Inorganic Reaction Mechanisms Volume 2

A Review of the Literature Published between September 1970 and November 1971

By J. Burgess

The Royal Society of Chemistry

Copyright © 1972 The Chemical Society
All rights reserved.
ISBN: 978-0-85186-265-1

Contents

Part I Electron Transfer Processes,
Introduction, 3,
Chapter 1 Reactions Between Two Metal Complexes,
Chapter 2 Metal Ion–Ligand Redox Reactions,
Chapter 3 Reactions involving Oxygen and Hydrogen Peroxide,
Part II Substitution and Related Reactions,
Chapter 1 Non-metallic Elements,
Chapter 2 Inert Metal Complexes: Co-ordination Number Four,
Chapter 3 Inert Metal Complexes: Co-ordination Numbers Five and Higher,
Chapter 4 Labile Metal Complexes,
Chapter 5 Reactions of Co-ordinated Ligands,
Part III Reactions of Biochemical Interest,
1 Introduction, 227,
2 Metal Ion Transport, 227,
3 Metal Complex Formation, 229,
4 Reactions involving Metals in Porphyrins and Related Ring Systems, 236,
5 Redox Reactions involving Metals in other Biological and Model Systems, 241,
Part IV Organometallic Compounds,
Chapter 1 Substitution,
Chapter 2 Metal–Alkyl, –Aryl, and –Allyl Bond Formation and Cleavage,
Chapter 3 Homogeneous Catalysis,
Chapter 4 Insertion Reactions,
Chapter 5 Reactions of Co-ordinated Ligands,
Chapter 6 Oxidative Addition and Reductive Elimination,
Chapter 7 Isomerization: Intramolecular Processes,
Author Index, 379,

CHAPTER 1

Part I

ELECTRON TRANSFER PROCESSES

BY A. McAULEY

Introduction

BY A. McAULEY

The format of this Part closely follows that in Volume 1. Although a degree of selection has had to be imposed owing to the large number of papers involving electron transfer processes, an attempt has been made to cover as comprehensively as possible all the areas in which studies are currently being undertaken. Compilations of data have also been assembled to allow easier comparison of the rate constants and thermodynamic parameters of reactions of a similar type.

Electron transfer processes between two metal ions continue to be examined, and a recent monograph sets out clearly the various modes of interaction between the reactant species. The effects of non-bridging ligands on these processes have been discussed, the main influence of such groups, both on oxidant and reductant, being to change the activation free energy by altering the overall free energy of the reaction. Although the donor ligands involving either N- or O-co-ordination have been most utilized, the effects of thioether ligands asnon-bridging ligands have also been studied. In the reactions of chromium(VI), the three-unit change in oxidation state contrasts with the (usually) single electron transfer step in most metal ion reactions. The reactions of this ion with both aquo-metal ions and complex species have been reviewed.

Several other useful reviews of reactions involving metal ions have also been published. Redox reactions of chromium(III)-amine species have been described and a survey has been made of the solution chemistry together with reaction paths involved in the redox reactions of various plutonium species. Oxidation reactions of thallium(III) have also been described. Developments in the redox chemistry of peroxides have been reviewed, the nature of the reactions which involve iron(III) in various complexed forms providing a fascinating example of the manner in which geometry and co-ordination to the metal centre greatly affect the reactivity of the system. Redox properties of cobalt chelates, with delocalized electronic structure have been described, and transfer processes involving metalloporphyrins have been reviewed. In the case of dithiolene complexes of transition metals, the mode of reaction is influenced by solvent, the extent of conjugation of the ligands, and substituents on the donor atoms.

The oxidation of organic substrates by metal-ion species continues to be a fruitful source of study. The question of bonded and non-bonded interactions in one-electron transfer processes has been explored, and several papers have been devoted to the nature of possible metal ion–substrate intermediates. The question of covalent bond formation between oxidant and reductant as a pre-requisite to the electron transfer step has also been investigated for a large number of reactions. The mechanisms of reactions of lead(IV) with organic species have been reviewed l7 and the reactions of oxygen-containing radicals and hydrogen atoms with metal ions in various oxidation states have been surveyed. The introduction of such ions into a system where radical polymerization is occurring results in retardation of the chain polymerization, both oxidizing and reducing ions being involved in the termination processes. The role of electrons in radical systems has been examined, and heterolytic oxidation reactions of hydrogen peroxide in acid solutions have also been reviewed.

1

Reactions Between Two Metal Complexes

BY A. McAULEY

1 Reducing Agents

Chromium(II). — Quantitative data for this and the other reducing agents are tabulated in Table 1, p. 19.

Many reactions have been studied using this reductant, and with inert oxidants, such as the ammine complexes of cobalt(III), there is the possibility of characterizing more fully the inner-sphere route in redox systems.

In the reduction of maleatopenta-amminecobalt(III), the (quantitatively) inner-sphere process yields an initial ratio of 4 : 1 of chelated to unidentate complexes of chromium(III). The amounts of the two maleato-complexes formed are however dependent on acidity, and it has been shown that the interconversion of the two species,

[FORMULA NOT REPRODUCIBLE IN ASCII]

is catalysed by chromium(II). The nature of the activated complexes in both the cobalt(III) and isomerization reactions is considered to be similar, with a chelated chromium(II) environment. The rate of reduction of the methyl-maleato-complex does not differ substantially from that of the maleatospecies. In the chromium(II) reductions of thiocyanato- and isothiocyanatopenta-amminecobalt(III) complexes, both adjacent and remote innersphere attack is observed. In an excess of chromium(II), two reactions with (H3N)5CoSCN2+ are detected, the first yielding a mixture of CrSCN2+ and CrNCS2+. In the second stage, the species CrSCN2+ is isomerized [chromium(II)-catalysed] to CrNCS2+. The initial step is considered to involve the parallel reactions

[FORMULA NOT REPRODUCIBLE IN ASCII]

the fraction of the S-bonded complex produced being 0.29. In the case of the reaction with (H3N)6CoNCS2+, it is considered that the mechanism involves quantitative remote attack,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The reaction is substantially slower than that involving the sulphur-bonded penta-ammine complex, a factor which is in part accounted for by the high electron-mediating ability of a sulphur atom bound to the oxidizing centre for reaction via an inner-sphere mechanism. Kinetic studies on the Cr reduction of isocyanopenta-amminecobalt(III) have also been carried out, the redox reaction being accompanied by a parallel hydrolysis step. The expected hydrolysis product [Co(NH3)63+] is not, however, found under the experimental conditions, and it is suggested that an intermediate is formed (Scheme 1). This might either decompose in an acid-catalysed reaction to yield Co(NH3)63+, or the N-bonded complex could act as a bridging ligand in an electron transfer. The latter reaction is considered to be at least 60 times faster than the former. The predominantly inner-sphere reductions of a series of carboxylatopenta-amminecobalt(III) complexes have been re-examined at high acidities, where for the acetato-complex the decrease in the reduction rate on increasing the hydrogen-ion concentration is suggestive of association between the complex and H+. Although inductive effects are small, steric influences could be important. The reduction of the trimethylacetato-complex is much slower than that of the other complexes, reflecting the strong hindrance, but the reaction is nevertheless considered to proceed via an inner-sphere route for at least 70% of the reaction.

It has been proposed that the groups in the bridging molecule attached to the oxidizing and reducing agents in these reactions should have a lone pair available for bonding and must also form part of a conjugated system. Such criteria cannot be simultaneously satisfied in the case of the NH2group. Studies have thus been undertaken to examine the mode of interaction of complexes where such a group is the potential bridging site. The reductions of penta-ammine cobalt complexes of urea, carbamate, N-cyanoguanidine [N [equivalent to] C — N=C(NH2)2], and cyanamide (N=C — NH2), by chromium(II) involve an outer-sphere path with the exception of the carbamato-species, where, however, there is no evidence of a nitrogenbonded chromium(III) species as product. The cyanamide-complex reaction follows the rate law

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

which is consistent with the reaction scheme involving a rapid pre-equilibrium:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

In this case, bridged electron-transfer occurs. The fact that urea and N-cyanoguanidine complexes are reduced with low rate constants, and that only when the conjugate base of the cyanamide complex is formed does the inner-sphere route become operative confirms the suggestion that two non-a-bonding electron pairs are required on the ligand attaching to the reducing agent in the transition state.

The relative rates of reduction of cobalt(III) complexes containing groups bridging in the activated complex have been determined by the method of competing reactions, where two potential bridging groups (in the same or in separate ions) may react with the chromium(II). For the systems Co(NH3)5L2+ + Cr2+, where L = I, Br, Cl, F, and N3, the relative rates are 2.75 : 1.89 : 1.51 : 1.00 : 0.88. These ratios are, however, dependent on ionic strength. If the two halides are present in the same complex, as in cis-and trans-Co(en)2ClBr+, the relative rate parameters are decreased to 1.14 and 1.00 respectively. For a series of trans substituted chloro-bis(ethylene-diamine)cobalt(III) complexes, it would appear that the activating effect of the group trans to the bridging ligand is in the order

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

A possible cis effect may also be operative. It is of interest to note that these reductions are susceptible to free-ligand effects, the order in anions decreasing in the sequence

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Sykes and his co-workers have made a detailed study of the metalion reduction of µ-amido-µ-(ligand)-bis[tetra-amminecobalt(III)] complexes where ligand = sulphato, selenato, phosphato, oxalato, and hydroxo. In the reaction between CrII and the µ-amido-µ-sulphato-complex, there is a two-stage reaction (CoIII2 = dimer),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

with an attack of a chromium(II) on the cobalt(III) centre in each case. The intermediate CrIII — CoIII (Figure 1), formed at the end of the first stage, has been characterized spectroscopically using the stopped-flow method, and is considered to be the complex ion µ-sulphato-[penta-amminecobalt(III)]-[penta-aquochromium(III)]. The rate law for this reaction may be expressed as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The second stage of the reduction is substantially slower and is hydrogen-ion dependent, the rate law being

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The hydrogen-ion inverse dependence has been taken as consistent with hydroxy-bridging, and the transition states (1) and (2) for the reaction paths k2 and k3 respectively may be described as shown. In the corresponding

[FORMULA NOT REPRODUCIBLE IN ASCII] (1)

[FORMULA NOT REPRODUCIBLE IN ASCII] (2)

µ-selenato reaction only the first step could be investigated, but in this case the reduction is faster by a factor of 50 than that for the µ-sulphato-system, a result which is explicable in terms of the ease with which the free oxoanions are reduced.

The µ-phosphato-complex reacts in rather a similar manner to the sulphato-species, the two-stage reaction again involving intermediate formation. The µ-amido-µ-phosphato-species may however undergo protonation at low pH (the ammonia groups have been omitted for clarity). K1 and K2 represent the acid dissociation constants. Studies on the first stage indicate that the species (4) is the reactant although complex (3) is the form predominating in solution. In the second reaction, the rate of disappearance of the cobalt(III)–chromium(III) intermediate is dependent

[FORMULA NOT REPRODUCIBLE IN ASCII] (3)

[FORMULA NOT REPRODUCIBLE IN ASCII] (4)

[FORMULA NOT REPRODUCIBLE IN ASCII] (5)

on acidity and it may be that in this case the activated complex involves deprotonation and multiple bridging. In the reaction of the corresponding µ-amido-µ-oxalato-species, two stages of reduction are observed, of which the first involves both a protonated and a deprotonated form of the bridged complex ion. The second path, however, is acid independent.

The reduction of the µ-amido-µ-hydroxo-complex is further complicated by aquation and hydroxo-bridge cleavage reactions which are rate determining. The rate law

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

involves the reactions (k1), the displacement of a co-ordinated ammonia by a water molecule, and (k2), the composite rate constant for the formation of the singly bridged µ-amido-bis-aquo-dimer. Both the mono- and bisaquo-complexes are considered to react rapidly with the reductant. The attack of Cr on the doubly bridged µ-amido-µ-hydroxo-species is negligible. A further pathway involving the protonated µ-amido-µ-aquo-complex has been detected, and this must be outer-sphere.

A two-electron inner-sphere reduction of chloropenta-ammineplatinum(IV) has been described, the final products in the pH range 0 — 2 being CrCl2+, Cr(H2O)63+, and Pt(NH3)2+. In the presence of excess reductant, an intermediate is observed with a subsequent rapid decay. Since the chloride ion is incorporated into the chromium(III) product, it is suggested that the initial two-electron process

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

is followed by a slower reaction of the chromium(IV) species with the reductant to yield the products. Transfer of a ligand in an inner-sphere process is also observed in the rapid reaction between chromium(II) and Co(en)2(SCH2CO2)+ where the mercaptoacetate is found in the oxidized product.

The outer-sphere reductions of slowly reacting species such as Co(NH3)63+, (NH3)5Co(DMF)3+, and (NH3)5Co(py)3+ are greatly accelerated by certain pyridine derivatives which possess an alkenyl or carbonyl substituent γ to the nitrogen. The increase in rate is most pronounced when 4-pyridine-carboxylic acid and its carboxy-bound-penta-aquochromium(III) derivative are present. The corresponding β-derivatives show no catalytic action, and that of the α-analogues is only marginal. In the presence of the (γ-) chromium(III) complex (1.8 × 10-3 M) the reduction rate of Co(NH3)63+ is increased by a factor of 2.6 × 104. The mechanism is considered to involve two hydrogen-ion-related highly reactive dinuclear radical ions (Scheme 2). The high reactivity of the dinuclear radicals may be promoted by the fact that electron loss from an extended conjugated complex-ion involves less bond distortion and solvent reorganization than that from a small hydrated cation.

The kinetics of oxidation of mixed 2,2′- and 4,4′-bipy-chromium(II) complexes by cobalt(III) species have also been investigated. Addition of 4,4′-bipy to acidic aqueous solutions of Cr(2,2′-bipy)32+ produces the mono-protonated complex [Cr(2,2′-bipy)2(4,4′-bipy-H)(H2O)]3+, which is found to react 20 — 40 times faster than the tris-(2,2′-bipy)-CrII species. The factors governing the rates include solvation differences between the reactants and the activated complexes, and the excitation energy of the cobalt(III) complexes from the t2g ground state to the t2geg configuration. Although there is no direct proof of the geometry of the activated complex involved in the reaction of the mixed (2,2′-4,4′-H) species, the considerably more negative value for ΔS≠ in comparison with those for Cr(2,2′-bipy)32+ suggests that specific orientation of the reactants is required for the attack by the oxidant at one of the heteroligands of the chrqmium(II), probably the 4,4′-bipy-H centre.

Iron(II). — Evidence has been established for a dinuclear intermediate preceding the electron-transfer process in the oxidation of iron(II) by penta-amminecobalt(III)nitrilotriacetate, in which only one of the carboxylates in the latter ligand is bound to the cobalt centre. In acid solution, the predominating form of the oxidant is (H3N)5CoLH2 [where L = N(CH2COO)33-], but in the pH range 1.3 — 3.4, the kinetics are consistent with a bridged mechanism involving the chelation of the reductant prior to electron transfer, the precursor complex being considered to be of the form (6) and having a life-time of the order of 10 s, and an equilibrium quotient of ~ 106 mol-1.

[FORMULA NOT REPRODUCIBLE IN ASCII] (6)

The 1 : 1 reductions of µ-amido-µ-superoxo-cobalt(III) complexes involving bipyridyl and 1,10-phenanthroline (LL) have been shown to be first order in reductant and in metal complex but independent of acidity in the range 0.05 — 0.20 M. The products of the reactions are the corresponding µ-peroxo-species,

[FORMULA NOT REPRODUCIBLE IN ASCII]

the net reaction being an electron transfer from the metal ion to the superoxo-bridge of the dicobalt complex. In the case of the ligands quoted compared with NH3 or en, the rates are considerably faster, the effect being associated with the delocalization of the electron into the bipy and phen ligand systems.

Superoxo-bridged dinuclear complexes of rhodium(III) with pyridine or γ-picoline ligands (L), [(H2O)L4Rh·O2·RhL4(H2O)]5+, have been prepared, in which acid–base equilibria involving the complex water molecules have been demonstrated. The corresponding dichloro-complex has been shown to be a strong oxidant, the reaction

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

yielding the peroxo-bridged analogue which may in acid media be reoxidized using CeIV or MnO4- to the original superoxo-ion. Iron(II) salts also interfere with the method of detection of peroxocobalt(III) complexes by means of the chemiluminescent oxidation of luminol, the iron(II) being oxidized by atmospheric oxygen.

The effect of thioether donor-atoms as non-bridging ligands in the reduction of cobalt(III) complexes has been investigated, the reaction of iron(II) with S-cis-dichloro(1,8-diamino-3,6-ditbiaoctane)cobalt(III) being 103 times more rapid than that for cis-Co(en)2Cl2+ under comparable conditions. The geometry of the ion (7) has been shown to involve two cis thioether donor-atoms, and the chloride participation in the bridging mechanism is evidenced by the lack of reaction with complexes with the corresponding 1,10-phenanthroline and 2,2′-bipyridyl groupings in the inner co-ordination sphere.

(Continues…)Excerpted from Inorganic Reaction Mechanisms Volume 2 by J. Burgess. Copyright © 1972 The Chemical Society. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.