The substance is 3,4-diflurothiophene, CAS # 19259-15-5. Let there be no confusion; the compound is small, but the picture is big.


When you look in the literature, you find

  • A lot of computational chemistry
  • Some who magically had it as a starting material
  • One or two lengthy preparations and miserable atom economy

Long story short: We need a couple of grams. We don’t have a lot of time on our hands and we don’t wanna cough up 5,000 USD for it.

How hard can it be?! The molecular weight is 120, dammit. There is no stereochemistry, neither thiophenes nor fluoroarenes are new under the sun, and the exact structure is known. (I have a feeling it is pretty volatile, though.)

So, how about this? He or she who comes up with the neatest* synthesis suggestion down below will receive: The honor of having it performed in real life – soon. And if it works, as a token of our appreciation, any chemistry book for up to 150 USD on Amazon sent to your door.

What are you waiting for? Go!

* Starting from stuff you can buy. One or two steps. No Sandmeyer crap, for Heaven’s sake. Think “robust and cost-efficient,” and what your first choice would be if you personally had to deliver.

Update Dec 19th, 2012

I’m totally blown away by your support and genius. Thank you! The winner will be announced later and separately, and I could only wish that that part, including performing the actual experiments, was as swiftly managed as communicating online. I have gotten a couple of really tempting leads already after one day, but should you think of something not already mentioned, don’t hold back! There is still plenty of time.

In a broader perspective, see this as a good example of the power and might of open science. I would like to highlight especially to you guys in industry that not everything we do needs to be hush-hush.


60 Responses to Small compound, big challenge. You in?

  1. Well I’ve got a route, although the first step isn’t brilliant.

    Take 2,5-dimethoxy-2,5-dihydrofuran ( and fluorinate across the double bond. I’ve found an Orgsyn page ( which brominates across the bond using Br2 in DCM, but doing that with F2 is going to be a bit more difficult, so this is where I’m stuck at the moment.

    If you can find a better way to fluorinate, once you have 3,4-Difluoro-2,5-dimethoxytetrahydrofuran, you can react it with H2S, P2S5, or Lawson’s Reagent (supposed to be the best, it’s Ar2P2S4 Ar=paramethoxyphenyl) and that should yield your product.

  2. Simon Higgins says:

    We have used the chemistry in JACS 123 4643, where the authors make perfluorosexithiophene. They make fluorothiophene intermediates by lithiation of bromothiophenes (protected by TMS groups where halogen dance might be an issue) and quenching with the F+ source (PhSO2)2NF, which is a nice stable crystalline solid, commercially available. We adapted it to make 3-alkyl-4-fluorothiophenes, for reasons which I am sure you can guess!

  3. O Hammond says:

    Starting with succinaldehyde, react with stoic. (2.1 equiv?) Accufluor/MeCN (preferably under anhydrous conditions, succinaldehyde cyclises in MeOH and H2O) to yield the difluoro-dialdehyde (probably quite unstable!), handle with care and then Paal-Knorr to theoretically yield the difluorothiophene.

    Sources: S. Stavber, M. Jereb, M. Zupan, Synthesis, 2002, 2609-2615.

    • I don’t think the cyclisation of succinaldehyde would be a problem, the Paal-Knorr will probably still work, I know it does in the synthesis of pyrroles. Accufluor is an awful lot better than my F2 in DCM!

  4. drfreddy says:

    Amazing response already after one hour. Don’t stop!

    I would in addition appreciate it if you could tell me if you think the compound is

    a) intrinsically unstable or not and
    b) how to best proceed practically to isolate a presumably volatile compound

    • Well for what it’s worth, the ACD/Labs software predicts:
      Boiling Point: 99.3±20.0 °C @ 760 Torr
      Enthalpy of Vaporization: 32.46±3.0 kJ/mol @ 760 Torr
      Flash Point: 13.9±21.8 °C

      • There’s also an NMR of your product on AIST’s SDBS, taken in cyclohexane, so I guess it’s soluble in cyclohexane from that?

