Epoxidation of Propenylbenzenes using Dimethyldioxirane or Oxone


For totally kickass epoxidation, SWIM has a sweet spot for dimethyldioxirane as a solution in acetone.

ref Chem Ber 2377, 124 1991 - Using Oxone, (Ald 22,803-6 1Kg/$24) and NaHCO3, in acetone

So simple and safe a chimp could do it.

Need dry ice, mechanical stirrer and dist apparatus, but in 45 min you might have 750 mL of 0.1 M Dimethyldioxirane. If yall use 1.1 eq/alkene then would be enough (at that scale) for 11 g alkene. I agree, it isn't ideal for idustrial scale syn. Very clean though and simply strip off solvent (recycle?) to yield an epoxide, so with rearrangement to ketone it could be interesting for the skilled and well equipped.

Please note that dimethyldioxirane is an explosive. unstable organic peroxide, and should never be allowed to form a concentrated solution. Read more about it and its dangers in Organic Synthesis.

K.C. Nicolaou:

These questions are in reference to a possible propenylbenzene -> epoxide -> ketone synthesis using dimethyldioxirane as the epoxidizing agent under slightly basic conditions(pH 7.5-8.0). >95% yields of the epoxidation product have been reported for a wide variety of alkenes(including styrene, stilbene, an 1-phenylpropene) using this reagent. Also, since the epoxide *should* not hydrolyze to the diol under these conditions, it may be possible to perform the epoxide -> ketone rearrangement using a mild lewis acid catalyst instead of the usual refluxing with H2SO4. If properly developed, this method could be an easy, fairly OTC, high yielding procedure for conversion of propenylbenzenes (isosafrole, asarone, etc) to 1-phenyl-2-propanones.

1. Osmium mentioned a method of rearranging the epoxide formed from epoxidation of propenylbenzenes to give the 2-propanones using a Li salt as a lewis-acid catalyst. Can anyone direct me to any literature references for this procedure and/or suggest a set of reaction conditions(solvent/temperature/catalyst amount/rxn time/etc)? The most convenient solvents in this particular case would be CHCl3, acetone, or toluene. Any alternatives as far as different lewis acids(AlCl3, ZnCl2, FeCl3, etc) might also be useful.

2. One of the journal refs I have for this rxn gives a procedure for a biphasic reaction(H2O and CHCl3) which uses (nBu)4NHSO4 as a PTC to shuttle peroxymonosulfate ions(HSO5-) into the organic phase where they react with acetone to generate dimethyldioxirane in situ. However, they use what seems to me like a riduculously large amount of PTC(.12 molar equivalents) in their procedure. Does anyone know a general rule for choosing a more modest amount of PTC for a scaled-up run based on either molar equivalents or solvent volumes? All of the reactants and products *should* be relatively stable under the rxn conditions, so rxn time is not a critical issue.


1) http://www.rhodium.ws/chemistry/guest.phenylacetone.txt

Deals with the electrolytic oxidation of propenylbenzenes (--->epoxides). These are rearranged in high yield by refluxing with LiI or LiBr in EtOAc.

2) When reaction time is not critical, use less PTC. I guess it will work with only half the PTC or even less.


A proposal was made like 3 monthes ago, about dimethyldioxirane. However, the simple prep listed was to get a 0.1 M solution in acetone. This solution would react (when used in molar excess say greater then 1.1 at the very least) to epoxidize cleanly the olefin. Strip it off and you got it, basically. However, the solvent: olefin ratio is rather high. (I figured 11 g to 750 mL of the solution). But it looks like LiI/EtOAc would give good conversion to the carbonyl compound of interest.

K.C. Nicolaou:

Dammit, I thought that was an original idea . I had the same problem as Acme with the procedure where the acetone/dimethyldioxirane solution is isolated and then used, although one of the refs I have indicates that concentrations of up to .185M could be obtained.

One of the refs I had (J. Org. Chem. 1985, 50, 2847) gives a procedure for doing a dimethyldioxirane oxidation (It's not an epoxidation, but that shouldn't really make a difference) in a CHCl3/water system. Basically, the compoung to be oxidized was dissolved in CHCl3 and then acetone, an aqueous solution of bicarb(to control pH), and the PTC were added. Then an aqueous solution of Oxone(2 KHSO5/1 KHSO4/1 K2SO4) was added. Transport of peroxymonosulfate ions into the aqueous layer by the PTC allows for in-situ generation of dimethyldioxirane, which should rather quickly epoxidize the alkene. Also, since the pH of the aqueous solution is held between 7.5 and 8.5 during the entire rxn, and the epoxide is in the organic layer, I would assume that the epoxide would not hydrolyze(never worked with epoxides much, so if I'm wrong, tell me).

My assumption would be that this rxn could be run with 100g alkene dissolved in 500-1000mL of CHCl3, making it a hell of a lot more efficient than the acetone procedure. As for the aqueous phase, I was kind of hoping that instead of using Oxone, which has a lot of extra bisulfate and sulfate salts lying around that you need to dissolve, I could prepare a solution of monoperoxysulfuric acid from sulfuric acid and H2O2, and slowly add that to the two phase system. I would have to up the amount of bicarb by 4/3, but I still should be able to get away with a much smaller volume of water that way.

As for the rearrangement using a Li salt, the link Osmium posted looks nice, but I wonder about the emphasis abou using "alkyl acetates" as solvents(ethyl acetate isn't really that easy for me to get, although I have come to love it's smell when using it as an extraction solvent). I'll probably try to do the rearrangement in CHCl3 or CHCl3/acetone before I start fucking around with trying to get EtOAc.

The chemical formula of dimethyldioxirane is C3H6O2. The structure can be visualized by considering the addition of an oxygen atom across the C-O double bond in acetone in the same manner as an epoxide. You can also find the structure on page 700 of March's Advanced Organic Chemistry (4th ed.) if you have that. The description of this reagent given by a professor of mine who is a very accomplished synthetic chemist reads as follows:


Review: Chem. Rev. 89, 1187, (1989)

Prepared from acetone and NaHCO3 by addition of oxone, and used in solution in acetone. Solution of the reagent is yellowish. Reactions are fast and take place at -40-(+)25C. Epoxidation proceeds under mild neutral conditions. Reagent of choice for the synthesis of sensitive epoxides of enol esters, enol lactones, and enol ethers.

A lot of emphasis on enols, but when he presented this in class, it seemed that he was saying that it was an all-purpose alternative for epoxidizing acid sensitive compounds that wouldn't survive epoxidation with peroxy acids.

As far as I know, persulfate is not a good epoxidizing agent by itself. However, under slightly basic conditions(pH 7.5-8.0), persulfate reacts with acetone to form dimethyldioxirane(three membered cyclic peroxide with both oxygens bonded to what used to be the carbonyl carbon of acetone), which is a very good epoxidizing reagent. I am looking into that as an epoxidation system for asarone and there is a thread under my name in the chemistry discourse about it. Persulfate is VERY easy to get. You can either make persulfuric acid from sulfuric acid and H2O2, or you can buy Oxone, which is a Dupont trade name for a mixture of 2KHSO5/KHSO4/K2SO4, at any home depot or pool supply place as a non-chlorine pool shock. The main difficulty with most dimethyldioxirane methods is that you can only use fairly dilute solutions(~.15M) of the reagent, so I am looking into some alternative procedures using a two-phase CHCl3/water system that generates the dimethyldioxirane in-situ. That procedure requires a PTC though, and you're still talking about having a total volume of ~2L to oxidize 100g asarone.

Actually, I have quite a few refs for the epoxidation rxn/generation of dimethyldioxirane part of this. And armed with the LiI rearrangement ref Osmium was so kind to give the link for, I plan to check out that end next time I'm at the library. I actually have most of the information I need, but I'm trying to tie up a few loose ends. What I am really looking for here is the following.

  • A general rule for amounts of PTC to use per solvent amount and/or per mole of reactant. Something akin to the rule of thumb you use for determining the amount of adsorbent/gram of material when running columns(ie, 50:1 for good separation, 300:1 for tight separation).
  • Whether solvent effects are very important for the rearrangement step. I would like to be able to use a solvent other than EtOAc, as this would be the hardest material for me to get that I might concievably need for this rxn.
  • The location of the equilibrium for the H2SO4 + H2O2 <--> H2SO5 + H2O rxn. I would assume that the peracid is kinetically favored because H2O2 is a much better nucleophile than H2O, but I'm not really sure about how to best generate the persulfate ion needed for this rxn. Using H2SO4 and H2O2 to generate peroxymonosulfuric acid would be much nicer than using Oxone if that rxn is anywhere near quantitative. I'll look at the performic procedures, as I'm sure the issues involved are similar.


SWIM has used powdered OTC potassium peroxymonosulfate (oxone) purchased at any pool supply in any amount desired in a modification of the performic reaction on our favorite 2-alkene. SWIMM got back 30% ketone -- but only because SWIM allowed the reaction mix to heat up CONSIDERABLY beyond the safety margin.

SWIM has total faith in this reaction but just couldn't afford to experiment on any more precious 2-alkene, so SWIM switched to the modified performic, a method with which SWIM has more experience.

The reference that details the macroscale reaction SWIM just described can be found in somewhere in JOC 1993, SWIM thinks. Just look up "epoxidation" or "epoxide" as keywords in the index. You'll find the article. The yield for B- methyl-styrene (propenylbenzene) was above 90%.

Perform the reaction at 40 deg C for 5 hrs but DON'T let the temp climb above 50 deg C as this caused EXCESSIVE decomposition in SWIM's first trial. And even then, SWIM still got (after dehydration of the diol) 30% of the ketone. Room temp gave low yields even after 12 hrs. Also, be sure to filter the aqueous potassium peroxymonosulfate solution BEFORE you add the 2-alkene, as the cloudiness that exists is caused by a "clarifier" that if not removed will cause a BITCH of an emulsion later on.

This method is VERY VERY OTC. All you need is 1-alkene reacted with KOH in EtOH to effect double bond migration to the 2-alkene. This can then be oxidized into the 1,2-diol by stirring at 40 deg C for 5 hrs in a water solution of potassium peroxymonosulfate. The diol is then dehydrated by 15% H2SO4 to the ketone.

Potassium peroxymonosulfate (oxone; pool bleach) is the way to go. In a JOC article that I don't have on hand, potassium peroxymonosulfate is used to either epoxidize or hydroxylate (depending on the pH) B-methylstyrene, better known as propenylbenzene. The reaction was performed in a biphasic mixture: water containing the oxidation salt was one phase, while the other was the alkene. No solvent was used. In my opinion, to get the reaction to work for ring substituted propenylbenzenes, heat may have to be applied. The method however, is definitely OTC all the way.

No acetone was used. Just water (quite a bit actually), the oxidizing salt, and the alkene ONLY. The article can be found in JOC 56, p.7022, 1991. It indicates that co-solvents and even PTC catalysts do not affect the reaction advantageously. Just stirring and in some cases, heat, were required. Looks promising to me. I did mention in the original thread, however, that some soot might need to be removed from the water solution before the reaction is commenced. It seems harmless, but makes the solution cloudy. Letting the solution sit for about 12 hours, maybe less, allows the soot to settle. Decantation might work for it's removal. Or filtration. Questions?

J. Org. Chem 1991, 56, 7022-7026 "Oxidations of Alkenes with Aq. Potassium Peroxymonosulfate and no Organic Solvents."

"Aqueous KHSO5 oxidizes many alkenes to diols in acid mixtures and to epoxides in neutral mixtures with no organic solvents."

The mechanism is similar to the epoxidation of alkenes by organic peroxy acids. The researchers feel that this is more cost efective and more practical for industrial processes than your standard peroxy acid epoxidations.

"Epoxidation of cyclooctene: In a vigorously stirred two-phase mixture (1800rpm), 1.7 molar equiv. of aqueous KHSO5 epoxidizes cyclooctene to >80% epoxide with no significant by-products in 5hrs. at room temp (23C). The oxide crystallizes and is isolated easily by filtering the reaction mixture."


3g cyclooctene 185ml of .26M KHSO5 (17.2g Oxone) Stir 1800rpm for 5hrs. Oxide produced: 2.85g

All the standard alkenes tested with similar results: (40-90%)