Alkene Reactions

By James Ashenhurst

OsO4 (Osmium Tetroxide) for Dihydroxylation of Alkenes

Last updated: January 20th, 2024 |

Osmium tetroxide, OsO4

  • Osmium tetroxide (OsO4) is a useful reagent for the dihydroxylation of alkenes
  • The products of these reactions are 1,2-diols (“vicinal” diols), where the two C-O bonds are formed on the same face of the alkene via a concerted mechanism.
  • Dihydroxylation of alkenes with OsO4 is functionally equivalent to dihydroxylation with cold, basic KMnO4.
  • OsO4 does not dihydroxylate alkynes!
  • The vicinal diols can subsequently be cleaved with NaIO4 providing products that are eqivalent to those obtained through ozonolysis.

summary of osmium tetroxide oso4 for the dihydroxylation of alkenes

Table of Contents

    1. Osmium Tetroxide, OsO4 And The Dihydroxylation of Alkenes
    2. The Mechanism for Dihydroxylation of Alkenes With OsO4
    3. Predicting the Stereochemistry of Dihydroxylation Products
    4. OsO4 vs KMnO4 As A Reagent for Dihydroxylation
    5. Catalytic OsO4 Using Stoichiometric Oxidant
    6. Reactions of 1,2-Diols – Oxidative Cleavage With NaIO4 
    7. Summary
    8. Notes
    9. Quiz Yourself!
    10. (Advanced) References and Further Reading

1. Osmium Tetroxide, OsO4 And The Dihydroxylation of Alkenes

Osmium tetroxide (OsO4) is a very useful reagent for the dihydroxylation of alkenes [Note 1] . In this reaction,

  • A C-C (pi) bond is broken
  • Two C-O bonds form on adjacent carbons
  • The two new C-O bonds are delivered syn , which is to say, on the same face of the alkene.

For example, the reaction of cyclohexene with OsO4 gives exclusively cis-cyclohexan-1,2-diol, with none of the trans diol formed.

osmium tetroxide oso4 is a very effective reagent for dihydroxylation of alkenes syn addition

Since we are breaking a C-C bond and forming two C-O bonds, this is an example of an oxidation reaction (See article: Oxidation and Reduction in Organic Chemistry)

Two alkenes that differ only in their configuration (e.g. the stereoisomers cis– and trans– pent-2-ene) will result in products that are themselves stereoisomers.

This fits the definition of a stereospecific reaction, as per IUPAC.

the reaction of oso4 with alkenes qualifies as a stereospecific reaction resulting in diastereomeric products

Chiral molecules with exactly opposite (R,S) designations are enantiomers. Chiral molecules that share the configuration at at least one chiral center and differ at the configuration of another chiral center will be diastereomers. For more, see article: Types of Isomers)

Note that in each case the two new C-OH bonds form on the same face of the alkene.

2. The Mechanism for Dihydroxylation of Alkenes With OsO4

The mechanism of alkene dihydroxylation is a concerted cycloaddition reaction where the C-C pi bond combines with two Os=O bonds to give a five-membered ring structure known as an osmate ester. (Note that in the osmate ester the Os is in the +6 oxidation state as opposed to the +8 oxidation state found in OsO4)

This concerted mechanism nicely accounts for the cis stereochemistry observed in the dihydroxlyation of cyclohexene.

mechanism for dihydroxylation of alkenes with oso4 is a concerted cycloaddition explains syn addition

Osmate esters are fairly stable products and can be isolated. [Note 2] However, since we are generally much more interested in the diol, a reagent such as potassium bisulfite (KHSO3) or sodium bisulfite (NaHSO3) is commonly used to break the Os-O bonds and liberate the diol.

Just a heads-up – in introductory courses, this second reagent may or may not be included. It purpose is just to get rid of the osmium.

(It is much more common nowadays to use catalytic OsO4 and a stoichiometric amount of an oxidant such as N-methylmorpholine N-oxide (NMO) or H2O2 to regenerate OsO4 from the Os(VI) species. In these cases, KHSO3 is not needed. See section below.)