        • drfreddy says:

          “Enthalpy of Vaporization: 32.46±3.0 kJ/mol @ 760 Torr” — does that translate to “very volatile”?

          I’ve seen some pretty weird things working with fluoro-compounds before. I once fluorinated a larger heterocycle even with two bromines on it, SM being a rock solid. The product was a colorless liquid that evaporated if you looked at it. Rotavap was not applicable; learned the hard way.

  5. Adam Zajdlik says:

    Use hetero Diels-Alder with 1,3,4-thiadiazole (CAS 289-06-5 ) and difluoroacethylene (CAS 689-99-6). Mechanism:

    • Adam Zajdlik says:

      I suppose technically this is not hetero Diels-Alder chemistry. Just plain old Diels Alder. I got a little excited with the nomenclature.

    • Simon Higgins says:

      Trouble is, difluoroacetylene is extremely unstable and spontaneously polymerises at room temperature with a T(1/2) of about 15 minutes even at very low pressure. See Chem Comm 1991 456.

  6. Chemist from Ohio says:

    (1) Perbrominate thiophene (easy, no column, good yield, cheap starting materials) to give tetrabromothiophene
    (2) Debrominate alpha positions (easy, simply zinc/acetic acid reflux and collect the product in high purity in a Dean-Stark trap–it azeotropes!) to give 3,4-dibromothiophene
    (3) Lithium-halogen exchange and fluorination with electrophilic fluorinating agent (NFSI?)

    First two steps have lots of lit. precedent. For (1) and (2) see SI in ref.

    • drfreddy says:

      We did (1) many times in the undergrad lab course. Nice reaction. I’ve done (2) myself, works as described.

      Regarding (3): Has 3,4-dilithiothiophene ever been used (with a different electrophile)? If so, we’re getting VERY close to something I like!

      That case, we’re definitely buying 3,4-dibromothiophene. It’s quite affordable.

      • Simon Higgins says:

        If you do Br/Li exchange on 3,4-dibromothiophene other than at *extremely* low temperatures which will probably obviate the second fluorination step, you will get halogen dance chemistry, resulting in 2-lithio derivatives. You will need to protect the 2,5 positions with TMS or something; this is the route I cited in my post above.

    • Alessandro Bedini says:

      An alternative can be:

      (i)the bromination of both alfa position using AcOH/NBS
      (dibromothiophene can be purified by distillation under vacuum)

      (ii) Fluorination of 3,4 position using fluorinating agents like perfluoroalkylsulfonimides (for a review on fluorinating electrophilic agents look at : Chemical Reviews, 1996, Vol. 96, No. 5, pg 1737)

      (iii) debromination of alfa position with zinc and acid

  7. dielsalderase says:

    Chemist from Ohio, you beat me to it, although the tetrabromothiophene isn’t that expensive from Aldrich ($50/25G) if you need to save a little more time.

    One other option might be nucleophilic aromatic substitution with F- (KF or TBAF in DMSO or DMF), Org. Lett. 2004, 6(19), 3381 (substitution with copper cyanide in DMF)

    • drfreddy says:

      Yes, aromatic nuc sub is likely the first reaction I’ll try. Commercial 3,4-dibromothiophene, NaF, acetonitrile or DMF, microwaves. Blast for 15 mins, check crude GCMS. For work-up, I’m thinking Et2O or pentane and no rotavap.

      • Simon Higgins says:

        If you do try this, let us know how it goes. I *think* my student tried that reaction with 3-bromothiophene (which may be less reactive), but got nowhere with it. However, if he did, he didn’t include it in his thesis.

        • drfreddy says:

          I have lithiated 3-bromothiophene back in the days. You had to add the electrophile almost immediately at -78 deg C. Else, swift rearrangement to the 2-position.

          I doubt 3,4-dilithiothiophene could exit, but I’d love to be proven wrong.

      • SN says:

        NaF in MeCN with the dibromide won’t do anything – no solubility whatsoever, and bromides don’t react. You’d at least need need scrupulously dry, finely ground KF (‘spray dried’ KF from aldrich + vacuum oven overnight) in thoroughly dry DMSO or sulfolane at >180 degC with the dichloride. A spot of phase transfer catalyst may help as well. Ph4PBr is relatively common. Adding in 18-crown-6 helps too. CsF is supposedly more reactive, but it’s less common in the literature, probably because aromatic substitution with fluoride is mainly used in industrial processes so the cost is prohibitive.
        Synthetic Communications 1994 vol. 24 p. 95 has SNAr of 3-chloro-2-cyanothiophene, but that’s quite activated.

        My bet is on Tom’s method for steppy but reliable, James’ method (buying it..) for actually saving you time and money in the long run.

  8. undergrad says:

    1)Nuc. aromatic sub of 3,4-dibromothiophene via -NH2/NH3 to yield 3,4-dinitrothiophene 2)NaNO2/HBF4 to yield 3,4-diflurothiophene. I’m a mere undergrad though, so this may sound ridiculous and be completely incorrect.

  9. mvs says:

    3,4-dimethoxy thiophene – electrochemical fluorination (

  10. mvs says:

    dicarboxylic acid*

  11. Tom says:

    0. Start with diethyl tartrate
    1. Substitute the OHs by F using DAST (or SF4 if you need a lot)
    2. DIBAL/LAH to the alcohol
    3. Swern to the dialdehyde
    4. Paal-Knorr to the thiophene

    It’s 4 steps, but there is no strange chemistry involved.
    Maybe you can even selectively reduce the diester to the dialdehyde (DIBAL, -78 °C?), so you don’t need to do the Swern oxidation. See the last reaction on page 8965 for such a reduction and in situ Paal-Knorr (pyrrole) synthesis:

  12. Dmitry B says:

    cross-coupling of 3,4 -dibromothiphene with F- source ( than evaporation of your product in the nitrogen trap. or electrophilic fluorinating of 2,5 disubstituted thiophene: R=TMS (as was mentioned above), carboxy (than thermal decarboxylation), Br (than Pd catalyzed reduction) or else

  13. James says:

    Buy 3,4-Difluoro-2,5-bis(trimethylsilyl)thiophene from TCI or VWR (~$220/gram, catalog # D3701-1G, CAS# 347838-12-4). Treat with TBAF, ala Journal of Organic Chemistry, 75, 1652-1658 (2010), or any number of similar procedures found with a simple literature search. Done!

  14. lithiator says:

    1) dilithiate tetrabromothiophene (~$4/g, acros), quench with TMSCl (95%: Chem. Mater. 2007, 301). 2) one-pot sequential Li-Br exchange, quench with (PhSO2)2NF (73%: JACS 2001, 4643). Then what James said.

  15. JBC says:

    Another vote for nucleophilic aromatic subsitution. I’d try a crown ether and KF to make the F- really nucleophilic, and react it with 3,4-dibromothiophene (which seems to be easy to make: add bromine to thiophene to get the tetrabromo- and then react with Zn to get 3,4-). You can get away with a less polar solvent in this way too, which might make isolation a bit easier.

  16. TDA says:

    Protodesilylate 3,4-Difluoro-2,5-bis(trimethylsilyl)thiophene in a high-boiling solvent and distill the product directly from the reaction mixture?

  17. JH says:

    If you’re aromatic nuc sub doesn’t work with 3,4 dibromothiophene you can also try 3,4 dichloro thiophene or maybe tetrachlorothiophene (readily available) with 2 eq of KF and remove the 2,5 dichloro substituents radically.

  18. EP says:

    2,5-di-TMS-3,4-dibromothiophene and its one pot conversion to 2,5-di-TMS-3,4-difluorothiophene are known (JACS 2001, 123, 4643 – Procedure is available for free in the supplementary information (uses n-BuLi and (PhSO2)2NF). The starting material can be made by lithiation of 3,4-dibromothiophene or tetrabromothiophene followed by reaction with TMSCl.
    Maybe you could do the whole conversion in one pot by controlling the stoichiometetry. Finish by deprotecting the 2,5-positions by protiodesilylation (acid or fluoride ion).

  19. […] an interesting challenge: over at Synthetic Remarks, there’s a need for a couple of grams of 3,4-difluorothiophene. But you can’t buy that much, and the […]

  20. Yoyo says:

    Maybe a bit academic/expensise but I’ll give a shot anyway !
    What about RCM ? 2,5 dihydrothiophene synthesis from diallyl sulfide using RCM is known (Org. Lett. (2002), 1767 among others…)
    Similar approach using MFCD20492668 (POA but synthesis known as well)to the difluorodiallyl sulfide and RCM to the 2´,5dihydro derivatives followed by oxidation (several options there I suspect you could do make it a one pot tandem by (mild) oxydation of the Ru cat. that is known too).

  21. TLC says:

    Superstoicometric amounts of silver(I) oxide and dilute HF in water/acetone mixture should rapidly exchange the bromides on 3,4 dibromothiophene with fluorides. Finkelstein inspired chemistry.

  22. woon says:

    Reacting oxidized thiophene (SO2) with MF (1,4 addition) twice under certain reaction conditions.
    Then oxidation of the ring and reduction of -SO2- to -S-.
    Not sure about the electrophilicity of the compound in the first step.

  23. Nick K says:

    Dimethyl 3,4-dihydroxythiophene-2,5-dicarboxylate is a known substance. Treat it with DAST or similar to give the 3,4-difluorodiester, then hydrolyse the ester and decarboxylate to the final compound. Alternatively, make the dimesylate and treat with TBAF.

  24. retro-Claisen says:

    start from 2-fluoroallyl chloride (1) (CAS # 6186-91-0, $245 for 5 grams from Matrix Scientific)

    Step 1: (1) + hydrogen sulfide = 2,2′-difluoro-diallylsulfide (2)
    step 2: (2) + Grubbs’ catalyst = 2,3-difluoro dihydrothiophene (3)
    Step 3: (3) + DDQ = difluorothiophene!!! WAHOO!!!!

    very simple reagents (e.g. hydrogen sulfide, grubb’s cat. and ddq) and well established chemistry (e.g. nucleophilic substitution, RCM, and DDQ oxidation).

    This is the Best route to this compound!

  25. Jay says:

    Potentially a short/cheap route if the early intermediates don’t polymerize.

    Alpha fluorination of acetaldehyde with proline and NFSI in THF, add MnO2 and heat to dimerize to the 2,3-difluorosuccinaldehyde. Filter to recover the crude aldehyde, add P4S10 and reflux to generate the thiophene which you should be able to fractionally distill from the pot.

    3 steps, but only 2 pots and one purification and cheap reagents.

  26. milkshake says:

    1) I think the Br metallation/F(+)quench based procedures are going to be lots of trouble: the 2 position of thiophene is quite acidic (BuLi lithiation of unsubst thiophene to 2 position goes very nicely even at low temperature, we used 2-thienyl lithium as a dummy cuprate ligand) so lithiations are out, the only possible exchange is Mg/Br or Zn/Br and it is going to be very tricky.
    2) There is a recent development in decarboxylative fluorination of arylcarboxylic acid in water + bicarbonate with Selectfluor, catalyzed by Ag salt, that gives good yields of fluoroarenes, especially with el rich heterocycles like thiophene. (I have made 2-fluoro-5-bromofurane from 5-bromofuroic acid in this way, using an older method without Ag catalysis, and it worked quite well). 3,4-thiophenedicarboxylic acid (CAS 4282-29-5) available from Fisher-Acros but somewhat pricey, you may get a better deal with some other supplier – run the available chemical directory search. But I would definitely give this decarboxylative fluorination a try.

    3) By the way, the product is going to be pretty volatile (boiling point similar to bp of thiophene) and the molecular weight is quite low so you will need lots of staring materials and reagents to make 5g, and to get it pure (you cannot use normal column/rotovap unless you use ether-pentane for the column, and even then the losses will be horrendous. You may be better served by using very high boiling solvents and rather distill the product from them). Either way, $5000 for 5g of pure stuff is a steal and take it if someone quotes you this price, you wont be able to make it on your own for that little in Sweden unless you employ lots of little Umpa Lumpas

    • drfreddy says:

      I was waiting for you to step in and sort out the facts, TYVM! I’m well aware of how difficult it is to convince the lithium to stay put in the beta-position. (-78 deg C, adding the electrophile after 5 mins is how I used to do it. Elemental sulfur being a favorite electrophile, usually followed by a second electrophile, in the form of a haloalkane, to give thioethers.)

      Decarboxylative fluorination — didn’t even know it existed. Will definitely look into.

      Yes, volatility is presumably a major obstacle during work-up and handling. Also, the propensity for this little one to auto-polymerize is uncharted territory.

      And hey Milky — if we indeed find someone who is willing to sell it straight to us for around 1000 USD per gram, and we go with it, are you still claiming the prize? 🙂

      • milkshake says:

        I am not claiming any prize, and USD 1000/g sounds reasonable if you only need few grams, fairly soon. Better than investing several weeks of person-lab time and buying starting materials, and having uncertain outcome. If the business in this picks up you can always develop your own route later on.

        Also, do you have, by any chance, need for high-quality and nearly monodisperse hetero-bi-functionalized PEG polymers, (terminal azides, acetylenes, amines, carboxylic acids and the like)?

    • Chemist from Ohio says:

      Re: metalation at the 3-position — it’s certainly possible to do it with minimal halogen dance but it depends how rigorous you’re willing to be with the conditions. Purification would be the nasty part… try separating the mixture of 3,4-difluoro, 3-bromo-4-fluoro, 3,4-dibromo, 2-fluoro-4-bromo, 2,5-dibromo, etc. that you would inevitably get.

      Probably a better shot with nucleophilic aromatic substitution. This works way better at the C-3 position of thiophene than at the C-2.

  27. milkshake says:

    also there is 2,2,3,3-tetrafluoro-1,4-butanediol which is quite affordable and could be perhaps turned (via bis triflate) into 3,3,4,4-tetrafluoro-tetrahydrothiophene. HF elimination with a crapload of LDA could give you the 3,4-diF-thiophene

  28. Tapan Maji says:

    Step 1. RCM of Diallyl sulfide (CAS 592-88-1, 25 mL, $ 30).
    Step 2. Fluorination of double bond (JACS, 1945, 1639)
    Step 3. DDQ aromatization.

  29. Steve says:

    Use Knochel chemistry:

    Start with 2,5-dichlorothiphene (cheap from Aldrich – $78/100g), treat with the Knochel-Hauser base TMPMgBr-LiCl (commercially available) to provide the 3-metallated cmpd, then add that to an electrophilic fluorination reagent such as F-TMP-BF4 to give 2,5-dichloro-3-fluorothiophene.
    Repeat the sequence again to give 2,5-dichloro-3,4-difluorothiophene.
    Reduce off the two chlorines with Pd/C, NH4formate to give the title cmpd.

    Knochel chemistry here:

    Piller, F. M.; Knochel, P. Org. Lett. 2009, 11, 445.

    Fluorination chemistry here:
    Beller, Matthias et al, Angew Chem Int Ed, 2010, 49, 2219.

  30. Hap says:

    Dammit, Milkshake.

    The tetrafluorobutanediol is available from Aldrich at about $90/5g. The ditriflate will undergo substitution/cyclization (with an amine), but maybe the ditosylate would work (and might be more stable and more easily separated). Nucleophilic substitution with sodium sulfide at dilution would give the tetrafluorodihydrothiophene (which will probably stink a lot). LDA might eliminate two fluorides to generate your product. Of course, 1) Milkshake figured that out and 2) separations might still be a problem (the intermediate THT will probably be low-boiling, stinky, and might be hard to separate from the desired difluorothiophene in the next step).

    I wonder if you could bromofluorinate 2,5-dihydrothiophene with NBS/TBAF; elimination would give the monofluorodihydrothiophene. Repeating would provide the difluorodihydrothiophene; Pd/C/cyclohexene might give the desired product. That sequence seems messier and less probable than others, though.

    Another possibility might be if you could ortho-metalate 2,5-thiophenedicarboxylic acid and capture the product with an electrophilic fluorine source – at least the thiophene products would be separable from reagent and not too volatile. Thermal decarboxylation or electrolysis would give the difluorothiophene. Again, Milkshake’s decarboxylative fluorination sounds better. Knochel chemistry might work, too. All of the separations for this seem like a PITA.

  31. milkshake says:

    On another though, I think there is a decent chance that your old friend 3,4-dimethoxythiophene, behaving very much like a vinyl ether, could be transformed directly into 3,4-difluorothiophene by heating it with Deoxyfluor (or DAST) + 1 drop of EtOH, to 60-80C and distilling the resulting mix

  32. Leandro Miranda says:

    May the chemistry just reported by by Meng and Li work on 2,5-Dibromothiophene (10.1021/jo302099d)? If so deprotection of the 2,5 position with Pd/Silane would be necessary.
    Best wishes

  33. JB says:

    An elegant approach would be the following:
    1) Make the difluoro-divinyl sulfide.
    – First react Fluoro-ethyne with sulfur (S8) in water/DMSO under basic conditions (KOH) at 80-120 degrees celcius.
    See Trofimov et al., tetrahedron 1982, 38, 713-718 for the synthesis of divinylsulfide out of acetylene gas (80% yield)
    – The same compound in principle can also be constructed out of 1-fluoro-2-chloro-ethene and sodium sulfide in DMSO under basic conditions at 100 degrees celcius.
    See Trofimov et. al., Journal of organic chemistry USSR (english translation), 1985, 21, 2123-2127 for the synthesis of divinylsulfide out of vinylchloride (86% yield)

    2) 6-electron Electrocyclic photochemical ring-closure of the difluoro-divinyl sulfide.
    Under photochemical conditions an electrocyclic ringclosure can be performed which will give you the thiophene. I’m not a photchemistry expert but my guess is that the C-F bond will not be cleaved under the given conditions. For the photochemistry see: Schultz, Acc. Chem. Res., 1983, 210-218.

    I’m not sure how general the above mentioned methodology is but it might give you a good lead. I also did not perform a detailed literature research on this but the synthesis starts out of commercially available materials with high yielding procedures for similar compounds, and there is some elegance (also important!). The disadvantage is that yo have to work with gasses and probably very volatile compounds but for the trained organic chemist this should not be the resean not to do it.
    Let me know what you think!

    Greetings from Groningen (The Netherlands)

  34. From diacetylene:
    HCl addition to get 2,3dicloro-buta-1,3-diene
    Heating with some SbF3 to to the corresponding fluoride (2,3-difluro-buta-1,3-diene)
    And reaction with SCl2 to produce the goal compound. Easy and well working.

    Diacetylene could be easily obtained from TMS-acetylene and Cu(OAc)2 in pyridine with high yields.

  35. […] the quest posted before regarding the most straightforward, but not necessarily the most elegant, synthesis of 3,4-difluorothiophene. Bear in mind, our goal was to get a decent amount of the material as swiftly as possible; a […]

  36. Anonymouse says:

    A paper on all sorts of polyfluorinated thiophenes.

  37. Klick says:

    Greetings from Idaho! I’m bored at work so I decided to
    check out your blog on my iphone during lunch break. I love the knowledge you provide here and can’t wait to take a look when I
    get home. I’m surprised at how quick your blog loaded on my mobile ..

    I’m not even using WIFI, just 3G .. Anyhow, good blog!

  38. mobile games says:

    hi!,I love your writing very much! percentage we keep in touch extra approximately your
    article on AOL? I require an expert on this house to unravel my problem.
    Maybe that’s you! Looking forward to peer you.

  39. This just came out in JACS:

    They dilithiate 3,4-dibromothiophene (TIPS-protected in 2,5-position) and then quench with N-fluorobenzenesulfonimide, claiming a stunning 88% yield.

    I must say I will try this sooner or later…

Leave a Reply

Your email address will not be published. Required fields are marked *