Org Prep Daily

October 6, 2010


Filed under: procedures — milkshake @ 2:51 pm


Chlorosulfonic acid 115 mL (1.73 mol) in a 1L wide-mouth round flask (with a 45/50 joint, equipped with a gas outlet tube) was cooled on ice bath and solid oxindole 25.60g (192.2 mmol) was gradually spooned in with vigorous stirring and cooling on ice, over a 20 min period. (A corrosive fog evolution!). After the complete addition, the flask was removed from the cooling bath and the mixture was stirred for additional 15 min. The flask was then placed on a 70C oil bath and the mixture was stirred at 70C for 2 hours. The resulting dark reaction mix was cooled on ice, then very cautiously poured in a thin stream onto crushed ice 1.4kg that was pre-chilled in a freezer, in a 3L beaker, with stirring. [Note 1] The quenched mixture was stirred until all ice melted, the precipitated solid was collected by filtration on a very large glass Buchner funnel, washed with 0.05 M HCl, semi-dried by filtration and then carefully dried on highvac. [Note 2] Y=42.23g (95% th) of a tan solid.

The sulfonylchloride from the previous step, 42.23g (182.3 mmol) was suspended in anhydrous dichloromethane (100mL) in a 1L round flask. The mixture was cooled on ice slush bath and a solution of triphenylphosphine 167.5g (638 mmol, 3.5 eq.) in anhydrous dichloromethane (300mL) was dropwise added from an addition funnel under nitrogen with cooling over a 45 min period. After complete addition the flask was removed from the cooling bath and the mixture was stirred at room temperature for 3 hours. The reaction was quenched by water addition, 200mL. The flask with the biphasic mixture was placed on a 50C water bath and the mixture was refluxed under nitrogen for 1 hour, then cooled on ice. The precipitated product was collected by filtration, washed thoroughly with ice-chilled dichloromethane and ice water, then dried by suction and on highvac, to provide 19.58g of a pure product.  The biphasic filtrates were de-oxygenated by argon/vacuum purge (3 times). The mixture was made strongly alkaline by addition of 50% aq. NaOH solution, shaken under Ar and then rapidly separated, the organic phase was re-extrated with water. [Note 3] The aqueous phases were promptly washed with fresh dichloromethane (200mL). The combined aqueous extracts were made acidic by addition of 6M HCl, the mixture was cooled on ice, the precipitated product was collected by filtration, washed with ice-cold water, dried by suction and on higvac, to provide a second crop of the product, 7.15g. The combined yield was 26.73g (88.5% th) of a light tan solid.

1H(d6-DMSO, 400MHz): 10.339(s, 1H), 7.151(s, 1H), 7.108(m, 1H), 6.707(d, 8.0Hz, 1H), 5.109(s, 1H), 3.436(s, 2H)

Note 1: Chlorosulfonic acid is viciously corrosive and has a huge quench exotherm – the quench has to be done in a fume hood with the sash pulled down and a full protection as there is a good chance of the reaction mixture splashing out. Plain latex gloves  are no match for ClSO3H – use thick long-sleeved ones.

Note 2: Wet chlorosulfonyl oxindole has thixotropic properties – a wet solid that suddenly starts flowing as a sludge when shocked (this is not a sign of decomposition). The wet sulfonyl chloride after filtration was transferred into a glass dish and thoroughly dried on highvac. Drying this quantity of material took 2 days (over weekend). A thoroughly dry sulfonyl chloride is required for the success of the next step – incomplete drying or substituting anhydrous DCM for a non-anhydrous DCM grade (stabilized with ethanol 1%) resulted in a product containing a large quantity of the corresponding symmetric disulfide.

Note 3: The product is soluble in aqueous NaOH as a thiolate, the alkaline solutions will gradually oxidize on air to the disufide but the extraction can be actually done without a protective atmosphere if one works without delay.


  1. I’m surprised an aqueous quench of the chlorosulfonylation step doesn’t hydrolyze the sulfonyl chloride. As undergrads we’ve always been told acyl/sulfonyl chlorides are supremely reactive, and should be kept far, far away from any traces water.