As a fairly electron-poor reagent, reactions with OsO4 increase in rate as the alkene becomes more electron-rich.

For practical purposes, this means that

  • reaction rates generally increase with increasing substitution on the alkene ( tetrasubstituted (fastest) > trisubstituted > disubstituted > monosubstituted (slowest)
  • reaction rates generally decrease if the alkene is attached to electron-withdrawing groups such as carbonyls

It’s possible to selectively dihydroxylate an electron-rich alkene in the presence of other alkenes. For a few examples, see Note 3 below.

3. Predicting The Stereochemistry Of Dihydroxylation Products

cis– and trans- alkenes can be each be prepared from alkynes, depending on the reagent(s) used for reduction.

  • Alkynes treated with sodium (Na) in ammonia (NH3) gives trans-alkenes. (See article – Partial Reduction of Alkynes Using Na/NH3)
  • Alkynes treated with Lindlar’s catalyst (palladium made less active through the addition of lead and quinoline) in the presence of hydrogen gives cis-alkenes (See article –  Lindlar’s Catalyst)

Why is this important right now?

Well, since cis– and trans– alkenes give dihydroxylation products that are stereoisomers of each other, dihydroxylation reactions provide great fodder for exam questions that challenge your understanding of stereochemistry.

See if you can answer this classic quiz question:

Click to Flip

For a refresher on solving these kinds of stereochemistry problems, see article – Enantiomers, Diastereomers or the Same?

Just as important as determining the stereochemistry of products is being able to work backwards from the products of dihydroxylation to the starting alkenes.

This is more challenging with linear (as opposed to cyclic) products, since it will require that you successfully perform a bond rotation.

See if you can work backwards from this diol to the starting alkene:

Click to Flip

Here is a similar example:

Click to Flip

4. OsO4 versus KMnO4 As A Reagent for Dihydroxylation

A reagent similar to OsO4 that is also capable of performing dihydroxylation is potassium permanganate, KMnO4.

Treatment of alkenes with cold, alkaline KMnO4 will also result in vicinal diols.

comparing the dihydroxylation of alkenes using KMno4 vs OsO4 both are syn additions

Like the reaction with OsO4, this also proceeds through a cyclic, concerted transition state that results in a cyclic metal species (this time called a “manganate ester”).

The key difference here is that unless  the manganate ester will react further to give the products of oxidative cleavage unless hydroxide ion HO(-) is present to hydrolyze the Mn-O bonds and liberate the vicinal diol. This is not generally a problem with OsO4.

This is also why the temperature is kept low for KMnO4 oxidations.

Yields with KMnO4 tend to be lower and KMnO4 is also much less tolerant of sensitive functional groups like alcohols and aldehydes. 

Dihydroxylations with KMnO4 are often used with a phase transfer catalyst. 

5. Catalytic OsO4 Using Stoichiometric Oxidant

It’s one thing to write a reaction down on a sheet of paper that uses a stoichiometric amount of osmium tetroxide.

It’s another thing entirely to carry it out in the lab.

For one thing, OsO4 is expensive$332/g last time I checked, slightly cheaper if you buy in bulk. The other thing is that it is a highly toxic liquid with a low vapor pressure that should be treated with extreme care.

Particularly noteworthy is its potential to cause blindness – all that retinol in your cornea is full of juicy double bonds that OsO4 would love to hydroxylate. 

The report on the first synthesis of cortisol from 1952  (see Note 3 below) has a reaction that used 68.48 g of OsO4 . That clocks in at, let’s see…  $22,768 worth of OsO4 at today’s prices.

Surely there must be a better way? Thankfully, yes.

the upjohn process is catalytic dihydroxylation of alkenes using OsO4 and stoichiometric nmo as stoichiometric reoxidant

The Upjohn process uses a catalytic amount of OsO4 (usually about 1-2 mol% ) in the presence of a stoichiometric amount of oxidant that converts the Os(VI) product back to OsO4. The oxidant of choice is generally N-methylmorpholine N-oxide (NMO) although various other oxidants can be used.

