Org Prep Daily

April 1, 2013

It curdles if you don’t stir it

Filed under: mechanisms — milkshake @ 5:07 pm


Trityl group on sulfur is unstable to LiAlH4 reduction. It falls off as triphenylmethyl anion – that’s where the gorgeous blood-red color is coming from. (Unlike trityl cation, which is canary yellow). I did not know about this S-trityl instability – my Greene book (3rd edition) for example mentions only the electrochemical reduction at highly negative potentials – and so I presume it is not as widely known.

In my hands, sulfur de-tritylation with LAH happens both with primary and secondary thiols protected as trityl thioethers. The rate of trityl loss seems structure-dependent: metal coordinating groups (such as OH, amino) in the vicinity of sulfur accelerate the LAH-promoted de-tritylation to a point that it cannot be avoided even under mild reaction conditions. In such cases all that remains to be done is completing the de-tritylation by overnight reflux and isolating the free-thiol product from the Al basic salt cake after the usual Fieser workup. The thiol actually ends up stuck within the salt cake as a thiolate; the filtrates contain only triphenylmethane.

Trityl-S group seems to be reasonably stable to borane-THF at room temperature.

Update: Trityl-O reductive cleavage to trityl anion is also facile, it takes place at room temperature with potassium naphthalenide solution in THF



  1. They’re also unstable to H2/Pd (found that out the hard way without looking it up first)

    Comment by NUchemist — April 3, 2013 @ 10:45 am

    • you can also inadvertently hydrogenate off Fmoc and other beta-phenethyl-based groups if you have a good Pearlman and push hard enough…

      Comment by milkshake — April 3, 2013 @ 2:59 pm

  2. How do you liberate your thiol from the cake?

    Comment by PotStirrer — April 4, 2013 @ 3:40 am

    • yes, it is unpleasant. In my specific case, I dissolve everything in HCl. My piece becomes part of a polymer. I remove all the inorganics by ultrafiltration. (It would have been so much better to use a different protecting group, in hindsight. Anyway, I hope I am done with this particular chemistry)

      Comment by milkshake — April 4, 2013 @ 9:16 am

  3. And there is the tie-in with your previous post.

    Comment by PotStirrer — April 4, 2013 @ 9:34 am

  4. Dear Milkshake-Org prep community, i would like to share some synthetic trouble, if you dont mind. i’m running some UGI 4CR on solid phase, best condition so far, formation of the imine with neat CH2O 10′ (35% wt sol in H2O) then wash away, and reaction with isonitrile and acid; so let’s say is a variation of the UGI, because i have found that preforming the imine gave me less trouble.
    Reaction behaves well, but there’s an interesting side product i can’t get rid of. it’s the double addition to the formaldehyde and isonitrile attack on the solid supported amine, instead of amide formation via rearrangment of the carboxylic acid intermediate. I believe that there’s some formaldehyde in the mix that comes from the imine equilibrium, do you guys have any idea on formaldehyde scavengers?I thought about ethylene diamine but this clearly can create trouble to my solid supported imine.

    Many thanks as always

    Comment by madforit — April 11, 2013 @ 6:37 am

    • I think this sideproduct will be hard to avoid because amine is on solid phase and so formaldehyde is there as the reagent used on excess. Formaldehyde imines are kinetically unstable – they equilibrate fast to aminals (mono and bis hydroxymethyl amines) and gem-diamines. And formaldehyde has an unpleasant tendency to stick to amides as aminal also. You could possibly try to add a formaldehyde scavenger like L-cysteine methylester hydrochloride or cysteinamine.HCl into your reaction but chances are that it will interfere and scavenge so well that it would actually decompose your imine.

      I think there is a better alternative: Since you are making the imine in a separate step anyway, it would be worth to stabilize the imine and wash away all unreacted formaldehyde. Aminoacids on solid phase with free NH2 (as free base) provide at room temperature aza-Diels-Alder adduct with CH2O and cyclopentadiene – azanorbornene – and the reaction is very fast and clean – 30 min at room temp – if you use a freshly distilled cyclopentadiene and neat trifluoroethanol as a solvent. This azanorbornene addduct behaves as a protected/stabilized version of formaldehyde imine, it has been used in the past for reductive amination to get mono N-methyl product (free of N-dimethyl).
      The azanorbornene is stable at room temperature as a free base but just a tiny amount of acid or heating reverses it back to imine. (Even silica is acidic enough to do this retro Diels-Alder so if you want to observe the adduct on TLC instead of the starting amine you need to deactivate the silica by pre-eluting it with chloroform-methanol-ammonia 100:10:1 mix).
      So, you could try to make this stabilised formaldimine surrogate, wash away the formaldehyde excess and even possibly store the intermediate on solid phase (in the fridge). I would use the azanorbornene adduct directly for Ugi, and see what happens. (Cyclopentadiene: there is a procedure in OrgSyn. Make sure to crack-distill dicyclopentadiene through a short column and then re-distill the produced cyclopentadiene under Ar and store it in a freezer, to keep it from dimerizing)

      Comment by milkshake — April 11, 2013 @ 12:45 pm

      • Thanks for sharing your endless knowledge.
        So, on the chemistry side: i do wash away the CH2O after 10 minutes (followed by DMF DCM wash), and checking by LC-MS after TFA cleavage, that imine is hard as a rock, i was impressed, and the formation is extremely fast (less impressive, though).
        The idea of azadiels alder the imine is brilliant, but for my purposes that’s not gonna be duable (i’m doing this in a peptide synthesizer for automated oligos synthesis) and that makes it bad, because the impurity grows at every cycle, and even a 5-mer oligo looks pretty crappy.
        I just set up a synthesis using Triptophane methyl ester HCl to try to trap the Formaldehyde in equilibrium, let’s see if that helps.
        WHat do you think about using a carbohydrazide?seems like a useful reagent for some idustrial purposes to scavenge excess of CH2O.