    Comment by UG4 — October 6, 2010 @ 10:22 pm

  2. Back in grad school, I purified an ancient bottle of tosyl chloride according to a method from Amarego. The first part of the procedure required dissolving the TsCl/TsOH mixture in ether and extracting with aqueous sodium bicarbonate. I’ve also distilled MsCl that had droplets of water in it. Sulfonyl chlorides that lack electron-withdrawing groups really don’t hydrolyze fast. Seem to remember a professor saying that nucleophilic attack on second-row atoms is “slow”. Triflic anhydride on the other hand…

    Comment by AnotherChemist — October 7, 2010 @ 8:50 pm

  3. Also I’m trying to figure out the mechanism for the reduction step but failing badly. All I can think of is that 2 equivs of PPh3 are needed for deoxygenation, probably generating O=PPh3 as a byproduct, and the 3rd equiv goes on to make the phosphonium salt, hence 3 equivalents are needed.

    Comment by UG4 — October 8, 2010 @ 10:48 pm

    • I think you are right. I was also surprised by this reduction – to get the thiol from sulfonyl chloride, originally I planed to use something like Zn/acetic acid (since I could not use more vigorous reducing agents like LAH) but then I found old Astra Zeneca process paper with a good precedent so I gave it a try. Apparently PPh3 is strong enough to take aryl sulfonyl chlorides all the way down to thiols. The only annoying thing about this reduction is the huge molecular weight of the reducing agent (PPh3) and one uses over 3 equivs of phosphine. Tri-n-butyl phosphine works just as well in this reduction but there was no advantage for me in using it instead of PPh3. I bet PMe3 solution would be even better reagent on a small scale, if one could purchase is cheaply enough.

      Comment by milkshake — October 9, 2010 @ 12:01 pm

  4. I thought oxindoles readily get oxidized to indigo-type stuff in air. Any pretty colors in the filtrates?

    Comment by LiqC — October 13, 2010 @ 7:35 pm

    • not really. Oxindoles are colorless and well behaved (although sometimes poorly soluble). Are you sure that you are not talking about 3-indolinones? Also, indoles oxidize easily into colorful crap – especially on light – if they have el donating substituents.

      Comment by milkshake — October 14, 2010 @ 12:47 pm

  5. Hi Milkshake,
    i am trying to alkylate a nitrogen atom of a very unreactive heterocycle (substituted carbazole), usually solution of azole in dry DMF, NaH 60% in oil, then 2-bromoethylamine protected as a phtalimide. The only product is the one from elimination of the bromine on the electrophile, as the starting electrophile disappears without consumption of the starting azole.
    thanks a lot.

    Comment by madforit — October 15, 2010 @ 8:12 am

    • this does not sound too good – your elimination is related to the basicity of the mix. You may try a milder base like cesium carbonate or cesium fluoride, or 40% KF on alumina – because these should be strong enough to deprotonate carbazole. You can add catalytic iodidie (NaI or Bu4NI) to improve the reactivity of your alkylating agent. And you can even try a better solvating solvent like 1,3-dimethyl-2-imidazolidone (DMEU) or NMP to make the anion more reactive – but I think you will always have to fight the elimination problem with the electrophile that you are using.
      Is there any chance to use stuff like bromoacetonitrile, and do a cyano reduction afterwards? (CoCl2 + NaBH4, or Pd-C/H2 in TFA).
      Also, tosylates and mesylates derived from phenethyl alcohols are lot less prone to elimination to styrenes than the corresponding phenethyl bromides with hindered secondary amines.So you may also want to try the sulfonate ester version of your reagent.

      Comment by milkshake — October 15, 2010 @ 8:46 am

  6. i was thinking about the possibility of using 2-iodoethanol as alkilating agents, i have tried bromoethylacetate, but also this wont work well, even KOH in DMSO!Also tosyl o mesyl derivatives of the corrisponding alcohol could be a good idea.Thank you so much, this is a serious problem with of this class of damned comnpounds!
    ThanKs a lot

    Comment by madforit — October 15, 2010 @ 9:18 am

  7. @6 Have you tried ethyl triflate or ethyl Meerwein salt with proton sponge?

