[7782-44-7]  · O2  · Oxygen  · (MW 32.00)

(oxidizing agent for many organic systems, including, most commonly, organometallic compounds, carbon radicals, and heteroatoms such as sulfur)

Physical Data: mp -218 °C; bp -183 °C; d 1.429 g L-1 (0 °C), 1.149 g L-1 (-183 °C).

Solubility: sol to some extent in most solvents. Selected data, expressed as mL of O2 (at 0 °C/760 mmHg) dissolved in 1 mL of solvent when the partial pressure of the gas is 760 mmHg, are as follows: Me2CO (0.207/18 °C), CHCl3 (0.205/16 °C), Et2O (0.415/20 °C), EtOAc (0.163/20 °C), MeOH (0.175/19 °C), petroleum ether (0.409/19 °C), PhMe (0.168/18 °C), H2O (0.023/20 °C).

Form Supplied in: dry gas; dilutions in Ar, He, or N2; 18O2; 17O2.

Handling, Storage, and Precautions: of itself, oxygen gas is essentially nontoxic. However, it will support and vigorously increase the rate of combustion of most materials. It may ignite combustibles and can cause an explosion on contact with oil and grease. The potential for autoxidative formation of explosive peroxides (e.g. with Et2O) should always be borne in mind.

Oxygenation of Carbanions and Organometallic Compounds.

Many organometallic species react with triplet oxygen to form the corresponding hydroperoxides,1,2 although the products are more usually reduced in situ or during workup to afford alcohols as the isolated products. A number of other sources of electrophilic oxygen have been developed (e.g. Oxodiperoxymolybdenum(pyridine)(hexamethylphosphoric triamide) (MoOPH), sulfonyloxaziridines) and compete for this niche, but no single reagent is universally preferable. As carbon anion equivalents, Grignard reagents are optimal for simple hydrocarbons,2 but organolithiums are more frequently employed. The potential for radical-mediated oxidative dimerization can constrain utility (particularly for aryl organometallics). Useful oxygenations of alkyl,3 vinylic,4 allylic,5 benzylic,6 and aryl (eq 1)6 organolithium compounds have been reported. 1,1-Diorganometallics give the corresponding carbonyl derivatives (eq 2).7

Effective oxygenation of enolate anions is generally restricted to tertiary centers where over-oxidation is not possible.8 Nonetheless, ketones (eq 3),9 esters/lactones,8 amides/lactams (eq 4),10 and carboxylic acids8 can all be usefully a-hydroxylated. In most cases the intermediate a-hydroperoxides are reduced in situ (usually with Triethyl Phosphite),11 although they can be isolated if desired.12 The small size of the electrophile and the potential for radical involvement8 do not encourage stereochemical chastity in these processes, but where sufficient bias exists, good discrimination can be observed (see eqs 3 and 4). A slightly different approach uses enolates derived from aqueous base treatment. These species have been usefully hydroxylated where there was little or no ambiguity in the direction of enolization8 and this process forms the basis for a surprisingly effective catalytic, enantioselective oxygenation (eq 5).13

In cases where the activating group is also a leaving group, oxygenation can provide the corresponding carbonyl compound. Thus oxidative decyanation can be effected under either phase transfer14 or anhydrous conditions (eq 6).15 The latter procedure is more general, although it does require treatment with Tin(II) Chloride and base to reduce and fragment the a-hydroperoxide, and this method is not effective for the primary nitrile to aldehyde conversion. a,b-Unsaturated nitriles generally react at the a-position to give a,b-unsaturated ketones.15

Oxygenation of sulfone anions and the consequent desulfonylation is frequently effected with the MoOPH reagent. In some cases, however, molecular oxygen has proved effective where MoOPH failed.16 This perhaps illustrates a functional advantage over more bulky reagents and a counterpoint to the stereochemical disadvantages (see above). It should be noted that one such attempt resulted in a minor explosion17 although, of course, any reaction involving peroxides bears this possibility. A similar process, preceded by nickel transmetalation, demonstrated the oxidation of the C-Ni bond, but the synthetic advantage is not clear.18 However, this report did demonstrate the conjugative oxygenation of an allylic sulfone anion to give a g-hydroxy sulfone.

Oxygenation of phosphorus-stabilized anions also produces the corresponding carbonyl compounds. The anions derived from phosphonates19 (eq 7)20 (including a-heteroatom substituted phosphonates)19 and phosphine oxides21 react smoothly with oxygen. Similarly, phosphorane ylides are readily oxidized.22 In all these cases, however, the reaction of primary substrates suffers from competing self-condensation, giving alkenes.22 It should be noted that a two-stage procedure involving the reaction of phosphonate anion with chlorodimethyl borate followed by oxidation with m-Chloroperbenzoic Acid has been advocated as a more efficient method23 (and, interestingly, allows isolation of the intermediate hydroxy phosphonate).

Oxygenation of Carbon Radicals.

Not surprisingly, triplet oxygen reacts rapidly with carbon centered radicals.24 Classical autoxidation is the most obvious example of this behavior. Traditionally, autoxidation refers to hydroperoxide formation from alkanes, aralkanes, alkenes, ethers, alcohols, and carbonyl compounds, where the initiating homolysis is induced thermally or photochemically.25,26 There is an extensive literature concerning these processes dating back many years. While very important commercially, they are generally too promiscuous to be of wide synthetic value, particularly when dealing with complex molecules. Interestingly, however, a deformylative hydroxylation of an allylic neopentyl aldehyde has been observed that bypasses the classical autoxidative fate of aldehydes (eq 8).27

Radical oxygenation is most valuable where there is more strict control over the site of radical formation and subsequent oxygenation. Good stereochemical control is, of course, not usually achieved, although exceptions can be found in most cases. The mild and controlled methods of radical generation that have seen much use in synthesis are readily applicable to oxygenation. The thermal or photochemical decarboxylation of the esters of thiohydroxamic acids,28 or their room temperature decomposition in the presence of tris(phenylthio)antimony29 (i.e. Barton's methodology), can be intercepted by triplet oxygen to generate the nor-alcohols. The addition of heteroatom radicals to alkenes can provide the source of carbon centered radicals for trapping. An interesting example of oxygenation initiated by phenylthio or phenylseleno radical addition to vinylcyclopropanes showcases the use of this methodology (eq 9).30 Here, instances of moderately successful stereocontrol in the C-O bond forming step were noted. This transformation also demonstrates the potential of the initially formed hydroperoxy radical to participate in further steps (where higher levels of stereochemical discrimination are observed as a consequence of the intramolecular nature of the radical trapping). Samarium(II) Iodide induced radical processes have been quenched with oxygen to provide hydroxyl functionalized products.31

One of the most prevalent uses of molecular oxygen in modern synthesis is for oxidative demercuration.32 Carbon radicals generated by the reduction of organomercurials with borohydride are efficiently trapped by oxygen, most frequently in DMF solution, to give hydroperoxides which are reduced under the reaction conditions to generate the corresponding alcohols directly.33 The alkene oxymercuration-oxidative demercuration sequence is commonly practised (usually through a b-alkoxymercury species, since b-hydroxy fails33), particularly where the oxymercuration is an intramolecular cyclization (eq 10).34 Typically, any stereocontrol observed in the oxymercuration (or other C-Hg bond forming step) is effaced in the oxygenation (as in eq 10).

Oxidation of Organoboranes.

Boranes, most frequently accessed by hydroboration of alkenes, can be oxidized by triplet oxygen.35,36 If the oxidation is carried out in fairly concentrated solution (~0.5 N) at 0 °C, intermolecular redox reaction of the intermediate diperoxyborane is facilitated and workup provides the corresponding alcohol.36 While this is quite efficient, Hydrogen Peroxide is more commonly used in synthetic applications. This is partly for convenience, but also a consequence of stereochemical issues. The oxidation with H2O2 occurs with retention of configuration at the carbon center. The radical characteristics of the dioxygen reaction generally lead to at least partial racemization. That this stereochemical corruption is not always complete is an indication of the uncertainty about the mechanism.35 Interestingly, rhodium(III) porphyrin has been shown to promote stereoselective oxidation in the dioxygen procedure (eq 11).37 In dilute solution (0.01-0.05 N) the intermolecular redox process is suppressed and diperoxyboranes are produced. Oxidation of the third B-C bond with H2O2 or peroxy acid and workup allows isolation of the corresponding alkyl hydroperoxides.36 Alternatively alkyl hydroperoxide formation is facilitated by the use of alkyldichloroboranes.36 This is one of the most convenient approaches to this functionality. The oxygen mediated approach to alcohols may be more convenient than H2O2 for radiolabeling; 17O (eq 12) and 15O alcohols have been prepared in this way.38,39

Heteroatom Oxidation.

Oxidation of nitrogen functionality with oxygen, while well precedented40 and of continuing interest,41,42 does not generally represent the method of choice for those processes of synthetic significance. However, a report of a mild procedure for the oxidation of silylamines to carbonyl compounds bears some synthetic potential (eq 13).43 Oxidation of phosphorus functionality by oxygen can be quite facile.44 For example, tertiary phosphines are very readily oxidized to their phosphine oxides, and secondary chlorophosphines can give the phosphinic acids.44 Perhaps the most common heteroatom air oxidations are those of Group 16 RX-H bonds to their corresponding dimers ((RX)2) and particularly the thiol to disulfide oxidation.45 This, of course, is related to the importance of the disulfide bond to peptide and protein secondary structure. One example that reflects current interest in the control of multiple disulfide bond formation in synthetic peptides is given in eq 14.46 Oxidation may be promoted by heavy metal ions.45 Higher oxidations (for example sulfide to sulfoxide) are best performed with other reagents (oxone, peroxy acid, etc.).

Other Uses.

The oxidative dimerization of organometallics, alluded to above, is particularly prevalent for organocuprates,47 although not totally unavoidable.48 In fact it is efficient enough to be regarded as a synthetic strategy and has been used as such (eq 15).49 Baeyer-Villiger oxidations generally employ peroxy acids, but a recent report indicates that 1 atm of oxygen can effect the rearrangement even in the absence of either metal catalysts or light.50 Epoxidation by oxygen is possible51 but, of course, is not usually the method of choice for laboratory synthesis.

There is a vast literature concerning metal-catalyzed oxidative processes involving molecular oxygen,52 of which only a fraction have seen synthetic use. Many metal catalysts behave as oxygen fixing species, that deliver oxygen to the substrate through a peroxo complex. Reports frequently concern experimental systems, probing substrate reactivity and/or asymmetric induction. The function of oxygen in metal-catalyzed oxidations is not necessarily that of a reagent. Thus, for example, in the Wacker oxidation of terminal alkenes it operates as a re-oxidant for copper(II) chloride which in turn is a re-oxidant for the PdII species. All of these applications are best regarded as functions of the metal component and, for this reason, are not discussed here.

Related Reagents.

Copper(I) Chloride-Oxygen; Diethylzinc-Bromoform-Oxygen; Iron(II) Sulfate-Oxygen; Oxygen-Platinum Catalyst; Singlet Oxygen.

1. Sosnovsky, G.; Brown, J. H. CRV 1966, 66, 529.
2. Wardell, J. L. In Comprehensive Organometallic Chemistry; Wilkinson, G., Ed.; Pergamon: Oxford, 1982; Vol. 1, Chapter 2.
3. Warner, P.; Lu, S.-L. JOC 1976, 41, 1459.
4. Panek, E. J.; Kaiser, L. R.; Whitesides, G. M. JACS 1977, 99, 3708.
5. Takahashi, T.; Nemoto, H.; Kanda, Y.; Tsuji, J. JOC 1986, 51, 4315.
6. Parker, K. A.; Koziski, K. A. JOC 1987, 52, 674.
7. Knochel, P.; Xiao, C; Yeh, M. C. P. TL 1988, 29, 6697.
8. Jones, A. B. COS, 1991, 7, Chapter 2.3.
9. Paquette, L. A.; DeRussy, D. T.; Pegg, N. A.; Taylor, R. T.; Zydowsky, T. M. JOC 1989, 54, 4576.
10. Kim, M. Y.; Starrett, J. E., Jr.; Weinreb, S. M. JOC 1981, 46, 5383.
11. Gardner J. N.; Carlton F. E.; Gnoj, O. JOC 1968, 33, 3294.
12. Bailey, E. J.; Barton, D. H. R.; Elks, J.; Templeton, J. F. JCS 1962, 1578.
13. Masui, M.; Ando, A.; Shioiri, T. TL 1988, 29, 2835.
14. Donetti, A.; Boniardi, O.; Ezhaya, A. S 1980, 1009.
15. Freerksen, R. W.; Selikson, S. J.; Wroble, R. R.; Kyler, K. S.; Watt, D. S. JOC 1983, 48, 4087.
16. Yamada, S.; Nakayama, K.; Takayama, H.; Shinki, T.; Suda, T. TL 1984, 25, 3239.
17. Little, R. D.; Myong, S. O. TL 1980, 21, 3339.
18. Julia, M.; Lauron, H.; Verpeaux, J-N. JOM 1990, 387, 365.
19. Mikolajczyk, M; Midura. W.; Grzejszczak, S. TL 1984, 25, 2489.
20. Tsuge, O.; Kanemasa, S.; Suga, H. CL 1986, 183.
21. Davidson, A. H.; Warren, S. CC 1975, 148.
22. Bestmann, H. J. AG(E) 1965, 4, 830.
23. Kim, S.; Kim, Y. G. Bull. Korean Chem. Soc. 1991, 12, 106.
24. Maillard, B.; Ingold, K. U.; Scaiano, J. C. JACS 1983, 105, 5095.
25. Swern, D. In Comprehensive Organic Chemistry; Barton, D. H. R., Ed.; Pergamon: Oxford, 1979; Vol. 1, Chapter 4.6.
26. Crabtree, R. H.; Habib, A. COS 1991, 7, Chapter 1.1.
27. Kigoshi, H.; Imamura, Y.; Sawada, A.; Niwa, H.; Yamada, K. BCJ 1991, 64, 3735.
28. Barton, D. H. R.; Crich, D.; Motherwell, W. B. CC 1984, 242.
29. Barton, D. H. R.; Bridon, D.; Zard, S. Z. CC 1985, 1066.
30. Feldman, K.; S. Simpson, R. E. JACS 1989, 111, 4878.
31. Molander, G. A.; McKie, J. A. JOC 1992, 57, 3132.
32. Kitching, W. COS 1991, 7, Chapter 4.2.
33. Hill, C. L.; Whitesides, G. M. JACS, 1974, 96, 870.
34. Broka, C. A.; Lin, Y.-T. JOC 1988, 53, 5876.
35. Pelter, A.; Smith, K. COS 1991, 7, Chapter 4.1.
36. Brown, H. C.; Midland, M. M. T 1987, 43, 4059.
37. Aoyama, Y.; Tanaka, Y.; Fujisawa, T.; Watanabe, T.; Toi, H.; Ogoshi, H. JOC 1987, 52, 2555.
38. Kabalka, G. W.; Reed, T. J.; Kunda, S. A. SC 1983, 13, 737.
39. Takahashi, K.; Murakami, M.; Hagami, E.; Sasaki, H.; Kondo, Y.; Mizusawa, S.; Nakamichi, H.; Iida, H.; Miura, S.; Kanno, I.; Uemura, K.; Ido, T. J. Labelled Compd. Radiopharm. 1986, 23, 1111.
40. Boyer, J. H. CRV 1980, 80, 495.
41. Riley, D. P.; Correa, P. E. JOC 1985, 50, 1563.
42. Gangloff, A. R.; Judge, T. M.; Helquist, P. JOC 1990, 55, 3679.
43. Chen, H. G.; Knochel, P. TL 1988, 51, 6701.
44. Gilchrist, T. L. COS 1991, 7, Chapter 6.1.
45. Uemura, S. COS 1991, 7, Chapter 6.2.
46. Ponsati, B.; Giralt, E.; Andreu, D. T 1990, 46, 8255.
47. Posner, G. H. OR 1975, 22, 253.
48. Lambert, G. J.; Duffley, R. P.; Dalzell, H. C.; Razdan, R. K. JOC 1982, 47, 3350.
49. Coleman, R. S.; Grant, E. B. TL 1993, 34, 2225.
50. Bolm, C.; Schlingloff, G.; Weickhardt, K. TL 1993, 34, 3405.
51. Rao, A. S. COS 1991, 7, Chapter 3.1.
52. Sheldon, R. A.; Kochi, J. K. Metal-Catalysed Oxidations of Organic Compounds; Academic: New York, 1981.

A. Brian Jones

Merck Research Laboratories, Rahway, NJ, USA

Copyright 1995-2000 by John Wiley & Sons, Ltd. All rights reserved.