3 Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions By D. P. N. SATCHELL Department of Chemistry University of London King’s College Strand London WC2R 2LS and R. S. SATCHELL Department of Chemistry Queen Elizabeth College Campden Hill Road London W8 7AH 1 Introduction There are essentially just four ‘electronic’ types of elementary reaction.’ Out of them in different combinations virtually all chemical reactions however complex are built up. The four types of electronic reaction are (i) those involving no change in electron ownership (e.g. conformational or order-disorder transformations) (ii) those involving complete electron transfer (iii) those involving bond homolysis and its reverse (free radical reactions) and (iv) those involving bond heterolysis and its reverse (Lewis acid-base reactions).Free or ligated metal ions can enter into all four types and a metal ion route for some overall change will sometimes prove to be a catalytic route. The tendencies of different metal ions to engage in these four different types of process depend upon the detailed electron configurations of the ions and upon their resident ligands. A masterly and penetrating start on the rationalization of these matters has been made by Halpern2 and it is difficult to conceive of anything that illustrates better the value of thinking of reactions in terms of electronic types. In a recent review Hipp and Busch3 have made a valiant effort to consider all the various types of metal ion-catalysed reactions in a systematic way by outlining many of the different possible functions of the metal.In this Report we are confining attention to reactions which involve only Lewis acid-base steps. In Lewis acid-base reactions nucleophiles either co-ordinate with or separate from electrophiles. The overall role of a catalyst is therefore to assist nucleophilic attack or departure or perhaps both. A proton the simplest Lewis acid catalyst acts in this respect either by attachment to a suitable site on an electrophile so making it more electrophilic (role I) or by attachment to a departing nucleophile ’ D. P. N. Satchell Nuturwiss.. 1977,64 113. J. Halpern Chem. Eng. News 1966,44,68; Adv. Chem. Ser. 1968 70 1. C. J. Hipp and D. H. Busch in ‘Coordination Chemistry’ ed.A. E. Martell Vol. 2 A.C.S. Monograph 174 Washington 1978 Ch. 2. 25 D. P. N. Satchell and R. S. Satchell thus rendering the latter less nucleophilic (role 11). Metal ions acting as Lewis acid catalysts can and do function in the same two general ways but for them other mechanisms of catalysis are possible. In one of these the metal acts as a centre for the simultaneous attachment of both an electrophile and the attacking nucleophile thus converting an inter- into an intra-molecular process (template catalysis role 111). In another catalytic mechanism of some importance the metal ion leads to the produc- tion of more or occasionally3 improved nucleophile usually by induced ionization of protons from ambient nucleophiles (role IV). Related to this is metal ion-induced ionization of protons from the electrophile which can result in a poorer electrophile and therefore in the easier departure of an attached nucleophile.(The proton as catalyst cannot achieve these last three effects for fairly obvious reasons.) Metal ion-induced conformational changes can also lead to catalysis. This role is probably mainly confined to enzymatic systems; these are excluded from this Report. Apart from the mechanisms of catalysis not available to protons there are certain other aspects in which catalysis by metal ions differs from that by protons. Compared with all metal ions the proton is small and highly polarizing. As a result a kinetically significant concentration of protonated reactant is often readily formed whilst the presence of even a substantial concentration of di- or tri-positive metal ions fails to produce detectable catalysis.Clearly only substrates carrying groups which permit the formation of a kinetically significant concentration of metal ion-substrate adduct will be susceptible to catalysis. In practice this often means that the substrate must act as a chelating ligand and perhaps largely be converted into the adduct. This is particularly the case with Class A (hard4) metal ions and with substrates which are 0-bases. Findings related to the relatively weak inherent acidity of metal ions are (i) the frequent ineffectiveness of monopositive (especially Class A) metal ions and (ii) the infrequency of mechanisms of metal ion catalysis analogous to slow proton transfer? with metal ions pre-equilibrium adduct formation with a reactant is the norm.Another important peculiarity of metal ion catalysis is that often it is not really catalysis at all in that the metal species is not quantitatively regenerated but remains attached to a product. Strictly speaking these examples are metal ion-promoted reactions but they are normally considered together with genuinely catalytic systems. (Even using protons promoted reactions are more common than is usually realised.) As is well known metal ions can be categorized as either inert or labile.6 By choosing an inert ion as catalyst a researcher can begin with a substrate-metal ion adduct of known structure and with it therefore demonstrate more certainly the presence or absence of specified catalytic effects.In the hands particularly of Buckingham Sargeson and co-workers this approach has greatly strengthened belief in the reality of types of mechanism postulated much earlier using labile ions. Labile ions of course continue to be studied; they represent the majority and contain those especially interesting biologically.’ R. G. Pearson Science 1966 151,172. ’ M. L. Bender ‘Mechanismsof Homogeneous Catalysis from Protons to Proteins’,Interscience New York 1971. R. G. Wilkins ‘The Study of Kinetics and Mechanisms of Reactions of Transition Metal Complexes’ Allen and Bacon Boston 1974. ’ Metal Ions in Biological Systems’ ed. H. Sigel Dekker Basle 1976 Vol. 5. Kinetic Studies ofMetal Ion Catalysis ofHeterolytic Reactions In the following sections we deal with the recent literature back to about 1974.For some of the topics the period prior to this has been thoroughly reviewed; for others it has not. It will be found that the general points made in the foregoing paragraphs are supported by current work and in particular the four main roles (I-IV) available to metal ions as catalysts for heterolytic reactions are all well represented. Many types of reaction fall into the heterolytic category. Three (heterolytic hydrogenation carbonylation and skeletal rearrangement) have especially close ties with their radical and electron-transfer counterparts3 and are therefore more conveniently discussed separately. They are omitted from the present Report.2 Carboxylic Ester Hydrolysis Ever since Kroll’s discovery’ that the hydrolysis of amino-acid esters is catalysed by metal ions ester hydrolyses have contributed much to metal ion studies. Hay and Morris’ have reviewed some aspects of the topic up to about 1974. Work with inert complexes has established that an ester co-ordinated by its electrophilic half can undergo hydrolysis in aqueous alkali by three different routes typified by equations (1)-(3) from which all other ligands have been omitted for clarity. Which route(s) MM++NH2CH2C02R+ OH-Mn++NH2CH2CO; + ROH (1) will obtain in given circumstances depends upon the metal ion upon the ester structure and for equation (3) upon the availability of a cis-OH group. When complexes such as [cis-CO(~~)~(H~I)(NH~(CH~),CO~R)]~+ are hydrolysedg-12 in the absence of other metal catalysts there occurs (principally) first a relatively rapid SNICB dissociation6 of halide (via dissociation of one of the en protons) to give a five-co-ordinate intermediate which when n = 1(glycine esters) very rapidly leads to a mixture of compounds (1)and (2) the ester carbonyl group competing with external hydroxide for the vacant Co site.There then follows a relatively slower H. Kroll J.Amer. Chem. SOC.,1952,74 2036. R. W. Hay and P. J. Morris in ref. 7 Ch. 4. lo D. A. Buckingham D. M. Foster and A. M. Sargeson,J. Amer. Chem. SOC.,1969,91,4102. l1 R. W. Hay R. Bennett and D. P. Piplani J.C.S. Dalton 1978 1046. l2 R. W. Hay R. Bennett and D. J. Barnes J.C.S.Dalton 1972 1524.D. P. N. Satchell and R. S. Satchell hydrolysis of the ester function via routes like equations (2) and (3),both leading to chelated glycine. When n =3 the hydrolysis follows the same pattern except that the ester carbonyl does not successfully compete with external hydroxide and the ester hydrolysis is entirely via a route like equation (l),leading to a hydroxypentamine compound (3). Kinetic studies have shown that for routes like (2) and (3) rate accelerations (compared with the rate of hydrolysis of the uncomplexed ester by OH-) are ca. lo6 and >lo7 respectively. For routes like (1) the acceleration depends upon the distance of the ester group from the metal;"*'2 for n =3 it is ca. seven-fold at 25°C. Compared with the OH- rate for the protonated ester (N'H,(CH2),C02R) the Co route (1) involves a three-fold deceleration when n = 3.The very great reactivity of a directly co-ordinated carbonyl group in these reactions means that any ambient base can be a successful nucleophile and general base catalysis of nucleophilic attack is unnecessary." The majority of recent studies have involved Cu2+ complexes. With this ion at high pH dipeptide esters give terdentate complexes for Etglygly (4 R =H) for Etgly-L-leucine (4 R = CH2CHMe2) and for Etgly-P-alanine (5). Kinetic st~dy'~ + 'OEt (4) (5) shows that hydrolysis of the ester function occurs for both the aquo and hydroxy (neutral) versions of these complexes at rates of ca. 2 x 103-fold and ca. 3 x 102-fold faster respectively than for the uncomplexed esters at 25 "C.These values suggest (i) moderately weak carbonyl co-ordination and (ii) external OH- attack [equation (2)]. All the esters give similar rates which suggests that for Cu2+ a six-membered ring is as convenient as a five-membered ring. Angelici Nakon and their colleagues have contin~ed'~-'~ their kinetic studiesg of Megly hydrolysis promoted by highly chelated Cu2+ Ni2+ and other divalent ions e.g. by [CUL]'~-")' where L"-is (~-P~CH~)~NH, (NH2CH2CH2)2NH (C0;CH2)2NH (CO;CH2)3N (pyr), etc. These tri-and tetra-chelated complexes generally co-ordinate one Megly molecule in aqueous solution in an essentially unidentate fashion with the possibility of some weak bidentate interaction. The rate of hydrolysis is normally given by Rate = k[MLglyMe'2-"'+][OH-].The sequence of l3 D. A. Buckingham J. Dekkers A. M. Sargeson and M. Wein J. Amer. Chem. SOC.,1972,94,4032. l4 R. W. Hay and K. B. Nolan J.C.S. Dalton 1974,2542. R. Nakon P. R. Rechani and R. J. Angelici J. Amer. Chem. SOC.,1974,96,2117. l6 D. E. Newlin M. A. Pellack and R. Nakon J. Amer. Chem. SOC.,1977,99,1078. S. A. Bedell and R. Nakon Inorg. Chem. 1977.16 3055. l8 R. D. Wood R. Nakon and R. J. Angelici Inorg. Chem. 1978,17 1088. l9 J. K. Walker and R. Nakon J. Amer. Chem. SOC.,1978,100,1151. Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions effectiveness of metals is generally Cu >Zn >Ni >Co (Irving-Williams) and save for yrtain of the Cu2+ complexes the observed rates are slightly slower than for NH3CH2C02Me.Clearly little activation of the carbonyl group is occurring and the suggested mechanism is like equation (1). The effects of the various ligands L"-on k for a given metal ion are apparently determined by (a)the resulting overall charge on the complex and (5) the effective basicity of the ligand. The more strongly the ligand is held the poorer is the complex at promoting ester hydrolysis. This effect is a recurring theme in this group's work and is clearly demonstrated for a series of [CuLMegly12' complexes (of fixed charge)." For the Cu2' complexes which are more effective than the proton it is suggested that weak C=O co-ordination (at an apical position) is involved. A similar conclusion is drawn from a comparative study of ester hydrolysis and water exchange" in complexes (6).Here the same sequence of metal ion effectiveness (Pb >Cu >Zn >Co >Ni) is observed for both reactions and the magnitudes of AH' for the two processes are similar. This result suggests that in the ester hydrolysis the carbonyl group first displaces a water molecule from an apical position and that the contribution of this step dominates the overall AH' value AH*and AS* for the subsequent intermolecular OH- attack having little influence on AG*. For those foregoing examples in which carbonyl co-ordination is thought negligible AS' for the attack of OH- is considered to be the principal factor determining AG'. Divalent metal ions have been shown2' to be powerful promoters of the hydrolysis of methyl-8-hydroxyquinoline-2-carboxylate(EH) by OH-.Both complexes (7 MEH2+) and (8,ME') lead to reaction and it is believed that the carbonyl group i + 0-M+O must be engaged by the metal (M2'). Attack by external hydroxide is suggested and supported by measurements using D20. The activity of M2' is in the sequence Cu>Co> Ni> Zn>Mn. In aqueous solution at 25"C the complexes CUE' and MnE' are ca. 106-fold and ca. 103-fold more rapidly hydrolysed respectively than is E-. '"R. W. Hay and C. R. Clark J.C.S. Dalton 1977 1866. D. P. N. Satchell and R. S. Satchell The foregoing studies are all examples of metal ion activation of the electrophilic part of the ester in hydrolysis (role I) and sometimes also of template action (role 111).In contrast we turn now to cases involving a metal's influence on the attacking nucleophile (role IV). Obviously if a metal ion is attached to the attacking nucleo- phile in a bimolecular Lewis acid-base reaction that nucleophile is going to suffer reduced nucelophilicity. The question is how much? We have already seen for Co promotion that if both the ester and the attacking nucleophile are attached to the same metal centre in suitable juxtaposition any reduction in nucleophilicity can be more than off -set by the gain in AS'. It appears that even when the free ester reacts with a metal-bound nucleophile in a positive neutral or negatively charged complex the nucleophile can show surprising strength; attachment to a metal does not reduce nucleophilicity so much as attachment to a proton.Thus OH- in [Co(NH3),(OH)I2' and deprotonated imidazole (Im-) in [CO(NH~)~I~]~' have pK values of ca.6.4 and 10 respectively and react withp-nitrophenylacetate in aqueous solution at rates2' expected for bases of such pK values. Similar results are found for OH-complexes of Pd Pt Hg and other metal^.^^-^^ The use of such complexes therefore permits the employment of substantial concentrations of useful nucleophile in circumstances (e.g.non-aqueous solvents or low pH) where little of the corresponding free nucleophile is available. For instance in aqueous solution at pH6.5 and 25"C about half of any [Pt(en)(H20)I2+ or [CO(NH~)~(H~O)]~' will be in the OH form while the concentration of free OH-will be <lo-'.Some especially interesting complex metal hydroxide catalysts have been reported by Werber and Shaliti~~~~'~~ who used phen bipy dien and pentamethyl- dien hydroxy metal complexes (9) to hydrolyse p-nitrophenyl esters of various substituted acetic acids. Species (9) is surprisingly reactive in simple bimolecular reaction with esters. I+ r Me Finally in the ester section there are the examples where a metal ion assists the departure of the leaving nucleophile (role 11). It is of course only in this sort of promotion that an Al-like mechanism' is possible. In the other three sorts of promotion (I 111 and IV) either an inter- or an intra-molecular A2- scheme' must always operate. Leaving groups sufficiently basic to attach a kinetically significant amount of metal ion are found in various S-esters when the metal ion is very soft.21 J. MacB. Harrowfield V. Norris and A. M. Sargeson J. Amer. Chem. Soc. 1976,98,7282. 22 M. M. Werber and Y. Shalitin Bioinorg. Chem. 1973 2 275. 23 M. M. Werber and Y. Shalitin Bioorg. Chem. 1975,4 149. 24 M. C. Lim and R. B. Martin J. Inorg. Nuclear Chem. 1976,38,1911;see also M. A. Wells G. A. Rogers and T. C. Bruice J. Amer. Chem. SOC.,1976,98,4336. Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions Thus the hydrolyses of thiol [equation (4)] thion esters26 [equation (5)],and thiolbenzimidate esters2' [equations (6)and (7)]are all powerfully promoted by Hg2+ p-RC6H4COSEt+Hg2++2H20 + p-RC6&C02H +H30' +HgSEt' (4) p-RC6H,CSOEt +Hg2' +2H20 + p-RC6H4C02Et+2H30++HgS (5) PhC(=&HZ)SEt +Hg2++2H20 -+ PhCN +2H30++HgSEt' (6) PhC(=&HR')SEt +Hg2++3H20 -+PhCONHR'+ 2H30+ +HgSEt' (7) ions in aqueous solutions of pH <3 at 25 "C(conditions under which the proton catalysed reactions are negligible).When R =OMe in equation (4) the reaction follows an A 1-like route equation (8);when R =NO an A2-like scheme prevails. Hg2+ f p-MeOC6H4COSEt+Hg2+ .%-COS slow --+ -CO+HgSEt' '::+ + \ Et p-MeOC6H4C02H+H30++HgSEt+ (8) An Al-like mechanism is also found in reaction (6) which has a complex rate equation owing to the necessary ionization of the N-bound protons. Three slow unimolecular product-forming steps have been identified steps (9),(lo),and (11).In slow __* Ph&H +HgSEt+ H20 b PhCN +H30++HgSEt' (9) PhC+ Et NH Et / slow S' -/k PhCN +HgSEt' (10) PhC Et vHg+ Hgz+ slow 's /\ ___* PhCN +Hg2++HgSEt+ (11) PhC Hg+ N all these Hg2+-promoted reactions of S-esters only a small extent of pre-equilibrium complex formation oc~urs.~~-~~ This is in contrast to the reactions discussed in the preceding paragraphs where usually one is working with a pre-formed complex or under conditions where saturation complexation is attainable at moderate metal ion concentrations.All the Hg2'-promoted S-ester hydrolyses are kinetically first order in the S-ester and in the Hg2+ concentrations and route (1 1) underlies a kinetic term in [Hg2'I2 found in the S-imidate reactions.It is common in metal-assisted nucleo- phile departures to find when the metal ion-substrate adduct has only a single ''D. P. N. Satchell and I. I. Secemski J. Chem. SOC.(B), 1970 1306. 26 D. P. N. Satchell M. N. White andT. J. Weil Chern. andInd. 1975 791. *' A. J. Hall and D. P. N. Satchel] J.C.S. Perkin ZZ 1976 1274 1278. 32 D. P. N. Satchell and R. S. Satchell positive charge that routes involving the help of a second metal ion are indicated. For example a similar effect is found for the A2-like Ag'-promoted hydrolyses of thiol esters" for which the rate equation is -d[S-esterlldt = (kl[Ag'] +k2[Ag']')[S-ester]. Here again two metal ions are very probably attached to one sulphur atom. A last point concerns the relative efficiencies of soft metal ions in these S-ester reactions.Usually Hg2' and Ag' are ca. 106-fold and lo3-fold respectively more effective than the proton. T13' and AuCl have also been shown to be of comparable activity to Hg2' in the S-imidate hydrolyses and are doubtless also very efficient with other types of S-esters. Other supposedly rather soft metal ions such as Cd" Cu2+ and Pb" lead to comparatively negligible promotion at pH 3 and are probably less effective than the proton. Metal ion promotion of reactions of various organo- sulphur compounds has been reviewed.28 A cobalt-ester adduct [~is-Co(en)~(Hal){NH~(CH~)~0COMe)l2', analogous to those considered above but with the cobalt attached to the leaving nucleophile has been studied by Hay and CO-workers." Hydrolysis by OH- in aqueous solution occurs as expected in two stages first there is a relatively rapid SNICB loss of halide which is followed by a slower hydrolysis of the ester group by free OH-via an intermolecular route like that in equation (1).The product is [cis-Co(en)JOH)- (NH2(CH2)'0H)I2+. At 25 "C the ester group hydrolysis occurs ca. 18-fold faster than for the uncomplexed ester and ca. 3-fold faster than for the N-protonated ester. No intramolecular or chelation effects intrude; these are clearly unfavourable when the carbonyl group is this far away from the cobalt atom. Cases of metal ions affecting a reaction by chelation at N-containing leaving groups are discussed by Fife and Squillacote3' (see also Section 3). One of these may conceivably involve intramolecular OH- attack via a metal attached to an ester's leaving group (10).Little complex forms in this system and no catalysis is detectable at pH <6; this is probably because at low pH (i) there exists less MOH' and (ii) protons compete more effectively for the N atom. A clearer example of this type of effect has been provided31 by Hay and Clark as shown in Scheme 1. In this system powerful (almost enzymatic) acceleration is found which supports the notion of an intramolecular route. Zn2' is more effective than Cu2' which being a stronger Lewis acid binds OH- more strongly. Available experiments on ester hydrolysis therefore suggest that various metal ion-promoted inter- and intra-molecular A2-like routes are possible and for esters co-ordinated via their leaving nucleophiles there is also the possibility of an A 1-like mechanism.3 Carboxylic Amide Solvolysis Parts of this topic have been reviewed' up to about 1974. Buckingham et al.32have reported on a system which by itself exhibits for amide hydrolysis analogues of all three mechanisms found for ester hydrolysis when 28 D. P. N. Satchell Chem. SOC.Rev. 1977,6 345. 29 K. B. Nolan B. R. Coles and R. W. Hay J.C.S. Dalton 1973 2503. 30 T. H. Fife and V. L. Squillacote J. Amer. Chem. SOC.,1978,100,4787. 31 R. W. Hay and C. R. Clark J.C.S. Dalton 1977,1993. 32 D. A. Buckingham A. M. Sargeson and F. R. Keene. J. Amer. Chem. SOC.,1974,96,4981. 33 Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions 0 L 0-M-0 t OCOMe 1 OH Scheme 1 a metal is attached to the acyl part of the substrate.In aqueous alkali [cis-C~(en)~(Br)glyglyOC~H~]~+ is hydrolysed in two stages (Scheme 2). As for the similar species discussed in Section 2 the first stage in an SN1CB loss of Br- leading to two products (11)and (12) both of which undergo amide hydrolysis in the second stage; (11)by external OH-attack on the chelate (a known route') and (12) by both intra- and inter-molecular routes which are about equally efficient at pH 10. The final Co-bound products are thus the hydroxypentamine (14) and the glycine chelate (13). General base catalysis of the intramolecular route is observed and in the presence of sufficient added base (e.gl phosphate) this route can be ca.10'O-fold faster than hydrolysis of the unco-ordinated substrate. Buckingham and co-worker~~~ have also found that dimethyl formamide is a sufficiently strong 0-base to co-ordinate to cobalt in [CO(NH~)~O=CHNM~~]~+ 33 D. A. Buckingham J. MacB. Harrowfield and A. M. Sargeson J. Amer. Chem. SOC.,1974,96 1726. D. P. N. Satchell and R. S. Satchell J[(en)2Co(OH)NH~CH2C02]++ RNHz 11 (14) NHR NHR H+&-OH- RNHz + (13) (11) Scheme 2 without chelation. It then undergoes hydrolysis via external OH-attack ca. 104-fold faster than does the free amide mainly as a result of a favourable AS* value. The conformations of Cu2+and Ni2' pentapeptides suggest that weak interaction of their terminal carbonyl groups with metal apical positions is possible whereas such interaction is impossible for tetra pep tide^.^^ In agreement with this it is found that the hydrolysis of pentapeptides is slightly (ca.10-fold) accelerated by co-ordination to these ions whereas tetrapeptides are unaffected. A somewhat qualitative of the glycolysis of benzamide at 139 "Creveals catalysis by the acetates and similar derivatives of various metals (including Na'). Fractional reaction orders are observed and the system is not easy to interpret mechanistically. A number of studies have appeared in which the metal ion facilitates leaving group departure. In one equation (12) attack of HzO on the (unchelated) Co-bound substrate is found to be general acid and base catalysed.21 This is an A2-like scheme.m m [(NH3),CoNwNCOMe]'+ + H,O --+ [(NH3),CoNwNHl3+ + MeC02H (12) The presence of Co increases the rate of H20 (or OH-) attack by ca. 20-fold rn compared with [HN-NCOMe]' but as expected greatly reduces the contribu- tion of routes involving catalytic protonation of the leaving nucleophile. The bimolecular hydrolysis (and aminolysis) of penicillin36 is powerfully (ca. 10') catalysed by Cu2' ions owing to the possibility of leaving group chelation (15). Certain buffer components are found to deactivate the Cu2' ions. 34 J. J. Czarnecki and D. W. Margerum Inorg. Chem. 1977,16 1997. 35 J. Malek and E. ZelCna Coll. Czech. Chem. Comm. 1976,41,395. 36 N. P. Gensmantel E. W. Gowling,and M. I. Page J.C.S. Perkin II 1978 375. Kinetic Studies of Metal Ion Catalysis ofHeterolytic Reactions RCoNH Kinetic studies have been made of the promoted hydrolysis and decomposition of thioamides by various soft metal ions in aqueous sol~tion.~~-~~ Regardless of the metal used N-unsubstituted amides (RCSNH2) lead to RCN and the metal sulphide by usually somewhat complicated routes the details of which depend upon the metal ion.With most of the metals used ionization of N-bound protons from meta1-S- amide adducts is involved prior to a unimolecular decomposition of species such as RC(=NH)SM'"-"' e.g. equation (13). With N-substituted S-amides the product is ,S-& PhC -+ PhCiH + AgS-% PhCN + H30++ AgS-(13) %H necessariIy the analogous 0-amide formed via A2-type routes e.g. equation (14). /S-Hg+ PhC +H2O + PhC'6H2 +HgS % PhCONHR+H30'+HgS (14) %R \NR With Hg2' T13' AuCl; and Ag' stoicheiometric adduct formation between S-amides and the metal ion is observed but with Cu2+ Cd2' Pb2+ and Co2' little adduct is formed.At pH e3 the sequence of reactivities is Hg2+ -Ti3+-AuCI > Ag'>>Cu2'>>Pb2' Cd2+ Ni2' Tl' and Cs'. The Hg2' group of ions provide substantial accelerations (ca. lo6) over the corresponding proton-catalysed hydrolysis whereas the ions beyond Cu2' in the sequence have only comparable or less reactivity than the proton. Similar results are found in S-ester hydrolysis (Section2). At ca. pH 6 in buffer solutions some catalysis is observed for Pb2+ Cd2+ and Co2' ions but interpretation is complicated by the presence of M2'-buffer and possibly of MOH' com~lexes.~~ In all these studies Cd2+ is anomalously ineffective in view of its supposed softness.Two interesting features of the reactions are (i) the greater reactivities (ca. 10-fold) of AuC1;- and AuC130H2 compared with AuC1; and (ii) the independence of the Au"'- and T13+-promoted reactions of [H,O+] which means that ionization of M-bound water or amide protons is kinetically unimportant with these ions. Therefore tertiary S-amides (RCSNR2') can be hydrolysed by them as readily as primary and secondary derivatives (which is not the case with the other soft ions). Lastly two studies by Fife and Squillac~te~~*~~ suggest that when a metal species can co-ordinate to both the electrophile and nucleophile parts of a substrate its 37 A.J. Hall and D. P. N. Satchell J.C.S. Perkin 11 1975 778 953 1273 1351. '' A. J. Hall and D. P. N. Satchell J.C.S. Perkin 11 1977 1366. 39 0.M. Peters N. M. Blaton and C. J. DeRanter J.C.S. Perkin 11 1978 23. 4o T. H. Fife and V. L. Squillacote J. Amer. Chem. Soc. 1977 99 3762. D. P. N. Satchell and R. S. Satchell presence will not accelerate decomposition. For example the presence of Co2+ Cu2+,Zn2' and especially Ni2' leads to marked inhibition of the intramolecular nucleophilic route to the hydrolysis of N-(2-phenanthroyl)phthalamicacid without providing any additional catalysis (16); clearly the metal effectively holds the two halves of the substrate together. 4 Reactions of Nitriles and Imines Once again we can do no better than to begin with the recent work of the Australian s~hool.~' Following earlier with the same systems it is now proposed that CN)]" the M"'-promoted hydrolysis at acid pH of [ci~-Co(en)~(Hal)(NH~(CH~) (where Hal = C1 or Br x = 1 or 2 and M"' = Hg2+ Ag' Hg22+ Zn2' or Cd2') proceeds principally by Scheme 3 to give 0-chelated amide.With an excess of M"' r NH2(CH2),CNI2+ 11H'O (er~)~Co NH2(CH2).CN] 2' 'OH \ Scheme 3 present the rates of these reactions are usually first-order in [M"'] and inversely proportional to [H30+]. The details of the final slow (intramolecular) step are unknown except that (interestingly) with Ag' an important kinetic term second- order in [Ag+] is also found suggesting that two silver ions can beneficially be 41 D.A. Buckingham P. Morris A. M. Sargeson and A. Zanella. Znorg. Chem. 1977 16 1910. 42 D. A. Buckingham A. M.Sargeson and A. Zanella J. Amer. Chem. SOC.,1972,94 8262. 43 K. B. Nolan and R. W. Hay J.C.S. Dalton 1974 914. Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions attached to the CN group. Very large (>lo8)accelerations are observed compared with the rate of hydrolytic cyclization of the M"'-free hydroxy derivative (HO(CH,),CN) especially when x = 2. The relative efficiencies of the soft ions are in the sequence Hg2' 3Hg22' >> Ag' > Cd2' > Zn2'. As found for many S-substrates,28 Hg2+ is ca. 103-fold superior to Ag' at 25 "C; a surprising result is the finding that Cd2+ and Zn2+ are only ca. 10 and 102-fold less effective than Ag' in this system whereas with S-substrates their contribution is relatively very When x = 1 a second hydrolysis route is present (Scheme 4) which involves intermolecular attack of water on the chelated nitrile.The relative importance of this route which leads to the N-chelated amide depends upon the nature of Hal and' not upon that of M"'. Scheme 4 In alkaline solution competition to hydrolysis arises from an intramolecular attack on the nitrile by a deprotonated NH group of en equation (15). + slow Hal Hg (15) The kinetics of some analogous intermolecular hydrolyses (by OH-) of Co-bound nitriles in [Co(NH3),NCRI3' (where R = 4-CNC6H5 4-CHOC6Hs etc.) have been described.44 The hydrolyses are first-order in each reactant and large accelerations (>lo6)are found.The dominant effect is that of Co3+:substituents in the benzene ring have little effect on kOb. The great importance of the Lewis acidity of the metal centre is also illustrated by similar of benzonitrile and acetonitrile hydrolyses by OH- at pH 8-9 in aqueous solution at 25 "C using the complexes [M(NH3)5NCR]"+. For MeCN it is found that Ru'" >> Ru" and Ru"'> Rh'" = CO'~ while for PhCN Ru"' > Co"' 3Rh'''3 I?". The Ru"' complexes lead to accelera- tions of ca. 108-fold compared with the metal-free nitriles. The sequences are 44 R. J. Balahura P. Cock and W. L. Purcell,J. Amer. Chem. SOC.,1974,96,2793. 45 A.W.Zanella and P. C. Ford J.C.S. Chem. Comm.. 1974,795. 46 A.W.Zanella and P. C. Ford Inorg.Chem. 1975,14 700. D. P. N. Satchell and R. S. Satchell discussed in terms of d-electron content and ion size. MeCN is also hydrolysed4' in the presence of Pt'" and a qualitative kinetic account of its hydrolysis promoted by Hg2+ in concentrated MeCN-H20 mixtures is also available.48 Added ions (AcO- NO3-) which tie up the metal inhibit reaction. Two interesting related reactions which involve the metal ion-promoted hydra- tion of oximes and imines have recently been reported. The first shows that the hydrolyses equation (16),of the Cu2+ complexes of (17)and (18)are rapid but n H2O-Cu2+ NHz '7 H20Cu2+ n ____) s)J NH NH -S-aldehyde -S-aldehyde (17) (181 are not due to relief of strain in quadri- and ter-dentate ligands since only the N atoms are co-ordinated.The second report" suggests that the rearrangement of oximes catalysed by Ni" and Pd" complexes proceeds uia oxime adducts which undergo dehydration followed by intramolecular hydration e.g. equation (17). CHR 1+ These metal-promoted imine and nitrile hydrolyses differ from those of esters and amides in that activation of the CN multiple-bond towards nucleophilic attack always appears to require direct attachment to the metal. In other systems direct attachment to a leaving nucleophile often provides unimolecular slow steps but since with nitriles it is the opening of the multiple-bond which corresponds to leaving group departure unimolecular Al-like schemes may be expected to be less common." 5 Dehalogenation Dehalogenation of both organic and of inorganic compounds by metal ions has long been used for preparative purposes.Its kinetic study provides a link between these two branches of chemistry and it is encouraging to find that a number of common features exist. Best results will clearly be obtained with soft metal ions (which form very stable halides) and such ions are commonly used. The reactions as a whole form the clearest category of processes in which the metal ion assists departure of the leaving nucleophile. Very often A 1-like mechanisms obtain. 47 A. K.Johnson and J. D. Miller Inorg. Chirn. Acru 1977 22 219. 48 Y-K. Sze and D. E. Irish Cunud. J. Chern. 1975 53,427. 49 A. C. Braithwaite C. F. E. Rickard and T. N. Walters J.C.S. Dulron 1975 2149.A. J. Leusink T. G. Meerbeck and J. G. Noltes Rec. Truu. chirn. 1977,96 142; ibid. 1976.95 123. " See also E. N. Zilberman V. I. Trachenko S.M. Danov and N. R. Shipmova Izuesr Vyssh. Uchebn. Zuued. Khirn. Tekhnol.. 1977 20 1141 (Chem.Abs.,87 183830). Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions 39 Examples have already been given (e.g.Schemes 3 and 4) in which the aquo- and ring-closed form of a complex cobalt halide has been produced via metal (normally Hg2') ion-promoted loss of halide.52 Usually these reaction^^'^'^' involve an A 1-like mechanism rapid pre-equilibrium attachment of the metal ion to the halide is followed by the rate-determining formation of a five-co-ordinate transient inter- mediate which then quickly undergoes intra- or inter-molecular nucleophilic attack.[This intermediate is not identical with that formed in the corresponding base catalysed unimolecular substitution (SNICB) e.g. Scheme 2 and the two do not necessarily lead to the same distribution of inter- and intra-molecularly formed products.'] A number of recent studies have looked at the relative effects of different soft metals in promoting the aquation of various inorganic halides. For [ReCl6I2- and [ReBr612- with M"' in aqueous solution the simple kinetic form kobs= kl + k2[Mm+] is k representing the unpromoted rate. For Hg2' and T13'k2[Mn'] >> kl,but for Cd2+ the two rates make comparable contributions and for In3'k2-0. For [CO(NH~)~C~]~+ efficiencies are54 in the sequence Hg2+ -HgCl' > T13' -T10H2+> T1C12' the overall factor being ca.60. Normally Hg2'>T13' but this order can be reversed for anionic cobalt complexe~.~~ When one or more halogen atoms is cis to the others in a substrate bridged metal ion adducts can be detected as intermediates. With only one halogen atom little adduct is detectable. Recent work with organic halides has mostly involved promotion by Ag'. Besides removing halogen atoms in promoted solvolyses Ag' ions can also catalyse the rearrangement of suitable unsaturated or strained hydrocarbons which contain no halogen. The balance between these roles is a concern of two studies of halocyclo-propanes in methan01.~~*~' As for the inorganic compounds the removal of halogen from organic halides usually involves slow unimolecular heterolysis.The resulting carbocation can then suffer attack at carbon by any ambient nucleophile (e.g. solvent) or by losing a proton to the nucleophile form olefin. For a variety of solvents (MeOH MeN02 MeCN) a reasonably self-consistent pattern of behaviour RHal + Ag+$RHalAg+ slow R+Hal-Ag' +R++ AgHal (route A) II.-IS Y-R+Hal-Ag+ alkene + SH' + AgHal (route B) 1 Y- RY+AgHal 4 Scheme 5 s2 See also V. Tinner and W. Marty Helv. Chim. Acta 1977 60 1629; V. I. Belevantsev E. I. Evdokimova and B. I. Peshchevitskii Zzvest Sib.Otd. Akad. Nauk. S.S.S.R.Ser. Khim. Nauk. 1978,38. s3 J. Burgess and S. J. Cartwright J.C.S. Dalton 1976 1561. 54 S. W. Foong B. Kipling and A. G. Sykes J. Chem. SOC.(A),1971 118." S. F. Chan and S. L. Tan Znorg. Nuclear Chem. Letters 1975,11,435. 56 G. M. Blackburn and C. R. M. Ward J.C.S. Chem. Comm. 1976,79. " D. B. Ledlie J. Knetzer and A. Gitterman J. Org. Chem. 1974.39 708. 40 D. P. N. Satchell and R. S. Satchell is beginning to This is shown in Scheme 5 in which S =solvent and Ag'Y-is any particular silver salt. The relative importance of the different paths depends upon (i) the dielectic constant and other properties of the solvent (ii) the tendency of R' to give olefin and (iii) the nature of Y-. Usually the reactions are first-order in organic halide but have terms of the first- and of the second-order in [Ag+]stoich. Since [Y-] = constant X [Ag']stoich the second-order term in silver ion can represent the contribution of the A2-like route B,but it is believed that it also and often mostly reflects promotion via a route involving two silver ions attached to the halogen atom.The kinetic parallel here with Ag'-promoted reactions of S-compounds and co-ordinated nitriles is as striking (see Sections 2 and 4)as that between the sequences of metal ion efficiency found for inorganic complex halides and for S-compounds (see above). Not only typical soft metal ions but other ions too can speed-up the reactions of halogen especially in solvents of low dielectric constant. Two interest- ing examples of powerful accelerations by LiC104 in diethyl ether have been recently provided by Pocker and Ellsworth.64 The exact role of the Li' ion in these reactions is perhaps uncertain.6 Dealkylation A metal ion-promoted reaction analogous to solvolysis via dehalogenation in that it (i) involves only assistance of leaving nucleophile departure and (ii) requires soft promoters is the solvolysis of organometal complexes via dealkylation. Reaction (18)is typical. This process is found6' to be first-order in the chromium complex and [(H20)&RI2' +Hg2+(or R'Hg+) %[Cr(H20),l3'+ RHg+ (or R'HgR) (18) in the mercury promoter. Various species R'Hg' are all ca. 102-fold less reactive than Hg2+ and electron withdrawal by R slows down the reaction. Steric effects are observed and other factors suggest that the reaction involves slow carbanion transfer followed by rapid attack of water on [Cr(H20)J3'. This process equation (19) is [(H20)5CrR]2' +[Hg(0H2),l2' [(H20)5Cr13' +[RHg(OH2)x-il' +HzO therefore a metal ion analogue of slow proton transfer to carbon.The relative reactivities of promoting species in this type of system depend to some extent on whether R carries a net charge (e.g.as in CH2-) but in general show that Y. Pocker and W. H. Wong J. Amer. Chem. SOC.,1975 97,7097,7105. 59 D. N. Kevill V. V. Likhite and H. S. Posselt J.C.S. Perkin IZ 1975 911. 6" R. D. Bach and C. L. Willis J. Amer. Chem. Soc. 1975,97 3844. 61 D. N. Kevill and R. F. Sutthoff J.C.S.Perkin 11 1977 201. 62 V. V. Zamashchikov E. S. Radakov I. R. Chanysheva and S. L. Litvinenko Dopov.Akad. Nauk. Akr. R.S.R. Ser. B. Geol. Khim. Biol. Nauki 1978.2 125 (Chem. Abs. 88 189528). " V. N. Plathotnick Katal.Katal. 1975,13,63;V. N. Plathotnik L. V. Boguslavskaya and V. V. Varekh ibid. 1977 15 41 (Chem. Abs. 88 177768). " Y. Pocker and D. L. Ellsworth J. Amer. Chem. SOC.,1977 99,2276,2284. 65 J. P. Leslie and J. H. Espenson J. Amer. Gem. SOC.,1976 98 4839. Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions Hg2+> T13+. The addition of aniomic ligands normally reduces promoting power in the sequences Hg2+ > HgCl' > HgC1,-> HgC142- and T13+> T10H2+> T1C12+> T1Clz+>TlCl > TlC14- but when R carries a positive charge the reactivity spread is greatly reduced or even reversed.66 Soft ligands can greatly lower the effectiveness of a metal ion for one demethylation it is observed67 that Hg(OAc)z >> HgC12> HgBrz>> Hg(SCN)zHg(CN)z.An interesting intramolecular slow metal transfer to carbon6* is provided by reaction (20). c1I1 +d+PY+PY c1 -+ P;CHzCHz-Pt+PYI Ic1 I CI 7 Phosphate Hydrolysis This topic has been reviewed6' up to about 1974. Phosphates are rather difficult to hydrolyse and rate accelerations are often modest. Parts of the field especially the hydrolysis of ATP and similar compounds are still in a state of flux. Recent ~~rk~~-~~ on the cleavage of the terminal phosphate group from ATP (e.g.reaction 21) catalysed by Cu2+ Zn" and Ni2+ has re-emphasized (i) the importance of both ATP4-+ OH-M2+ w ADP2-+ P3-(21) dimeric 1 1-and 2 1-complexes [i.e. (MATP)z4- and (M,ATP),] as the reactive species and (ii) the role of the N-7 atom of the adenosine moiety in engaging the first M2+ ion.This type of phosphate has an embarrassing number of potential sites for a metal ion. However although charge neutralization by the metal ion is obviously an important part of its function it seems unlikely that attachment at the &-positions will be ideal since this will tie the cleaving fragments together (19). Perhaps the 000 II. llp II Ad-O-P-O-P-O-PY-O-I I 0-0-0-hd M2+ (19) reason for the importance of the (MzATP)2 species is that one M2+ ion can be attached to the terminal phosphate group and the other to the a@ phosphate groups and the adenosine nitrogen atom so providing a two-way stretch. Too great a concentration of OH- or other ligands (e.g. bipy) are considered to reduce the hydrolysis rate by competition for the metal 66 D.Dodd M. D. Johnson andD. Vamplew J. Chem. Soc. (B),1971,1841. b7 Y.Yamamoto T. Yokoyama J. Chen and T. Kwan Bull. Chem. SOC.Japan 1975,48 844. " I. M.Al-Najjar and M. Green J.C.S. Chem. Comm. 1977 926. 69 B. S.Cooperman in ref. 7,Ch. 2. 70 D. H.Brisson and H. Sigel Biochem. Biophys. Acta 1974,343,45. 71 H.Sigel and P. E. Amsler J. Amer. Chem. SOC.,1976,98 7390. 72 P.E.Amsler D. H. Brisson and H. Sigel 2. Naturforsch. 1974,29,680. 73 See however M. M. T. Khan and M. S. Mohan Indian J. Chem. 1976,14A,945. D. P. N. Satchell and R. S. Satchell of the hydrolysis of the phosphate esters (20) and (21) show that chelation of a metal ion to either the ester or the phosphate half of the substrate leads as expected to an increase in hydrolysis rate.The extent of involvement of the phosphate oxygen atom with the Zn2' ion in (20) is probably slight. A particularly H interesting example76 of phosphate ester hydrolysis involves (unsurprisingly) [CO(~~)~~~(P=O)OC~H~NO~]+ where tn =trimethylenediamine and the phos- phate group is chelated to cobalt (with ring strain). It is found that simultaneous ester hydrolysis and aquation (oia ring-opening) occur according to the outline in Scheme 6. The rate of the direct hydrolytic route provides very powerful (ca. lo9)promotion compared with the free phosphate ester. 0 00 OH-' /\4 __* (tn)*Co \P '0-+ (tn)2~o P \/\ 0 0-OH 0 I 'OPNP 0-0-Scheme 6 Ligand effects have been reported77 for the Cu2-catalysed hydrolysis of acetyl phosphate.At pH 5 whereas amines terpy and amino acids decrease the catalytic activity of Cu2+ bipy and phen increase its activity; Zn2+ Co2+,and Ni2+ are not affected in the same way by phen and bipy. 74 J. E. Loran and P. A. Naylor J.C.S. Perkin 11 1977,418. 75 C-M. Hsu and B. S. Cooperman J. Amer. Chem. SOC.,1976,98,5652,5659. 76 B. Anderson R. M. Milburn J. MacB. Harrowfield G. B. Robertson and A. M. Sargeson. J. Amer. Chem. SOC.,1977,99,2652. 77 M. Murakata Kinki Daigaku Rihogakuba Kenkyu Hokoku 1977.53 (Chem.Abs. 87,167 135). Kinetic Studies ofMetal Ion Catalysis ofHeterolytic Reactions 8 Decarboxylation Important contributions to our understanding of the Zn2'- and Cu2+-catalysed decarboxylation of oxaloacetic acid have recently come from Leussing and This reaction has often been studied and has been reviewed,'l together with related decarboxylations up to about 1974.It well illustrates two points made earlier (i) that metal ion promotion of substrates which are 0-bases usually require (at least in aqueous solvents) a chelating substrate and (ii) that tying the cleaving halves of a substrate together is unlikely to facilitate decomposition. Thus in water only 0-keto-acids containing another potential ligand (e.g. another carboxy-group) can be decarboxylated using metal ions. The essentials of the reaction are well-established:81 it proceeds from the 1:1-Py-chelate of the keto form of the oxsloacetic dianion (22)to the chelated pyruvate enol [23] which then ketonizes Scheme 7.[The negatively charged enolic chelate (24) Scheme 7 does not decarboxylate.] Normally two consecutive reactions one appreciably faster than the other and both involving changes in visible absorption are measure- able but there has been debate as to exactly what parts of Scheme 7 these spectral changes reflect. These matters are clarified by Leussing's kinetic work which also deals with effects of Zn2' and Cu2' ions on the establishment and position of the initial keto-enol equilibrium. It is known that inactive 2 1-adducts (25) are present in the reaction mixtures especially at high pH and high metal ion concentrations. Leussing suggests that the 'I3 W. D. Covey and D. L. Leussing J. Amer. Chem. SOC.,1974,96,3860.79 D. L. Leussing and N. V. Raghavan J. Amer. Chem. Soc. 1974,% 7147. N.V.Raghavan and D. L. Leussing I. Amer. Chem. SOC.,1977,99,2188. 81 R.W.Hay in ref. 7 Ch. 3. 44 D. P. N. Satchell and R. S. Satchell effects of added ligands (L),such as amino-acids,82 bipy and phen which are found to increase the effectiveness of M2+(especially Cu”) is not normally due to an increase in the acidity of ML2+ (compared with free M2+) towards the substrate but is due usually to the effect of L in reducing the concentration of (25). A kinetic of the Cu+-catalysed decomposition of aromatic acids in pyridine has appeared. The reaction is thought to be heterolytic. 9 Schiff Base Formation Schiff base formation equation (22),is promoted by metal ions if both the amino and carbonyl components are attached to the metal it is an example of template action (role 111).Using a series of M2+ ions Leussing and c011eagues~~*~~ discovered that Pb2+ Mn2+ and Zn2’ were effective but that Co2+ Cu2+ and Ni2+ were ineffective as promoters. Leussing has argued that this is because the reacting ligands are more rigidly held by some ions the most strongly acidic ions and therefore less free to react with each other than when more loosely held on other ions. However the issue is a complicated one since many factors are involved the loss of freedom and loss of nucleophilicity on being bound the gain in electrophilicity on being bound the effect of binding one basic ligand on the metal’s acid strength towards the other the relative amounts of the two ligands attached and the detailed stereochemistry of reaction.All these effects must be involved and it appears unlikely that a single rationale will be applicable to all template catalyses. Leussing’s most recent work86 emphasizes that the acceleration provided by template action by metal ions arises chiefly from the conversion of an intermolecular into an intramolecular process and not from directly acidic (polarization) effects on the substrates. Studies of pyridoxyl catalysed decompositions of P-hydroxy amino-acids and similar compounds show that they are promoted by metal ions and involve Schiff base intermediates formed from the amino-acid and pyridoxyl on the metal as a template.87*88 10 Acetal Hydrolysis and Related Reactions The kinetics of metal ion-promoted acetal hydrolysis have been examined very little.One or two studies have appeared recently. Simple U-acetals although potentially chelating are weak bases and even highly charged ions (e.g.,Fe3+) are little better89 in aqueous solution than the proton as catalysts. The hydrolysis of the O-acetal(26) ’* Y. Yasuhiro Y. Nobuyuki 0.Tadashi and M. Motoichi Yakugaku Zasshi 1977,97,70,76(Chem. Ah.,86 171 826). 83 T. Cohen R. W. Berninger and J. T. Wood J. Org. Chem. 1978,43,837. 84 D.Hopgood and D. L. Leussing J. Amer. Chem. SOC.,1969,91,3740. 85 B.E.Leach and D. L. Leussing J. Arner. Chem. SOC.,1971,93,3377. 86 R.S.McQuate and D. L. Leussing J. Amer. Chem. Soc. 1975,97 5117. ’’ Y. Murakami and H. Kondo Bull.Chem. SOC.Japan 1975,48,541. ’’ K. Tatsumoto and A. E. Martell J. Amer. Chem. Soc. 1978 100 5549. 89 G.Wada and M. Sakamoto Bull. Chem. SOC.Japan 1973,46,3378. Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions 45 which has improved chelating possibilities is promoted by Cu2+ Ni2+ and Co2+ at pH 6 and 70 "C. The observed" rates are a substantial improvement on those found in the absence of the metal ions whose relative efficiencies are Cu" :Ni2+:Co2' 1380 :15:1. Only small amounts of adduct are formed and an Al-like route Scheme 8 is proposed. HO + cu2+ d fast lH2* HO Scheme 8 In contrast to 0-acetals S-acetals are very readily hydrolysed in the presence of soft metal ions and large rate accelerations (e.g. ca. lo6compared with the proton) are found.With S-acetals such as (27) only a small extent of pre-equilibrium adduct formation occursg1 even with Hg2+ but for substrates such as (28) stoicheiometric SR / Ph,C \ SR (28) 1:1-adduct formation can occur even with the weaker Ag+ and this is another system in which kinetic terms reflecting paths involving two Ag' ions co-ordinated to the substrate are obser~able.~~ Both A 1and A2-like schemes have been proposed for these S-acetal reactions. The topic was reviewed" in 1977. Somewhat related to the acetal hydrolyses is the M2'-promoted hydrolysis and halogenation of 2-pyridyloxiran reaction (23). In aqueous solutions at pH 5 the sequence of metal ion effectiveness is Cu2' >Co2+>Zn2+>NiZ+ with Cu2+ accelerating the rate by ca.lo4-fold. Little chelate is formed especially at low pH 90 C. R. Clark and R. W. Hay J.C.S. Perkin 11 1973 1943. L. R. Fedor and B. S. R. Murty J. Amer. Chem. SOC.,1973,95,8407. 92 D.P.N. Satchel1 and T. J. Weil Inorg. Chim. Acta Letters 1978 29 L239. D. P. N. Satchell and R. S. Satchell when protons compete for the nitrogen atom. Cleavage is probably only at the p-C-0 bond in an A2 H H 11 Miscellaneous Reactions Examples of a number of other types of metal ion-promoted reactions have also been reported recently. These include hydrogen exchange and racemization of optically active the mutarotation of glucose,96 the hydrolysis of quinoline ~ulphate,~' (in some the glycolysis of organic acids and related e~terification~~~-'~' of which M' ions appear to be effective catalysts) and the ethanolysis of N-sulphinylanilines.lo2 Two particularly interesting reactions are (i) the cleavage of P-diketones with methanol (to give a monoketone and an ester) for which a reactivity sequence Zn2+>Co2' >Ni2+>Cu2' has been e~tablished,"~ and (ii) the hydration of a~etaldelyde"~ catalysed by complexes of Zn2' designed to provide general base catalysis vis template action e.g.(29). Me (29) 93 R. P. Hanzlik and W. J. Michaely J.C.S. Chem. Comm. 1975 113. 94 L. G. Stadther and R. J. Angelici Inorg. Chem. 1975 14 925. 95 P. R. Norman and D. A. Phipps Inorg. Chim. Acta Letters 1978 28 L161. q6 S.Kirschner R. V. Moraski and G. Dragulescu J. Indian Chem.SOC. 1977,54,29. 97 R. W. Hay,.C. R. Clark and J. A. G. Edmonds J.C.S. Dalton 1974 9. 98 N. E. Khomatov B. Ya. Eryshev B. P. Yatsenko and T. A. Smirnova Zhur. Vses. Khim. O-ua 1978 23,118 (Chern. Abs. 88 1692 852). 9q 0.M. 0.Habib and J. Malek Coll. Czech. Chem. Comm. 1976,41 2724. loo N. S. Antonenko E. P. Kovsman G. N. Freidlin G. A. Tarakhanov and A. I. Gravschenko Zhur. priklad. Khim. 1975. 48,692 (Chem. Abs.. 82 138 868). J. Vejrosta E. ZelCna and J. Malek Coll. Czech. Chem. Comm. 1978 43,424. "* W. K. Glass I. J. King and A. Shiels Inorg. Chim. Acta 1977 25 157. K. Uchara F. Kitamura and M. Tanaka Bull. Chem. SOC. Japan 1976,49,493. P. Woolley J.C.S. Chem. Comm. 1975. 579. Kinetic Studies of Metal Ion Catalysis of Heterolytic Reactions 12 Conclusions In the introduction we mentioned the four principal roles played by metal ions in catalysing heterolytic reactions (at least in non-enzymatic systems) and noted some of the ways in which metal ion catalysis is similar to and differs from catalysis by protons.The recent literature can be considered to exemplify these opening remarks. Apart of course from the existence of roles I11 and IV the chief difference between metal ion and proton catalysis (via roles I and 11) is probably the lack of examples of mechanisms which involve successive catalyst-substrate equilibria in which the catalyst first activates the electrophilic part of the substrate and subsequently shifts to stabilize the departing nucleophile. This type of mechanism is generally considered to be very common in proton-catalysed reactions of for example carboxylic acid derivatives.' The reason underlying this difference is that usually only one half of the substrate (the electrophile or the nucleophile part) is capable of forming a kinetically significant amount of adduct with the metal.Substrates which can chelate via both parts are not as we have seen normally decomposed effectively since the products of decomposition are still joined together. This latter effect is sometimes expressed somewhat differently and perhaps over-simply by saying that the attachment of the metal ion to the reactant makes formation of the transition state for decomposition more difficult. The 'tying-up' of a reactant by the metal ion in a less reactive form (so leading to inhibition) is a general possibility which is sometimes observed but is probably no more common than the deactivation of reactants by protons (e.g.the removal of OH-as H,O); this is the impression given by the current literature.One further aspect of metal ion catalysis which differs from that by protons is that for metal ions the rate of reaction is frequently not only dependent on metal ion concentration terms but also upon terms involving [H30+],kobssometimes being a complicated function of [M"'] and [H30']. This effect can arise not only when the concentration of attacking nucleophile is controlled by [H30+],but when any metal-induced ionization of ligand or substrate protons affects the catalysis. Complications of this nature are rarely central to the understanding of the mechanism of catalysis and have not been stressed in the foregoing discussions of individual systems.The reader needs to be prepared however for the occasional elaborate rate equation. One desirable aim in this field is to be able to predict the best metal ion to promote any given heterolytic reaction. For processes promoted by soft-soft interactions which to date have mainly concerned reactions involving assistance to nucleophile departure (role 11) the sequences of M"' reactivities obtained in different experi- mental contexts are very largely self-consistent as are the effects of ligands on the metal's reactivity nearly always ligands deactivate the softest the most. Such promoted reactions usually involve a straightforward acid-base effect.For hard (Class A) metal ions the picture is cloudier. Again if just role 11 or role I is involved usually the relative acidities of the ions (e.g. Irving-Williams series) towards that particular class of substrate seem paramount in determining reactivity but for roles I11 and IV competing effects of the metal are present. The effects of ligands (substituents) both unidentate and chelating on a hard metal's reactivity seem again often to be straightforward (the most basic leading to the greatest lowering of D. P. N. Satchell and R. S. Satchell reactivity) but not infrequently the co-ordination of a neutral base or even a negative ion leads to an apparent increase in catalytic reactivity. The reasons for this will sometimes be nothing directly to do with modified Lewis acidity but sometimes apparently they are.There seems to be a need for more understanding of the effects of substituents on the Lewis acidity of metal ions and generally for more systematic information on the relative acidities of a wide range of metal ions towards different classes of base especially unidentate bases. There are grounds therefore for believing that we still have some way to go before our predictions of M"' reactivity will be at all reliable over most of the field. The main current motivation behind studies of metal ion catalysis is their relevance as background chemistry to enzymatic catalysis. Many of the papers reviewed finish with a paragraph or two in which possible enzymatic implications are mooted.