    Comment by AnotherChemist — October 15, 2010 @ 12:18 pm

  8. Hello Folks,

    I am trying to make the formyl derivative from an iodo-substituted imidazole by metalating it (IPrMgCl.LiCl or BuLi) and then, formylating with DMF or N-CHO-morpholine.

    I always get hydrolysis of the organometallic species to a big extent. I suppose the organometal is not very reactive but very basic.

    Solvent seems to play a role too, with toluene with better results than THF.

    Any advice there to increase reactivity and decrease the amount of I-H substitution by-product?

    So far I have a 1:1 mixture.

    Thank you

    Comment by vasili — October 16, 2010 @ 5:02 am

    • Here’s a ref about imidazole grignards where they use DCM as a solvent: J. Org. Chem., 1991, 56 (20), pp 5739–5740.

      Comment by JH — October 16, 2010 @ 6:55 am

  9. I have never done a metallation on unprotected imidazoles – but perhaps can you protect the imidazole NH by something like 1-ethoxy-1-ethyl group?

    Comment by milkshake — October 16, 2010 @ 10:11 am

  10. Thank you for the comments.

    S. Ley paper is interesting. I’ll take a look on monday at work, no access now.

    Milkshake: Imidazole nitrogen is alkylated, so no free N-H. C between the N’s is also substituted.

    Comment by vasili — October 16, 2010 @ 10:26 am

  11. Vasili:

    We have done a bunch of these metalation reactions with imidazoles either with N-alkyl substituents or N-sulfonyl ureas, including formylating them. We have found that metallating with EtMgBr in DCM (based on the Lindell chemistry) followed by N-methyl formanilide works well in our hands (several different grad students in my lab have used this), or in some examples N-methyl 2-pyridylformamide (I think this is the name) works in every case we have tried. It is a little expensive commercially, but it can be prepared fairly easily. This reagent was reported by Commins and Meyers in a Synthesis paper in the 80’s – if I recall correctly.


    Comment by Carl Lovely — October 16, 2010 @ 4:54 pm

  12. Thank you Carl,

    I’ll try your suggestions and let you know.

    Comment by vasili — October 18, 2010 @ 2:50 am

    • Hey, good news!
      Metalation with iPrMgCl and formylation with N-CHO-Morpholine all in CH2Cl2 increased the ratio desired/I-H substution up to 8:1.
      Solvent has a crutial role here.

      Comment by vasili — October 20, 2010 @ 2:57 am

  13. Anyone have any good ideas on how to get rid of tetrabutylammonium crap left over after a reaction when your compound is essentially copolar? Washing 100 times with water doesn’t seem to be the best solution. Thanks.

    Comment by PotStirrer — October 18, 2010 @ 9:10 am

    • what is in your molecule? Maybe you can use a column of acidic ion exchanger resin (netralized with ammonia or sodium carbonate and washed), to scavenge out the Bu4N+ (to exchange it for an inorganic cation)

      Also giving it few washes with diluted sulfuric acid may help because Bu4N+ HSO4- has quite poor solubility in organics.

      Comment by milkshake — October 18, 2010 @ 9:57 am

      • If you used Bu4NF for alcohol de-sylilation, try Dowex H+ resin instead, in MeOH, just filter and you have your pure alcohol with no nasty BuN+ salts.

        Comment by vasili — October 18, 2010 @ 1:55 pm

        • My molecule has an imidazole core and the salt comes from Bu4NBr from a Jeffrey Heck rxn. I found a Kishi reference for removing TBAF with DOWEX and CaCO3 (OrgLett, 2007, 723-726). Maybe I’ll get lucky with that. I’d be worried about extracting my compound into an acidic aqueous solution, but will give these suggestions a try. Thanks a lot.

