too heavy to stink strongly, but I don’t know.
    I guess it’s a natural consequence of trying to make an electrophile out of something that really doesn’t want to be one, but it seems like the electrophilic trifluoromethylating reagents involve reagents roughly the size and weight of a barn door.

  3. This is a newbie question: if these have been used widely, how come this is only just coming to light?

  4. Anonymous says:

    #3 because this is how we learn about them. They are widespread enough that people think about using them at large scale and hence perform feasibility studies. None of this matters when working with a few mg of the stuff.

  5. curious says:

    What is Black powder?

  6. Okie Chemist says:

    Curious – “Black Powder” = Gunpowder.

  7. Anon says:

    Black Powder is the original explosive used to make guns and cannon go boom… a mixture of mostly potassium nitrate, charcoal, and sulfur

  8. Mike P says:

    If these have been used widely, how come this is only just coming to light?
    Black powder can be safely handled by pouring, weighing, etc. Chemists don’t handle Togni’s Reagent any more roughly than that.

  9. Patrick Eisenberger says:

    We used to refer to this method of trifluoromethylation as “fluorine chemistry for sissies”, since you did not require harsh, corrosive fluorinating reagents. It seems this is no longer the case. But hey, DMP (and its precursor IBX) makes a pretty nice bang and is used routinely in organic chemistry. These things are called reagents for a reason: they are designed to be reactive.

  10. Real issue? says:

    The tests as described in the Organic Process R&D paper are carried out routinely in any professional process chemistry setup nowadays. Furthermore, the results are not a big surprise. The issue is not as dramatic as to become the subject of a major discussion.
    Obviously, these compounds need some attention when it comes to their safety on scale. But based on a detailed evaluation in a risk assessment they may be used safely with appropriate measures in place.

  11. Anonymous says:

    Wow, I’d been using the hell out of this on a (shelved for now) alkane trifluoromethylation project about a year ago. It never went off on me of course, and from what it sounds like it doesn’t seem so bad, just should come with a warning for being potentially explosive.
    I hope it’s not harder to get if I get back on that…

Comments are closed.