[The original paper is here  – Org Synth. 1978, 58, 43 – and has helpful tables that compare oxidants and also its performance to KMnO4]

Yields are generally high and the reaction is mild. Furthermore there’s no need to add KHSO3 since the osmate ester is cleaved under these conditions.

It’s even possible to perform a hydroxylation on an alkene without affecting an alkyne, as OsO4 does not react with alkynes.

alkynes are unreactive with oso4 and alkenes will undergo selective oxidation

An enantioselective version of dihydroxylation known as the Sharpless asymmetric dihydroxylation has been developed. It also uses catalytic osmium (potassium osmate)  in the presence of a stoichiometric amount of oxidant. For more details see [Note 4].  

6. Reactions of Vicinal Diols – Cleavage with NaIO4

vicinal diols can undergo oxidative cleavage with various reagents to break a C-C bond and form two new C-O (pi) bonds.

The most commonly used reagents for these purposes are sodium periodate (NaIO4) and lead tetraacetate Pb(OAc)4, although earlier we also touched on the fact that this is a prominent side reaction when performing dihydroxylations with KMnO4.

Sequentially treating a double bond with OsO4 to give a diol followed by oxidative cleavage with NaIO4 or Pb(OAc)4 gives the functional equivalent of ozonolysis (reductive workup).

-treatment of alkenes with OsO4 followed by NaIO4 is the functional equivalent of ozonolysis - lemieux johnson

(Later in Org 2, you will learn that diols will react with aldehydes and ketones to form acetals – See article: Hydrates, Hemiacetals, and Acetals)

7. Summary

Let’s summarize the key points we’ve covered about OsO4.

  • OsO4 will convert alkenes into vicinal diols (1,2-diols) via a concerted syn addition
  • A reducing agent such as KHSO3 is often added to liberate the diol from the osmate ester.
  • The diols can undergo oxidative cleavage using a reagent such as NaIO4 or Pb(OAc)4 to give aldehydes/ketones.
  • Using the oxidant N-methylmorpholine N-oxide (NMO) allows for the catalytic use of osmium.
  • In the presence of multiple alkenes, OsO4 will react with the most electron-rich alkene.
  • A related reagent is cold, basic KMnO4 that will also make vicinal syn diols. With KMnO4, however, there is an increased risk of the resulting diol undergoing oxidative cleavage.

Notes

Note 1. OsO4 is prepared through burning metallic osmium in an atmosphere of pure oxygen. From Brauer’s Handbook of Preparative Inorganic Chemistry (Academic Press, 1963, New York). (page 1603)

preparation of OsO4 from H L Grube Preparative Inorganic Chemistry

“Pure OsO4 is best prepared by a dry method. Osmium powder is heated in a boat placed in a glass or quartz tube through which a  stream of dry oxygen is passed. The metal burns to OsO4, which deposits beyond the heated zone of the tube or, better, in a bulb fused to the tube and cooled in ice. The deposit consists of white shiny crystals, though at first it may be a liquid (occasionally pale yellow in color), which forms a crystalline solid on cooling.”

Note 2. Many crystal structures of osmate esters are known. Here is an example of an osmium ester with adenosine. [Ref – J. Am. Chem. Soc. 197496, 7152]

Note 3. What if a molecule contains multiple alkenes? Which alkene will OsO4 react with?

OsO4 will react with the most electron-rich alkene (i.e. generally the one attached to the most carbon substituents).

For example in this classic synthesis of cortisone from 1952, note that OsO4 preferentially attacks the isolated alkene and doesn’t touch the electron-poor diene that is conjugated to the C=O.

dihydroxylation reaction in woodwards synthesis of cortisone using 68 grams of OsO4 - selectivity for most electron rich alkene

Woodward and his team employed 68.5 g of OsO4 on 61.5 g of substrate. At current market prices this one reaction would cost [checks Aldrich] $23,105 in today’s dollars.  Thankfully, using so much OsO4 has been made completely unnecessary by using the co-oxidant N-methylmorpholine N-oxide (NMO).

OsO4 can be made to be even more reactive and even react with very electron-poor alkenes through adding pyridine, which coordinates to osmium and renders it more electron-rich.  [Ref]

Note 4. The dihydroxylation reaction has been made even more useful through the work of Prof. K. Barry Sharpless’ research group at Scripps. Using chiral amines to coordinate to osmium and a stoichiometric oxidant, Prof. K. Barry Sharpless’ group at Scripps successfully developed a useful catalytic enantioselective dihydroxylation reaction.

The Sharpless asymmetric dihydroxylation (Sharpless AD) is effective for a wide range of alkenes. For convenience, the oxidant, osmium salt, and chiral ligand are all sold as kits known as “AD-mix α” and “AD-mix β”. Using the mnemonic below, one can choose which of the two reagent kits to use in order to get the desired chiral diol.

Mnemonic for sharpless asymmetric dihydroxylation reaction alpha and beta AD mix

For far more detail see this handout from Prof. Andrew Myers’ Chem 115 course at Harvard or consult Sharpless’ Nobel lecture.


Quiz Yourself!

[quizzes]


(Advanced) References and Further Reading

For examples of reactions employing OsO4, see:

  1. On Two Metals, Found In The Black Powder Remaining After The Solution of Patina
    Smithson Tennant
    Philosophical Transactions of the Royal Society, 1804, 411
    LINK: (JStor)
    The first description of what was to become known as OsO4 was made in 1804, where Smithson Tennant observed that the oxide of osmium “stains the skin of a dark color, which cannot be effaced”. In this remarkable paper he also gives names to what came to be called iridium and osmium.
  2. Zur Kenntnis des Osmiums
    Makowka, O.
    Chem. Ber. 1908,  41, 943
    DOI: 10.1002/cber.190804101182
    Believed to be the first application of OsO4 for dihydroxylation of alkenes, from 1908.

  3. Osmium and Its Compounds
    W. P. Griffith
    Q. Rev. Chem. Soc., 1965,19, 254-273
    DOI: 10.1039/QR9651900254
    Overview of osmium and some of its reactions. A similar review (from the Johnson-Mathey site) is found here.
  4. CATALYTIC OSMIUM TETROXIDE OXIDATION OF OLEFINS: cis-1,2-CYCLOHEXANEDIOL
    V. VanRheenen, D. Y. Cha, and W. M. Hartley
    Org. Synth. 197858, 43
    DOI: orgsyn.058.0043
    The “Upjohn method” for dihydroxylation of alkenes using catalytic OsO4 and stoichiometric NMO (N-methyl morpholine N-oxide) in Organic Syntheses.
  5. The Total Synthesis of Steroids
    R. B. Woodward, Franz Sondheimer, David Taub, Karl Heusler, and W. M. McLamore
    Journal of the American Chemical Society 1952 74 (17), 4223-4251
    DOI: 10.1021/ja01137a001
    One step of this paper involves hydroxylation of 61.5 g of substrate with 68.5 g of OsO4.
  6. Nobel Lecture
    K. Barry Sharpless
    LINK
    Prof. Barry Sharpless won the 2001 Nobel Prize in chemistry for the development of asymmetric catalysis. His 2001 Nobel Lecture describes the path toward making asymmetric dihydoxylation a useful process (starts on page 11).
  7. Osmium tetraoxide cis hydroxylation of unsaturated substrates
    Martin Schroeder
    Chemical Reviews 1980 80 (2), 187-213
    DOI: 10.1021/cr60324a003
    Comprehensive review on the dihydroxylation of alkenes with OsO4 up to 1980.
  8. Experimental and Theoretical Kinetic Isotope Effects for Asymmetric Dihydroxylation. Evidence Supporting a Rate-Limiting “(3 + 2)” Cycloaddition
    Albert J. DelMonte, Jan Haller, K. N. Houk, K. Barry Sharpless, Daniel A. Singleton, Thomas Strassner, and Allen A. Thomas
    Journal of the American Chemical Society 1997 119 (41), 9907-9908
    DOI: 10.1021/ja971650e
    Study on the mechanism of the dihydroxylation of alkenes that supports a [3+2] versus a [2+2] mechanism.

Comments

Comment section

40 thoughts on “OsO4 (Osmium Tetroxide) for Dihydroxylation of Alkenes

  1. In your diagrams for dihydroxylation of cis- and trans-pent-2-ene, you have the two products from the cis isomer correctly labelled as 2R,3S and 2S,3R … but for the trans isomer, the enantiomer to the 2R, 3R product is labeled as 2R,3R but it’s really 2S,3S

  2. Dear Dr
    James

    I think I found a way how to this, without protecting the aldehyde. I would appreciate to get your opinion

    Stage 1: reaction with peroxide to get epoxide
    Stage 2: acidificaion

    1. Without knowing the structure, it’s hard to say. Is it an alpha, beta unsaturated aldehyde? Epoxidation may work, but then again peroxide may end up oxidizing your aldehyde.

      Aldehydes are some of the most sensitive functional groups to oxidation and other side reactions. I would generally suggest they are protected unless you have some strong reason for believing it would not be affected by the OsO4 or its co-oxidant.

  3. Dear James

    If we have a molecule which contain both alkene and aldehide’and we want to prepare diol. will osmium tetraoxide oxidize the aldehide? If yes can you suggest me how to get the diol without oxidizing the akdehide?

    1. Hi Ptachia – I would not do a dihydroxylation with an unprotected aldehyde. Too many things can go wrong. Either keep it at the protected alcohol stage and then oxidize up afterwards, or protect the aldehyde as an acetal and then perform the dihydroxylation.

  4. I do not believe that products in Example 1 are meso compounds that result from the syn addition. They certainly look like the products of anti-addition. Both -OH groups should be either wedged or dashed, if added on the same side. Here they are the same but on different sides. If you rotate, they definitely look like the products from anti-addition (the example on hydroxylation of a linear alkene).

    1. Hi, it doesn’t say that they are meso compounds in this specific example, it says they are a mixture of enantiomers.

      However, what about example 2?

  5. James I am looking to recover osmium in a mixture of strong acids at high temperature from a HNO3, H2SO4, HCl digestion. Had no luck complexing it with thiourea since acids + heat = destroyed complexing agent. I’m basically looking to prevent OsO4 formation and recover low level Os spikes. Any help would be greatly appreciated. Regards

  6. Hey James,
    Reaction of alkenes with cold KMnO4 and OsO4 results in d production of cis -Diol .is there any reaction for production of trans-diol????

  7. Hi James,
    How exactly does the H2O2 hydrolyze the intermediate 5-member ring? Does it donate a proton onto the Oxygen involved in the C-O bond? If so, why would the oxygen want to form that O-H bond instead of remaining with the rest of the ring?

  8. Hey thanks for all your help and dedication to furthering education. Really appreciate your help with making these pages. Between this and the people at Clutchprep I’m acing organic chem and can’t wait for orgo II, bio chem, and pharmacy school next august.
    Cheers :)

    1. When the reaction is done, “H2O2” re-oxidizes the OsO2(OH)2 back to OsO4. It’s a reoxidant, which allows OsO4 to be used catalytically. A more common reagent for this purpose is N-methyl morpholine N- oxide (NMO)

  9. Hi, just wondering in what cases does OsO4 fail? Certain alkenes? What about enones (I’ve seen that it does some enones but wondering if some cases would fail)? And how are the reactivities between OsO4 and K2OsO4 differ, besides the latter having hydrates and not useful for moisture sensitive reactions? Thanks!

  10. Why isn’t the stereoisomer for the cyclo hexene drawn (side question)? I feel like the answer’s at the tip of my tongue but I can’t quite get it.

      1. Since one of those reactions gives a mix of products (enantiomers) while the other gives only one product, the reaction in general is stereoselective?

        1. Whether you get one product or a mixture of enantiomers/diastereomers will depend on the alkene you start with.

          Dihydroxylation of ethene would only give one product, for example.

          The reaction itself is stereoselective. Starting with cyclohexene you get one product – cis 1,2-cyclohexane diol, and none of the trans. If you obtained a random *mixture* of cis- and trans- 1,2-diols from cyclohexene, then it would not be stereoselective.

          This is going beyond what you asked, but not only is this reaction stereoselective, it is stereospecific.
          Here’s why.
          Take two alkenes that are diastereomers. cis but-2-ene and trans but-2-ene.
          Now treat each with OsO4.
          The cis but-2-ene will give a single product (S,R)-2,3-butanediol (meso).
          The trams but-2-ene will give a racemic mixture of diols. (S,S)-2,3-butanediol and (R,R)-2,3-butanediol. These are each diastereomers of (S,R)-butanediol.

          So two diastereomeric alkenes give two diastereomeric products. That’s the IUPAC definition of stereospecific.

  11. Wow, I had no idea it’d ever been used on that scale in a synthesis! Unfortunately for those guys the Upjohn catalytic procedure wasn’t reported until 20 years later. I seem to remember that Kumamoto’s synthesis of methyl-kinamycin C and Corey’s neotripterifordin also use it stoichiometrically, although not on anywhere near that scale. I once saw the selective dihydroxylation of one bond in a 1,4-cyclohexadiene reported in a very old Tet. Lett. using stoichiometric OsO4. When I tried it with NMO as the terminal oxidant aromatisation was a major problem so I changed my route. I never had the courage to try with the 1.1 equiv. reported so I’ll always wonder if it would have worked. If nothing else, OsO4 is also damned expensive (~300$/g from Aldrich).

    At the university where I currently am (I can tell you it’s in the UK, if you couldn’t guess from my spelling) 3rd year undergrads get to do Sharpless AD, although obviously that’s with potassium osmate, which is far less volatile/nasty. Still, they do generate OsO4 during the reaction, and they only have to quench it wrong (which is not impossible) for potentially bad things to happen. I get very nervous when this experiment runs.

    1. Wow! Sharpless AD in 3rd year? That’s fantastic. The AD-mix makes it pretty easy I guess. Kudos to your school. In the “advanced organic chemistry” labs I’ve been in or TA’d, there’s always seemed to be a fine, 30-year or more patina of dust on the procedure we were using. Nothing wrong with that per se, but it didn’t give you the exciting feeling of doing something cutting edge.

  12. You can make the use of OsO4 catalytic if you throw in a sacrificial oxidant, N-morpholine oxide (NMO). I’ve had several lab mates use catalytic NMO without incident.

    1. Oh yeah. My point (perhaps not so well made) was that undergraduate teaching laboratories are unlikely places to learn how to use OsO4. But maybe there are places out there who do?
      Your point brings up another question – I”m wondering how many instructors include NMO when they teach the dihydroxylation reaction.

      1. I’ve worked with students from dozens of schools, and I can’t find another of an example of an instructor who does (although Jim Tour teaches cat. OsO4 and H2O2). You, my good man, are one of the very few! Good on you.

      2. They do where I go to college (Uni. of Minnesota). My prof mentioned using NMO because OsO4 is ridiculously expensive, and NMO turns it into a catalyst so you don’t need as much. She failed to mention the burning out the eyes part, though.

          1. NMO is in McMurray, 8th Edition (used by 1 of 3 sections at SUNY Geneseo).

    2. Ohio State Organic chem teaches using it and Miami University of Ohio

      example problem from class..

      an alkene —–OSO4 (cat.), NMO——> enantiomers

      (side note: if possible adding an image attachment option to the comments)

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.