          Comment by PotStirrer — October 19, 2010 @ 5:44 am

          • Kishi workup: CaCO3 is probably there just to eat fluorides (CaF2 is very insoluble). You need to neutralize your Dowex so that it does not bind your imidiazole as a salt – but ammonia or sodium carbonate will do for you just fine

            Comment by milkshake — October 19, 2010 @ 11:30 am

          • Yes, CaCO3 is there to eat fluorides and they found it necessary to “drive the reaction to completion”. Could you suggest exactly how you would go about neutralizing the resin? Thanks!

            Comment by PotStirrer — October 20, 2010 @ 4:54 am

          • I would slurry Dowex or Amberlyst-15 sulfonated ion exchanger resin in boiling deionized water, several times, and wash it thoroughly to get the dark crap of degraded polymer off the resin. (This will take care of swelling the resin at the same time). Then I would add some sodium carbonate aq. solution or some concentrated ammonia to the slurry at room temperature, stir for a while, filter, wash the resin with plenty deionized water – and pack the slurry into a short column, in a methanol-water mix. I would dissolve your crude mix in methanol + water and pass it through the column. It is like running a silica column (but I would collect far fewer fractions than you normally would with silica chromatography) and you need to check that your stuff came off the column – sometimes compounds can tail.

            Comment by milkshake — October 20, 2010 @ 11:54 am

  14. Thanks for the details Milkshake. Kishi method didn’t work for TBAB, perhaps not so surprising. We will try your suggestion today. I have absolutely no experience with ion exchange resins…do you think the percentage of the methanol/water mixture will be important?

    Comment by PotStirrer — October 21, 2010 @ 2:53 am

    • Methanol concentration is not as crucial – as long as there is enough to dissolve your reaction mix. If you add too much organic co-solvent the resin may not swell properly, and the beads will visibly shrink, and the ionex capacity will decrease. This shrinkage is more a problem with Dowex (which works only with water-methanol mix but not straight MeOH). Amberlyst is designed to be more compatible with organics (having polyamide instead of polystyrene matrix) but Amberlyst also has more dark resin-decomposition crap coming off so it needs to be washed more thoroughly. By the way, maybe you can use diluted NaOH instead of sodium carbonate, to avoid CO2 bubble formation inside the beads… but ammonia should work too

      Comment by milkshake — October 21, 2010 @ 11:37 am

  15. Hi MS,

    I am trying to invert hydroxy group by Mitsunobu reaction via O – Formate derivative, but this reaction is not clean for some reasons. I am using DIAD, PPh3 and HCOOH for this reaction.
    I am getting some inconclusive product. Please suggest me how to improve this conversion.



    Comment by marto — June 20, 2013 @ 1:46 pm

    • formic acid is not the best one – it is hard to make perfectly anhydrous, and it spontaneously dehydrates to CO and water, and more so in presence of DEAD+PPh3.

      Try p-nitrobenzoic acid instead. See for example

      Click to access CV9P0607.pdf

      Comment by milkshake — June 20, 2013 @ 1:57 pm

  16. HELLO SIR,
    i had tried the reaction of chlorosulfonic acid with oxindole and 2-substituted oxindole, firstly the oxindole (1 mmol) cool at 0 C the add drop wise excess of
    chlorosulfonic acid (1.2 mmol) and stirred for at 60 C for 2 h. the dark solution was formed. when this soln was pour in to ice the dark red color clear solution was formed nut not precipitation was occur. please give me the solution and procedure for synthesis.

    Comment by manjinder singh — May 7, 2014 @ 5:27 am

    • first, make sure your chlorosulfonic acid is still good (not hydrolyzed). Second, follow the procedure above exactly, without any modifications – it is a good tested procedure – and if you read it carefully, it says oxindole powder should be added into chilled chlorosulfonic acid (not the other way around). So follow the procedure. Also, when the chlorosulfonation is finished, add the reaction mix in a very thin stream onto a very large excess of crushed ice that was pre-chilled in freezer to -20C. The quench is quite dangerous, and hugely exothermic – it is possible that you are decomposing the product at this point. It needs to stay cold, as ice sludge, throughout the quench.

      Comment by milkshake — May 7, 2014 @ 12:02 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: