Organic Chemistry of Periodates By B. SMarz UNIVERSITY OF BIRMINGHAM 1 Introduction Malaprade’s observation1 that mannitol is destroyed by periodic acid and Fleury’s closer definition of the reaction as a specific oxidative fission of 1,2-diols,* launched periodic acid and its salts on a distinguished career among the oxidising agents of organic chemistry. Closely related fissions of a-ketols a-diketones and a-amino-alcohols having been discovered in the 1930’s the reagent became an indispensable tool for the structure determination of the most varied natmal products particularly the carbohydrates and more recently the nucleic acids. The reviews by Jackson3 and by Bobbitt4 fully describe these developments while more recent applications in the carbohydrate and nucleic acid fields will be found in standard w~rks.~n~ Malangeau’s review’ is limited in scope and is in- accessible.The well-established theory and practice of oxidation with periodates particularly in carbohydrate chemistry is described in practical manuals.8 Only a brief outline including recent innovations will be attempted here. In the last decade the organic chemistry of periodate has grown mainly in two directions. The classical glycol cleavage has been studied in detail and a consistent picture of its mechanism is emerging. This has shed light on the oxidation of simple sugars where detailed re-examination of the reactions has revealed some of the intermediates. On the other hand several new well-defined reactions have come to light which extend the use of the reagent in organic chemistry. This Review is devoted principally to these two aspects.The term ‘oxidation’ will refer to reaction with periodic acid or its salts unless otherwise qualified. L. Malaprade Bull. SOC. chim. France 1928 683. P . Fleury and J. Lange Compt. rend. 1932,195 1395. E . L. Jackson Org. Reactions 1944 2 341. J. M. Bobbitt Adv. Carbohydrate Chem. 1956 11 1. W. Pigman (ed.) ‘The Carbohydrates’ Academic Press Ohio 1956; E. Percival ‘Structural Carbohydrate Chemistry’ J. Garnett-Miller London 1962. A. M. Michelson ‘Chemistry of the Nucleosides and Nucleotides’ Academic Press London 1963; see also P. R. Whitfield Biochim. Biophys. Acta 1965 108 202 and references therein. P. Malangeau Miscs au Pointde la Chimie Analytique pure ct appliqued et de l’dnalyse bromato- Zogiquc 1961 9 81. (a) G. F. Smith ‘Analytical Applications of Periodic Acid and Iodic Acid’ G.F. Smith Chemical Co. Ohio 1950; (b) J. R. Dyer Methods Biochem. Analysis 1956,3 11 1 ; (c) R. D. Guthrie ‘Methods in Carbohydrate Chemistry’ ed. R. L. Whistler and M. L. Wolfrom Academic Press London 1962 p. 432 et seq. 3 1. Quarterly Reviews Methods.-Periodate oxidation lends itself particularly to analytical experiments which are often followed by preparative oxidations. Periodic acid and its sodium and potassium salts are the usual reagents. In analytical experiments mixtures containing excess of 0.1-0.01 M-periodate are allowed to react at room temperature or below along with a blank and for long reaction times should be kept in the dark. The photochemistry of periodate solutions which slowly form ozone in daylight has been Organic solvents can be added but may retard some reaction^.^ Tetraethylam- monium periodate which is highly soluble in water and considerably soluble also in several organic solvents including chloroform may be of value in over- coming solubility problems.1° The choice of pH depends on the r e d ~ c t a n t .~ ~ ~ ~ The reaction is usually followed by estimation of the remaining periodate in aliquot portions withdrawn at intervals from the mixture and the blank. The various iodometric titrations are described in the above reviews.8 Methods will also be found there for the estimation of small-molecular oxidation products such as formaldehyde formic acid carbon dioxide and ammonia. The elegant methods of polarographyll have been extended to periodate reactions and in favourable caseslla permit simultaneous estimation of organic products.Periodate can also be estimated by its absorption at 223 mp8c or in more concentrated solution at 260 mp.12 When oxidations are carried out on a preparative scale,86 it may be necessary to remove periodate and iodate ions by precipitation or by ion-exchange if the organic products cannot be extracted. Excess of reagent is sometimes destroyed by adding ethylene glycol or pinacol etc. Several simple compounds frequently used as solvents or appearing as oxida- tion products are hardly attacked by periodate in the dark though more appre- ciably in sunlight .13 They include methanol ethanol and their derived aldehydes and acids. The limited solubility of the commonly used periodates their high molecular weight and the frequen? appearance of iodine are the principal drawbacks to their preparative use.2 Glycol Fission and Related Reactions Introduction.-The oxidative cleavage of vicinal diols is the classical and most widely used reaction of periodate. Oxidation is fast at room temperature except for heavily substituted diols aldehydes and ketones being formed (a; Scheme 1). Analogous cleavage occurs more slowly with a-hydr~xy-carbonyl~~~~ and a-dicarbonyl compounds,1* in which the carbonyl groups appear as carboxyl R. D. Guthrie Chem. and Ind. 1960 691. lo A. K. Qureshi and B. Sklarz J. Chem. SOC. (C) 1966 412. l1 (a) P. Zuman and J. Krupicka CoZZ. Czech. Chem. Comm. 1958 23 598; (b) P. Zuman ‘Organic Polarography’ Pergamon Oxford 1964. l9 J. S. Dixon and D. Lipkin An&. Chem. 1954 26 1092. l3 F.S. H. Head and G. Hughes J . Chem. Sac. 1952,2046. l4 P. W. Clutterbuck and F. Reuter J. Chenr. SOC. 1935 1467. 4 Sktarz functions in the products (b). a-Hydroxy- and a-keto-acids are oxidised rather sl0wly,1~J~ with some exceptions such as glycollic and glyoxylic acids.15 a-Amino-alcohols and a-diamines are likewise oxidised to carbonyl fragments (c) unless one of the amino-groups is tertiary or acylated.17~la These points are discussed in detail later (p. 13). All these reactions are two-electron oxidations requiring one molecule of periodate in which the iodine atom is reduced from the $7 to the + 5 valency state with the formation of iodate ions. Lead tetra-acetate has long been used as a glycol-cleaving reagent.lg It is used in glacial acetic acid or in other organic solvents and oxidises also a-hydroxy- and a-keto-acids and many other structures.20 It is thus a useful complement to rather than ‘a serious c~mpetitor’l~a of periodate.Iodosobenzene diacetate21 resembles lead tetra-acetate as does sodium bismuthate,22 which can be used also in aqueous acid solution. Chromium trioxide has been used for the cleavage of di-tertiary glyc01~,2~a and the mechanism has been Mechanism-A mechanism for the periodate oxidation of glycols suggested as early as 1933 by Criegee% involved the formation of a cyclic periodate complex in analogy with lead tetra-acetate. Much subsequent work has borne out and amplified this idea. The evidence consists of direct observations on the complexes and more extensively of kinetic evidence. The interaction of periodic acid with glycols in water has been observed spectrometrically by an increase in the absorbance of the periodate ion at l6 D.B. Sprinson and E. Chagraff J . Biol. Chem. 1946 164 433. l6 J. E. Courtois and M. Guernet Ann. pharm. franc. 1958 16 119. B. H. Nicolet and L. A. Shinn J. Amer. Chem. SOC. 1939 61 1615. P. F. Fleury J. E. Courtois and M. Grandchamp Bull. SOC. chim. France 1949 88. 19(a) R. Criegee L. Kraft and B. Rank Annalen 1933 507 159; (6) A. S. Perlin Adv. Carbohydrate Chem. 1959 14 9. *O R. Criegee Angew. Chem. 1958,70 173. *l L. K. Dyall and K. H. Pausacker J . Chem. Soc. 1958 3950 and refs. therein. 22 W. Rigby J . Chem. SOC. 1950 1907. 23 (a) M. Uskovic M. Gut E. N. Trachtenberg W. Klyne and R. I. Dorfmann J. Amer. Chem. SOC. 1960,82,4965; (b) J. Rocek and F. Westheimer J.Amer. Chem. SOC. 1962 84 2241. 24 R. Criegee Sitzungsberichte der Gesellschqft fur die Be0 frderrrng der gesamten Wissen- schaften Marburg 1934,69,25 (Chem. Abs. 1935 29 6820). 5 Quarterly Reviews 222.5 mp,a5 while increased absorbance at 226 mp in isopropyl alcohol also suggests some interaction.28 A fall in pH has been observed on addition of ethylene glyc01,2~ other vicinal diolsF8 and certain 1,3-di0ls~~ to periodate solutions. Malaprade30 had already observed this effect and deduced the formation of an 'addition compound' but as to its r81e in the mechanism of the oxidation he could only speculate. Investigating the periodate oxidation of some cyclic monosaccharide com- pounds and cycloalkane polyols Barker and Shaw31 found that several of these compounds rapidly took up one molecule of the reagent which reappeared slowly after the titrations while considerable amounts of substrate (ribose) were recovered.All these compounds had a 1,2,3-cis-triol system capable of existing in a favoured diaxial-monoequatorial conformation from which a terdentate periodate ester (I) was thought to be formed. The other polyols took up periodate rapidly with oxidation. The terdentate complex from 1,2-O-isopropylidene-a-~-glucofuranose has been studied by nuclear magnetic resonance (n.m.r.).32 Its stability in alkaline solution is such that the monosaccharide is not oxidised. Probably the triesters cannot break down directly to oxidation p r o d u ~ t s ~ ~ ~ ~ ~ and any oxidation involves prior hydrolysis to the diester. Despite this stability the terdentate esters have not been isolated.The cyclic diesters are much less stable and the principal evidence for their structure and function in glycol fission comes from kinetic Early experiments had shown that the oxidation of ethylene glycol and of pinacol have different kinetic forms.35 A two-stage reaction course ( d ) was later s ~ g g e s t e d ~ ~ . ~ ~ in which the glycol (G) reacts reversibly with periodate (P) to give an intermediate (X) which breaks down to the products. The rate constants ka k (d) G + P +X-+ products kb k, kb and k and the equilibrium constant K for formation of X have been evaluated for a series of simple glycols in an elegant set of studies by Bunton and his colleagues. Taking the complex equilibria of ionisation and hydration of periodate3' into account they showed that the complex X undergoing breakdown must 25 G.J. Buist C. A. Bunton and J. H. Miles J. Chem. Soc. 1957 4575. 26 J. Klaning and M. C. R. Symons J . Chem. SOC. 1960,977. 27 G. J. Buist and C. A. Bunton J. Chem SOC. 1954 1406. 28 G. J. Buist C. A. Bunton and J. H. Miles J. Chem. SOC. 1957 4567. 29 J. L. Bose A. B. Foster and R. W. Stephens J. Chem. SOC. 1959 3314. 30 L. Malaprade Bull. SOC. chirn. France 1934 883. 31 G. R. Barker and D. F. Shaw J. Chem. SOC. 1959 584. 32 A. S. Perlin and E. von Rudloff Canad J. Chem. 1965 43 2071. 33T. P. Nevell Chem. and Znd. 1959 567. 34 C. A. Bunton Ann. Reports 1959 56 185. 35 C. C. Price and H. Kroll J. Amer. Chem. SOC. 1938 60 2726; C. C. Price and M. Knell ibid. 1942 64 552. 36 F. R. Duke and V. C. Bulgrin J. Amer. Chem. SOC. 1954 76 3803.37 C. E. Crouthamel A. M. Hayes and D. S. Martin J. Amer. Chem. SOC. 1951 73 82. 6 Sklarz be a mono-anion written as (11) or its dehydrated form (III).27 The complex is a stronger acid than periodic acid (cf. borate and tellurate complexes). The stability of X is subject to electronic and steric influences. K increased in going from ethane- through propane-1,2- to (-)-butane-2,3-diol i.e. with increasing methyl substitution and consequent electron availability at oxygen but the sharp drop in the equilibrium constant for meso-butane-2,3-diol is a steric effect.28 A model (Fig. 1) based on atomic dimensions was suggested in which methyl groups interfere sterically with the octahedral periodate oxygen atoms when placed at the hindered positions H,H’ but not at the free positions F,F’.The rate of formation k was determined spectrometrically for some diols of the series and decreased with increasing substitution for steric reasons. Under certain conditions the initial esterification was relatively slow followed by fast cy~lisation.~~ The rate of collapse of X was also estimated and increased with methyl sub- stitution so that steric crowding is probably the dominant factor although electronic effects (hyperconjugation) may also operate.28 The importance of this work lay in the separate evaluation of the various constants for the two reactions steps. With the help of these the difference in over-all kinetics between ethylene glycol and pinacol could be explained. In the former and other lightly substituted diols K is large i.e. the intermediate accumulates and its collapse (k) is rate-determining.The rate of oxidation is greatest near neutrality when the concentration of mono-anion is maximal but decreases at higher pH where the stable dianion is formed but cannot break down. For hindered diols such as pinacol K is small and formation of the complex appears to be rate-determining (ka) leading to overall second-order kinetics. The influence of pH is also more complex and earlier reports are in c o n f l i ~ t . ~ ~ ~ ~ * According to a detailed study by Bunton and his colleagues the oxidation of pinacol is subject to general acid-base catalysis of the ring-closing step39 ammonia being a good catalyst.40 2-Methylbutane-2,3-diol lies on the mechanistic borderline with kinetics which depart from first order between pH 4 and 5 and can be interpreted on@ in terms of an intermediate X.4l Acyclic threo-diols are very generally oxidised more quickly than the erythro- 38S.Senent and P. Escudero Andes real SOC. espaii. Fis. Quim. 1961 57,B 153; Chem. Abs. 1961 55 21761. 39 G. J. Buist C. A. Bunton and J. Lornas J. Chem. SOC. (B) 1966 1094 1099. 40 C. A. Bunton and M. D. Carr J. Chem. Soc. 1963 5854 (p. 5860). 41 G. J. Buist and C. A. Bunton J. Chem. SOC. 1957,4580. 7 Quarterly Reviews isomer and relative configurations have been assigned on this basis.42 Polyols are likewise oxidised first at a threo-diol group glucitol (V) for instance giving glyceraldehyde and D-erythrose when oxidised with limited amounts of perio- date.qs A more recent study has revealed the detailed sequence of oxidation of all the diol pairs in [14C]glucitol [(3,4) > (2,3) > (43) > (5,6) > (1,2)].44 A neat rationalisation of these results was possible in terms of the cyclic complex (IV).Studies with cyclic 1,Zdiols have further clarified the structure of the com- ~ l e x . 4 ~ ~ cis- and trans-Cyclopentanediols (VI; R = R’ = H or CH,) undergo second-order (‘pinacol-type’) oxidation which is faster for the cis compounds. The trans-diol (VI; R ing in the complex. = R’ = CH,) is not attacked at all owing to severe crowd- CHO __c CH,OH cis-Cyclohexane-l,2-diols are oxidised faster than the trans- (diequatorial) isomers and the trans-diaxial diol group is not oxidised.46 In a detailed analysis of the ‘ethylene glycol’ kinetics of cis- and trans-cyclohexanediols Bunton and P 0 Oxygen 0 Carbon FIG. 1. [Reproduced by permission from ref.281 Model of the intermediate complex (two oxygen atoms omitted). 42 P. Zuman J. Sicher J. Kiupicka and M. Svoboda Coll. Czech. Chem. Comm. 1958 23 1237. 43 J. C. P. Schwartz J. Chem. SOC. 1957,276; J. E. Courtois and M. Guernet Bull. Sot. chfm. France 1957 1388. 44 D. H. Hutson and H. Weigel J. Chem. SOC. 1961 1546. 45 (a) V. C. Bulgrin and G. Dahlgren J . Amer. Chem. SOC. 1958 80 3883; (6) C. A. Bunton and M. D. Carr J . Chem. SOC. 1963 770. 46 J. Honeyman and C. J. G. Shaw J. Chem. SOC. 1959 2454. 8 Sklarz his colleagues47 showed that the trans-fused complex is actually more stable than the cis-fused at least at pH 9 and that the higher overall oxidation rate of the cis-diol is due to the faster breakdown of the cis-fused complex (VIII). The flex5bility of the five-membered ester ring is such that the trans-diequatorial junction with the cyclohexane ring is preferred (VII; Fig.2). At the same time FIG. 2. oxygen atoms ‘a’ to iodine are in the plane of the paper. are out of the plane of the paper ‘a” being away from the observer. [Reproduced by permission from ref. 471 (VII) Intermediate complex of the trans-diol (equational conformation); bonds joining (VIII) Intermediate complex of the cis-diol; bonds joining oxygen atoms ‘a’ ‘a” to iodine In both models two of the oxygen atoms joined to iodine are omitted. there is less non-bonding interaction between the periodate oxygen and ring- carbon atoms. Methyl groups increase these interactions and lead to second- order kineti~s.~~b a-Glucopyranose with a cis-diol function at C( 1)-C(2) is oxidised at this position appreciably more quickly than the ,L?-an~mer.~~ Rigid trans-diaxial diols such as truns-decalin-9,1 O - d i 0 1 ~ ~ G are not oxidised ; a cyclic intermediate is clearly impossible.Ditertiary alcohols of this type are cleaved by lead tetra-a~etate,4~~$~~ an observation which among others prompted the suggestion by Levesley Waters and Wright51 that monoesters undergo breakdown in glycol fission. This is certainly excluded for periodate since the monoester of an erythro-diol or of a trans-diaxial cyclic diol should then favour cleavage by trans elimination which is contrary to ~bservations.~~ The stereo- chemical requirements of lead tetra-acetate differ somewhat from those of periodate,5O as seen in the differing initial action on sucr0se.5~ The oxidative cleavage of a-diketones has received relatively little attention.Shiner and Wasmuth5* have shown that it is base-catalysed and of second-order and have postulated a cyclic intermediate formed by nucleophilic attack of the various periodate anions on the carbonyl groups. The glycol-periodate esters arise from electrophilic attack on the diols. 47 G. J. Buist C. A. Bunton and J. H. Miles J . Chem. SOC. 1959 743. 48 S. J. Angyal and J. E. Klavins Austral J. Chem. 1961 14 577. 49 S. J. Angyal and R. J. Young J . Amer. Chem. SOC. 1959 81 (a) 5251 ; (6) 5467. 5 o R . Criegee E. Hoger G. Huber P. Kruck F. Marktscheffel and H. Schellenberger Annalen 1956 599 81. 51 P. Levesley W. A. Waters and A. N. Wright J . Chem. SOC. 1956 840. 52 K. B. Wiberg and K. A. Saegebarth J . Amer.Chem. SOC. 1957 79 2822. 53 A. K. Mitra and A. S. Perlin Canad. J. Chem. 1959 37 2047. 54 V. J. Shiner and C. R. Wasmuth J . Amer. Chem. SOC. 1959 81 37. 9 Quarterly Reviews Support for these views has come from isotopic labelling experiment^.^^ When pinacol and 2-methylpropane-l,2-diol were oxidised in 180-enriched water no label was found in the acetone. Thus oxygen atoms of periodate (which exchange rapidly with water5s) do not become linked to carbon (e; Scheme 2). Experi- ments with [180]biacetyl and with methylacetoin were more difficult to interpret but showed that here periodate oxygen atoms were being linked to carbon. The w * -* 0 * o 0,11 HO-CMe -0 I O-CMe2 Me,CO 8 I 0-CMaR MQ€OR 0 11 HO-CMeR - \\A 1 ___) * '\ * //I I O* *OH Me2C0 6 H 6 $ * - I y 0 II C*.O-CMe -* II,O-CM% - 0-1 I _Ic C=% II'g-T-OH MeCGo* 9 Me 'OH Me (4 O=f\o..3 Scheme 2 mixed nucleophilic and electrophilic r6les of the reagent were strikingly observed with methylacetoin for the acetic acid produced was labelled while the acetone was not (f).The breakdown of the cyclic intermediates in glycol oxidation with periodate and chromatez3 b have been discussed from a theoretical viewpoint.57 Stable glycol chelates have been obtained with some antimony (v) compounds which on pyrolysis give the glycol fission products.58 Contrary to an earlier suggestion glycol oxidation does not involve free radicals methyl methacrylate not being polymerised in the reaction mixture.59 The graft-polymerisation of acrylonitrile on cellulose which is undergoing periodate oxidation is thus surprising but may be due to secondary free-radical reaction of the many aldehyde groups which do in fact terminate polymerisa- tion.60 A study of glycol fission by electron spin resonance is lacking.Intermediates in the Oxidation of Polyo1s.-Because of the importance of periodate oxidation in the carbohydrate field the intermediates arising in the oxidation of various polyols have been examined in detail. We have already referred to the preferential cleavage of certain bonds for steric reason (p. 8). 210; CH20 2HC02H - 55 C. A. Bunton and V. J. Shiner J. Chem. SOC. 1960 1593. 56 Cf. M. Anbar and S . Guttmann J. Arner. Chem. Soc. 1963,83 781. 57 M. C. R. Symons J. Chem. SOC. 1963 4331. 58 F. Nerdel J. Buddrus and K. Hoher Chem. Ber. 1964 97 124. 59 H. Tanabe Chem. Pharm. Bull. (Tokyo) 1960 8 365 (Chem.Abs. 1961 55 10307). 6o T. Toda J . Polymer Sci. 1962 58 411. 10 Sklarz In the complete degradation of glucose five molecules of the reagent are required and formic acid (five molecules) and formaldehyde (one molecule) are formed but the reaction does not proceed at a uniform speed. At pH 3-5 distinct stages in the uptake of reagent and liberation of fragments were observed,61t62 and p-formylglyceraldehyde (IX) was isolated from the reaction mixture.g3 Glucose is oxidised mainly in the pyranose form but exists to about 15 % in the acyclic and furanose forms as estimated from the initial release of formaldehyde by oxidation of the 5,6-b0nd.~~ The formyl ester (IX) is relatively stable at pH 3.6 but is quickly hydrolysed at pH 7 where the oxidation is smooth and complete.Consideration of these esters can be useful in structural studies. Amylose consists of largely unbranched chains of 1,4-a-linked glucose units (X). Periodate oxidation (Scheme 3) releases one molecule of formic acid from each terminal residue by direct fission and a further molecule by subsequent hydrolysis of the formyl ester at the reducing end. The total amount thus represents one un- branched molecule of amylose whose length can then be calculated given the sample weight. Conflicting results were obtained as to the release of the third molecule of formic acid from a simple model compound maltose (X; n = O).64 To avoid the uncertainty Wolff et al.,S5 working with a corn amylose first estimated the formic acid released immediately and then after destroying excess of periodate hydrolysed the intermediate ester (XI) and estimated the additional formic acid.This procedure gave a separate estimate of the number of reducing residues in the amylose molecule which was thus shown to consist on average of two branches. The chemistry of the polyaldehydes produced in the oxidation of poly- saccharides has been reviewed.ss Carbohydrates tend to reduce more periodate than expected from mere glycol fission.67 The new reaction was shown to be a hydroxylation of CH- groups activated by adjacent carbonyl and is discussed further later (p. 15). An a-alkoxy-malondialdehyde moiety (XII) arises from the 4-O-substituted terminal residue of polysaccharides such as (X). The activated CH-group of the largely enolised dialdehyde (XII) is hydroxylated by periodate to give a hemi- acetal (XIII).The formation and breakdown of such intermediates has been studied with model compounds (XII; R = CH or C6H5.CH2).69 At pH 3.6 where (XIII) is most stable to hydrolysis oxidative cleavage predominates 61 F. S. H. Head Chem. and Ind. 1958 360; S. A. Warsi and W. S. Whelan ibid. p. 71. 62 L. Hough T. J. Taylor G. H. S. Thomas and B. M. Woods J. Chem. SOC. 1958 1212. 64 K. H. Meyer and P. Rathgeb Helv. Chim. Acta 1943 31 1545; A. L. Potter and W. Z. Hassid J . Amer. Chem. SOC. 1948 70 3489. 65 I. A. Wolff B. T. Hofreiter P. R. Watson W. L. Deatherage and M. M. Machfasters J. Amer. Chem. SOC. 1955 77 1654. 66 R. D. Guthrie Adv. Carbohydrate Chem. 1961 16 105. 67 T. G . Halsall E. L. Hirst and J. K. N. Jones J . Chem. SOC. 1947 1427. 68 C. F. Huebner S. R. Ames and E. C.Bubl J . Amer. Chem. Soc. 1946 68 1621. 69 (a) J. C . P. Schwartz and M. MacDougall J . Chem. Soc 1956 3065. (b) M. Cantley L. Hough and A. 0. Pittet Chem. and Ind. 1959 1126 1253. C. Schopf and H. Wild Chem. Ber. 1954 87 1571. 11 QuarterZy Reviews -0. ?> H O n -H CHO I CHO RO*YH (XI I) $H,OH HCO,H H-CO,H + C ~ R 3% etc. I 10; CHO i ,!?$& CHO I 04- (XI I) - RO-?-OH YHO (XIII) CHO CHo ROH + 70 L- atc. Scheme 3 giving formic acid and an ester which is slowly hydrolysed further and oxidised. At other pH's hydrolysis of (XIII) occurs followed by oxidation of mes- ~xalaldehyde.~~ In the later stages of carbohydrate oxidation the above hydrolyses and hy- droxylations varying independently with pH are superimposed to produce a complicated pattern of reactions. This has been resolved for some simple mono- and di-saccharides by kinetic cstimation of the various fragments at different aciditie~.~~*~l Under the acidic conditions used in early studies over-oxidation was found to be associated with the liberation of iodine.g7 A method for the detection and estimation of 1,4-links in polysaccharides was even devised on the basis of this fact.72 It has since been shown that sodium iodate in acid solution oxidises benzyloxymalondialdehyde the iodide ion formed being re-oxidised to Over-oxidation is usually undesirable since it confuses the number of true 70 M.Cantley L. Hough and A. 0. Pittet J . Chem. SOC. 1963 2527. J. Chem. Soc. 1954 603. 73 K. Ahlborg Svensk Kem. Tidskr. 1942 54 205; Chem. Abs. 1944 38 4254. L. Hough and B. M. Woods Chem. and Ind. 1957 1421 ; F.S. H. Head and G. Hughes 12 Sklarz glycol-cleavage steps from which structural deductions are most easily made. Low temperatures (5 and reduced concentrations of periodatesb have been recommended to minimise it. In oxidation mixtures exposed to sunlight less specific hydroxylations involv- ing singly-activated CH-groups can occur and oxidations of polyols should therefore be carried out in the dark.74 Amino-alcohols.-The periodate cleavage of N-primary and N-secondary a-amino-al~ohols~~J* (p. 5) is fastest at pH 7-8 and is suppressed in mildly acid solution. (The oxidation of hydroxy-amino-acids is discussed separately below.) NH p 10,- N% CH2W R ( 3 c H 2 M HrOs- R-C~;CH~LR' - c% - + R % COR' (x I v> Scheme 4 4-Hydroxymethyloxazolines (XIV) are oxidised in a two-stage reaction (Scheme 4).75 Acid-catalysed ring opening produces an a-amino-alcohol which is cleaved in slightly alkaline solution.The tertiary amino-group does not prevent cleavage entirely.ls Oxidation of desosamine (XV; R = H) gave in turn the tetrose (XVI) and acetaldol and the methyl acetal (XV; R = CH,) also took up one molecule of the reagent.76 However erythromycin and erythralosamine parent glycosides of 0 gave" the N-oxide (p. 23). An electron-withdrawing substituent on nitrogen such as the acetyP7 or 2,4-dinitrophenyl retards the cleavage which can however proceed in the latter case if there is a second hydroxyl group vicinal to the first. Thus the arylaminoethanol (XVII) resists oxidation but the 3-arylaminopropane-1,2-diol (XVIII) is cleaved giving 2,4 dinitr~aniline.'~ Studies on the oxidation of 2-amino- and 2-acetylamino-2-deoxyglucose 73 Ref.16 in M. Cantley et al. ref. 70. 74 F. S. H. Head J. Text. Znst. 1953 44 ~ 2 0 9 ; Chem. Abs. 1953 47 8378. 76 H. L. Wehrmeister J . Org. Chem. 1961 26 3821. 76 R. K. Clarke Antiobiotics and Chemotherapy 1953 3 663. 77 E. H. fly^ M. V. Sigal jun. P. F. Wiley and K. Gerzon J. Amer. Chem. SOC. 1954 76 3121. K. Hattori H. Harada and Y . Hirata Bull. Chem. SOC. Japan 1962,35 312. 13 Quarterly Reviews (XIX; R = H or Ac) (Scheme 5 ) and the corresponding methyl glucosides confirmed that N-acetylation prevents the aminoalcohol c l e a ~ a g e . ~ ~ ~ ~ ~ The subsequent reactions involving hydrolysis of (XX) and the hydroxylation of 2,4-(02N)2C,H3*NH.CH,*CH20H 2,4-(0,N),C,H3.NH*CHe*CH.CH,0H I (XVII) OH (XVIII) aminomalondialdehyde derivatives (XXI; R = H or Ac) have been elucidated.61 CHO CH,OH CHO CHO -% H t N H R 3.H O t H (xx 0 NHR NHR (x 1x1 (xx) Scheme 5 There have been few studies on the kinetics and mechanism of the fission. It is accderated by increasing pH and probably involves the unprotonated amine.s2~83 With rare exceptions only second-order kinetics have been observeds4 which permit of no deduction as to an intermediate. The kinetic form may be due to the very dilute solutions used by the Czech workers.28p36 Evidence for a cyclic intermediate is again stereochemical. In a series of acyclic a-amino-alcohols the threo-isomers were consistently oxidised more quickly than the erythuo-compounds. This difference was enhanced by N-methylation which when followed by periodate oxidation at pH 6.5-7.0 was recommended as a method for establishing the relative configurations of diasteroisomeric cu-amino-alcohols.84 N-Benzylation retarded the oxidation of the threo-compounds to such an extent that the usual order could be reversed.These effects are interestingly rationalised by the authors in terms of hydrogen bonding.84 cis-2-Aminocyclopentanol is oxidised somewhat more quickly than the trans isomer but the difference is much smaller than for the corresponding diols,82 and the order is reversed in the aminocyclohexanols.84 Differences in hydrogen bonding presumably contribute to these effects. H PhCH-NHEt 10,- P hCH= N Et I - I- (11 Ph CH*NHEt PhCHO H (xx I I) @XI 11) 70 R. J. Jeanloz and E. Forchielli J. Biol. Chem. 1951 188 361. 8o L. Hough and M. I. Taba J .Chem. SOC. 1956,2042. 81 M. Cantley and L. Hough J . Chem. SOC. 1963 2711. ** G. E. McCasland and D. A. Smith J . Amer. Chem. SOC. 1951 73 5164. 83 G. Dahlgren and J. M. Hodson J . Phys. Chem. 1964 68,416. 86 J. Kovar J. Jary and K. Blaha Coll. Czech. Chem. Comm. 1963 28,2199. 14 Sklarz a-Diamines.-Periodate oxidation is similar to that of a-amino-alcohols with regard to rate and pH.ls Piperazine (XXII) gave besides ammonia and formalde- hyde a little formic acid,85 while benzaldehyde and benzylidene-ethylamine were formed from the diamine (XXIII).s6 The second product is probably the precursor of the aldehyde. 3 Oxidation of Enolic Compounds Activated CH-Groups and Eno1s.-The hydroxylation of malondialdehyde intermediates in the oxidation of sugars has already been mentioned.A studys8 with a series of acyclic model compounds (XXIV) showed that one of the activating groups must be an aldehyde or carboxyl function (Rf = H or OH) but that there is no oxidation with a keto-group (as in acetylacetone) or a cyano- group (as in cyanoacetic acid). The intermediate (XXV) undergoes normal cleavage giving ultimately formic acid (R1 = H) or carbon dioxide (R1 = OH) and other expected fragments depending on R2 and R3 (alkyl alkoxyl or hydro- gen). Further hydroxylation of (XXV) competes with the slower cleavage when R1 = OH malonic acid giving some oxalic acid and for alkoxymalondialde- hydes hydrolysis of the hemiacetal (XXV; R2 = OAlk) can occur (p. 12). Cyclic p-diketones are oxidised smoothly.s7 The reductone (XXVII) (p. 17) and the triketone (XXVIII) were oxidised (Scheme 6) just as readily as 1,3- cyclohexanedione (XXVI; R = H) and are the probable intermediates.Predict- ably 2-alkyldiketones gave a carbolcylic acid (XXIX) instead of carbon dioxide and 2-dialkylketones were not oxidised. One of the activating groups may be an aromatics8 or heteroarornati~~~ ring as in benzyl ketones (slow oxidation)88 and in riboflavin,89b and rapid hydroxyla- tion probably occurs also in the p-dinitrone (LXXV p. 25).86 The mechanism of hydroxylation is unknown. Enolisation cannot be the sole factor as pointed out by Bose et aZ.,90 since some weakly enolised compounds such as malonic acid are readily oxidised while certain strongly enolised compounds (e.g. acetyl- acetone) are hardly attacked. The cyclic mechanism (XXX) suggested for malon- dialdehyde and its largely enolised derivativesg0 cannot be extended readily to malonic acid or for steric reasons to cyclohexanedione.A. Wickstrom and A. Valseth Ann. pharm. franc. 1954 12 777. 88 V. M. Clark B. Sklarz and Sir A. R. Todd J. Chem. SOC. 1959 2123. 87 M. L. Wolfrom and J. M. Bobbitt J . Amer. Chem. SOC. 1956 78 2849. 8B(u) C. F. Huebner R. Lohmar R. J. Dimler S. Moore and K. P. Link J. B i d . Chern. 1945,159,503; L. J. Haynes N. A. Hughes G. W. Kenner and Sir A. R. Todd J. Chem. SOC. 1957 3727; (b) H. S. Forrest and A. R. Todd J. Chem. SOC. 1950,3295. H. Felkin Bull. SOC. chim. France 1951 915. J. L. Bose A. B. Foster and R. W. Stephens J . Chem. SOC. 1959 3314. 15 Quarter& Reviews (XXVI) 2 10; R*C02H (XXIX) 2x0,- t + *10; &R &; '0;- k0 C C H *2H (XXVI I) (XXVI I I) Scheme 6 A transient malonic-periodic acid anhydride can be envisaged in which enolisation is promoted but there is no precedent or evidence for this at present.As shown experimentally for catechol (p. 19) a labile enol-periodate complex (XXX) (XXXI) J u may be formed from the enolised cyclic p-diketone and written hypothetically as (XXXI). The contrasting resistance of the acyclic diketones is perhaps related Me Me Me (xxx I I) (XXXIII) to the anomalous properties of their e ~ o l s . ~ l However the above discussion must remain speculative without further experimental evidence. G. S. Hammond in 'Steric Effects in Organic Chemistry' ed. M. S. Newman Wiley New York 1956 p. 452. 16 Sklarz Free radicals were not detected in the oxidation of malonic acid.D2 However a study of the hydroxylation by electron spin resonance would be valuable.Methylcyclopentane-2,3-dione which exists largely as the enol -I) in an exothermic oxidation afforded acidic products one of which was probably (xxxIII) presumably via hydroxylation as above and cleavage.93 The iodo- compound (XXXIV) was isolated when limited amounts of periodate were used. Reductones.-The mductone structure -C( :O)C(OH) = C(0H)- is oxidised rapidly and the triketone subsequently cleaved as in triose r e d u c t ~ n e ~ ~ ~ ~ * and in (XXVII p. 15)J5J7 The dimethyl ether (Xxxvd) of reductic acid was not oxidised but the parent reductone (XXXVa) and the monomethyl ethers (XXXVb) and (XXXVc) gave a-oxoglutaric acid.D3 Iodine was formed under the acid conditions. This recalls the behaviour of the enolic a-alkoxymalondialdehydes already mentioned to which the monoethers (XXXVb and c) are in fact related structurally.As in the oxidation of catechol the reductone oxidation probably proceeds by removal of the second enolic proton rather than by attack of water on carbon (g). Other mechanisms are suggested by Hesse and Mixg3 Phenols.-Several research groups concerned with lignin chemistry have studied the oxidation of phenolic compounds by peri~date.~~ They confined themselves to recording the uptake of the reagent and did not isolate products from the often coloured solutions. In a recent set of elegant studies Adler and his col- leagues have described the complex products which can arise and the essentially simple reactions leading to them.96-100 Adler’s oxidations were carried out with sodium periodate in aqueous or 80 % acetic acid solution.The phenols studied were of two types the dihydric phenols with their mono-ethers and some methylphenols. The study was prompted by the observation that methanol is formed rapidly and quantitatively in the periodate oxidation of guaiacol (XXXVI) and an estimation of guaiacol residues in lignin was based on this.96 The red solution 9a M. C. R. Symons J. Chem. SOC. 1955,2794. gg G. Hesse and K. Mix Chem. Ber. 1959 92 2427. B4 J. C. P. Schwartz Chem. and Ind. 1955 1588. *= J. P. Feifer M. A. Smith and B. R. Willeford J . Org. Chem. 1959,24,90 and refs. therein. s6 E. Adler and S. Hernestam Acta Chem. Scand. 1955 9 319. g7 E. Adler and R. Magnusson Acta Chem. Scand. 1959 13 505. (a) E. Adler R. Magnusson B.Berggren and H. Thomelius Acfa CIiem. Scand. 1960 14 515; (b) E. Adler and B. Berggren Acfa. Chem. Scand. 1960 14 p. 529; (c) E. Adler R. Magnusson and B. Berggren Acta. Chem. Scand. 1960 14 p. 539. O9 E. Adler I. Falkenberg and B. Smith Acta Chem. Scand. 1962 16 529. loo (a) E. Adler L. Junghahn U. Lindberg B. Berggren and G. Westin Acta Chem. Scand. 1960 14 1261 ; (6) E. Adler J. Dahlen and G. Westin Acta. Chem Scand. 160 14 p. 1580. 17 Quarterly Reviews t MuOH (XXXVI) (XXXVI I) (XXXVI 11) from the oxidation of (XXXVI) gave o-benzoquinone (XXXWI) and cis-cis- muconic acid (XXXVIII) obviously derived by further cleavage of the q u i n ~ n e . ~ ~ The reaction is general for catechol and quinol ethers the latter yielding p - benzoquinone but resorcinol monomethyl ether is attacked only very slowly giving methoxy-p-quinone.The quinones are also formed in the very fast oxidation of catechol and quinol themselves. The relative facility of these various reactions is apparent from the major products in the following cases when two pathways are possible. Simple de- hydrogenation is preferred to deinethylation (h i) and the latter reaction is faster for para- than for ortfzo- placed groups (j).97 U OMe It Sodium bismuthate also demethylated guaiacol but lead tetra-acetate and Fremy's salt gave 2-methoxy-p-benzoquinone. para-Hydroxylation by periodate was observed in the by-products (XL) and (XLI) (coerulignone) of the oxidation of 2,6-dimethoxyphenol (XXXIX),9sa but the mechanism is probably ionic. W (XXXIX) (XLV) Besides the expected reaction a + OMe 0 rearrangement occurred (Scheme 7) in the oxidation of the phenyl ether (XLII) with the formation of a biphenyl (XLIII) formed also by the normal oxidation of 2-hydroxydibenzofuran (XLIV).Q7 Dimeric products were obtained in concentrated solution and at room tem- perature from the oxidation of (XXXIX).3,8-Dimethoxy-1 ,Znaphthaquinone 18 Sklarz 0 ' 0.6 / "R,. (XLI v) 0 Scheme 7 (XLVII)98c was formed together with small amounts of an isomer. Adler showed that the Diels-Alder dimer (XLVI) of 3-methoxy-o-quinone (XLV) was oxidised by periodate to (XLVII) in a pathway involving diketone fission and decarb- oxylati~n.~~ b 2 (XLV) - __c (XLVI) (XLVI I) The details of the mechanism by which the dihydric phenols and their ethers are oxidised are not yet understood. Kinetic studies by the stopped-flow method have revealed for catechol the second-order formation of an intermediate which breaks down by a first-order reaction but no intermediate was detected for guaiacol.lol The oxidation of quinol is of the second order.lo2 Using 180-labelled water Adler Falkenberg and Smithg9 demonstrated a difference between the oxidation mechanism of the dihydric phenols and their monoethers.The water-soluble guaiacol derivative (XLVIII) was oxidised with sodium periodate in l80-enriched water but the liberated methanol contained no label. The o- and p-benzoquinones isolated from similar oxidation of the monomethyl ethers were 50% labelled. Since the quinones did not exchange l80 with water at the experimental pH (3-4) the label must have entered during the oxidation envisaged as the attack of water on a periodate ester (XLIX) or a transient cation (L).Attack by water at the methyl carbon would have given labelled methanol . When catechol and quinol were oxidised in H,180 the quinones obtained were not labelled. Here the water molecule removes the hydroxylic proton rather than attacking at carbon (k). lol E. T. Kaiser and S. W. Weidman Tetrahedron Letters 1965 497. lo2E. T. Kaiser and S. W. Weidman J . Amer. Chem. SOC. 1964 86 4354. 19 Quarterly Revfews EtCH-S03Na OObk HO (XLVII I) (XLIX) ( L) The periodate oxidation of alkylphenols is surprisingly fast leading to various products of ortho- and para-hydroxylation which have been elucidated by Adler and his group.1oo Phenol itself is only very slowly atta~ked.~' The oxidation of 2,4-dimethyl- phenol illustrates the various reactions that have been found to occur.Dimeric products (LV and LVI) arise from the o-quinol (LI) itself and from its Diels- Me OH 2(LI) - - O w e (LI) + (LIII) - Me Me (LV) (LVI) Adler addition to the o-quinone (LIII). The latter arises from the oxidation of (LII) the alternative ortho-hydroxylation product.loOa The p-quinols e.g. (LIV) which are minor products do not dirnerise.lo0* The reaction of oestrogens with periodate reported without details may involve hydroxylation of the steroid at C(1Q)).103 The mechanism is probably analogous to that of the first hydroxylation step in the oxidation of guaiacol [cf. (XLIX or L) 1. Other hydroxylations of phenols have been reviewed.lo4 Flavanols.-The flavanols (LVII; R = H or OCH,) were oxidised by periodic acid in aqueous dimethylformamide to give a tautomeric mixture of the 2- lo3 R.A. Harkness and K. Fotherby Experentia 1961 5 253. lo* J. D. Loudon Progr. Org. Chem. 1961 5 46; W. A. Waters ibid. p. 35. 20 hydroxyflavandiones (LVLII ; analogue of (LVIII; R = H) flR 0 (LVI I) Sklarz R = H or OCH3).lo5 In methanol the methoxy- was obtained as its methyl hemiacetal (LX).lOB Q P R 0 (LVIII) O It (LIN Smith has pointed out the analogy of this oxidation with that of phenols and reductones. In those reactions and in the oxidation of simple enols one may consider the attack of water on a transient cation or enol-periodate ester (I). If R is a hydroxyl group a diketone is formed by loss of a proton; otherwise C-hydroxylation occurs. In either case further reactions may ensue as illustrated throughout this section.R HO 6 4 Other Functional Groups Alcohols Olefins and Epoxides-Methanol and ethanol are attacked only in the light.13 A selective oxidation of the 11p-OH group in steroids to the 11- ketone has been reported and at higher concentrations of periodic acid 3a- and 17#l-16p-hydroxyl groups are also attacked.lo3 W) HO (UI I I) HO (W HO Scheme 8 lo5 M. A. Smith J . Org. Chem. 1963,28 933. lo6 M. A. Smith R. A. Webb and L. J. Cline J. Org. Chem. 1965,30,995. 21 Quarterly Reviews Olefins are inert to periodate at least at room temperature and steroids again provide the only well-established exception.lo7 With a three-fold excess of periodic acid cholesterol (Scheme 8) gave 3P7Sa,6P-cholestanetriol (LXI) while with ten-fold excess the 6-ketone (LXTI) was also formed.The trio1 is not an intermediate being unaffected under the conditions (trans-diaxial vic-glycol !). The epoxide (LXIII) was presumed to be involved. With the high concentration of periodic acid acid-catalysed ring-opening by periodate ion cannot be excluded the periodate ester then collapsing to the ketone. A simple hydrolysis of epoxides catalysed by periodic acid was reported earlier by Fieser and Rajagopalan.lo8 Styrene and stilbene oxides were reported (without details) to take up periodate slowly.88 Sulphur Compounds.-The oxidation of sulphur compounds with periodate was studied by Sykes and Todd in connection with the penicillin problem.loQ Thiols are oxidised via disulphides to sulphonic acids though the second stage is effected also by iodic acid. Thus spontaneous hydrolysis of the thiazolidine (LXIV) gave a thiol oxidation of which gave penicillaminic acid (LXV).A periodate estimation of penicillin has been described.l1° Thioethers (sulphides) are oxidised to sulphoxides. Thus excellent yields of (LXVI) were obtained by oxidation of the sulphide at O" although other oxidis- ing agents failed.lll At higher temperatures sulphones are forrned.l1lJl2 Although there was no cleavage of 2-aminoethanethiol it has been observed for thioethers. The ethanol derivatives (LXVII) and (LXVIII) were both oxidised (LX I v> I CO,H / CO H ( U V ) Et S AcO (E t S) &H C H *CH ,CH *CH; NHAc (LXVIII) (LXVII) lo' R. P. Graber C. S. Snoddy H. B. Arnold and N. L. Wendler J. Org. Chcm. 1956 21 1517. lo8 L. Feiser and S. Rajagopalan J. Amer. Chem. Soc. 1949,71 3938.lob P. Sykes and A. R. Todd in 'The Chemistry of Penicillin' ed. H. T. Clarke J. R. Johnson and R. Robinson Princeton 1949 p. 927. lloL. Mazor and M. K. Papay Acta Chim. Acad. Sci. Hung. 1961 26 473; Chem. Abs. 1961 55,20330. 111 N. J. Leonard and C. R. Johnson J. Org. Chem. 1962 27 282. lla W. A. Bonner and R. W. Drisko J. Amer. Chem. SOC. 1951,73 3699. 22 Sklarz by sodium periodate the former alone yielding formaldehyde. Detailed study of the oxidation of thio-derivatives of sugars and amino-sugars showed that cleavage of the group R-S-C-C-X can occur when X is OH NH, or NHAc.I13 In all the examples adduced the carbon atom adjacent to C-X actually carries two thioether groups or one together with a second electronegative function as in (LXVIII). The importance of this is seen in the remarkable cleavage of the acetaldehyde derivative (LXIX) to give methanol and of 2-deoxyglucose diethyl- (but not dibenzyl-)dithioacetal (LXX) to give as an intermediate glycolalde- hyde.l13 The fate of the sulphur-bearing fragment was not reported and the reaction merits further study.Amines.-Formation of an N-oxide from a tertiary amino-group was observed with erythromycin (p. 13),77 and has been found for N-(2-hydroxyethyl)- and N-propyl-piperidine.ll* Triphenylphosphine readily gave the 0 ~ i d e . l ~ ~ Complexes apparently formed between potassium periodate and some primary and second- ary aliphatic arninesllj merit further study in connection with the mechanism of these reactions. A new pattern of periodate oxidation thus emerges summarised by the expression R,Z -t R,Z-0 where 2 is N P or S.Its scope and utility par- ticularly for the first two classes remains to be studied. Mono- and di-alkylanilines toluidines and more slowly halogenoanilines are o ~ i d i s e d . ~ ~ J l ~ The rate of the extensive oxidation of phenylenediamines decreases in the order meta > ortho > para,17 in contrast to the phenols (p. 18). Electron-withdrawing groups such as -CHO -COCH,,117 and -N0,78$116 retard or suppress the oxidation particularly when placed ortho or para to the amino- group- Reaction products have been studied only for aniline and vary considerably with pH. Emeraldine is formed at pH 1,118 various amino- and anilino-quino- neimines and anils at pH 4*5,119 and unidentified products at pH 9. Free radicals are involved.120 Hydrazine Derivatives.-Hydrazobenzene is rapidly oxidised by periodate to a~0benzene.l~~ Monoalkylhydrazines give nitrogen and alkanes in good yield121a via an alkyldi-imine (m) the mechanism and stereochemistry of whose break- down depends on the amount of base present.The reaction has been used in a synthesis of 3-deoxyglucose derivatives,121 and an improved preparation of ll9 L. Hough and M. I. Taha J. Chem. SOC. 1957 3994. 114 B. Sklarz and A. K. Qureshi unpublished observations. 116 K. L. Jaura K. K. Tewari and R. L. Kaushik J. Indian Chem. SOC. 1963 40 1008. 118 J. Kawashiro J. Pharm. SOC. Japan 1953 73,943 (Chem. Abs. 1954,43 10630). 117 H. Tanabe J . Pharm. SOC. Japan 1956 76 1023 (Chem. A h . 1957 51,2598). 118 H. Tanabe J. Pharm. SOC. Japan 1958 78 410 (Chem. Abs. 1958 52 14562). lL0 H.Tanabe Chem. Pharm. Bull. (Tokyo) 1958 6 645 (Chem. Abs. 1960 54 16417) and refs. therein. lao H. Tanabe Chem. Pharm. Bull. (Tokyo) 1959,7 177 316; (Chem. Abs. 1960,54,22425). 121 (a) D. J. Cram and J. S. Bradshaw J. Amer. Chem. SOC. 1963,85 1108; (b) D. M. Brown and G. H. Jones Chem. Comm. 1965 561. 23 Quarterly Reviews nicotinaldehyde by the periodate oxidation of nicotinic acid hydrazide at 0" is a related reaction.122 Strongly acid solutions of various hydrazine derivatives have been titrated with potassium ~eri0date.l~~ Hydroxylamine Derivatives.-Hydroxylamine is instantly oxidised by sodium periodate with formation of nitrous oxide and iodine.lM Analogous oxidation of phenylhydr~xylaminell~ and rnethylhydroxylamine126 gave the nitroso- compound in the latter instance as the cis-dimer without tautomerisation to the oxime.Such oxidation proceeds also in chloroform with tetraethylammonium periodate.lO However a-hydroxyamino-acids undergo instantaneous oxidative decarboxyla tion (n) .126 Primary hydroxamic acids are oxidised to nitrous oxide and the parent acid ( o ) . ~ ~ ~ The formation in fair yields of amides in the presence of primary a m i n e ~ l ~ ~ Th 10,- Th (m) Me-q-Et - Me-<-Et -+ MeCHPhEt + N NH~NH N=NH IO+' (4 RCH (NH. OH) co2 H RCHO + co2 + (HNO) - H ~ N ~ O ~ - N~O points to the existence of an acylating intermediate,124 assumed to be (LXXI).127 N-Alkylhydroxamic acids e.g. (LXXII)86 similarly give the acid or amide,127 together with a nitroso-compound,86J25 e.g. (LXXIII). N-Hydroxypyrrolidines with an a-hydrogen atom are rapidly oxidised to the nitrone (p).10v86 This reaction is also seen in the first two stages of the oxidative OH 0- 120 H.N. Wingfield W. R. Harlan and H. R. Hanmer J. Amer. Chem. SOC. 1952,74 5796. 103 B. Singh and S. S. Sahota J . Sci. Ind. Res. India 1958 17,B 386 (Chem. Abs. 1959 53 7863). lap T. F. Emery and J. B. Neilands J. Org. Chem. 1962 27 1075. 1s T. F. Emery and J. B. Neilands J. Amer. Chem. SOC. 1960 82,4903. les G. A. Snow J . Chem. SOC. 1954 2588; J. B. Neilands and P. Azari Acta Chem. Scand. 1963,17 S190. 1%' B. Sklarz and A. F. Al-Sayyab J. Chern. Soc. 1964 1318. 24 Sklarz degradation of the 2,3'-bispyrrolidinyl (LXXIV) to the nitroso-acid (LXXIII) and the keto-nitrone (LXXVQs6 Hydroxylation at the activated C(3') is a probable step in the sequence. Several Al-pyrroline 1-oxides unsubstituted at C(2) (Lxxvn) were cleaved smoothly by sodium periodate at the double bond with formation of a nitroso- and a carboxyl group (LXXX).Io There is indirect evidence for a reaction path (LXXIV) I -O (LXXV) R ' G R ' m 5 " j o o ~ eR'C).lo; R r c O z H " t 0- OH R y R $ I R T 0- OH OH Fi' R (wxx> -0 (LXXVI 1) (LXXVII I) ( U X I x) via the hydrate (LXXVIII) and the hydroxamic acid (LXXIX). In agreement with this the nitrone-acid (LXXVII; R = CH, R' = C0,H) gave lavulalde- hyde presumably via oxidative decarboxylation of the a-hydroxyamino-acid (LXXVIII; R = CH, R' = CO,H)J0 The oxidation of nl-pyrroline to pyrro- lidone may be of similar nature. Amino-acids and Peptides.-Several studies on the periodate oxidation of amino- acids have been re~orded,1~*-~~~ usually as preliminaries to the oxidation of proteins.In a recent and extensive the periodate uptake and formation of carbon dioxide at various pH's were measured. Products have only occasion- ally been identified. Cleavage of the a-amino-acid function is extremely slow (cf. a-hydroxy-acids) but is promoted by higher temperature 129 and pH 132 and by N-alky1ati0n.l~~ Thus proline is oxidised even at pH 2.2 to A l - p y r r ~ l i n e ~ ~ ~ J ~ which is oxidised further at pH 7.2 to give 2-pyrr0lidonel~~ (cf. this page). Expectedly the polyamide chain of peptides is not susceptible to periodate and oxidation occurs only at the side-chains of certain constituent amino-acids. Free serine threonine and hydro~yprolinel~~ undergo normal cleavage permitting their estimation,l= followed by extensive further oxidation.In peptides they are cleaved only when in the N-terminal position when the laa P. Desnuelle S. Antonin and A. Casal Bull. SOC. Chim. biol. 1947 29 694. lS9 K. Arakawa J. Biochem. (Japan) 1957 44 217. 130 P. D. Bragg and L. Hough J . Chem. Soc. 1958,4050. 131 H. Hormann K. Hannig and G. Fries 2. physiol. Chem. 1959 315 109. 132 J. R. Clamp and L. Hough Biochem. J. 1965,94 17. 133 L. Skursky Z . Naturforsch. 1959 14b 473. 13p See ref. 132 for collected references. 25 Quarterly Reviews amino-group is free>% Thus brief treatment of corticotropin with periodate destroyed the N-terminal serine and the borohydride-reduced product had altered biological properties.lS In yet other amino-acids oxidation of the side-chain is independent of the a-amino-group.Hydroxylysine undergoes normal fission,’37 cysteine and cystine are oxidised to the sulphonic acid cysteic acid,log and the methionine residue of peptides gives the s~1phone.l~~ H (LXXX I) Tryptophan tyrosine (a phenol) and histidine are extensively oxidised to unknown coloured products. The slow dissolution of collagen in aqueous peri~date’~~ probably involves a specific cleavage at tyrosine of the type reviewed by W i t k ~ p l ~ ~ and the cleavage of the model amide (LXXXI) to give ethyl glycinate is also of this type.139b Recently substituted indoles (LXXXII; R = H or Me) have been shown to undergo mild rapid and specific oxidations the products depending on the acidity.140 Sodium periodate cleaves the 2,3-double bond of 3-alkylindoles (LXXXII ; R’ = H) giving ortho-acyl-N-acylanilines.Periodic acid leaves this bond intact but oxidises an alkyl methylene group attached at the 2-position to a carbonyl group (Scheme 9). A recent analysis131 of the effect of periodate treatment on the amino-acid composition of procollagen confirms that oxidation occurs largely at the amino- acids of the last type and is in consonance with earlier less refined studies on ovalbumin,12* ribon~clease,’~~~ chymotrypsin,l&b and 1ys0zyrne.l~~ (LXXXI I) HOCH,*CHR*CHO R*CO,H (LXXXIII) Scheme 9 la5 S. Fujii K. Arakawa and N. Aoyagi J. Biochem. (Japan) 1957 47 471. lS6 H. B. F. Dixon Biochem. J. 1962 83 91. 13* H. Zahn and L. Zurn 2. Naturforsch. 1957 12,B 788. 139 B. Witkop Adv. Protein Chem. 1961 16 (a) 221 ; (b) 252. 140 L. J. Dolby and D. L. Booth J . Amer. Chem. SOC.1966 88 1049. 141 (a) W. F. Goebel and G. E. Perlman J. Exp. Med. 1949,89,479; (6) E. F. Janscn A. L. Curl and A. K. Balls J. Biol. Chem. 1951 189 671. D. D. van Slyke A. Hiller and D. A. MacFayden J . Biol. Chem. 1941 141 681. L. Maekawa and M. Kushibe Bull. Chem. SOC. Japan 1954 27 277. 26 Sklarz Miscellaneous Reactions.-The oxidation of the mixed aldols (LXXXIII ; R = C,H or C,H,,) proceeds readily in cold aqueous dioxan containing bicarbonate the acids RC0,H being formed.l&a The reaction is comparable with the hydroxylation of other activated CH-compounds (p. 15). Cameron and his co-workers have suggested the intervention of free radicals arising from impurities in the dioxan. They observed two hydroxylations in the aphid pig- ment series which required the presence of benzoyl peroxide and thus involve free radi~a1s.l~~ However their phenolic-quinonoid substrates are not strictly comparable with aldol (LXXXIII) and /3-diketones (XXIV XXVI) and in their conditions (periodate in refluxing aqueous dioxan) the specificity of the reagent is lost even acetone being 0xidi~ed.l~~~ Glucose polymethyl ethers are appreciably 0 x i d i ~ e d .l ~ ~ ~ Iodine is liberated slowly from primary alkyl iodides. Prior hydrolysis appears to be promoted by the periodate ion (cf. p. 22) which oxidises the liberated iodide ion.lq4 Periodic acid has been used in a potentiometric estimation of uric acid,145 and dilute sodium periodate in the detection of ferrocene derivatives on paper chroma tog ram^.^^^ 5 Periodate as a Co-oxidant A mild specific reagent for the oxidative fission of olefinic double bonds was developed by Lemieux and von Rud10ff.l~' It consists of aqueous sodium periodate and potassium permanganate at pH 7.7 in a molar ratio of about 60 1.Intermediate aldehydes are oxidised further to acids which with ketones are the oxidation products. Permanganate first oxidises the olefin to the two ketols which are cleaved by periodate. The latter also reoxidises MnV to MnV" at this pH so that the process amounts to a permanganate-catalysed oxidation by periodate. The reagent has been particularly valuable in the analysis of oils and fats butanol or pyridine being used as co-s~lvents.~~~ The fragment acids from an unsaturated lipid are esterified in situ and identified by gas-liquid chromato- g r a p h ~ . ~ ~ ~ This method should also be useful in the oxidative degradation of terpenes,150 steroids,151 etc.The action of the reagent on various functional groups has been examined to define its specificity.152 For example oxidation of isolated alcoholic groups 148 (a) A. J. Birch D. W. Cameron Y . Harada and R. W. Rickards J . Chem. Soc. 1959 889; (b) D. W. Cameron R. I. T. Cromartie Y . K. Hamied E. Haslam D. G. I. Kingston Lord Todd and J. C. Watkins ibid. 1965 6923; (c) P. Fleury and R. Boisson Compt. Rend. 1939 208 1509; (d) G. D. Greville and D. H. Northcote J . Chem. Soc. 1952 1945. 144 A. B. Foster M. Stacey and R. W. Stephens J. Chem. SOC. 1959 2681. 145 A. Berka Analyt. Chim. Acta 1961 25,434. 146 A. N. de Belder E. J. Bourne and J. B. Pridham Chem. and Ind. 1959 996. 14' R. U. Lxmieux and E. von Rudloff Canad.J . Chem. 1955,33 1701. 148 E. von Rudloff Canad. J. Chem. 1956 34 1413. lP9 T. C. L. Chang and C. C. Sweeney J . Lipid Res. 1962 3 170. 150 R. U. Lemieux and E. von Rudloff Canad. J. Chem. 1955,33 1710 1714; 1965,43,2660. 151 M. E. Wall and S. Serota J . Org. Chem. 1959 24 741. ls2 E. von Rudloff Canad. J. Chem. 1965 43 1784. 27 Quarterly Reviews is slow but activating groups accelerated the oxidation (banzyl alcohol giving benzoic acid) and may lead to further degradation as with ally1 and tetra- hydrofurfuryl alcohols. Aldehydes are oxidised smoothly to acids some /3- dicarbonyl compounds not attacked by periodate alone are oxidised and acyclic ethers also suffer a slow oxidation. By the eflective use of very small quantities of permanganate greater selectivity is achieved than is normally associated with this reagent.Disubstituted olefins are oxidised cleanly by a mixture (210:l molar) of sodium periodate and osmium tetroxide in aqueous dioxan but with more substitution oxidation is slow.153 Aldehydes can be isolated. The method is thus selective and effects a valuable economy in the expensive and toxic osmium tetroxide. A combination of ruthenium tetroxide and sodium periodate (1 1 5 molar) has been used in the mild oxidation of steroid C(3)- and C(6)- alcohols in neutral conditions.lM I thank Professor V. M. Clark for suggesting my first periodate experiment. To Professor L. N. Owen and to Dr. G. J. Buist I am indebted for valuable comments and to Professor C. A. Bunton and the Chemical Society for per- mission to reproduce diagrams (IV) (VII) and (VIII). The facilities for experi- ment and study at Cambridge and at Imperial College are gratefully acknow- ledged. 153 R. Pappo D. S. Allen R. U. Lemieux and W. S. Johnson J . Org. Chem. 1956 21 478. 150 H. Nakata Tetrahedron 1963 19 1959. 28