Skip to main content

Carving Out Nitrogens: Pick Your Conditions

I wanted to do a follow-up to this post, about a reaction that removes nitrogen atoms from heterocyclic rings. Some readers took issue with the paper in the comments, pointing out that a mechanistically similar transformation had been published in 2017. The group that did the earlier work has a new paper on the subject out in Angewandte Chemie, so I thought it might be a good time to take a look.

Both of these papers eventually go through an isodiazene intermediate; that’s the species that turns into nitrogen gas and takes the existing ring nitrogen with it. But they get there in different ways. The paper I blogged on earlier (Mark Levin’s group at Chicago) uses a N-benzyloxy N-pivaloyloxy amide reagent, while the other group (Hongjian Lu at Nanjing) uses sulfuryl azide reagents to get there. The Levin paper references the 2017 work in passing, noting how they were “circumventing the laborious and undesirable features of existing protocols“. As mentioned, some readers took exception to what they saw as shade-throwing.

The problem is, any reaction protocol that calls for sulfuryl diazide is something I’m going to have to think about very hard before trying. Here’s a 2011 paper from Helge Willner’s group in Wuppertal that describes it in detail, but from the structure you probably already know what you’re looking at. Chemist Nick Chiappini noticed the new Angew. Chem. paper on Twitter, and directed people to this 2011 work as well. Sulfuryl diazide is described as having “exceedingly explosive, unpredictable properties”, and I can believe it. It’s not a compound that appears in the literature very often, but it always comes with phrases like this attached to it. Now, I don’t know what the pivaloyloxy reagent from Levin’s group is like, either (you don’t see many N,N-dioxy amides), but at least so far I have no reason to think that it’s as hazardous as sulfuryl azide.

And as Chiappini mentions in another tweet in his thread, another problem is that the azide work in the new paper is being done in dichloromethane. That’s not a good combination: if you have any azide anion around, it can react with dichloromethane to form diazidomethane, and you truly do not want any amount of that whatsoever. Now, the Nanjing authors describe using and storing the sulfuryldiazide in solution and say that they experienced no explosions, and I’m glad to hear it. But they do strongly recommend safety precautions and I wish what they’d said something about the azide/dichloromethane combination as well.

So overall, even though this is an interesting reaction and does something that’s very hard to do otherwise, I don’t really want to do it with with the sulfuryl azide protocol. There are quite a few reactions in this category, though!

23 comments on “Carving Out Nitrogens: Pick Your Conditions”

  1. das_f says:

    Apologies for being THAT guy, but the correct journal abbreviation is Angew. Chem.

    1. Derek Lowe says:

      True, true. . .

  2. Hap says:

    The problem is that, unless there’s not another way, there’s no reason to take really high risks to do a reaction – there’s no reaction that worth your life. Even if there is no other way to do the reaction, you have to ask yourself if the ends are worth the risks involved. Having another way to do it, even if expensive, raises the bar very high on a hazardous method.

    You can recognize both the novelty of a reaction (“this is a neat idea for synthesis that no one’s done before”) and the development of a better method (“Someone had this idea before but the method is really hazardous so no one wanted to use it. Here’s a less hazardous way to do it.”). For me, making a method that people can actually use (without high risk of death) is a big deal (depending on the reaction) – the ability to use the method is more important than the concept for me. You have to be honest and note the concept, but a concept that no one will use doesn’t help, and something that makes the concept useful is important.

    Why did they use CH2Cl2? Azides and CH2Cl2 is a known hazard. Convenient, slightly polar solvents (that you can pull off easily) aren’t common enough (chloroform and DCE are out for this reaction, and PhCF3 is much higher-boiling and expensive) but convenience ought to lose out to my desire to keep my limbs.

    1. Some idiot says:

      Just a quick one… as a process chemist, if someone shows me those sorts of reagents/intermediates, then my first reaction tends to be in the direction of “when what freezes over???”

      Having said that, if it really does solve a lot of other problems, then there is always the possibility of doing it all in flow… one of the bigger companies in (I think) Holland do something like that on a pretty serious overall scale, by using a mass of smaller scale flow reactors (meaning that the actual volume of anything which has “ouch” written on it is on the scale of a few ml at any given time).

      But yes. You would have to convince me that there was a damn good reason for doing it that way first…!

    2. anon the II says:

      Just wanted to say, “Amen!”.

      Give me a reaction I can use. Even better if you can get Aldrich (or somebody) to sell the reagent.

      It was much more than a cheap extension. Congrats on the Nature Pub.

  3. Hap says:

    Also, any reagent that starts with a reference from Helge Willner’s group is going to get the side-eye (in the same drawer as reactions citing Klapotke’s group).

    1. Derek Lowe says:

      It’s a sign, for sure!

  4. Ken says:

    This suggests the amusing possibility of someday seeing a new reaction paper, where Derek’s review consists entirely of links to “Things I Won’t Work With” articles.

  5. Jim says:

    Thanks for the follow up, Derek! If not for this post, I would have never seen the prior art.

    With so many “new” reactions being repackaged versions of older ones (in lower-tier journals), I’m curious where organic chemistry is going? There’s a wave of new synthetic chemists focusing on rebranding old reactions instead of moving the field forward. Are there no future directions? Why is the field so adept at attracting and training master salespeople?

    1. Rh(baby) says:

      It’s the special sauce™

    2. AVS-600 says:

      Not sure that a switch from sulfuryl diazide to an actually usable reagent counts as “repackaging”. Kumada and Corriu established the existence of palladium-catalyzed cross-coupling reactions well prior to the discovery of the Suzuki coupling, but people generally consider the latter to be a legitimate contribution to the literature on its own merits.

  6. Gold Titanium Samarium says:

    I think the issue here is not the reagent itself but where the respective papers were published, which ties in with the “rebranding” of old reactions mentioned in an above comment. We can agree that the reagent used by the Levin group is the safer option (it’s not explosive but I’d be worried about it being a mutagenic as per Org. Biomol. Chem. 2003, 1, 2238). However, I disagree that improving the safety of a reaction (and maybe shaving off a step) warrants a Nature article, especially if the reaction is not widely used. In the current case the reaction is not even widely known as evidenced by a lot of people not being aware of the 2017 JOC article – I wasn’t either until I went over the references in the Levin paper. The reviewers of said paper presumably did the same amount of reading so I’m wondering why they accepted the paper given that it is objectively not as ground-breaking as the text and figures in said paper make it out to be. At the very least, the reviewers should have demanded a better visual and textual representation of the prior art – a great example of which can be seen in Scheme 1 of the Angew. Chem. paper by Lu.

    So the question here is – how much new chemistry do you need to do to get an article in Nature?

    1. Anon says:

      Not much. You just need good PR and marketing skills.

      1. Gold Titanium Samarium says:

        Unfortunately, I agree – but this shouldn’t be the case. Science was supposed to be about objectivity…

        1. Ogamol says:

          The ideal of science is corrupted by pragmatism. Not “who benefits”, per se, but “how does X benefit” where X is {funder, publisher, writer, researcher, likely beneficiary, (etc)}.

          (I’d use a phrase like “preaching to the crowd”, but I remember being unaware of external influences and wondering why science wasn’t “knowledge for knowledge’s sake”.)

    2. Jim says:

      The problem has less to do with answering “is this a nature/science paper?” and more about what this work contributes. Small advances can be worthy of top tier journals. Small advances can squeak into top tier journals too. That’s not a significant problem, in my opinion. Giving inadequate (an understatement) credit and hyping up a rebranded version of the same reactivity / general concept as your own is deeply concerning.

      There’s both damage to the researchers that originally published the work and also to the students that were rewarded for their success repackaging the same work. With professors across organic chemistry preaching this style of research, I am unnerved thinking about the training our next generation of scientists are receiving.

    3. Hap says:

      Mutagenicity can be mitigated, though, while explosiveness is a lot harder to mitigate (once you break containment, lots of mitigations are done, and explosiveness is the strong desire to break containment). Lots of things are toxic when propelled deep inside your body that aren’t toxic otherwise.

      YMMV, but having a method that people can use easily is more important than having an interesting idea that no one will use. The best analogy might be to azide-alkyne cycloadditions (and maybe fluorosulfates) – they weren’t invented by Meldal or Sharpless, but their utility was realized by them. Before them, they weren’t used, and now, they’re almost the kudzu of chemistry. That seems to indicate that the choice of publication was not misguided. Time and chemists will tell if the method is hype or useful.

  7. anonymous says:

    Didn’t we have, a while back, an article or quote from George Whitesides to the effect that if an invention in chemistry is really groundbreaking investors will take out their checkbooks and pay for it? Have there been any new organic reactions that people really invested in, in the way that was done for advances in biology?

    Ziegler-Natta for sure, but anything else?

    It would seem that in an “applied” science like organic synthesis, this would be a better metric than a publication in Science/Nature.

  8. angrygecko says:

    Azides and diazides and explosions Oh My! Glad to see you’re back.

  9. a says:

    Safety stuff notwithstanding…..

    Originally I thought the Nature/Levin paper’s difference might have been that they did a huge scope test (a la Baran lab style) whereas the original JOC/Lu Paper only did a dry mechanistic analysis, which might have indicated they didn’t understand the potential for the work. But looking at Lu’s JOC, there is the *beginning* of that kind of scope test, but they didn’t do anywhere near the breadth of testing that Levin did.

    After the Levin paper came out, did Lu then go “oh shit, we should test all of those conditions too to retcon the importance of our work”? Or was Lu ripped off?

    My guess is it’s somewhere in the middle of these two trajectories; new prof at U.Chicago, a product of elite groups (Toste/Jacobsen) was taught how to Ring Big Bells In Synthesis These Days.

    Lu at Nanjing might not have had that training/perspective.

    1. anon says:

      Didn’t Jacobsen also take a similar approach with his recent work on hypervalent iodonium? He took reactions and catalysts that were in the literature, made them more useful and studied them thoroughly. There is a ton of value in his papers even though it’s not “his” reaction or catalyst. I feel like some people get too hung up on who did it first.

  10. David Edwards says:

    I suspect sulfuryl azide will make an appearance in its own TIWWW feature pretty soon ..

  11. Olandese Volante says:

    Interesting note: this blog apparently spawned a new wikipedia article, on sulfuryl diazide 🙂
    (created July 11th)

Leave a Reply

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

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.