8 Reaction Kinetics in Solution By S. 6. BROWN and P. JONES Department of Physical Chemistry University of Newcastle upon Tyne IN AN article written in celebration of the centenary of Nature Ferris’ comments, ‘Now they (scientists) are driven to read specially prepared collections of titles, from which they hopefully select the papers they actually want to see.’ 1969 has seen the demise of Current Chemical Papers the Chemical Society’s own ‘collec-tion of titles’ and it seemed appropriate to mark its passing. Section 8 (Kinetics and Mechanism) of Current Chemical Papers for 1969 listed articles according to the categories : Reviews 164 Radical Reactions 2082 Other Reactions 2613 Total 5612 Catalysis 753 ~ -Whilst the separation of reviews is sensible the choice of the remaining categories has proved less than satisfactory.The assignment of a radical mechanism is not uncommonly based on inference rather than demonstration and in many cases is debatable. The sub-section ‘Other Reactions’ is a non-category which has contained papers from very diverse areas of investigation and its over-tones of apartheid may well have disturbed the sensitive. Papers in the final section have ranged from studies with enzyme extracts to experiments with technically important heterogeneous catalysts. To readers interested in reactions in solution this classification scheme has offered no direct assistance. We have carried out the experiment of counting the papers cited in Current Chemical Papers 1969 which are recognisably concerned with homogeneous reactions in solution (47 % of the total) and applying a rudimentary classification scheme.The results are as follows : Category Number of Papers % of Total 1. Reviews 52 2. General 177 3. Substitution reactions : (a) Organic 482 (b) Inorganic 346 2-0 6.7 18.1 13-0 P. Ferris The Observer (Colour Supplement) 28th December 1969 p. 30 108 S. B. Brown and P. Jones 4. Redox reactions: (a) Organic 185 (b) Inorganic 171 (c) Autoxidation 43 5. Catalysis : (a) Acid-base 144 (b) Metal complex 132 (c) Enzyme 184 6. Unclassified 71 1 7.0 6.4 1.6 5 *4 5.0 6.9 26-8 Papers concerned with photochemical and radiolytic reactions have been excluded. 72 % of the papers are available in English. Of the remainder 15 % are in Russian 5.3 % German 2.9 % French and 1.9 % Japanese.The classification is clearly not without ambiguity particularly in the assignment of sub-categories. The second category includes new techniques work devoted to the development of general theory (e.g. kinetic and solvent isotope effects) and applications of recognised techniques which are not as yet commonplace (e.g. 3C isotope effects). We have tried to avoid double-entry and where possible have classified according to the emphasis indicated by the title of an article. This exercise has left the impression that advice to authors from editors of periodicals on the construction of titles could be a most useful aid in the subsequent problem of literature retrieval. About three-quarters of the literature may be classified according to this scheme.In quantitative terms the study of reactions in solution is dominated by investigations of substitution reactions almost one-third of all papers falling into this category. Within this category the largest sub-group is concerned with solvolysis reactions ; among inorganic reactions studies on transition metal (mainly Cr"' and Co"') complexes predominate. The remainder of this articIe is devoted to aspects of the kinetics and mechanisms of homogeneous redox reactions in solution some of which have received relatively little review attention in recent years. Perhaps the most general state-ment one can make concerning studies of these processes is that they reveal the limitations of unsupported kinetic investigations for determination of mechanism, particularly the natures of the redox acts themselves.This is partly because in many cases non-redox rate-limiting steps precede the redox process partly because of difficulties in characterising the participant species and partly because the redox behaviour of a particular species may depend on the nature of its reactant partner. The Faraday Society Discussion on 'Homogeneous Catalysis with Special Reference to Hydrogenation and Oxidation' has now been published. Besides containing many interesting and diverse papers this volume is particularly note-worthy for a masterly essay by Halpern,2 which forms the General Introduction to the Discussion and which gives a valuable perspective on achievements and problems in this area. Halpern concludes in an optimistic vein in referring to the J .Halpern Discuss Faraday SOC. 1968 46 7 Reaction Kinetics in Solution 109 problems of homogeneous catalytic activation of saturated hydrocarbons and the catalytic fixation of molecular nitrogen. The latter topic was clearly on the minds of a number of workers e.g. Pratt3 who commented that the problem in the biological fixation of N is probably not connected with the poor donor properties of nitrogen (molecular nitrogen complexes are now almost common-place) but with the highly endothermic nature of the first step in the reduction to the di-imine level. The subsequently published experiments of van Tamelen4,5 and co-workers are an important development in that they have achieved the reduction of N2 to ammonia or hydrazine in room temperature atmospheric pressure processes which can be operated in a cyclic overall catalytic fashion.The methods rely on the N,-fixing ability of titanium(r1) species and on the regenerability of Ti" and reduction of titanium-bound N2 both brought about by naphthalide radical anion. McQuillin6 and co-workers have demonstrated important kinetic correlations between homogeneous and heterogeneous hydrogenation reactions. From an analysis of the complex dependencies of rates of homogeneous catalytic hydro-genation of olefins upon olefin concentrations they advance the scheme : L,MH + S e L,MSH (1) (2) L,MSH + S L,M(S)SH (3) (4) L,MSH + H2 e L,MH + SH2 L,M(S)SH + H2 + L,MSH + SH2 where S is the olefin SH2 the alkane product L,MH is an appropriate hydrido-complex bearing ligands L where x 2 y > z.The stage L,MSH which in one case has been characterised e~perimentally,~ corresponds with the half-hydrogenated state postulated in heterogeneous hydrogenation. Reaction (3) has analogy in the phenomenon of self-inhibition due to hydrogen exclusion which is encountered in the heterogeneous hydrogena-tion of strongly adsorbed substances. Whereas in (1) olefin reacts with preformed hydride in (4) hydrogen reacts with pre-complexed olefin. This situation has close parallels in heterogeneous catalysis depending on whether the catalyst is pretreated with hydrogen or olefin which do not then mutually establish adsorp-tion equilibrium under the conditions of hydrogenation. Jardine and McQuillan' report studies on the homogeneous hydrogenation of a group of cycloalkenes using the complex [py2(DMF)RhC12(BH4)] in dimethyl-formamide (DMF) solution where the rate of hydrogenation shows the typical saturation behaviour with increasingalkene concentration found in heterogeneous J .M. Pratt Discuss Faraday SOC. 1968 46 93. E. E. van Tamelen R. B. Fechter S. W. Schneller G. Boche R. H. Greeley and B. Akermark J . Amer. Chem. SOC. 1969,91 1551. E. E. van Tamelen R. B. Fechter and S. W. Schneller J . Amer. Chem. SOC. 1969 91, 7 196. I. Jardine R. W. Howsam and F. J. McQuillin J . Chem. SOC. (C) 1969 260. I . Jardine and F. J. McQuillin Chem. Comm. 1969 502. ' J. Trocha-Grimshaw and H. B. Henbest Chem. Comm. 1968 757 110 S . B. Brown and P. Jones hydrogenation and where the rate order is the same in the homogeneous and heterogeneous cases.The same authorsg also discuss parallels in stereoselectivity between homogeneous and heterogeneous catalytic hydrogenation. Papers by Pregaglia" and co-workers and by Edwards' ' and co-workers again demonstrate the use of the simplest method of studying the kinetic role of a solvent [cf Goodall Annual Reports (A) 1968,65 1561 by looking for a term in solvent concentration in the rate law. In an attempt to avoid the 'ageing reaction' with water which results in a decrease with time in activity of CO(CN),~- for the activation of hydrogen Pregaglia et aE." examined the reaction in ethanol. The reaction proceeds without 'ageing' in ethanol but the rate of hydrogen absorption is slow in the anhydrous solvents and increases with increasing water concentra-tion.In fact the rate is accurately second order in H 2 0 up to 5-25 moll-'. The authors do not consider mechanistic implications except to comment that the effect could be related to ion pair formation but that the ratio kH20/kD20 = 1.8 cannot easily be explained by an ion-pair effect alone. Edwards' ' and co-workers have studied the kinetic role of water in the oxida-tion of organic sulphides to sulphoxides by hydrogen peroxide by using as solvents dioxan containing various concentrations of water. They concluded that the transition states contain water molecules which they suggest provide a proton transfer circuit. These reactions are formally closely related to the dehydration of carbonyl hydrates : >clO-" -+ >C=O + H20 0 - H 0 - H 0-H > S + I -+ >S=O + HzO and it is interesting to note that Bell and co-workers,12 using a very similar approach reached a similar conclusion about the kinetic role of water in the hydration of 1,3-dichloroacetone.EndicottI3 has discussed the possible effects of magnetic exchange interactions on the rates of electron-transfer reactions. Magnetic exchange interactions between adjacent paramagnetic metals in solids and in known binuclear com-plexes of paramagnetic ions are often very large even at room temperature. Similar interactions would be expected to occur in the 'activated complexes' of at least some electron-transfer reactions so that magnetic restrictions on the probability of electron transfer are to be expected.Limitations on the rate of I. Jardine and F. J. McQuillin Chem. Comm. 1969 503. l o G. Pregaglia D. Morelli F. Conti G. Gregorio and R. Ugo Discuss. Furuduy SOC., 1968 46 1 10. R. Curci R. Diprete J. 0. Edwards and G. Modena in 'Hydrogen Bonded Solvent Systems' eds. P. Jones and A. K. Covington Taylor & Francis London 1968 p. 303. R. P. Bell J. P. Millington and J . M. Pink Proc. Roy. SOC. 1968 A 303 1 . l 3 J . F. Endicott J . Phys. Chem. 1969 73 2594 Reaction Kinetics in Solution 111 electron transfer by Franck-Condon restriction^'^*' and by weak interaction between donor and acceptor orbitals16-’ have been discussed previously. Endicott ’ gives an essentially qualitative discussion of magnetic exchange restrictions based upon Anders~n’s’~ review of magnetic interaction in solids.He concludes that the antiferromagnetic coupling which can result from magnetic exchange in the interaction of paramagnetic complexes can account for the relative slowness of some electron-exchange reactions notably ‘inner-sphere’ reactions involving aqua-ions. The effect can also account for some of the specific ligand effects observed in reactions of this type. The theme of complications arising from the formation of binuclear and poly-nuclear metal complexes shows signs of rivalling ‘ion pairing’ in the vocabulary of explanations for unexpected behaviour. Nevertheless there are a number of well-documented cases where polynuclear complexes are important sources of complexity in redox reactions. The review by Spiro and Saltman2’ both serves as a timely reminder of the existence of an extensive body of equilibrium studies2’ and describes the results of more recent structural and thermodynamic investiga-tions.Wilkins andYelin2 demonstrate the potential of stopped-flow and particularly, temperature-jump methods for the investigation of the kinetics and mechanism of binuclear complex formation. They have studied the monomer-dimer inter-conversion of iron(III)-edta and related chelates and show that the results are consistent with the mechanism FeL(H,O) + FeL(0H) FeL(0H) + FeYOH) Fe2L20 + H 2 0 The break-up of the dimer is described by a two-term rate law and is strongly acid-catalysed. The dimer can form more easily from one molecule of aquo-species and one molecule of hydroxo-species than from two molecules of hydroxo-mononuclear iron(m) species.It is suggested that the formation of dimers may be controlled by the water-exchange rate as previously propo~ed.~ 14R. A. Marcus Ann. Rev. Phys. Chem. 1964 15 155. W. L. Reynolds and R. W. Lumry ‘Mechanisms of Electron Transfer’ Ronald Press Inc. New York 1966. l 6 J. Halpern and L. E. Orgel Discuss. Faraday Soc. 1960,29 32. A. G. Sykes Adv. Inorg. Chem. Radiochem. 1967,10 153. I s P. George and J. S. Griffiths ‘The Enzymes’ Academic Press New York 1959 pp. 1 , 289. P. W. Anderson ‘Magnetism’ vol. 1 eds. G. T. Rado and H. Suhl Academic Press, New York 1963 ch. 2. T. G. Spiro and P. Saltman ‘Structure and Bonding’ Springer-Verlag Berlin Heidel-berg New York 1969 vol. 6. L. G.Sillen Quart. Rev. 1958 13 146. 2 2 R. G. Wilkins and R. E. Yelin Inorg. Chem. 1969 8 1470. 23 M. Eigen and R. G. Wilkins ‘Advances in Chemistry’ Series No. 49 American Chemi-cal Society 1965 p. 55 112 S. B. Brown and P. Jones Meyer and T a ~ b e ~ ~ and Wilkins and YelinZ5 have obtained sets of data which permit tests of the equation derived by Marcus26 governing relationships between the rate constants for outer-sphere electron-transfer reactions : k12 = (kllk22K12f)+ where In f = (In K12)~/4 In ( k ,kZ2/Z2) and Z is the collision frequency between uncharged molecules in solution k is the rate constant of the reaction for reactants 1 and 2 k l l and k22 are the self-exchange rate constants for the two oxidation states of systems 1 and 2 and K12 is the equilibrium constant for the reaction.kl , k 2 2 and KI2 have been experimentally determined and k12(calc.) compared with k12(obs.). Meyer and T a ~ b e ~ ~ have obtained all necessary quantities for the V2 + reduction of R u ( N H ~ ) ~ ~ + and for the Fe3 + oxidations of R u ( N H ~ ) ~ ~ + and R ~ ( e n ) ~ ~ + . The data of Wilkins and Y e l i r ~ ~ ~ refer to the edta and cydta complexes of Cr"*"' Mn"*"' Fe"*"' and Co"*"'. In both cases the trends of k12(obs.) follow the predictions of the Marcus equation well and the agreement between k12(calc.) and k l 2(obs.) is reasonably good. Less complete data have been reported by Stasiw and Wilkins2 (for ferri-ferrocyanide coupled with other iron complexes) and by Burgess2' [oxidation of substituted tris(1,lO-phenanthroline)iron(Ir) complexes by Tl"'].Stasiw and W i l k i n ~ ~ ~ note that some ofthe reactions they studied were complicated [e.g. those of Fe(CN),H202-] probably as a result of the formation of binuclear or polynuclear species. Meyer and T a ~ b e ~ ~ consider that although complexes of back-bonding ligands such as 1,lO-phenanthroline and cyanide might be expected to show 'outer-sphere' electron-transfer reactions their mechanisms may differ significantly from those of complexes involving only saturated ligands because of electron delocalisation over the ligands. Although tris( l;lO-phenanthroline)iron(II) is justifiably considered to be a species that undergoes oxidation by a one-electron 'outer-sphere' pathway, evidence is accumulating that this is not always the case.In the reactionz9 with P2OS4- and that3' with CIOz- the rate is independent of oxidant concentration and is controlled by the rate of dissociation of the complex : Fe(phen),'+ + Fe(~hen),~+ + phen Shakhashiri and Gordon3' have shown that the reaction with C102- produces the iron(III)-phenanthroline dimer [(phen),Fe-@-Fe(phen),] and consider that this product is characteristic of these non-outer-sphere'reactions. On this basis they suggest that the reduction of H202 by Fe(~hen),~+ does not involve an outer-sphere one-electron process. 2 4 T. J . Meyer and H. Taube Inorg. Chem. 1968 7 2369. 2 5 R. G. Wilkins and R. E. Yelin Inorg. Chem. 1968,7 2667. 2 6 R . A. Marcus J . Phys. Chem. 1963 67 8 5 3 . 27 R. Stasiw and R. G. Wilkins Inorg. Chem. 1969 8 156.2 8 J . Burgess J . Chem. SOC. ( A ) 1968 3123. 29 A. A. Green J. 0. Edwards and P. Jones Inorg. Chem. 1966 5 1858. 30 B. Z . Shakhashiri and G. Gordon J . Amer. Chem. Sac. 1969,91 1103 Reaction Kinetics in Solution 113 The permanganate oxidation of primary and secondary alcohols has previously been studied in basic and weakly acidic s~lution.~' The reaction is strongly accelerated by base and is thought to involve the ready transfer of hydrogen from alkoxide ion to the permanganate ion-but whether as hydride (a 2e-equivalent process) or as hydrogen atom (a le-equivalent process) remains uncertain. In other permanganate oxidations there is support for both one-electron equivalent and two-electron equivalent steps.2 Banoo and Stewart32 noted that little was known about the way in which strongly acidic permanganate reacts with alcohols moreover virtually nothing was known about the oxidation of tertiary alcohols.In pursuit of further informa-tion about the mechanisms of these oxidations they investigated the reactions of di- and tri-arylcarbinols in aqueous sulphuric acid. Unfortunately in acid solution the kinetic form for these reactions changes to Rate = k[alcohol]h, i.e. independent of oxidant concentration suggesting that the rate-controlling step is carbonium ion formation : ROH + H + S ROH2+ fast slow. ROH2+ -+ R+ + H 2 0 Whereas the isotope effect for permanganate oxidation of PhzCDOH is 7.3 at pH 7 in acidic solution (where the reaction becomes first order) this falls to 1.08. Banoo and Stewart32 consider it likely that a permanganate ester is the first intermediate formed in the fast step leading to product and for tertiary alcohols this decomposes via an aryl group rearrangement: For the secondary alcohols the analogous 1,2 hydride shift is clearly a possibility, but cannot be distinguished from alternatives including one-electron equivalent oxidation via the radical Ar2CHO'.The same workers33 have also compared chromic acid oxidation of di and tri-arylcarbinols (in 80 wt. % acetic acid containing sulphuric acid). Here the reactions are clearly second-order overall (first order in both alcohol and oxidant). This again suggests a 1,2 aryl shift in the case of the tertiary alcohol oxidation but in the case of the secondary alcohols e.g. 4-methylbenzhydrol where there is a possibility of competition between 1,2 aryl and hydride shifts no trace of 4-methylphenol could be found on product analysis.rhis together with argu-ments based on differences in acidity function correlations is adduced as evidence against a hydride shift in the oxidation of secondary alcohols and of the possible 3 1 R. Stewart 'Oxidation Mechanisms' W. A. Benjamin New York 1964 ch. 5. '' F. Banoo and R. Stewart Canad. J . Chem. 1969,47 3199. 3 3 F. Banoo and R. Stewart Canad. J . Chem. 1969,47 3207 114 S . B. Brown and P. Jones alternatives the authors favour reaction via the cyclic transition state : A paper by Norman and West34 represents a determined attempt to bring order to an area which has become increasingly confused. In 1962 Dixon and Norman3' reported experiments on the reduction of hydrogen peroxide by titanium(u1) ions in aqueous acid solution using a flow system in which the reactants were mixed just before entering the cavity of an e.s.r.spectrometer. An e.s.r. signal was observed which was a single line near g = 2 width 3 gauss. The signal was assigned as deriving from hydroxyl radical formed according to : Ti"' + HzOZ --+ TiIV + OH- + OH The possibility that the radical might be perhydroxyl produced according to : OH + H202 --+ H20 + HOz was considered but was discounted after comparison with results obtained36 in studies of the oxidation of hydrogen peroxide by cerium(1v) : Ce" + H202 -+ Ce"' + H+ + HOz When oxidisable organic compounds were included in the system the spectra of organic free radicals were observed and very elegant and detailed studies of these radicals and their subsequent reactions have since been made.The reader is referred particularly to the series of papers by Norman and co-workers which has now reached Part XXI.37 Further work on the primary species did not however support the idea that it was the hydroxyl radical which had been observed. The spectrum was found to consist of one or two singlets depending on condition^,^*-^^ and various g-factors were reported for each singlet. Under suitable conditions the intensities of the singlets were high enough for each to show weak satellite lines from inter-action of the unpaired electron with the magnetic isotopes of titanium.41 The addition of low concentrations of methanol resulted in an increase in intensity of both singlets ;40 at higher methanol concentrations the singlet intensities 34 R.0. C. Norman and P. R. West J . Chem. SOC. (B) 1969,389. 3 5 W. T. Dixon and R. 0. C. Norman Nature 1962,196,891. 3 6 E. Saito and B. H. J. Bielski J . Amer. Chem. SOC. 1961 83 4467. 37 A. L. Beckwith and R. 0. C. Norman J . Chern. SOC. (B) 1969 265. 38 W. T. Dixon and R. 0. C. Norman J . Chem. SOC. 1963 31 19. 3 9 F. Sicilio R. E. Florin and L. A. Wall J . Phys. Chem. 1966 70 47. 40 Y. S. Chiang J. Craddock D. Mickewich and J. Turkevich J . Phys. Chem. 1966, 4 1 H. Fischer Ber. Bunsengesellschaft Phys. Chem. 1967 71 685. 4 2 J . Stauff and H. J . Huster 2. phys. Chem. (Frankfurt) 1967 55 39. 43 C. R. E. Jefcoate and R. 0. C. Norman J .Chem. SOC. (B) 1968 48. 70 3509 Reaction Kinetics in Solution 115 passed through a maximum then fell to zero and were replaced by the spectrum Adding to the confusion Shiga and c o - w o r k e r ~ ~ ~ ~ ~ compared the organic radicals produced by the Ti111-H202 system and Fenton's reagent (they used Fe"-edta-H,O in neutral phosphate buffer). The latter system was expected to generate hydroxyl radicals according to : of CHZOH. Fe" + H 2 0 2 + Fe"' + OH- + OH. It was reported that in the oxidation of alcohols Fenton's reagent generated radi-cals formed by a-oxidation( e.g. cH2CH20H) in contrast with the behaviour of Ti'11-H202 systems where the a-carbon position was attacked (e.g. CH3 - eHOH) and the authors concluded that the primary oxidising species were different in the two systems.Smith and co-worker~~~ found that the position was not so clear-cut ; the ratio [CH3cHOH] [cH2CH20H] decreased as the ratio [Ti"'] [H202] decreased. Norman and West3j conclude that all observations are consistent with the view that the primary oxidising species generated in both the Ti"'-H202 and Fe''-H,O systems is the hydroxyl radical when the inherent complexity of the reaction systems is taken fully into account. They argue and indeed demonstrate, that complexity arises because organic radicals formed by reaction of organic substrate with hydroxyl radical can undergo further reaction with other oxidants present in the solution : R + H202 + R + + OH- + OH R + M("fl)+ j R+ + M"S R + O2 + R + + 02-the e.s.r. spectrum observed depending on the sensitivity of the different radicals to oxidation via these pathways.The experimental techniques employed are rather subtle and involve the use of two-mixer techniques which enabled studies to be made of the reactions of organic radicals generated at the first mixing point with reagents introduced at the second mixing point. In particular it was shown that the markedly different relative concentrations of radicals which are observed with Fe1'-H202 compared with Ti"'-H202 reflect the fact that Fe"' is a much stronger and more selective oxidising agent for organic radicals than Ti'". As for the 'primary' signals observed with the Ti111-H202 system they are assigned33 as relating to rather unreactive radical species which are derived from titanium(1v)-peroxo-complexes although their precise nature remains uncertain.It should be noted that the formation of free hydroxyl radicals is now inferred in these systems rather than demonstrated. The detailed nature of the primary redox 44 T. Shiga J . Phys. Chem. 1965 69 3805. 4 5 T. Shiga A. Boukhors and P. Douzou J . Phys. Chem. 1967,71 3559. 46 T. Shiga A. Boukhors and P. Douzou J . Phys. Chem. 1967 71,4264. 4 7 P. Smith J. T. Pearson and R. V. Tsina Canad. J . Chem. 1966 44 753 116 S. B. Brown and P. Jones act in these reactions is also uncertain. The results of Wells and may^^*-^' particularly suggest the widespread importance of the formation of metal ion peroxo-complexes as the first step in the oxidation of hydrogen peroxide by metal ions. Halpern2 comments that the mechanism of the decomposition of H202 under the catalytic influence of iron(@ complexes including the highly active enzyme, catalase is not as yet entirely clear.Most monographs (Ardon's is a notable exception) assert that the mechanism of the Fe"' salt-catalysed reaction is established as a free-radical chain r e a ~ t i o n ~ ~ . ' ~ initiated by Fe3+ + H02- -+ Fe2+ + H 0 2 and in which oxygen is produced in the reaction : Fe3+ + 02- -+ Fe2+ + O2 This mechanism ignores the formations4 of the complex Fe3+H02- ; the kinetic data available at the time provided no evidence concerning its role if any in the reaction. However it was suggested52 that H 0 2 radicals might be derived from this complex : Fe3+HOz- -+ Fez+ + H 0 2 and it was stated that provided that the complex does not participate in other reactions the only effect would be to change the kinetic form to a hyperbolic dependence on [H202] at high concentrations.Only the initiation reaction was considered in this argument and it appears to have escaped attention that the mechanism involves Fe3+ in the oxygen production step as has the implication that at high [H202] [Fe3+]-0 and hence rate-0. At high [H202] [Fe3'] ratios Kremer and Stein" and otherss6," have since found that the reaction obeys Michaelis-Menten kinetics consistent with the view that the rate-limiting step under these conditions is the breakdown of Fe3+H02-. At lower [H202] [Fe3'] the kinetic form is more complex ; Kremer and show that under these conditions both kinetic and spectrophoto-metric data are consistent with the accumulation of a second intermediate formed from Fe3+H02-.Kremer59,60 argues that the formation of the second inter-4 8 C. F. Wells and D. Mays Znorg. Nuclear Chem. Letters 1969 5 9. 4 9 C. F. Wells and D. Mays Inorg. Nuclear Chem. Letters 1968 4 43. 5 1 M. Ardon 'Oxygen' W. A. Benjamin New York 1965 ch. 4. 5 2 W. G. Barb J. H. Baxendale P. George and K. R. Hargrave Trans. Faraday SOC., 5 3 J. H. Baxendale Adu. Catalysis 1952 4 31. 5 4 M. G. Evans P. George and N. Uri Trans. Faraday SOC. 1949 45 236. 5 5 M. L. Kremer and G. Stein Trans. Faraday SOC. 1959 55 959. 5 6 P. Jones R. Kitching M. L. Tobe and W. F. K. Wynne-Jones Discuss. Faraday SOC., 5 7 T. J. Lewis D. H. Richards and D. A. Salter J . Chem. SOC.1963,2434. 5 8 M. L. Kremer and G. Stein 2nd Znt. Congr. Catalysis Paris vol. 1 p. 551. 5 9 M. L. Kremer Trans. Faraday SOC. 1963 59 2535. 6 o M. L. Kremer Trans. Faraday SOC. 1962,58 702. C. F. Wells and D. Mays J . Chem. SOC. (A) 1968,665. 1951 47,462 591. 1959 55 79 Reaction Kinetics in Solution 117 mediate does not involve a redox act and that the total reaction may be described by the scheme: Fe3+ + H 0 2 - Fe3+H02-Fe3+H02- --+ F e 3 + 0 + OH-Fe03+ + H02- + Fe3+ + O2 + OH-Thus the activation of H202 in this system is viewed as a two-stage heterolysis which results in a net dehydration of HzOz. The scheme shows an interesting conceptual analogy with the heterolytic splitting process which is probably2 the most widespread mechanism for the catalytic activation of H2 in solution.It is also closely analogous6 1,62 to the 'peroxidatic' mechanism first proposed by Chance and c o - ~ o r k e r s ~ ~ for catalase action and to the scheme proposed by Wang64-65 for the catalytic action of Fe"'-teta complex and related species. Wang's investigations were essentially phenomenological and did not consider the possibility of binuclear complex formation. The Fe"'-porphyrin complexes (the protoporphyrin-IX complex is the prosthetic group of catalase) have long been thought to be extensively dimerised in aqueous solutions.66 They have recently been shown to form 0x0-bridged d i m e r ~ . ~ ~ - ~ ~ An investigation of dimerisation equilibria and catalytic behaviour of proto- and deuterio-ferrihaem (nomenclature advocated by the International Union of Biochemistry for the Fe"' complexes of prctoporphyrin-IX and deuterioporphyrin-IX) in this labora-tory7' suggests that in the decomposition of H202 monomer species are the active catalysts and that they achieve activities comparable to that of catalase In 1961 Duke and Haas7' published an account of experiments which they considered referred to the thermal (uncatalysed) decomposition of hydrogen peroxide in alkaline solution.The results were explained in terms of the mech-at PH ' PK,(H,O2). anism : H 2 0 2 H+ + H 0 2 -H202 + H 0 2 - -P H 2 0 + O2 + OH-This work has been included in the secondary literature (e.g. Ardon," where the reaction is described as a base-catalysed decomposition) an unfortunate 6 1 M. E. Winfield in 'Haematin Enzymes' ed.J. E. Falk R. Lemberg and R. K. Morton, 6 2 P. Jones and A. Suggett Biochem. J. 1968 110 621. 6 3 B. Chance D. S. Greenstein and F. J. W. Roughton Arch. Biochem. Biophys. 1952, "J. H. Wang J . Amer. Chem. SOC. 1955 77 882. '' J. H. Wang J . Amer. Chem. SOC. 1955 77 4715. 6 6 J. E. Falk 'Porphyrins and Metalloporphyrins' Elsevier Amsterdam 1964. 6 7 S. B. Brown P. Jones and I. R. Lantzke Nature 1969 223 960. 6 8 E. B. Fleischer and T. S. Srivastava J . Amer. Chem. Soc. 1969,91 2403. 6 9 I. A. Cohen J . Amer. Chem. SOC. 1969,91 1980. '' T. C. Dean Ph.D. Thesis University of Newcastle upon Tyne 1969. ' I F. R. Duke and T. W. Haas J . Phys. Chem. 1961,65 304. Pergamon Oxford 1961 p. 245. 37 301 118 S. B. Brown and P. Jones occurrence since the experiments were shown to be unsound by Edwards and co-w o r k e r ~ ~ ~ in a paper published in 1963 which seems to have escaped attention.Using highly purified reagents and adding edta to sequester trace metal-ion impurities they found that the rate of hydrogen peroxide decomposition under the conditions of Duke and H a a ~ ~ ’ was vanishingly small (at least 100 times slower than the previous report). They pointed out that the rate of thermal decomposition assuming a bimolecular polar mechanism cannot be faster than the slowest observed rate. In contrast studies of the kinetics and mechanism of the thermal decomposition of peroxoacids in aqueous solution : 2ROOH + 2ROH + 0 2 have developed to a point where these must be rated as perhaps the most thoroughly understood peroxidic redox reactions.The first reaction studied was the decomposition of Caro’s acid (peroxomonosulphuric acid) by Ball and Edwards.73 in 1956. The reaction has also been studied by Goodman and R o b ~ o n . ~ ~ These investigations have been followed by studies on peroxomono-phosphoric acid,7 peroxoacetic acid,72 peroxochloroacetic acid,72 peroxo-benzoic acid and a number of substituted peroxobenzoic acids,7 monoperoxo-phthalic and peroxopivalic and a general mechanism has emerged. Kinetic studies of these reactions are inevitably bedevilled by the catalytic action of trace metal impurities but the addition of edta as a sequestering agent has proved a widely acceptable technique for elimination of this interference. The justification for this procedure was that above very low concentrations of edta, reproducible kinetic behaviour was observed and the reaction rate was insensitive to further increase in edta concentration.All the reactions were found to be second-order in peroxoacid ; the second-order rate constant varied with pH and passed through a maximum at a pH corresponding to the pK of the peroxoacid. The mechanism suggested is: ROOH ROO- + H i ROOH + ROO- -+ ROH + RO- + 0 2 in effect a nucleophilic attack of peroxoacid anion upon the undissociated acid. Goodman and co-workers7 conceived an interesting test of this mechanism, pointing out the implication that a mixture of two peroxoacids should decompose faster than would be expected if each acid decomposed independently. This was confirmed experimentally using mixtures of p-nitro- and p-methyl-peroxobenzoic 72 E.Koubek M. L. Haggett C. J. Battaglia K. M. Ibne-Rasa H. Y. Pyun and J. 0. 7 3 D. L. Ball and J. 0. Edwards J . Amer. Chem. Soc. 1956,78 1125. 7 4 J. F. Goodman and P. Robson J . Chem. SOC. 1963,2871. ’’ J. F. Goodman P. Robson and E. R. Wilson Trans. Faraday SOC. 1962 58 1846. 7 6 R. E. Ball J. 0. Edwards M. L. Haggett and P. Jones J . Amer. Chem. SOC. 1967 89, l 7 E. Koubek and J. E. Welsch J . Org. Chem. 1968 33,445. Edwards J . Amer. Chem. SOC. 1963 85 2263. 233 1 Reaction Kinetics in Solution 119 acid in which an additional pathway via the strongest nucleophile (MeC6H4C03 - ) and the most favoured acceptor molecule (NO2C6H4Co3H) is possible. An intriguing development in these investigations derives f r o m a r g ~ m e n t ~ ~ ? ~ ~ concerning the nature of the electrophilic centres in the reactions.There are two kinetically indistinguishable possibilities illustrated below for an organic per-oxoacid : R-C 40 0-0 ?'H -0 (1) (2) For reaction of a mixture of double-' 80-labelled and unlabelled peroxoacid, pathway (2) implies the formation of scrambled oxygen (3402) whereas (1) implies no ~crambling.~~ Experiments suggest that both pathways are important,72378 their relative contributions depending for example on electro~tatic,~~'~ and steric hindran~e.~ Edwards and Fleischauer7 have discussed the possibilities for more general application of double-isotopic labelling techniques in a recent review. Keith and Powell" have reported studies of an even simpler peroxoacid decomposition.Peroxonitrous acid decomposition in aqueous solution (pH 4-9) is consistent with the simple internal redox mechanism : HOONO S H+ + OONO-HOONO -+ NO,- + H+ The authors attempted to extend their studies to higher pH but were unable to find a suitable buffer system. They comment 'Borate buffers are intolerable' -a sentiment with which all students of peroxide reaction mechanisms will heartily concur. Studies of peroxodisulphate reactions continue to produce a substantial literature. Behrman and McIsaac" have reviewed papers appearing in 1966-67 and give references to reviews of the earlier literature. The catalysed and un-catalysed oxidations of a wide variety of organic compounds have been studied. The importance of sulphate radical ion (SO4-) in these reactions is generally acknowledged.A paper by Fronaeus and &tman82 offers an interpretation of studies of the thermal decomposition of peroxodisulphate in aqueous solution, '' E. Koubek G. Levey and J. 0. Edwards Inorg. Chem. 1964 3 1331. 7 9 J. 0. Edwards and P. D Fleischauer Znorg. Chim. Acta Rev. 1968 2 53. 'OW. G. Keith and R. E. Powell J . Chem. SOC. ( A ) 1969,90. E. J. Behrman and J. E. McIsaac Quart. Reports Sulphur Chem. 1968 (Mechanisms of Sulphur Compounds vol. 2) 193. '* S. Fronaeus and 0. Ostman Acta Chem. Scand. 1968,22,2827 120 S . B. Brown and P. Jones which if generally accepted has important implications for the construction of mechanisms in peroxodisulphate oxidations. This reaction yields a two-term rate law:83 where the kH term is important only for pH < 2. It has been widely a c ~ e p t e d ~ ~ ? ~ ~ that the primary and rate-determining step in the uncatalysed reaction is :86 S2082- -+ 2S04-Fronaeus and &tman82 have studied oxygen gas evolution in the reaction. In the presence of Ce"' the rate of oxygen evolution in the uncatalysed reaction was found to decrease by one-half. It was also confirmed that only one Ce'" ion is formed for each peroxodisulphate ion decomposing in the uncatalysed path. It is argued that these observations strongly support : S2OS2- + H 2 0 + HS04- + SO4- + OH' as the rate-determining initial step in the reaction. 8 3 I. M. Kolthoff and I. K. Miller J . Amer. Chem. SOC. 1951 73 3055. 8 4 M. Tsao and W. K. Wilmarth J . Phys. Chem. 1959 63 346. 8 5 W. K. Wilmarth and A. Haim in J. 0. Edwards 'Peroxide Reaction Mechanisms', 8 6 P. D. Bartlett and J. D. Cotman J . Amer. Chem. SOC. 1949 71 1419. Interscience New York 1962 175