Org Prep Daily

January 5, 2007

Trityl azide is lazy

Filed under: procedures — milkshake @ 11:55 am


Before the preparation, sodium azide 10.98g (168.9mmol) was powdered with a spatula and dried in a 0.5L round flask on 80C oil bath on highvac (0.25 Torr) for 90 min. The cooled flask was flushed with Ar, trityl chloride 31.19g (111.0mmol) and anhydrous acetonitrile (about 100mL) was added and the mixture was vigorously stirred at 80C for 4 hours. The reaction mixture was cooled to ambient temperature, the salts were removed by filtration and washed thoroughly with toluene. The combined filtrates were evaporated, the residue was dissolved in hexane 100mL and filtered through a pad of silica 50g. The silica pad was washed with additional hexane 100mL and then with 200mL of 1:1 hexane-dichloromethane mixture. The combined filtrates were evaporated, the residue was dissolved in pentane 100mL. The solution was seeded with a sample of crystalline material*, allowed to crystallize at RT for 1 hour and then in a fridge overnight (+4C, 8 hours). The supernatants were quickly decanted, the crystalline mass in the flask (27.051g) was rinsed with chilled pentane (2x10mL) and then dried on highvac. Evaporating the supernatants and re-crystallizing the residue from pentane (20mL, +4C) provided additional 3.701g of a pure product. The combined yield was 30.752g (97%) of large shiny pale yellow crystals.

1H(d6-DMSO, 400MHz): 7.408(m, 9H), 7.255(m, 6H); 13C(d6-DMSO, 100MHz): 142.48(3C), 128.44(6C), 127.88(9C), 76.98(1C)

Hexane works for crystallization as well. *A simple way to obtain seeds for crystallisation is to dry a small sample of the oily product on highvac, then let it sit in an open flask for a day or two. 

The reaction can also be run in toluene if 5mol% of Bu4NBr is added as solid phase transfer and the azide is powdered thoroughly. Toluene may be a better solvent for process scale-up (the reaction mix can be directly filtered through silica pad, the solvents are not mixed, etc)  but the reaction takes longer to complete and evaporating large volumes of toluene is not fun. Running the reaction in acetonitrile is preferable on lab scale.

Mtr-N3 can be made in the same way from p-methoxytrityl chloride, in nearly quantitative yield as a pale yellow oil.

Trityl azide is easy to work with – it forms gorgeous crystals, is not impact-sensitive and its decomposition occurs well above boiling point of toluene (110C). But it is rather unreactive azide – I tried to use it for making trityl triazoles and the Cu(I)-catalyzed cyclizations did not work on monoaryl alkynes.


  1. Nice to see a new entry. I have been a fan of this blog, and it’s still in my rotation. Cheers!

    Comment by Russ — January 7, 2007 @ 2:32 am

  2. The trouble is that when writing about stuff related to current work one has to be very conservative – it is much easier to select something from a project that has been dead for 3 years and nobody cares about anymore (but I run out of those already).

    Comment by milkshake — January 7, 2007 @ 5:15 am

  3. I frequently use sodium azide with ethyl bromo acetate in an attempt to substitute the bromine with the azide. I generally run the reaction in dry DMF at about 50C for 5-6 hours with a slight excess of azide to ethyl bromo acetate (1.1:1). I was curious why you dried the azide on high-vac for so long.. Is it really that hydroscopic? Enough so to impact a substitution reaction?

    Comment by tom — January 12, 2007 @ 2:06 am

  4. It’s just a precaution. I have done this preparation couple times without any azide drying – and it worked fine. But 90 min is not such a long time – I mean, it is not even enough for a quick lunch here in Florida… I think I would worry more about the quality of anhydrous acetonitrile.

    I wanted to supress trityl acolhol impurity in my crude product when I was doing this. (Trityl alcohol has tendency to co-crystallize with the azide product to some degree). But since there is always some trityl alcohol in Aldrich trityl chloride, and since the silica filtration removes most of it, it is not that crucial as you have guessed.

    Comment by milkshake — January 12, 2007 @ 3:03 am

  5. How to prepare sodium azide ?

    Comment by Nirag — October 17, 2010 @ 12:05 am

    • Diazotation of hydrazine with NaNO2 and HCl produces hydrazoic acid, which is the way one would make it in his garage. Also, NaN3 is the gas-producing charge in airbags. But organic chemist or a biologist would buy NaN3 from a chemical supply company. Dear friend, I suggest that you do not mess with azides, and at any rate you should not ask any more questions here because I do not wish to teach kids how to make explosives at home. Thanx

      Comment by milkshake — October 17, 2010 @ 12:37 pm


    Comment by PRATIK — August 16, 2020 @ 11:00 am

    • dear Pratik, please do not use ALL CAPS when commenting, people will take you more seriously that way (and spam filters will not trash-bin your comment).
      Trityl chloride solvolyses readily, by SN1 mechanism (it spontaneously forms trityl cation in solution) and the only way to stop this is to convert trityl chloride to something else, by derivatization. You can take some anhydrous non-hindered low boiling primary amine – for example n-propyl amine or n-butyl amine, add excess of it to your trityl chloride until it dissolves, then blow off the excess of amine by stream of nitrogen and analyze the tritylated amine vs trityl alcohol ratio.

      Also, HPLC may not be the best method. Try GC analysis (dissolve your sample in some hydrocarbon solvent like toluene). Maybe it will work without need for derivatization

      Comment by milkshake — August 19, 2020 @ 7:41 am

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 )

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: