Piancatelli–Margarita Oxidation and Its Recent Applications in Organic Synthesis

Abstract

Piancatelli–Margarita oxidation is a reaction where primary and secondary alcohols are converted to aldehydes and ketones, respectively. It utilizes TEMPO (2,2,6,6-tetramethylpiperidine 1-oxyl), a stable aminoxy radical, as the catalyst and BAIB (bis(acetoxy)iodobenzene), a hypervalent iodine compound, as the stoichiometric oxidant. The reaction proceeds at room temperature, without the need for strong acids, bases, or anhydrous conditions. Mild reaction conditions allow for the chemoselective oxidation of complex and sensitive substrates and the selective oxidation of primary alcohols in the presence of secondary alcohols. The reaction conditions can be controlled to favor the oxidation of primary alcohols to aldehydes or promote the overoxidation of aldehydes to carboxylic acids. This review highlights some recent applications (2020–2025), especially in total synthesis, with special emphasis on large-scale reactions. This review aims to honor the memory of Prof. Piancatelli (1936–2025) and Dr. Roberto Margarita (1970–2016), who developed this reaction.

1. Introduction

Back in 1997, funding and instrumentation were limited in the chemistry department at Roma Sapienza. No exact mass spectrometer was available; the NMR instrument was an old Gemini 200 MHz, and there was limited access to a 300 MHz instrument. However, something was not limited: scientific creativity, and most of all, scientific freedom. Professor Giovanni Piancatelli allowed his first-year PhD student Roberto Margarita to be free to test new ideas and reactions. While most of the projects they attempted led to non-useful reactions, one of them afforded a significant breakthrough. They developed a very interesting oxidation methodology using a catalytic amount of TEMPO 1 ((2,2,6,6-tetramethylpiperidin-1-yl)oxyl) or (2,2,6,6-tetramethylpiperidin-1-yl)oxidanyl) and BAIB 2 (bis(acetoxy)iodobenzene, also known as PIDA, phenyliodine(III) diacetate, or phenyl-λ³-iodanediyl diacetate), which acted as the stoichiometric oxidant. The combination of these two reagents leads to a unique methodology for the oxidation of primary and secondary alcohols to aldehydes and ketones (Scheme 1), with the possibility of further in situ overoxidation of the aldehyde to a carboxylic moiety if the reaction conditions are forced. Since the publication of the original paper in 1997 [1], this methodology has been applied several times on various complex substrates, on small and large scales. This reaction should not be confused with the Piancatelli rearrangement, a unique chemical transformation also developed by Prof. Piancatelli in 1976, which was nearly forgotten for decades and recently surged to popularity [2].

The initial Journal of Organic Chemistry article on the oxidation of alcohols, co-authored by Prof. Piancatelli and Dr. Margarita, has been cited over 750 times to date. Sadly, Roberto Margarita, after a brilliant career in the industry with BMS and CordenPharma, departed in 2016. In June 2025, Prof. Piancatelli also passed away.

Researchers have applied this transformation hundreds of times over the last decades, but there appears to be no existing review specifically on this subject. Those who were the most entitled to write it, Prof. Piancatelli and Dr. Margarita, published only another paper on this subject, an invited contribution on Organic Synthesis [3,4]. A measure of success for a new transformation in organic chemistry is how many times unrelated researchers cite this work, rather than relying on many self-citations, which is the case with the Piancatelli–Margarita oxidation.

The purpose of this review is to describe a useful tool in organic synthesis and remember two brilliant scientists (Figure 1) through the well-known reaction they developed together.

Additionally, the review will present some selected examples of complex molecules and total syntheses to illustrate its wide applicability. To keep the material to a manageable amount and give a timely report, I decided to select (with few exceptions) only papers published from the year 2020 and on. Special attention was dedicated to large-scale (tens of grams) reactions, because nowadays only transformations that use environmentally benign conditions are normally scaled up to these amounts. The fact that such large-scale reactions are reported is indirect evidence of the robustness of this reaction. I hope that this review can be of interest to academic and industrial scientists who need an easy, safe, and reliable protocol for the oxidation of alcohols.

2. The Original Piancatelli–Margarita Oxidation

2.1. TEMPO, BAIB, and Comparison with Other Hypervalent Iodine Compounds as Oxidants

TEMPO 1, a red-orange solid, is a stable aminoxyl radical, with several applications in synthesis, mechanism elucidation (as a radical trap), and biochemistry [5]. It is obtained by oxidation of 2,2,6,6-tetramethylpiperidine 3, and it is a stable N-O free radical due to the steric hindrance around the oxygen atom. TEMPO 1 and its derivatives have been used as catalytic organic oxidants for alcohols in several instances. The most well-known reaction is Anelli–Montanari oxidation [6], a biphasic (DCM-aqueous buffer) oxidation protocol that originally used bleach (NaClO) as the stoichiometric oxidant, although other oxidants have also been employed, for example, trichloroisocyanuric acid [7] or N-chlorosuccinimide [8]. The main difference between Anelli–Montanari oxidation and Piancatelli–Margarita oxidation is that the latter is a purely organic oxidation. While Anelli–Montanari oxidation has been widely used in process chemistry [9], the Piancatelli–Margarita protocol is generally more often employed in total synthesis and medicinal chemistry.

The unsubstituted TEMPO 1 is relatively expensive (it costs about 60 US dollars for 5 g) [10], but it must be considered that it is used only in catalytic amounts. Other derivatives of TEMPO (4-hydroxy-TEMPO, 4-acetamido TEMPO), which are used in industry, are much cheaper when obtained in bulk amounts [9].

BAIB 2 (or PIDA), a white solid, is a hypervalent iodine compound, since the iodine atom has a +3 oxidation state [11]. Notably, hypervalent iodine (III) compounds are iodosobenzene 4 (PhI=O), which is explosive under heating and impact, and IBX 5 (1-hydroxy-1λ⁵,2-benziodoxole-1,3-dione), which is an oxidized form of 2-iodobenzoic acid 6, which is also explosive and scarcely soluble in organic solvents. To avoid the issues linked to IBX 5, especially solubility and danger of explosion, a procedure for the tri-acetoxylation of IBX 5 has been developed [12], and the resulting compound, 3-oxo-1λ⁵,2-benziodoxole-1,1,1(3H)-triyl triacetate, better known as Dess–Martin periodinane 7, is widely used in organic synthesis [13]. The major drawback of this compound is that it is generally prepared from IBX 5, which itself must also be prepared from 2-iodobenzoic acid 6. Both compounds are commercially available: IBX costs around 750 USD for 50 g [14], and Dess–Martin periodinane 7 costs around 740 USD for 50 g [15] (see Scheme 2).

On the other hand, BAIB 2 is less expensive. It costs around 220 USD for 100 g, and 1 kg can cost as little as 800 USD [16]. The potential risks related to explosions should be carefully evaluated when using any oxidant, especially in large-scale chemistry. According to its published SDS (substance data sheet), in terms of hazard identification, BAIB 2 is classified as “not a hazardous substance or mixture”. In terms of GHS (Globally Harmonized System of Classification and Labelling of Chemicals), it is described as “no hazard pictogram, no signal word, no hazard statement(s), no precautionary statement(s) required” [17]. It can be considered generally safe if handled with precaution, while Dess–Martin periodinane 7 is classified as flammable and irritating, and IBX 5 is also corrosive and presents a significant danger of explosions. One (large) explosion of IBX 5 on a 200 g scale was directly witnessed by the author during his postdoc at The Scripps Research Institute, San Diego, CA, and the extent of damage to people and things was significant. Besides personal experience, BAIB 2 is safer and cheaper with respect to other hypervalent compounds used as oxidants.

2.2. Reaction Conditions and Substrate Scope in the Original Paper

In the original paper [1], many substrates were subject to the oxidation protocol, which uses 0.1 equiv. of TEMPO 1, 1.1 equiv. of BAIB 2, DCM as the solvent, and room temperature. A catalytic load of 0.05 equiv. of TEMPO 1 still leads to the formation of some of the desired products, while no reaction occurred with 0.01 equiv. Simple primary alcohols such as compounds 8a–c can be oxidized to the corresponding aldehydes 8′a–c within a couple of hours. The reaction also works on carbohydrate derivatives such as 8d, and more recent examples will be presented in a specific section of this review.

The reaction is faster on the allylic alcohols 8e–8h. Notably, no E-Z isomerization of the double bond occurs in the Z configurationally unstable compounds 8′f–g. The oxidation of nerol 8h to neral 8h′ is reported in the journal Organic Synthesis to set up a standard protocol [3] (see Scheme 3).

Primary and secondary benzylic alcohols 8i–j are readily oxidized to the corresponding carbonyl compounds 8′i–j. Also, secondary alcohols 8k–l are converted to ketones 8′k–l; however, longer reaction times are required. Alcohols 8m–n, which contain an easily oxidizable function as the sulfur or selenium atom, are selectively oxidized to carbonyl compounds 8′m–n, in which only the hydroxy group is oxidized. Finally, furyl carbinol 8o is smoothly transformed into ketone 8′o. The attempted oxidation of furyl carbinol using acidic conditions is the reaction that led to the serendipitous discovery of the Piancatelli rearrangement in 1976 [2]. Under these mild reaction conditions, no cyclopentanone rearranged product was observed (see Scheme 3).

Piancatelli and Margarita also studied chemoselectivity for their reaction concerning different alcoholic functions. Thus, the primary alcohol functionality of compound 8p can be selectively oxidized in the presence of secondary alcohol 8q, affording the aldehyde 8′p but not the ketone 8′k. The primary allylic alcohol 8b can be selectively oxidized to aldehyde 8′b in the presence of secondary allylic alcohol 8r (the formation of ketone 8′r is not observed). Benzyl alcohol 8s in the presence of secondary benzylic alcohol 8j is exclusively oxidized to benzaldehyde 8′s, without observing the formation of acetophenone 8′j (see Scheme 4).

Piancatelli and Margarita also investigated the reaction in different solvents. They found a correlation between the dielectric constant and the reaction rate. Non-polar solvents such as n-hexane led to a slow reaction; instead, polar solvents such as acetonitrile or DMSO give rise to a faster oxidation of alcohols. Although they did not investigate the formation of carboxylic acids, it is expected that using more polar solvents (for example, a mixture of water and acetonitrile), the resulting aldehydes can be oxidized to the corresponding carboxylates. Indeed, this was successively reported by other authors.

2.3. Key Features of the Piancatelli–Margarita Oxidation

-

The reaction is catalytic with respect to the oxidant; the reagents are non-toxic or explosive and can be considered environmentally benign.

-

For most reactions, only slightly more than one equivalent of BAIB 2 is needed. Solid BAIB 2 can be weighed and dosed in a better manner with respect to a solution of NaClO. Therefore, the Piancatelli–Margarita reaction can also be performed on a very small scale, which is the one employed, for example, in the last stages of a total synthesis.

-

The reaction is purely organic. No water is necessary.

-

With primary alcohols, the reaction can lead to the selective formation of aldehydes. The conditions for selective oxidation are short reaction times (one or two hours), room temperature, and the usage of a low-polar solvent such as DCM. If the reaction conditions are forced (longer reaction times, mixtures of acetonitrile and water as the solvent, higher reaction temperatures such as 70 °C), this transformation leads to the formation of carboxylic acids (see following examples for details).

-

In some instances, overoxidation cannot be avoided, especially when five-member or larger cyclic hemiacetals are formed. In these cases, only lactones are formed (see following Schemes, with related references).

-

The reaction is chemoselective: primary alcohols can be selectively oxidized in the presence of secondary alcohols, and easily oxidizable sulfur/selenium functionalities are not affected.

2.4. Proposed Reaction Mechanism

The free radical TEMPO 1 can react with alcohols in its form 1-A to afford the corresponding carbonyl compound and its reduced form 1-B; the role of BAIB 2 is to re-oxidize 1-B to continue the catalytic cycle. The rate-determining step of the reaction is often the oxidation of TEMPO 1 to TEMPO 1-A. This specific step can be catalyzed by bromine anions (Scheme 5).

3. Selected Examples of Piancatelli–Margarita Oxidation Applications in Synthesis: Large-Scale Reactions and Small Molecules

In this section, some examples from the recent literature (since 2020) will be presented. While reactions conducted on large molecules can be impressive, transformations of molecules with very few carbon atoms can also be difficult. Not always does the synthesis of “small” molecules equal “easy”. First, a small molecule usually has a boiling point that is close to the solvent employed. Second, some small molecules (e.g., those with a hydroxy- or amino- group) are often water-soluble; therefore, their purification via phase extraction can be difficult. Third, small molecules, especially if bearing one or more reactive groups, can be quite unstable, making reactions difficult to reproduce if conditions are not exactly controlled.

Keeping these aspects in mind, O’Reilly and coworkers attempted the oxidation of the 3-chloro propanol 9 to 3-chloro propanal 9′ with catalytic TEMPO 1 and BAIB 2 [18]. The yields are given as a range (50–70%), since during purification and concentration in vacuo some decomposition of the aldehyde 9′ to acrolein via HCl elimination occurred. This was not a major issue, since aldehyde 9′ can be telescoped in a DCM solution to the next reaction. However, in their synthesis of azetidines, they ultimately chose a different procedure based on the hydrolysis of acrolein diacetal, because it was found to be more convenient. Nevertheless, the TEMPO 1/BAIB 2 oxidation has been proven to be effective in preparing the elusive aldehyde 9′ (Scheme 6).

For their preparation of the anticancer agent BI 1810284, the industrial researchers of the group of Reevs reported an eighty-gram scale Piancatelli–Margarita oxidation of alcohol 10 to aldehyde 10′. In this reaction were employed 80 g of alcohol 10 (MW = 404.5 D, 90% purity, 0.18 moles), 2.8 g of TEMPO 1, and 68 g of BAIB 2. After work-up and concentration, 116 g of a mixture of aldehyde 10′ (MW = 402.5 D, 58% purity, 0.17 moles) and the residual iodobenzene were obtained. An aliquot was purified, and the purity of the aldehyde was evaluated to be around 95%. This means that 0.17 moles of aldehyde 10′ were obtained starting from 0.18 moles of alcohol 10 [19]. Since purifications are expensive on a large scale, it is not unusual that products are directly telescoped to the following steps when they are unnecessary. The by-product of Piancatelli–Margarita oxidation is iodobenzene, a relatively inert compound. This renders the reaction attractive for large-scale industrial reactions (Scheme 7).

In their total synthesis of pladienolide A, Rhoades, Rheingold, O’Malley, and Wang depicted a protective group-free sequence. One of the key steps was the oxidation of the primary alcohol functional group of the intermediate 11 to the aldehyde 11′, despite the presence of a secondary alcohol and of a reactive epoxide functionality. This task was successfully achieved by using the Piancatelli–Margarita oxidation, obtaining aldehyde 11′ in 80% yield [20] (Scheme 8).

Gayon, Lefevre, and coworkers, in their large-scale synthesis of the sex pheromone of the horse-chestnut leaf miner, planned the oxidation of alcohol 12 to the desired target molecule 12′ [21]. The final reaction was run on a fifty-gram scale, using 51 g of alcohol 12, 3.8 g of TEMPO 1, and 85 g of BAIB 2 (Scheme 9). According to the authors, this synthesis is easily scalable. It is worth noting that the oxidation of alcohols at the end of long hydrophobic chains (fourteen carbon atoms, in this case) is not easy with traditional oxidants (for example, with chromium (VI) species) because of the differences in polarity between the reagents. In this case, the reaction is successful since TEMPO 1 could be hydrophobic enough to reach more effectively the hydroxyl functionality. Biodegradable pheromones can be a valuable alternative method in insect control if compared to other synthetic insecticides. Nevertheless, the preparation of pheromones must employ eco-friendly conditions, and the development of a synthesis with these characteristics was the goal of the industrial researchers.

During their asymmetric synthesis of (-)-dehydrocostus lactone, Metz and coworkers employed oxidation reactions in several steps [22]. In some instances, these authors exploited Piancatelli–Margarita oxidation. They commenced their synthesis with the preparation of aldehyde 13′ from alcohol 13, on a five-gram scale. Aldehyde 13′ was obtained in quantitative yield either with Piancatelli–Margarita oxidation or Swern oxidation (Scheme 10).

A useful protective group for the alcoholic moiety is the PMB or p-methoxybenzyl group. This protective group can be removed either with catalytic hydrogenation or using a mild oxidant, giving back the free alcohol functionality. Gosh and Hsu, in their total synthesis of (+)-EBC-23, an anticancer agent from the Australian rainforest, showed that the PMB protective group of intermediate 14 is not affected by the Piancatelli–Margarita oxidation protocol, giving aldehyde 14′ in good yield [23]. This transformation is part of a four-step sequence in which the intermediate alcohol 14 is telescoped from the previous reaction in DCM solution. An 80% yield is reported for the entire sequence (Scheme 11).

In their total synthesis of the pseudopterosin A-F aglycone, Schmalz and coworkers employed the Piancatelli–Margarita oxidation on primary alcohol 15 to give aldehyde 15′ [24]. The reaction proceeded in good yield despite the presence of an electron-rich aromatic ring, which can, in some instances, be easily oxidized but in this case is not affected. The yield of this reaction (including the previous double bond hydrogenation) is 75% (Scheme 12).

An example of the chemoselectivity of Piancatelli–Margarita oxidation was reported by Hanessian and coworkers in their synthesis of insecticide metabolite yaequinolones J1 and J2. These authors needed an oxidation method for primary alcohol 16 that would afford an intermediate, the N-protected hemiaminal 16′, which can undergo further oxidation to a lactam, specifically N-Boc dihydroquinolin-2-one 16′. They tested several oxidants (PDC, Corey−Schmidt, and PCC), but none of them was as effective and selective as the TEMPO 1/BAIB 2 combination. The hemiaminal intermediate 16 could be detected by NMR analysis after 3 h before further oxidation to the lactam took place. The authors state that, to the best of their knowledge, the TEMPO/BAIB reagent combination was used here for the first time for the direct synthesis of an N-Boc lactam from a primary alcohol [25]. It is worth noticing that neither the enol ether functionality, the aryl methoxy group, nor the labile acid-sensitive benzylic tertiary alcohol was affected (Scheme 13).

To develop a synthesis of chloroalkene dipeptide isosteres, Tamamura et al. resorted to Piancatelli–Margarita oxidation for homoallylic alcohol 17 to aldehyde 17′ in 93% yield [26] (Scheme 14). The following step of their synthesis is the oxidation of the aldehyde functionality to a carboxylate, which was achieved by means of Pinnick oxidation. It is surprising that the authors used a two-step oxidation protocol, since forcing the reaction conditions would presumably lead to the same carboxylate. However, no explanation is given in the original paper.

Another example of oxidation of the primary alcohol in the presence of a double bond and a secondary alcohol was reported by Sarpong and coworkers in their studies toward the synthesis of diverse taxane cores, specifically in the oxidation of alcohol 18 to aldehyde 18′, to give, after TBS protection of the secondary alcohol functional group, the taxane C-ring fragment [27]. After this two-step sequence, aldehyde 18′ could be isolated in 78% yield (Scheme 15).

In the original paper by Piancatelli and Margarita [1], it was reported that the phenylthiol and phenylselenium functional groups are not affected by their oxidation methodology. This is noteworthy, since the phenylthiol and phenylselenium groups are usually introduced into molecules to generate a double bond upon oxidation, for example with NaIO₄. Their findings on the selective alcohol group oxidation are confirmed by the work of Gardiner et al. in their efforts toward the synthesis of heparan sulfate- and dermatan sulfate-related oligosaccharides. The Piancatelli–Margarita oxidation of alcohol 19 afforded [2.2.2] bicyclic lactone 19′ in 73% yield using the “forced” conditions (acetonitrile and water) [28] (Scheme 16).

Sarpong and coworkers solved an intriguing problem in their studies toward the synthesis of the longiborneol sesquiterpenoids [29]. Their goal was to oxidize the alcohol functionality of intermediate 20 to the corresponding aldehyde in compound 20′.

However, when they used Dess–Martin periodinane 7 or other oxidation methods (Ley–Griffith oxidation), their product was instead carboxylic acid 20″. They hypothesized that the overoxidized compound 20″ can be derived from the fast oxidation of internal hydrate acetal 21. Then, testing the standard conditions of Piancatelli–Margarita oxidation (Scheme 17, conditions A), none of the desired aldehyde 20′ was isolated. They also hypothesized that an analogue internal acetal of compound 21, specifically compound 21′, can prevent oxidation. They then tested two new approaches: the addition of a second equivalent of BAIB 2 (conditions B) and the addition of 3 equivalents of acetic acid (conditions C) to break the internal acetal. Both these modifications were successful, delivering the desired aldehyde 20′ in 99% yield (Scheme 17).

The formation of internal hemiketal was also an issue faced during the synthesis of the ABC ring system of kadlongilactones reported by Wang and Chen [30]. The ketone 22a existed as a mixture with emiketal 22b, which was detected using NMR. Also, in this case, the oxidation to the corresponding lactone 22′ required more equivalents of BAIB 2 (2.5 equiv. according to the supporting information of the article by Wang and Chen cited in ref [30]; see Scheme 18).

Being a versatile and robust reaction, the Piancatelli–Margarita oxidation can be incorporated in some standard synthetic sequences. One of these is the primary alcohol oxidation–Wittig reaction. An interesting example is reported by Cordero-Vargas, Sartillo-Piscil, and coworkers in their synthesis of (+)-lasionectrin [31]. Tetrahydrofuran intermediate 23 was first subject to simultaneous double bond reduction and primary alcohol deprotection; then, the oxidation of the alcoholic function gave intermediate 23′. This compound was then subject to the Wittig–Horner reaction with the resulting methyl (triphenylphosphanylidene)acetate and hydrolysis of the methyl ester to give, after DCC-mediated intramolecular lactonization, the desired compound 24 in 25% overall yield (Scheme 19).

A similar example was reported by Takamura and coworkers during their synthetic approach toward the preparation of scabrolide F [32]. Thus, double-protected intermediate 25 was first oxidized with the standard Piancatelli–Margarita protocol to give aldehyde 25′ and then subjected to the Wittig–Horner reaction to give alkene 26 in 91% overall yield (Scheme 20).

It is well-known that esters can be reduced to the corresponding aldehydes using DIBAL-H reduction. However, it can be difficult to exactly dose the amount of the reducing reagent, and since aldehydes are easily reduced with respect to esters, often the result is the formation of a primary alcohol or incomplete reduction of a mixture of compounds. From a practical point of view, it can be more convenient to use a two-step sequence: first, total reduction of the ester to primary alcohol and then partial re-oxidation to the aldehyde. Piancatelli–Margarita oxidation can be conveniently and successfully employed in the second step. A recent example can be found in the work of Saito, Shimokawa, and Yorimitsu in their studies of the dioxasilepanyl groups [33]. These authors first reduced ester 27 to primary alcohol 28 and then re-oxidized it to the desired compound 28′ in 85% and 91% yields, respectively (77% overall, Scheme 21).

Kaspar and Kudova investigated the selectivity of oxidizing agents toward axial and equatorial hydroxyl groups. Among the several oxidants tested, there was the combination TEMPO 1/BAIB 2. Methyl chenodeoxycholate 29 bears both an axial and an equatorial group in blocked positions. When this compound was reacted with catalytic TEMPO 1 and an increasing amount of BAIB-2, only the formation of the axial-oxidized compound 29′a was observed. No amount of the equatorial-oxidized compound 29′b or diketone 29′c could be detected [34]; see Scheme 22.

This selectivity was reversed with other hypervalent iodine (III) oxidants such as Dess–Martin periodinane 7, confirming that in Piancatelli–Margarita oxidation the oxidizing molecule is TEMPO 1-A and not the hypervalent iodine compound. The authors hypothesized that this selectivity arises from the steric hindrance around the C7 hydroxy group, leaving only the C3 hydroxy group accessible (see mechanism in Scheme 23).

The authors also studied the oxidation of cis 4-t-butyl cyclohexanol 30_ax, in which, due to the bulky t-butyl group, the hydroxy moiety is conformationally blocked in the axial position, and of trans 4-t-butyl cyclohexanol 30_eq, where the hydroxy group is instead fixed in the equatorial position. However, in this case they found that ketone 30′ was formed, consuming equal amounts of cis 4-t-butyl cyclohexanol 30_ax and of trans 4-t-butyl cyclohexanol 30_eq. They hypothesize that the reason why this reaction is nonselective now is that no significant steric hindrance is present around the axial or equatorial hydroxy group (Scheme 24). Their results are relevant because they suggest that the TEMPO 1/BAIB 2 combination can be quite sensitive to steric factors.

4. Noteworthy Applications of Piancatelli–Margarita Oxidation in Carbohydrate Chemistry: Examples from the Recent Literature

Oxidation in carbohydrate chemistry is an especially challenging transformation since it operates on polyoxygenated structures, which, in the case of polysaccharides, are also acid-sensitive moieties. A first example of the oxidation of a protected sugar moiety was reported in the original paper by Piancatelli and Margarita [1], specifically the oxidation of five-membered sugar 8d to aldehyde 8d′. As seen before, by forcing the reaction condition, the carboxylic acid moiety can also be obtained.

In a recent Paper published by Li et al. on Journal of the American Chemical Society, the authors needed an efficient oxidation protocol for the transformation of the primary alcohol function of protected sugar 31 into carboxylate 31′ [35]. This is one of the first reactions toward the synthesis of 6-deoxy-D-/L-heptopyranosyl fluorides and was conducted on a 1 g scale (see supporting information of this article) with catalytic TEMPO 1 (0.3 equiv.) and excess BAIB 2 (2.5 equiv.) to afford compound 31′ in excellent yield (98%; see Scheme 25).

A similar oxidation procedure was incorporated in a three-step sequence, as reported by Boons in 2020, for the preparation of a disaccharide unit for the synthesis of heparan sulfate oligosaccharides [36]. These consecutive reactions were run on a 1 g scale. The consecutive reactions involved first the TFA-mediated selective cleavage of the acetal on disaccharide 32; then, the TEMPO 1-catalyzed/BAIB oxidation of the primary alcohol of intermediate 33 to give carboxylate 33′ in a mixture of DCM; and finally, the formation of the methyl ester 34 using trimethyl silyl diazomethane (TMSCHN₂). The entire sequence has a satisfying yield of 61%, despite the presence of several functional and protective groups (see Scheme 26).

While working toward the total synthesis of the natural substance capuramycin, Xiao and coworkers needed to oxidize the primary alcohol function on compound 35 to a carboxylate moiety [37]. The usual combination of catalytic TEMPO and BAIB in acetonitrile and water (the more forcing conditions using a more polar solvent mixture; see introduction) proved effective, accessing intermediate 35′ in over 90% yield (Scheme 27).

An impressive example of TEMPO 1/BAIB 2 multiple oxidations on complex molecules was reported by the group of Codée while synthesizing a set of Staphylococcus aureus capsular polysaccharides [38]. These authors needed to oxidize a primary alcohol functional group to a carboxylic moiety. They modified the original protocol, using instead a mixture of ethyl acetate and t-butanol/water with the addition of sodium bicarbonate. They tested their improved condition reaction on some substrates and even on the complex molecule hexasaccharide 36, which also contains an easily oxidizable protective group (PMB = p-methoxybenzyl). As mentioned before, this moiety can normally be cleaved by mild oxidation with DDQ (2,3-dichloro-5,6-dicyano-1.4-benzoquinone). Despite this, hexasaccharide 36′ was obtained in 65% yield after 12 days, and, if the reaction time is shortened to 6 days and the reaction temperature is increased to room temperature, the yield rises to 75% (Scheme 28).

5. Selected Examples of Piancatelli–Margarita Oxidation Applications in Total Synthesis: Late-Stage Intermediates and Endgame

The strength, effectiveness, and usefulness of a methodology are ultimately proven by its application in total synthesis. Only a robust methodology can pass this crucial test, which involves selectivity (are other functional groups affected?), large- and small-scale tests, and efficacy on a complex molecule. While the reaction on a simple substrate can be incorrectly reported, a low-yielding transformation in a multistep sequence can dramatically drop the global yield to a level at which no significant amount of the final product can be detected. Therefore, when a reaction is incorporated in one or more complex successful total syntheses, this is the best possible guarantee that it works, at least to some extent.

Reactions on molecules with high molecular weight represent a unique challenge. Reactions on similar small molecules can give a similar outcome. As an example, it is expected that an oxidation on a five-carbon primary alcohol would be a similar reaction to a six-carbon primary alcohol, but the outcome might change dramatically if the same reaction is then run on a primary alcohol that has twenty carbon atoms or more. Organic molecules with few atoms can have a limited number of possible conformations. The number of conformations increases exponentially with the number of atoms; therefore, the functional group can be “hidden” in a specific conformation. This is especially true with molecules with long chains. Domain separation could occur between the strongly hydrophobic substrate and a water-soluble oxidant. Therefore, finding a successful reaction on long-chain molecules is not straightforward. As seen before (ref. [21]), TEMPO 1 could be hydrophobic enough to reach the “reaction pocket”. Chambron and coworkers were studying an enantiopure bifunctional chelator for ⁸⁹Zr-immuno PET. Specifically, they were looking for a tetradentate ligand or zirconium [39]. One of the reactions they needed was the oxidation of primary alcohol 37, whose molecular weight is more than 800 Dalton. Even on this complex molecule, the Piancatelli–Margarita oxidation works well, affording aldehyde 37′ in 32% yield if the usual reaction conditions are used (DCM as the solvent) and carboxylic acid 37″ in 51% yield if the solvent is changed to an acetonitrile/water mixture. These authors did not detect any of the TEMPO 1-substrate adducts (Scheme 29).

Carreira and Fadel reported in 2023 the first enantioselective total synthesis of (+)-pedrolide, a tigliane-derived diterpenoid, showing an unprecedentedly reported skeleton. One of their late-stage key transformations was the selective oxidation of the primary alcohol of intermediate 38 to give lactone 38′ in 76% yield using what they called “Piancatelli’s and Margarita’s TEMPO oxidation protocol” [40]; see Scheme 30.

A selectivity between secondary alcohol oxidations has been reported by Hao, Ding, and coworkers in their synthesis of (−)-crinipellins [41]. In this case, the different reactivity of the two secondary hydroxy groups on compound 39 was because the allylic hydroxy group can be easily oxidized. The reaction proceeded in 76% yield to give enone 39′ (Scheme 31).

A selective oxidation (primary alcohol vs. secondary alcohol) was also needed by Woo and coworkers in their enantioselective synthesis of sangliferin A. These researchers installed a hydroxy group in a stereo-controlled manner on intermediate 40 [42]. The configuration of the C36 stereocenter was correct. Unfortunately, the configuration obtained at the C35 stereocenter was instead the opposite with respect to the target compound. This was not unexpected, since hydroboration would most likely have occurred on the less hindered Re face of the C35-C36 double bond because the Si face of compound 42 is hindered by the methyl group in the C34 position. To solve this issue, the authors had to re-oxidize the secondary alcoholic function (and most likely also the primary one, although this is not explicitly stated in the paper) using Dess–Martin periodinane 7 and then operate a selective reduction on compound 43. In this case, both the methyl groups on the C34 and C36 positions direct the attack of the hydride on the less hindered Re face of the ketone on the C35 position of compound 43. Finally, the primary alcoholic function was oxidized with the standard Piancatelli–Margarita conditions to give aldehyde 43′ in 24% overall yield over four steps (Scheme 32).

Although protective group introduction and removal can be often (but not always) high yielding and straightforward, the addition of these two steps in a long total synthesis is better avoided, considering also the possible loss of precious material in the purification procedures. It is then more desirable to find instead a chemoselective reaction, rather than rely on protective groups.

The issue of selective oxidation of a primary alcohol group in the presence of a secondary alcohol functionality was encountered by Gao and coworkers in a late synthetic step of their sequence during their efforts toward the synthesis of norzoanthamine [43]. The problem was solved by employing the selective Piancatelli–Margarita oxidation. In this case, intermediate 44 had three different hydroxy groups: one primary hydroxy group and two secondary alcoholic groups, an axial and an equatorial one. Using the Piancatelli–Margarita reaction, it was possible to oxidize only the primary alcohol group to give aldehyde 44′ in 62% yield (Scheme 33).

Another impressive example of selective primary hydroxy group oxidation in the presence of a secondary one at an advanced synthetic step was reported by Paterson and coworkers in their approach toward the synthesis of actinoallolides [44,45]. Thus, the selective oxidation of alcohol 45 gave aldehyde 45′ in excellent yield (90%; see Scheme 34).

As the last example, it will be presented an important example of the work of Boger and coworkers in their synthesis of vancomycin and its analogue tetrachlorovancomycin [46,47] because it summarizes most of the aspects that have already been encountered in this review (mild condition, chemoselectivity, and applicability in telescopic and large-scale reactions and in the synthesis of complex molecules). A late intermediate in the synthesis of vancomycin (compound 46) was subject to oxidation to carboxylic acid 46′ on an impressive 25-gram scale. A t-butyl ester was then formed by using t-butyl 2,2,2-trichloroacetimidate in 82% overall yield (Scheme 35).

6. Conclusions

I hope that the examples reported can help the chemists to understand the usefulness as well as the limitations of the Piancatelli–Margarita oxidation. To the best of my knowledge, despite the reported protocol having been widely used over the years, the term “Piancatelli–Margarita oxidation” is still not yet of common use, despite having sometimes been employed in the scientific literature, as exemplified by the papers of Kudova [34] and Carreira [40]. I believe that it would be correct to call this reaction Piancatelli–Margarita oxidation to give recognition to these two outstanding scientists, and most of all, it would help the scientific community to better know a very useful chemical reaction, which should be an indispensable tool in any organic chemist’s toolbox.