        Once again, thanks for the help.

        Comment by madforit — April 11, 2013 @ 2:16 pm

        • I think carbohydrazide will be an effective scavenger but may interfere with Ugi. Also I do not believe that your imine is “hard as a rock” and survives TFA cleavage unmolested. What you see on LC/MS is mass of something that is isomeric with your imine (for example some kind of Mannich chemistry imine cyclization product). I think with all this massive formaldehyde excess around, you are forming (HOCH2)2N- and -CON(CH2OH)- species that you do not see on LC/MS because they fall apart in TFA but they act as semi-labile source of extra formaldehyde thats messing up with your Ugi. I don’t see how you can fix it with adding a formaldehyde scavenger.

          Also, the azanorbornene formation should be perfectly amendable to automation (room temperature, mix everything together, wait, wash, then follow with Ugi) as long as you make a fresh solution of cyclopentadiene and formaldehyde in trifluoroethanol every morning.

          But you should not be using the automation yet for a reaction condition development – especially for reactions that are done at room temperature. Do it by hand, something like 5 or 8 vessels at one time as maximum number [A disposable polyethylene syringe (with a polypropylene plunger without a rubber tip) equipped with a internal plastic filter (Bel-Art Scienceware Fritware Porous Polyethylene Sheets cut to size with a cork-borer) is a cheap disposable solid phase minireactor].
          Remember, first you have to make the reaction work once, on few simple substrates, before you try to repeat it in tandem multistep sequence, in a large format, by using automation. The only thing automation does for you is faster pipetting and washing in multiwell setup but the trade off is delays, uneven quantities dispensed due to cloging and vapor aspiration, a moisture exposure, the leaks and well cross-contamination, and overall longer setup times. So if your chemistry is not optimized before you automate it you will only get crap squared.

          Comment by milkshake — April 11, 2013 @ 4:14 pm

  5. Sure, it does make sense. Chemistry was “optimized” by hand and it works ok, but is not perfect; as you know, having a peptide-coupling conversion/purity using other chemistries starting from scratch is tough.
    Which acid would you use to retro diels alder the azanorbornene?
    many thanks, always helpful

    Comment by madforit — April 11, 2013 @ 4:55 pm

    • I would use the same acid that you are using for Ugi – maybe you can add little bit more but it should work even with the usual 3 equivalents or whatever you are using normally. What solid phase resin support are you using – is it polystyrene or is it TentaGel -like? If you encounter a swelling problem in pure trifluoroethanol with polystyrene, perhaps you can use something like 1:1 trifluoroethanol + dichloromethane instead. (Those endless optimizations and the impossibility to follow directly whats happening on the resin was always frustrating for me)

      Comment by milkshake — April 11, 2013 @ 5:13 pm

  6. it’s a rink linker PEG resin, it behaves well in everything except Et2O. For deprotection of Mtt group i use a mix of HFIP-DCE and the swelling is really good, so i think trifluoroethanol-DCM should be good as well.
    Yes, i normally (project is quite new though) use 4 eq of acid, and 4 of isonitrile

    Comment by madforit — April 11, 2013 @ 5:19 pm

  7. FWIV, I don’t have the ref handy where I picked it up, but I understood that formaldehyde imines are not thermodynamically stable compared to the corresponding hemiaminals or other geminal methylene forms, just as it is with formaldehyde itself, so that one cannot expect to see a methylene imine at all.

    Comment by JH — April 12, 2013 @ 3:05 pm

  8. Hi Milkshake, random chemistry question (again) if you don’t mind. How stable are t-butyl groups to acid? I need to cleave off an ester in the presence of a t-butyl ether, but basic conditions thus far have all failed spectacularly (not too surprising since I have 2 base sensitive groups). I was hoping that maybe something like “anhydrous HCl” such as HCl in dioxane would do the trick, but wanted to check to make sure I don’t end up wasting time and precious material. Thanks!

    Comment by Student — April 21, 2013 @ 1:55 am

    • it would be difficult to cleave tBu ester selectively in the presence of tBu ether – in fact the opposite might be possible because my colleague has been able to cleave selectively benzyl aryl ether in the presence of benzyl ester, with TFA plus a scavenger although it was far from perfect and it required a careful optimization to be of practical use.

      I would try to throw LiOH and H2O2 (the final concentration of H2O2 should be about 10%) on your tBu ester in THF-water; the combo is more powerful and milder than just hydroxide alone. The hydrolysis product will be a mixture of acid and peroxyacid so you may want to treat your organic extract after the workup with acid-stable reducing agent like NaHSO3, to decompose the peroxyacid. Something similar to this (last step):

      Comment by milkshake — April 21, 2013 @ 3:48 am

      • Sorry should have clarified, it’s an ethyl ester, not a t-butyl ester.

        Comment by Student — April 21, 2013 @ 4:01 am

        • Thats even better. Try the LiOH + H2O2 (0 to -5 C), use exactly one equivalent of LiOH.H2O and see what it does.

          Comment by milkshake — April 21, 2013 @ 4:11 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 )

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: