Recently, I was reading a paper about a total synthesis of (+)-rubriflordilactone A (the article is open access: Angew. Chem. Int. Ed. 2015, 54 (43), 12618–12621). In two different steps where an alcohol -> aldehyde oxidation was required, the authors favoured using the Parikh–Doering reaction (DMSO activated with pyridine-sulfur trioxide complex):

Part of the synthetic sequence

In practice, is there a reason why the Parikh–Doering might be favoured over the Swern (DMSO activation with oxalyl chloride), or is it simply a case of "we just got better yields with it"?

I checked in Kurti–Czako. No hints were offered, but they also did list two more Swern variants, namely Pfitzner–Moffatt (activation with a carbodiimide) and Albright–Goldman (activation with an acid anhydride). On top of that, as Jan mentioned in a comment, there is also the Corey–Kim oxidation where N-chlorosuccinimide is used to generate the chlorosulfonium salt from DMS (instead of DMSO).

In general: are there benefits or drawbacks that are specific to each variant that would influence a synthetic chemist's decision, and if so, what are they?

  • 1
    $\begingroup$ Will think about it properly later but in practice I think here's little difference— labs seem to have a preferred method. I have a feeling that the Parikh Doering doesn't have the issue of an unwanted Pummerer that I've seen once or twice with a Swern but practically its of little consequence. DMP all the way! $\endgroup$ – NotEvans. Jan 5 '17 at 19:25
  • 1
    $\begingroup$ We often had this in our seminar. Parikh-Doering has $\ce{SO3 . Py}$ as the activating agent meaning that $\ce{Py . HSO4}$ is eliminated and no formal $\ce{HCl}$ is formed that would need to be quenched additionally. Or something along those lines. $\endgroup$ – Jan Jan 5 '17 at 20:35
  • 1
    $\begingroup$ I imagine "we just got better yields" is frequent, but the nature of the byproducts could be one rational consideration. $\endgroup$ – jerepierre Jan 5 '17 at 21:31
  • 1
    $\begingroup$ Sometimes it is a matter of scaleability. The pyridine.SO3 conditions are easier to use than handling a lot of oxalyl chloride or TFAA $\endgroup$ – Waylander Mar 2 '17 at 17:52
  • 1
    $\begingroup$ I've been out of the loop for a while, but in general, I think for reaction variants, we tend to care about how strong/mild the reagents are, in terms of unwanted side reactions. Also, we care a lot of cleaning up side products for purifications. Too bad I can provide some detailed examples of which conditions fit which properties... :( $\endgroup$ – Zhe Mar 28 '17 at 20:45


Swern and Parikh–Doering are by far the most practically useful and widely used oxidations. Whilst both have slight differences, the reality is that most research groups have a preference. This may be because they just have the reagents on hand (distilled oxalyl chloride vs good quality sulfur trioxide–pyridine), or because someone had a bad experience with one over the other.

The Anderson group (in the post you linked) commonly uses the Parikh–Doering conditions, whilst other groups appear to favour the Swern conditions using oxalyl chloride in place of sulfur trioxide–pyridine.

A brief overview of various DMSO oxidations and their drawbacks:

One thing to mention about all of these DMSO based reactions is the formation of stoichiometric dimethylsulfide (stench) as a by-product, which increasingly makes them avoided, especially by process chemists. That said, the reagents are cheap (compared to a Dess–Martin, for example), and the reaction is generally very reliable and amenable to use on moderate scales. The DMSO and triethylamine must also be distilled for optimal results (which can be a drawback, especially if large scales are needed).

1. Moffatt oxidation

Reagents: DMSO, DCC, PPA.

Reference: Pfitzner, K.; E.; Moffatt, J. G. A New and Selective Oxidation of Alcohols. J. Am. Chem. Soc. 1963, 85 (19), 3027–3028. DOI: 10.1021/ja00902a036.

The Moffatt oxidation was one of the earliest examples of a DMSO-mediated oxidation. Two issues generally exist. Firstly, a strong acid (polyphosphoric acid, TFA, etc.) is required to get the reaction going, and, although the reaction becomes basic as it proceeds, the initial conditions using polyphosphoric acid/similar quite often have unwanted side effects (stripping of protecting groups, from experience). Secondly, and more importantly from a scale point of view is the stoichiometric N,N'-dicyclohexylurea produced as a by-product (similar to the Steglich esterification).

Due to the difficult-to-remove stoichiometric by-products, and the harsher conditions, Moffatt oxidations have largely been forgotten/ignored. Reagents such as EDCI have been introduced, however the high cost relative to oxalyl chloride/sulfur trioxide–pyridine have limited the appeal of Moffatt oxidations.

2. Albright–Goldman oxidation

Reagents: DMSO, acid anhydride

Reference: Albright, J. D.; Goldman, L. Dimethyl Sulfoxide-Acid Anhydride Mixtures. New Reagents for Oxidation of Alcohols. J. Am. Chem. Soc. 1965, 87 (18), 4214–4216. DOI: 10.1021/ja01096a055.

The Albright–Goldman oxidation commonly uses acetic anhydride, and goes at room temperature - although with significant reaction times which can give side reactions time to occur.

The biggest drawback, and reason why the Albright–Goldman is rarely used, is due to a competing Pummerer reaction (which is often significant), along with protection of the alcohol as its acetate ester if unhindered.

Some improvements have been made using TFAA in place of acetic anhydride. However, mixing TFAA and DMSO causes a violent exotherm unless the temperature is carefully and cautiously controlled.

3. Parikh–Doering

Reagents: DMSO, sulfur trioxide–pyridine, triethylamine

Reference: Parikh, J. R.; Doering, W. v. E. Sulfur trioxide in the oxidation of alcohols by dimethyl sulfoxide. J. Am. Chem. Soc. 1967, 89 (21), 5505–5507. DOI: 10.1021/ja00997a067.

Along with the Swern (see below), the Parikh–Doering is one of the most widely used DMSO based oxidations, with many examples in the synthetic literature.

The Parikh–Doering is able to be ran at mild temperatures (0 °C to room temperature), and the by-products are easily removed (crashing out and filtration). One advantage to the Parikh–Doering over the Swern is the fact that sulfur trioxide–pyridine is a bench stable solid, compared to oxalyl chloride which is a liquid that must be distilled and stored in the freezer to prevent degradation.

4. Swern oxidation

Reagents: DMSO, oxalyl chloride, triethylamine

Reference: Omura, K.; Swern, D. Oxidation of alcohols by “activated” dimethyl sulfoxide. a preparative, steric and mechanistic study. Tetrahedron 1978, 34 (11), 1651–1660. DOI: 10.1016/0040-4020(78)80197-5.

The Swern is the newest of the widely used DMSO oxidations, and often represents the best combination of reliability, cost of reagents, and ease of reaction.

Unlike the Parikh–Doering, the initial stages of the reaction must be ran at −78 °C to avoid unwanted decomposition. However, the conditions are equally as mild. One advantage over the Parikh–Doering is that oxalyl chloride decomposes into gaseous by-products, meaning that the only thing needed to be removed is the hydrochloride salt of triethylamine. Swern reactions can often be filtered, concentrated, and telescoped into the subsequent reaction without attrition in yield.


The majority of the chemical information here comes from Comprehensive Organic Synthesis, Vol. 7. A whole chapter is dedicated to DMSO based oxidations and their examples in total synthesis: see Lee, T. V. Oxidation Adjacent to Oxygen of Alcohols by Activated DMSO Methods. DOI: 10.1016/B978-0-08-052349-1.00191-8.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.