Org Prep Daily

September 5, 2006


Filed under: procedures — milkshake @ 6:45 am


Three-necked 2L flask with a mechanic stirrer, addition funnel and reflux condenser (equipped with an outlet tube adapter attached to a wide tubing vented into the hood intake – noxious gases).

Also needed: a 24/40 fritted outlet adapter (or filtration finger), a positive-pressure filtration funnel with the upper joint 24/40 or wider (a short chromatographic column with a sintered glass bottom also works).

Before the experiment, a mixture of solid KF 40.1g (Aldrich “spray-dried” grade, 690mmol) and potassium hexafluorophosphate 116.0g (630 mmol) was pre-dried by spinning at 160C/0.3 Torr/3hours using a large Kugelrohr “coffee can” apparatus. (Dry stirring on oil bath on highvac overnight also works)

Oxalyl chloride 70mL (800 mmol) was added in one portion to a solution of tetramethylurea 69.7g (72.6 mL, 600mmol) in toluene 1L with vigorous stirring and with cooling on ambient oil bath. (Two-layer mixture formed, non-exothermic addition). After complete addition, the heating was turned on and the bath temperature was gradually increased to 60C over 20 min. With gas evolution, the bottom layer gradualy solidified. The mixture was stirred at 60C for 2 hours and then cooled to ambient temperature. The addition funnel was replaced with a fritted adapter, the supernatants were decanted under mild positive pressure of nitrogen and the intermediate in the reaction flask was rinsed with toluene (2x300mL).

To this chloroformamidinium chloride in the reaction flask (heavy hygroscopic solid, containing some toluene) under blanket of nitrogen, the pre-dried KF+KPF6 mixture was quickly added in one portion. Dry acetonitrile 1L was added and the reaction mixture was stirred at 55C for 1 hour. The cooled mixture was quickly filtered through a large coarse filtration funnel (the solids were washed with acetonitrile (3×150 mL). The combined filtrates were evaporated from ambient bath and the obtained oily residue was dissolved in dichloroethane 0.7L with heating. The resulting cloudy solution was filtered while hot using a medium-porosity filtration funnel and the filtrates were concentrated to approx 450 mL volume. THF 600mL was added with stirring, the flask was then gently shaken (2 min) to promote crystallization. The mixture was allowed to crystallize at ambient bath for 1 hour. The precipitated product was collected by filtration under positive pressure of nitrogen and washed with additional THF (2x250mL). A stream of nitrogen was passed through the product for 20 min, the drying was completed on highvac (2h). The product was stored at +5C in parafilm-sealed teflon-lined wide-mouth flask.

Y=143.0g (90% th) of a white crystalline product, m.p. 112-113C 1H(DMSO-d6, 300MHz): 3.150 (apparent d)

Note: The solvents used in large-scale prep were not specifically anhydrous, a good grade (HPLC) from freshly-opened 4L bottle works well. Drying the salts is important, so is the filtration under positive pressure (fortunately the intermediate and the TFFH product are coarse heavy solids that filter well.) Dried TFFH is not hygroscopic but it does decompose slowly at RT with HF formation.

Note: Oxalyl chloride has to be added to a cold solution of tetramethyl urea in one portion. Oxalyl chloride plus tetramethyl urea form a non-volatile liquid adduct that controlably decomposes upon gentle heating with gas evolution to produce the chloroformamidinium chloride. The reaction rate (= the gas evolution rate) is controled best by adjusting the bath temperature. (Adding oxalyl chloride at 60C is a bad alternative – unreacted oxalyl chloride is readily carried into the condenser due to the gas evolution so addition rate would need to be very slow).

TFFH is a useful reagent for making acyl fluorides from acids in situ (dichloromethane, 2 eq. of Hunig’s base, RT, 20min), the only side-product is teramethylurea and amine salts. TFFH reacts with amines to form teramethylguanidines so the activation needs to be done in separate step, before the coupling.

TFFH is available from Aldrich. It is ridiculously expensive and it will be, until the Carpino’s patents expire.


  1. I’ve always wanted to find a reason to use an acid fluoride, but I haven’t found one yet. When is it most useful?

    Comment by JPB — March 21, 2013 @ 2:37 pm

    • It is slightly less reactive than acyl chloride (acyl fluoride can even survive a quick aqueous workup with chilled brine) and seems to be less prone to racemization (in case of protected aminoacid) and formation of dark ketene-related impurities when combined with a base like iPr2NEt or NEt3. And in this case with TFFH acyl fluoride could be made under almost neutral conditions in situ. When I was working at a combichem company, TFFH was the coupling agent of last resort, after HATU

      Comment by milkshake — March 21, 2013 @ 3:07 pm

  2. Kind of an off the wall question. But, have you or anyone else seen these pressure joint clamps? My institution has them and as I am shortly leaving and dearly love them would like to procure some of my own. However, the internet seems completely devoid of information on what these things are called or who sells them. Any help would be appreciated.

    Comment by galen — March 28, 2013 @ 4:15 pm

  3. We have tried to repeat this procedure and failed twice in each case at the stage of halogen exchange. First time we repeated it exactly (stirring with KF/KPF6 for 1 h) and obtained a mixture of TCFH and TFFH in 8:1 ratio. After that(a year later) we increase stirring time to 3 hours and obtain a mixture again in a 3:1 ratio. Isn’t any mistake in a procedure?

    Comment by Igor A Ivanov — December 13, 2017 @ 9:57 am

    • Maybe the grade of KF has made a big difference. In this case, we have been only using “spray dried” KF that has very porous particles. Normal crystalline KF may take longer to react because KF solubility in acetonitrile is not that great and if the surface area is low it may take a while to dissolve.

      KF is very hygroscopic and so it is not fun to try to pulverize it. I would suggest to get hold of old kitchen blender (that you won’t need for food anymore), put KF in it and dry mill it by “blending it” as a solid – this is surprisingly effective for making powdered chemicals, I used blender for making powdered KOH.

      Also, when you are vacuum drying the KF + KPF6 mixture, do not overbake it, fused particles could be a problem. If one of the components (KF, or KPF6) is old and already wet and likes to get sticky when heated, maybe pre-dry it separately in glassware drying oven at 150C overnight before pulverizing it and re-drying it, water contained in these salts is a real problem, more significant than water in acetonitrile

      Comment by milkshake — December 13, 2017 @ 5:52 pm

      • I just noticed that Aldrich stopped selling spray-dried KF (” The product 307599 has been discontinued”) . I am pretty sure this is the reason why the procedure failed. The procedure previously always worked very well in my hands – we published it in Org. Prep. Process Int. 29(4) 497-8, (1997) – and I have been making mol-scale batches of it, repeatedly. Enough to get a sneezing allergy and itch on my arms (TFFH is a contact irritant). But the procedure was very reproducible back then, and I know other people used it and liked it on the first try, so I think something must have changed since then: my bet is that it is KF grade. So you need to finely pulverize your KF and then re-dry it

        Comment by milkshake — December 13, 2017 @ 6:07 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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

Blog at

%d bloggers like this: