Org Prep Daily

September 4, 2007

Soliciting advice on diazomethane

Filed under: questions — milkshake @ 9:49 pm

diazo.jpg Credit: Jiri Sliva

I have been doing Arndt-Eistert homologation (example: OrgSyn 79, p.154, 2002) so I needed pure anhydrous diazomethane solution. I made diazomethane from nitrosomethyl urea (NMU) and to avoid the distillation, I transfered diazomethane from my KOH-dried diazomethane solution in toluene by passing a stream of Ar through it (at R.T. for 30 min) and condensing the liberated dry Ar-diluted diazomethane gas directly into the reaction mixture cooled to -78C. (I heard of this alternative of diazomethane solution distillation from Rapoport group – and it worked, though I cannot recomend doing this kind of gas transfer on large scale because diazomethane is so nasty and toxic.)

I have only a limited experimental experience with diazomethane. There is plenty literature on the subject. But I would like to ask you about your personal perspective: What is you favorite method of making high-purity dry diazomethane solutions (that can be used for sensitive applications such as Arndt-Eistert)? 

One reason why I am asking this is that it apears to me that all the commonly used precursors of diazomethane have some problems associated with their use: Diazald needs a presence of alcohol (preferably a high-boiling one) or a phase-transfer catalyst to work with aqueous hydroxide and the produced diazomethane solution should be distilled to remove the sideproducts. MNNG is highly toxic and difficult to buy nowadays. NMU is expensive, explosive, very carcinogenic and the NMU-produced diazomethane solution contains trace of methylamine. So I would be delighted to learn what you have tried and liked.

The other reason is that I have one rather simple-minded idea – that I would like to try, to see if it works –  for a method of preparing contaminant-free anhydrous diazomethane solution without the risky distillation step. I would like to know if you think it is a worthwhile use of lab time, if you think a new method is even needed – a method that would be convenient and work with benign commercial reagents. 


  1. Aoyama T, Shioiri T. (1980) Trimethylsilyldiazomethane, a new, stable, and safe reagent for the classical Arndt-Eisert synthesis.
    Tetrahedron Letters, 21, 4461-4462.

    Comment by TMS — September 4, 2007 @ 10:01 pm

  2. I agree with TMS. Try TMS diazomethane. I have used it many times without incident. It is a safe alternative to diazomethane and is commercially available as a toluene solution.

    Comment by RK — September 4, 2007 @ 10:16 pm

  3. Ah, I lived under the impression that TMS-diazomethane is no good for Arndt-Eistert – and I was wrong all these years. Thanks a lot for the reference, as this could really simplify the things.

    Comment by milkshake — September 4, 2007 @ 10:18 pm

  4. There is a company in Sacramento, CA (I think) that does loads of diazo chemistry… contract your project out 🙂

    Comment by Milo — September 5, 2007 @ 7:58 pm

  5. TMS-diazomthane works like a charm for the A-E reaction. I flashed the diazo-ketones, and the TMS nicely falls off on silica. Also, I’d recommend using silver benzoate- much better than any other Ag source in my experience.

    Comment by Jose — September 6, 2007 @ 1:02 am

  6. Is it possible to do Arndt-Eister (with Ag benzoate) in the presence of aniline, to obtain anilide directly?

    Comment by milkshake — September 6, 2007 @ 1:47 am

  7. this would suggest yes, but no details:

    Comment by kiwi — September 6, 2007 @ 6:12 am

  8. on a side note… but not completely off-topic…I believe that TMS-diazomethane can also be used with LDA to convert aldehydes to alkynes, thus bypassing nasty triphenylphosphine in the corey-fuchs reaction

    Comment by tom — September 6, 2007 @ 7:44 pm

  9. oops forgot to ask if anyone had a reference for it as I can’t seem to find it

    Comment by tom — September 6, 2007 @ 7:44 pm

  10. No, the reagent for making alkynes is MeCOCN2PO(OMe)2 and is not commercial but is one step from MeCOCH2PO(OMe)2

    Comment by milkshake — September 7, 2007 @ 4:48 am

  11. bestman-ohira reagent works as well.. but I believe TMS-diazomethane when treated with LDA can homologate an aldehyde to an alkyne

    Comment by tom — September 7, 2007 @ 12:37 pm

  12. 4-(4-Ethynyl-2-methoxyphenoxy)benzonitrile (14). A solution of lithium diisopropyl amide (2.0 M in heptane, diethylbenzene and THF, 0.374 mL, 0.748 mmol) in dry THF (4.97 mL) was treated with TMSCHN2 (0.374 mL, 0.748 mmol). This mixture was stirred for 30 min at -78 °C, then a solution of 8 (0.139 g, 0.621 mmol) in dry THF (1.56 mL) was added. This mixture was stirred for 1 h at -78 °C and heated to reflux for 3.5 h. The reaction mixture was quenched with water and extracted with Et2O (3x). The combined organic layers were dried (Na2SO4), filtered and concentrated under reduced pressure. Purification by chromatography on SiO2 (hexanes:EtOAc = 10:1) afforded 14 (0.104 g, 76%) as a white solid: 1H NMR (CDCl3) δ 7.59 (d, 2 H, J = 9.5 Hz), 7.17-7.14 (m, 2 H), 7.02 (d, 1 H, J = 8.6 Hz), 6.95 (d, 2 H, J = 9.5 Hz), 3.80 (s, 3 H), 3.11 (s, 1 H); LCMS (m/z) 250.2.

    Comment by tom — September 7, 2007 @ 12:39 pm

  13. Thank you for the info and the procedure!

    Comment by milkshake — September 7, 2007 @ 12:41 pm

  14. I found two references:

    Ohira, S.; Okai, K.; Moritani, T. J. Chem. Soc., Chem. Commun. 1992, 721.

    J. AM. CHEM. SOC. 2007, 129, 1760-1768

    Comment by tom — September 7, 2007 @ 12:46 pm

  15. About comment 12, this reaction is often called a “Colvin Rearrangement”. In addition to the the RCHO to Alkyne transformation, Lithio TMSdiazomethane also transforms ketones to the disubstituted alkynes, such as taking acetophenone to phenylpropyne. I have used both of the versions Colvin Rearrangements to great advantage over the years. I recommend them both.

    Comment by TMS — September 7, 2007 @ 6:58 pm

  16. ahhh neato.. This seems like a simple one step procedure.. i wonder why people still use corey-fuchs or bestmann ohira? There must be a catch? Can you use BuLi instead of making the LDA or perhaps even LiHMDS?

    Comment by tom — September 8, 2007 @ 1:47 am

  17. Sorry to take the post off track, but is there a decent procedure for the asymmetric epoxidation of terminal olefins? I understand some Shi catalyst variants can get quite high with styrene type systems, but I need something that can do 1,1-disubstituted terminal olefins. Any ideas? I was sure I had seen a very recent paper quoting >80% ee for ‘2-methylprop-1-ene’ systems but I cannot find it.

    Do they even exist?

    Cheers in advance

    Comment by TWYI — September 8, 2007 @ 4:49 am

  18. I don’t know of any – try to ask at Tot Synth

    Comment by milkshake — September 9, 2007 @ 10:20 pm

  19. Number 17: I’m not sure if this will help or not, but check out this paper from Shi a few years back:

    It seems this may be a way to access the type of product you are looking for.

    Comment by Andrew — September 13, 2007 @ 1:44 pm

  20. Thanks for that Andrew

    Comment by TWYI — September 14, 2007 @ 7:33 pm

  21. You should try the reagent we reported on in Synthesis, 2005, p19. The isonitrile reagent we reported is by far the best way to make diazoketones from acid chlorides. the isonitrile is a shelf-stable anhydrous reagent which routinely gave yields of diazoketone better than diazomethane or TMS-diazomethane.

    Comment by Matthew — September 17, 2007 @ 8:30 pm

  22. Ahh – just as I was writing down my Aldrich shopping list I found your comment and so I added formic hydrazide. Thank you for the helpful reference. Also, Matthew – is removal of PPh3O a problem?

    Comment by milkshake — September 17, 2007 @ 9:25 pm

  23. TMS-diazomethane is pretty damn pricey like a 100 bucks for 10 mls of something of a 2M solution

    Comment by tom — September 17, 2007 @ 11:28 pm

  24. actualy Aldrich now sells 2M TMS-diazomethane in ether 100mL for $113 (catalog# 527254-100mL) which is sort of acceptable, for 0.2 mol of the reagent, except for massive scale-ups

    Comment by milkshake — September 18, 2007 @ 1:41 am

  25. The triphenylphosphine oxide was well separated from the diazoketones we looked at either by chromatography or crystallization of the diazoketone. It could be an issue though.

    The intermediate iminoylchlorides are quite stable and can usually be flashed on silica (pretreat with dilute triethylamine) or by crystallization both of which will usually remove triphenylphosphine oxide because the iminoylchlorides are usually quite polar.

    We have used this process safely on kilogram scale. The isonitrile reagent is thermally stable and can be stored for months at room temperature without degradation. Good luck

    Comment by Matthew — September 18, 2007 @ 4:20 pm

  26. If you are only doing a one-off reaction, try TMS diazomethane. But in my experience, TMS diazomethane is not quite as clean as the real deal, and is expensive if you are working on medium to large scale. I used to routinely distill >100mmol quantities of diazomethane from diazald straight into my reaction using a procedure developed by the process group at my old company, who would do this procedure on a .5kg scale. This was 8 years ago, so I can’t give you an exact procedure, but here is more or less what I did…

    1. Get a 200-500ml (depends on scale of course – should be 1/4 full at the end of the addition) polypropylene or polyethylene bottle that seals reasonably well and has a cap with enough space to put 3 pass-through fittings for small-diameter polyethylene tubing (~1/8″ o.d.??? – we scavenged it from the HPLC). The first pass-through will accept a stream of nitrogen – set the end of the tubing near the bottom of the bottle so it bubbles through the diazald solution. The second pass-through is the outlet – set the end of the tubing near the top of the bottle, so no liquid passes into the reaction flask. The third pass-through is an inlet for the KOH solution added to the diazald – set the end of the tubing above where the solution level will be at the end of the addition, but below the outlet.

    2. Charge the flask with diazald and 2-(2-ethoxyethoxy)ethanol (~25mL per 5g of Diazald). Charge a syringe with aq. KOH (see Aldrich manual at for stoichiometry – I would use 1.5 equivalents of diazald to mixed anhydride and my reactions always went to completion)and attach it to the inlet tube such that you have good control over the addition and you can stop it immediately if need be. Stir the diazald solution in a 40C water bath. Slowly open the nitrogen valve attched to the inlet line such that there is ~1-2 bubble per second passing through the diazald solution. It is very important to maintain positive nitrogen flow through the apparatus at all times.

    3. Charge a RBF with a smooth glass joint with your acid chloride/anhydride. I used to make a mixed anhdride with isobutyl chloroformate in THF using NMM as the base at -20C, and would just filter the -20C rxn mixture to remove NMM HCl salt, and make the diazoketone using the crude mixture at 0C so I wouldn’t get any more NMM HCl crystalizing out. Attach a septum with the outlet from the diazald-containing bottle passing through and terminating at a depth near the bottom of the reaction flask, so it will bubble through. The septum will also have an a second piece of tubing as an outlet, which will go to a enlenmeyer set up as a suckback trap, the outlet of which is placed in the bottom of an erlenmeyer containing AcOH to quench any diazomethane that makes it through the reaction vessel. Nitrogen should be slowly bubbling through the diazomethane generator, reaction mixture, and the trap. Place a blast shield in front of everything.

    4. Slowly add the KOH solution to the Diazald solution in the generator. Since the syringe is attached to a length of tubing, you can stand with the blast shield and a lowered sash between you and the nasty business. At the beginning, a few drops will produce a fairly large amount of diazomethane, less so at the end. Add the KOH as fast as you feel comfortable, but be aware that there is a slight lag time between the addition of fresh KOH and a surge in diazomethane boil-off. Be ready to slightly increase the N2 flow if you get suckback from the reaction flask. After the addition is complete, let the N2 blow the residual diazomethane out of the generator for 30-45min. The reaction vessel should be yellow with excess diazomethane, and the generator should be white with K toluenesulfonate. My reactions would be done at this point. If I wanted the bromomethyl ketone, I would quench with 48% aq. HBr, if I wanted diazomethylketone, I would quench with AcOH and work up as normal. Wash all the glassware and tubing with AcOH to quench any lingering diazomethane.

    As for the Arndt-Eistert homologation, I had access to a photochemical reactor, so I would just photolyze the diazomethylketone in methanol. I tried silver oxide as well. It gave comparable yields, but mirrored the flask, so I wasn’t a fan.

    Comment by MikeS — September 19, 2007 @ 10:08 pm

  27. I was able to make a diazoketone following a prep by Baran, Nicolaou and some other people (OrgLett about 2002…sorry, it’s late). The prep involved mixing carboxylic acid + MsCl + NEt3….10 minutes at 0…..quench with diazomethane. Reaction goes through an acyl mesylate. Worked fine but yields were mediocre. Acid chloride failed completely.

    Comment by ... — September 23, 2007 @ 3:25 am

  28. Does anybody know if TMS-diazomethane can work in herbicide derivatization for a GC-ECD analysis?

    Comment by Adolfo — October 11, 2007 @ 1:12 pm

  29. *This question is in regards to your high vacuum manifold post*
    Milkshake, if I attach a high vacuum pump to my manifold (<1 Torr), what is the best way to regulate the vacuum pressure if I need to use the pump at only at 70% of its maximum vacuum? Sometimes I need 100% vacuum capabilities, while at other times I only need 50% of the maximum vacuum. Would you suggest I use a vacuum regulator, or another type of external needle valve? Would you suggest I bleed in air or gas (Ar or N2) to reduce the effect vacuum on the system? Can you make a recommendation for a purchase?

    Comment by DIBAL — March 14, 2017 @ 12:23 pm

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at

%d bloggers like this: