Ligand substitution reactions. Topic: Elementary stages involving coordination and organometallic compounds in solutions and on the surface of metals and oxides

The elementary stages of organic reactions catalyzed by acids, bases, nucleophilic catalysts, metal complexes, solid metals and their compounds in gas-phase or liquid-phase heterogeneous and homogeneous processes are reactions of the formation and transformation of various organic and organometallic intermediates, as well as metal complexes. Organic intermediate compounds include carbenium ions R + , carbonium RH 2 + , carbo-anions R-, anion- and radical cations, radicals and biradicals R·, R:, as well as molecular complexes of organic donor and acceptor molecules (DA), which are called also by complexes with charge transfer. In homogeneous and heterogeneous catalysis by metal complexes (metal complex catalysis) of organic reactions, the intermediates are complex (coordination) compounds with organic and inorganic ligands, organometallic compounds with an M-C bond, which in most cases are coordination compounds. A similar situation occurs in the case of “two-dimensional” chemistry on the surface of solid metal catalysts. Let us consider the main types of reactions of metal complexes and organometallic compounds.

Reactions of metal complexes can be divided into three groups:

a) electron transfer reactions;

b) ligand substitution reactions;

c) reactions of coordinated ligands.

Electron transfer reactions

Two mechanisms are implemented in electron transfer reactions - the outer-sphere mechanism (without changes in the coordination spheres of the donor and acceptor) and the bridging (inner-sphere) mechanism, leading to changes in the coordination sphere of the metal.

Let us consider the outer-sphere mechanism using the example of octahedral complexes of transition metals. In the case of symmetric reactions ( G 0 = 0)

rate constants vary in a very wide range of values ​​- from 10-12 to 10 5 l mol-1 sec-1, depending on the electronic configuration of the ion and the degree of its restructuring during the process. In these reactions, the principle of least movement is very clearly manifested - the least change in the valence shell of the reaction participants.

In the electron transfer reaction (1) (Co * is an isotope of the Co atom)

(symmetric reaction), Co 2+ (d 7) goes into Co 3+ (d 6). The electronic configuration (valence shell) does not change during this transfer

6 electrons at the triply degenerate bonding level remain unchanged (), and from the antibonding level e g level one electron is removed.
Second order rate constant for reaction (1) k 1 = 1.1 lmol-1 sec-1. Since Phen (phenanthroline) is a strong ligand, the maximum number is 7 d-electrons are paired (spin-paired state). In the case of a weak ligand NH 3 the situation changes radically. Co(NH 3) n 2+ (n = 4, 5, 6) is in a spin-unpaired (high-spin) state.

The stronger complex Co(NH 3) 6 3+ (stronger than Co(NH 3) 6 2+ ~ 10 30 times) is in a spin-paired state, like the complex with Phen. In this regard, in the process of electron transfer, the valence shell should be strongly reconstructed and, as a result, k= 10-9 lmol-1 sec-1. The conversion rate of Co 2+ to Co 3+, equal to 50%, is achieved in the case of the Phen ligand in 1 second, and in the case of NH 3 ~ in 30 years. It is obvious that a stage with such a rate (formally elementary) can be excluded from the set of elementary stages when analyzing reaction mechanisms.

Magnitude G for the electron transfer reaction during the formation of a collision complex, according to Marcus theory, includes two components and

The first term is the energy of reorganization of M-L bonds within the complex (the length and strength of the bond when the valence state changes). The value includes the energy of rearrangement of the outer solvation shell in the process of changing the M-L coordinates and the charge of the complex. The smaller the change in the electronic environment and the smaller the change in the M-L length, the lower; the larger the ligands, the smaller and, as a result, the higher the rate of electron transfer. The value for the general case can be calculated using the Marcus equation

Where. At = 0 .

In the case of the intrasphere mechanism, the process of electron transfer is facilitated, since one of the ligands of the first complex forms a bridging complex with the second complex, displacing one of the ligands from it

The rate constants of such a process are 8 orders of magnitude higher than the constants for the reduction of Cr(NH 3) 6 3+. In such reactions, the reducing agent must be a labile complex, and the ligand in the oxidizing agent must be capable of forming bridges (Cl-, Br-, I-, N 3 -, NCS-, bipy).

One of the most important stages in metal complex catalysis, the interaction of substrate Y with the complex, occurs through three mechanisms:

a) Replacement of the ligand with a solvent. This stage is usually depicted as the dissociation of the complex

The essence of the process in most cases is the replacement of ligand L with solvent S, which is then easily replaced by a substrate molecule Y

b) Attachment of a new ligand at a free coordinate with the formation of an associate followed by dissociation of the replaced ligand

c) Synchronous substitution (type S N 2) without intermediate formation

In the case of Pt(II) complexes, the reaction rate is very often described by the two-path equation

Where k S And k Y- rate constants of processes occurring in reactions (5) (with a solvent) and (6) with ligand Y. For example,

The last stage of the second route is the sum of three fast elementary stages - the elimination of Cl-, the addition of Y and the elimination of the H 2 O molecule.

In flat square complexes of transition metals, a trans effect is observed, formulated by I.I. Chernyaev - the influence of LT on the rate of substitution of a ligand that is in a trans position to the LT ligand. For Pt(II) complexes, the trans effect increases in the series of ligands:

H2O~NH3< Cl- ~ Br- < I- ~ NO 2 - ~ C 6 H 5 - < CH 3 - <
< PR 3 ~ AsR 3 ~ H- < олефин ~ CO ~ CN-.

The presence of the kinetic trans-effect and thermodynamic trans-influence explains the possibility of synthesizing inert isomeric complexes of Pt(NH 3) 2 Cl 2:

§ Reactions of electrophilic substitution (SE) of hydrogen with a metal in the coordination sphere of the metal and their reverse processes

SH - H 2 O, ROH, RNH 2, RSH, ArH, RCCH.

Even H 2 and CH 4 molecules participate in reactions of this type

§ Reactions of introduction of L through the M-X connection

In the case of X = R (organometallic complex), metal-coordinated molecules are also introduced into the M-R bond (L - CO, RNC, C 2 H 2, C 2 H 4, N 2, CO 2, O 2, etc.). The insertion reaction is the result of an intramolecular attack of nucleophile X on a molecule coordinated by - or - type. Reverse reactions - - and - elimination reactions

§ Reactions of oxidative addition and reductive elimination

M 2 (C 2 H 2) M 2 4+ (C 2 H 2) 4-

Apparently, in these reactions there is always preliminary coordination of the added molecule, but this cannot always be detected. In this regard, the presence of a free site in the coordination sphere or a site associated with a solvent, which is easily replaced by a substrate, is an important factor affecting the reactivity of metal complexes. For example, bis-allyl complexes of Ni are good precursors of catalytically active species, since due to the easy reductive elimination of the bis-allyl, a complex with the solvent appears, the so-called. “bare” nickel. The role of empty seats is illustrated by the following example:

§ Reactions of nucleophilic and electrophilic addition to - and - metal complexes

As intermediates of catalytic reactions, there are both classical organometallic compounds having M-C, M=C and MC bonds, and non-classical compounds in which the organic ligand is coordinated according to the 2, 3, 4, 5 and 6 type, or is an electron-deficient element structures - bridging CH 3 and C 6 H 6 groups, non-classical carbides (Rh 6 C(CO) 16, C(AuL) 5 +, C(AuL) 6 2+, etc.).

Among the scientific mechanisms for classical organometallic compounds, we note several mechanisms. Thus, 5 mechanisms of electrophilic substitution of the metal atom at the M-C bond have been established.

electrophilic substitution with nucleophilic assistance

AdE Addition-elimination

AdE(C) Addition to the C atom in sp 2 hybridization

AdE(M) Oxidative addition to metal

Nucleophilic substitution at the carbon atom in demetalation reactions of organometallic compounds occurs as an oxidation-reduction process:

Possible participation of an oxidizing agent in this stage

Such an oxidizing agent can be CuCl 2, p-benzoquinone, NO 3 - and other compounds. Here are two more elementary stages characteristic of RMX:

hydrogenolysis of the M-C bond

and homolysis of the M-C bond

An important rule that applies to all reactions of complex and organometallic compounds and is associated with the principle of least motion is Tolman's 16-18 electron shell rule (Section 2).

According to modern concepts, complexes and organometallic compounds similar to compounds in solutions are formed on the surface of metals. For surface chemistry, the participation of several surface atoms in the formation of such compounds and, of course, the absence of charged particles is essential.

Surface groups can be any atoms (H, O, N, C), groups of atoms (OH, OR, NH, NH 2, CH, CH 2, CH 3, R), coordinated molecules CO, N 2, CO 2, C 2H4, C6H6. For example, during the adsorption of CO on a metal surface, the following structures were found:

The C 2 H 4 molecule on the metal surface forms -complexes with one center and di-connected ethylene bridges M-CH 2 CH 2 -M, i.e. essentially metal cycles

On the surface of Rh, for example, during the adsorption of ethylene, the following processes of ethylene conversion occur as the temperature increases:

Reactions of surface intermediates include the stages of oxidative addition, reductive elimination, insertion, - and -elimination, hydrogenolysis of M-C and C-C bonds and other reactions of the organometallic type, but without the appearance of free ions. The tables show the mechanisms and intermediates of surface transformations of hydrocarbons on metals.

Table 3.1. Catalytic reactions involving the cleavage of a C-C bond.

Designations:

Alkyl, metallacycle;

Carbene, allyl;

Carbin, vinyl.

Table 3.2. Catalytic reactions involving the formation of a C-C bond.

Designations: see table. 3.1.

The formation of all of the above organometallic compounds on the surface of metals has been confirmed by physical methods.

Questions for self-control

1) How does the rule of the smallest change in the valence shell of a metal manifest itself during electron transfer reactions?

2) Why do coordination vacancies contribute to effective interaction with the substrate?

3) List the main types of reactions of coordinated ligands.

4) Give the mechanisms of electrophilic substitution in the reactions of organometallic compounds with NX.

5) Give examples of surface organometallic compounds.

6) Give examples of the participation of metal carbene surface complexes in the transformations of hydrocarbons.

Literature for in-depth study

1. Temkin O.N., Kinetics of catalytic reactions in solutions of metal complexes, M., MITHT, 1980, Part III.

2. Collman J., Higedas L., Norton J., Finke R., Organometallic chemistry of consumer metals, M., Mir, 1989, vol. I, vol. II.

3. Moiseev I.I., -Complexes in the oxidation of olefins, M., Nauka, 1970.

4. Temkin O.N., Shestakov G.K., Treger Yu.A., Acetylene: Chemistry. Mechanisms of reactions. Technology. M., Chemistry, 1991, 416 pp., section 1.

5. Henrici-Olivet G., Olive S., Coordination and catalysis, M., Mir, 1980, 421 p.

6. Krylov O.V., Matyshak V.A., Intermediate compounds in heterogeneous catalysis, M., Nauka, 1996.

7. Zaera F., An Organometallic Guide to the Chemistry of Hydrocarbon Moities on Transition Metal Surfaces., Chem. Rev., 1995, 95, 2651 - 2693.

8. Bent B.E., Mimicking Aspects of Heterogeneous Catalysis: Generating, Isolating, and Reacting Proposed Surface Intermediates on Single Crystals in Vacuum, Chem. Rev., 1996, 96, 1361 - 1390.

Conventionally, chemical reactions of complexes are divided into exchange, redox, isomerization and coordinated ligands.

The primary dissociation of complexes into the inner and outer sphere determines the occurrence of exchange reactions of outer-sphere ions:

X m + mNaY = Y m + mNaX.

Components of the internal sphere of complexes can also participate in metabolic processes involving both ligands and the complexing agent. To characterize substitution reactions of ligands or the central metal ion, use the designations and terminology proposed by K. Ingold for reactions of organic compounds (Fig. 42), nucleophilic S N and electrophilic S E substitutions:

Z + Y = z +X S N

Z + M"= z + M S E .

According to the mechanism of the substitution reaction, they are divided (Fig. 43) into associative ( S N 1 and S E 1 ) and dissociative ( S N 2 and S E 2 ), differing in the transition state with an increased and decreased coordination number.

Classifying a reaction mechanism as associative or dissociative is a difficult experimentally achievable task of identifying an intermediate with a reduced or increased coordination number. In this regard, the reaction mechanism is often judged on the basis of indirect data on the effect of the concentration of reagents on the reaction rate, changes in the geometric structure of the reaction product, etc.

To characterize the rate of ligand substitution reactions in complexes, 1983 Nobel laureate G. Taube (Fig. 44) proposed using the terms “labile” and “inert” depending on the time of the ligand substitution reaction, less than or more than 1 minute. The terms labile or inert are characteristics of the kinetics of ligand substitution reactions and should not be confused with thermodynamic characteristics of the stability or instability of complexes.

The lability or inertness of the complexes depends on the nature of the complexing ion and the ligands. In accordance with ligand field theory:

1. Octahedral complexes 3 d transition metals with distribution of valence ( n -1) d electrons per sigma*(e g ) loosening MOs are labile.

4- (t 2g 6 e g 1) + H 2 O= 3- + CN - .

Moreover, the lower the energy of stabilization by the crystal field of the complex, the greater its lability.

2. Octahedral complexes 3 d transition metals with free sigma* loosening e g orbitals and a uniform distribution of valence ( n -1) d electrons in t 2 g orbitals (t 2 g 3, t 2 g 6) are inert.

[Co III (CN) 6] 3- (t 2 g 6 e g 0) + H 2 O =

[Cr III (CN) 6] 3- (t 2 g 3 e g 0) + H 2 O =

3. Plano-square and octahedral 4 d and 5 d transition metals that do not have electrons per sigma* loosening MOs are inert.

2+ + H 2 O =

2+ + H 2 O =

The influence of the nature of ligands on the rate of ligand substitution reactions is considered within the framework of the “mutual influence of ligands” model. A special case of the model of mutual influence of ligands is that formulated in 1926 by I.I. Chernyaev's concept of trans influence (Fig. 45) - “the lability of the ligand in the complex depends on the nature of the trans-located ligand” - and propose a number of trans-influences of the ligands: CO, CN -, C 2 H 4 > PR 3, H - > CH 3 -, SC (NH 2) 2 > C 6 H 5 -, NO 2 -, I -, SCN - > Br -, Cl - > py , NH 3 , OH - , H 2 O .

The concept of trans influence allowed us to justify the rules of thumb:

1. Peyrone's rule- due to the action of ammonia or amines on tetrachloroplatinate ( II ) potassium is always obtained dichlorodiamineplatinum cis-configuration:

2 - + 2NH 3 = cis - + 2Cl - .

Since the reaction proceeds in two stages and the chloride ligand has a large trans influence, the replacement of the second chloride ligand with ammonia occurs with the formation of cis-[ Pt (NH 3 ) 2 Cl 2 ]:

2- + NH 3 = -

NH 3 = cis -.

2. Jergensen's rule - upon the action of hydrochloric acid on platinum tetrammine chloride ( II ) or similar compounds is obtained dichlorodi-ammineplatinum trans configuration:

[ Pt (NH 3 ) 4 ] 2+ + 2 HCl = trans-[ Pt (NH 3 ) 2 Cl 2 ] + 2 NH 4 Cl .

In accordance with the series of trans-influences of ligands, the replacement of the second ammonia molecule by a chloride ligand leads to the formation of trans-[ Pt (NH 3 ) 2 Cl 2 ].

3. Kurnakov's thiourea reaction - various products of the reaction of thiourea with geometric isomers of trans-[ Pt (NH 3 ) 2 Cl 2 ] and cis-[ Pt (NH 3 ) 2 Cl 2 ]:

cis - + 4Thio = 2+ + 2Cl - + 2NH 3 .

The different nature of the reaction products is associated with the high trans influence of thiourea. The first stage of the reactions is the replacement of thiourea chloride ligands with the formation of trans- and cis-[ Pt (NH 3 ) 2 (Thio ) 2 ] 2+ :

trans-[Pt (NH 3) 2 Cl 2 ] + 2 Thio = trans-[ Pt (NH 3) 2 (Thio) 2 ] 2+

cis - + 2Thio = cis - 2+.

In cis-[Pt (NH 3 ) 2 (Thio ) 2 ] 2+ two ammonia molecules in trans position to thiourea undergo further substitution, which leads to the formation 2+ :

cis - 2+ + 2Thio = 2+ + 2NH 3 .

In trans-[Pt (NH 3 ) 2 (Thio ) 2 ] 2+ two ammonia molecules with little trans influence are located in a trans position to each other and therefore are not replaced by thiourea.

The patterns of trans influence were discovered by I.I. Chernyaev when studying ligand substitution reactions in square-planar platinum complexes ( II ). Subsequently, it was shown that the trans-influence of ligands also manifests itself in complexes of other metals ( Pt(IV), Pd(II), Co(III), Cr(III), Rh(III), Ir(III )) and other geometric structure. True, the series of trans-influence of ligands for different metals are somewhat different.

It should be noted that trans influence is kinetic effect- the greater the trans influence of a given ligand, the faster it is replaced by another ligand that is in a trans position relative to it.

Along with the kinetic effect of trans influence, in the middle XX century A.A. Grinberg and Yu.N. Kukushkin established the dependence of the trans-influence of the ligand L from the ligand located in cis-position to L . Thus, the study of the rate of substitution reaction Cl- ammonia in platinum complexes( II):

[PtCl 4 ] 2- + NH 3 = [ PtNH 3 Cl 3 ] - + Cl - K = 0.42. 10 4 l/mol. With

[ PtNH 3 Cl 3 ] - + NH 3 = cis-[ Pt (NH 3 ) 2 Cl 2 ] + Cl - K = 1.14. 10 4 l/mol. With

trans-[ Pt (NH 3 ) 2 Cl 2 ] + NH 3 = [ Pt (NH 3 ) 3 Cl ] + + Cl - K = 2.90 . 10 4 l/mol. With

showed that the presence of one or two ammonia molecules in the cis position to the replaced chloride ligand leads to a consistent increase in the reaction rate. This kinetic effect is called cis influence. Currently, both kinetic effects of the influence of the nature of the ligands on the rate of ligand substitution reactions (trans- and cis-effect) are combined in a general concept mutual influence of ligands.

The theoretical substantiation of the effect of mutual influence of ligands is closely related to the development of ideas about chemical bonds in complex compounds. In the 30s XX century A.A. Greenberg and B.V. Nekrasov considered the trans influence within the framework of the polarization model:

1. The trans effect is typical for complexes whose central metal ion is highly polarizable.

2. The trans activity of ligands is determined by the energy of mutual polarization of the ligand and the metal ion. For a given metal ion, the trans influence of the ligand is determined by its polarizability and distance from the central ion.

The polarization model is consistent with experimental data for complexes with simple anionic ligands, such as halide ions.

In 1943 A.A. Greenberg hypothesized that the trans activity of ligands is related to their reducing properties. The shift in electron density from the trans-located ligand to the metal reduces the effective charge of the metal ion, which leads to a weakening of the chemical bond with the trans-located ligand.

The development of ideas about trans influence is associated with the high trans activity of ligands based on unsaturated organic molecules like ethylene in [ Pt(C2H4)Cl3 ] - . According to Chatt and Orgel (Fig. 46), this is due topi-the dative interaction of such ligands with the metal and the associative mechanism of substitution reactions for trans-located ligands. Coordination to the metal ion of the attacking ligand Z leads to the formation of a five-coordinate trigonal bipyramidal intermediate followed by rapid elimination of the leaving ligand X. The formation of such an intermediate is facilitated bypi-dative ligand-metal ligand interaction Y , which reduces the electron density of the metal and reduces the activation energy of the transition state with subsequent rapid replacement of ligand X.

Along with p acceptor (C 2 H 4 , CN - , CO ...) ligands that form a dative ligand-metal chemical bond have a high trans-influence andsdonor ligands: H - , CH 3 - , C 2 H 5 - ... The trans-influence of such ligands is determined by the donor-acceptor interaction of ligand X with the metal, which lowers its electron density and weakens the bond of the metal with the leaving ligand Y.

Thus, the position of the ligands in the series of trans-activity is determined by the combined action of sigma- donor and pi-properties of ligands - sigma- donor and pi-the acceptor properties of the ligand enhance its trans-influence, whereaspi-donor ones weaken. Which of these components of the ligand-metal interaction predominates in the trans effect is judged on the basis of quantum chemical calculations of the electronic structure of the transition state of the reaction.

One of the most important stages in metal complex catalysis—the interaction of the substrate Y with the complex—occurs by three mechanisms:

a) Replacement of the ligand with a solvent. This stage is usually depicted as the dissociation of the complex

The essence of the process in most cases is the replacement of the ligand with a solvent S, which is then easily replaced by a substrate molecule Y

b) Attachment of a new ligand at a free coordinate with the formation of an associate followed by dissociation of the replaced ligand

c) Synchronous substitution (type S N 2) without intermediate formation

In the case of Pt(II) complexes, the reaction rate is very often described by the two-path equation

Where k S And k Y are the rate constants of processes occurring in reactions (5) (with a solvent) and (6) with ligand Y. For example,

The last stage of the second route is the sum of three fast elementary stages - the elimination of Cl –, the addition of Y and the elimination of the H 2 O molecule.

In flat square complexes of transition metals, a trans effect is observed, formulated by I.I. Chernyaev - the influence of LT on the rate of substitution of a ligand located in a trans position to the LT ligand. For Pt(II) complexes, the trans effect increases in the series of ligands:

H 2 O~NH 3

The presence of the kinetic trans-effect and thermodynamic trans-influence explains the possibility of synthesizing inert isomeric complexes of Pt(NH 3) 2 Cl 2:

Reactions of coordinated ligands

Reactions of electrophilic substitution (S E) of hydrogen with a metal in the coordination sphere of the metal and their inverse processes

SH – H 2 O, ROH, RNH 2, RSH, ArH, RCCH.

Even H 2 and CH 4 molecules participate in reactions of this type

Reactions of introduction of L along the M-X connection

In the case of X=R (organometallic complex), metal-coordinated molecules are also introduced into the M-R bond (L–CO, RNC, C 2 H 2, C 2 H 4, N 2, CO 2, O 2, etc.). The insertion reaction is the result of an intramolecular attack of a nucleophile on a - or -type-coordinated molecule. Reverse reactions – - and -elimination reactions

Oxidative addition and reductive elimination reactions

M 2 (C 2 H 2)  M 2 4+ (C 2 H 2) 4–

Apparently, in these reactions there is always preliminary coordination of the added molecule, but this cannot always be detected. Therefore, the presence of a free site in the coordination sphere or a site associated with a solvent that is easily replaced by a substrate is an important factor affecting the reactivity of metal complexes. For example, bis--allyl complexes of Ni are good precursors of catalytically active species, since due to the easy reductive elimination of the bis-allyl, a complex with the solvent appears, the so-called. “bare” nickel. The role of empty seats is illustrated by the following example:

Reactions of nucleophilic and electrophilic addition to - and -complexes of metals

1. Reactions of organometallic compounds

As intermediates of catalytic reactions, there are both classical organometallic compounds having M-C, M=C and MC bonds, and non-classical compounds in which the organic ligand is coordinated according to the  2 ,  3 ,  4 ,  5 and  6 -type, or is an element of electron-deficient structures - bridging CH 3 and C 6 H 6 groups, non-classical carbides (Rh 6 C(CO) 16, C(AuL) 5 +, C(AuL) 6 2+, etc.).

Among the specific mechanisms for classical -organometallic compounds, we note several mechanisms. Thus, 5 mechanisms of electrophilic substitution of the metal atom at the M-C bond have been established.

electrophilic substitution with nucleophilic assistance

AdEAddition-elimination

AdE(C) Addition to the C atom in sp 2 hybridization

AdE(M) Oxidative addition to metal

Nucleophilic substitution at the carbon atom in demetalation reactions of organometallic compounds occurs as a redox process:

Possible participation of an oxidizing agent in this stage

Such an oxidizing agent can be CuCl 2, p-benzoquinone, NO 3 – and other compounds. Here are two more elementary stages characteristic of RMX:

hydrogenolysis of the M-C bond

and homolysis of the M-C bond

An important rule that applies to all reactions of complex and organometallic compounds and is associated with the principle of least motion is Tolman's 16-18 electron shell rule (Section 2).

The main substitution reaction in aqueous solutions, the exchange of water molecules (22), has been studied for a large number of metal ions (Fig. 34). The exchange of water molecules in the coordination sphere of a metal ion with the bulk of water molecules present as a solvent occurs very quickly for most metals, and therefore the rate of such a reaction could be studied mainly by the relaxation method. The method involves disturbing the equilibrium of the system, for example by a sharp increase in temperature. Under new conditions (higher temperature), the system will no longer be in equilibrium. The rate of equilibrium is then measured. If you can change the temperature of the solution within 10 -8 sec, then you can measure the speed of a reaction that requires more than a period of time to complete 10 -8 sec.

It is also possible to measure the rate of substitution of coordinated water molecules in various metal ions with ligands SO 2-4, S 2 O 3 2-, EDTA, etc. (26). The speed of this reaction

depends on the concentration of the hydrated metal ion and does not depend on the concentration of the incoming ligand, which makes it possible to use the first-order equation (27) to describe the rate of these systems. In many cases, the rate of reaction (27) for a given metal ion does not depend on the nature of the incoming ligand (L), be it H 2 O molecules or SO 4 2-, S 2 O 3 2-, or EDTA ions.

This observation, coupled with the fact that the rate equation for this process does not include the concentration of the influent ligand, suggests that these reactions proceed by a mechanism in which the slow step involves breaking the bond between the metal ion and water. The resulting compound likely then quickly coordinates nearby ligands.

In Sect. 4 of this chapter it was stated that more highly charged hydrated metal ions, such as Al 3+ and Sc 3+, exchange water molecules more slowly than M 2+ and M + ions; This gives reason to assume that the breaking of bonds plays an important role in the stage that determines the rate of the entire process. The conclusions obtained in these studies are not conclusive, but they give reason to believe that S N 1 processes are important in substitution reactions of hydrated metal ions.

Probably the most studied complex compounds are cobalt(III) ammines. Their stability, ease of preparation, and slow reactions make them particularly suitable for kinetic studies. Since studies of these complexes were carried out exclusively in aqueous solutions, we should first consider the reactions of these complexes with solvent molecules - water. It was found that in general, ammonia or amine molecules coordinated by the Co(III) ion are so slowly replaced by water molecules that the replacement of ligands other than amines is usually considered.

The rate of reactions of type (28) was studied and found to be of first order relative to the cobalt complex (X is one of many possible anions).

Since in aqueous solutions the concentration of H 2 O is always approximately 55.5 M, then it is impossible to determine the effect of changing the concentration of water molecules on the reaction rate. Rate equations (29) and (30) for an aqueous solution are not experimentally distinguishable, since k is simply equal to k" = k". Therefore, it is impossible to tell from the reaction rate equation whether H2O will participate in the rate-determining step of the process. The answer to the question whether this reaction proceeds by the S N 2 mechanism with the replacement of the X ion by an H 2 O molecule or by the S N 1 mechanism, which first involves dissociation followed by the addition of an H 2 O molecule, must be obtained using other experimental data.

This problem can be solved by two types of experiments. Hydrolysis rate (replacement of one Cl - ion per water molecule) trance- + is approximately 10 3 times higher than the rate of hydrolysis 2+. An increase in the charge of the complex leads to strengthening of metal-ligand bonds and, consequently, to inhibition of the cleavage of these bonds. The attraction of incoming ligands and the facilitation of the substitution reaction should also be taken into account. Since a decrease in rate was found as the charge of the complex increased, in this case a dissociative process (S N 1) seems more likely.

Another method of proof is based on the study of the hydrolysis of a series of complexes similar trance- + . In these complexes, the ethylenediamine molecule is replaced by similar diamines, in which the hydrogen atoms at the carbon atom are replaced by CH 3 groups. Complexes containing substituted diamines react faster than the ethylenediamine complex. Replacing hydrogen atoms with CH 3 groups increases the volume of the ligand, making it more difficult for the metal atom to be attacked by another ligand. These steric hindrances slow down the reaction via the S N 2 mechanism. The presence of bulky ligands near the metal atom promotes the dissociative process, since the removal of one of the ligands reduces their accumulation at the metal atom. The observed increase in the rate of hydrolysis of complexes with bulky ligands is good evidence that the reaction proceeds according to the S N 1 mechanism.

So, as a result of numerous studies of Co(II) acidoamine complexes, it turned out that the replacement of acido groups with water molecules is a dissociative process in nature. The cobalt atom-ligand bond is extended to a certain critical value before water molecules begin to enter the complex. In complexes with a charge of 2+ and higher, breaking the cobalt-ligand bond is very difficult, and the entry of water molecules begins to play a more important role.

It was found that the replacement of the acido group (X -) in the cobalt(III) complex with a group other than the H2O molecule, (31) first goes through its replacement with a molecule

solvent - water, followed by replacing it with a new group Y (32).

Thus, in many reactions with cobalt(III) complexes, the rate of reaction (31) is equal to the rate of hydrolysis (28). Only the hydroxyl ion differs from the other reagents in its reactivity with Co(III) ammines. It reacts very quickly with amine complexes of cobalt(III) (about 10 6 times faster than water) according to the reaction type basic hydrolysis (33).

This reaction was found to be first order with respect to the substituting ligand OH - (34). The overall second order of the reaction and the unusually rapid progress of the reaction suggest that the OH - ion is an exceptionally effective nucleophilic reagent for Co(III) complexes and that the reaction proceeds via the S N 2 mechanism via the formation of an intermediate.

However, this property of OH - can also be explained by another mechanism [equations (35), (36)]. In reaction (35), the 2+ complex behaves like an acid (according to Brønsted), giving the + complex, which is amido-(containing)-compound - base corresponding to acid 2+.

The reaction then proceeds via the S N 1 mechanism (36) to form a five-coordinate intermediate, which further reacts with solvent molecules to produce the final reaction product (37). This reaction mechanism is consistent with the rate of a second-order reaction and corresponds to the S N 1 mechanism. Since the reaction in the rate-determining stage involves a base conjugate to the original complex - the acid, this mechanism is given the designation S N 1CB.

Determining which of these mechanisms best explains experimental observations is very difficult. However, there is compelling evidence to support the S N 1CB hypothesis. The best arguments in favor of this mechanism are as follows: octahedral Co(III) complexes generally react via the S N 1 dissociative mechanism, and there is no convincing argument why the OH - ion should mediate the S N 2 process. It has been established that the hydroxyl ion is a weak nucleophilic reagent in reactions with Pt(II), and therefore its unusual reactivity with Co(III) seems unreasonable. Reactions with cobalt(III) compounds in non-aqueous media provide excellent evidence for the formation of five-coordinate intermediates provided by the S N 1 SV mechanism.

The final proof is the fact that in the absence of N - H bonds in the Co(III) complex, it slowly reacts with OH - ions. This, of course, suggests that the acid-base properties of the complex are more important than the nucleophilic properties of OH for the rate of reaction." This reaction of the basic hydrolysis of ammine Co(III) complexes illustrates the fact that kinetic data can often be interpreted in more than one way, and In order to exclude one or another possible mechanism, it is necessary to carry out a rather delicate experiment.

Currently, substitution reactions of a large number of octahedral compounds have been studied. If we consider their reaction mechanisms, the most common is the dissociative process. This result is not unexpected since six ligands leave little space around the central atom for other groups to attach to it. There are only a few examples where the occurrence of a seven-coordinate intermediate has been demonstrated or the influence of an intervening ligand has been detected. Therefore, the S N 2 mechanism cannot be completely rejected as a possible pathway for substitution reactions in octahedral complexes.

Reactions of coordination compounds always occur in the coordination sphere of a metal with ligands bound in it. Therefore, it is obvious that in order for anything to happen at all, the ligands must be able to fall into this sphere. This can happen in two ways:

• a coordinatively unsaturated complex binds a new ligand
• in an already completed coordination sphere, one ligand is replaced by another.

We have already become familiar with the first method when we discussed coordination unsaturation and the 18-electron rule. We'll deal with the second one here.

Ligands of any type can be substituted in any combination

But usually there is an unspoken rule - the number of occupied coordination places does not change. In other words, the electron count does not change during substitution. Substitution of one type of ligand for another is quite possible and often occurs in reality. Let us only pay attention to the correct handling of charges when changing the L-ligand to the X-ligand and vice versa. If we forget about this, then the oxidation state of the metal will change, and the replacement of ligands is not an oxidation-reduction process (if you find or come up with an opposite example, let me know - it will be automatically credited right away, if I cannot prove that you were mistaken, and even in In this case, I guarantee a positive contribution to karma).

Substitution involving hapto ligands

With more complex ligands there are no more difficulties - you just need to remember a fairly obvious rule: the number of ligand sites (that is, the total number of ligands or X- or L-type ligand centers) is maintained. This follows directly from the conservation of electron counting. Here are self-evident examples.

Let's pay attention to the last example. The starting reagent for this reaction is iron dichloride FeCl 2 . Until recently, we would have said: “It’s just salt, what does coordination chemistry have to do with it?” But we will no longer allow ourselves such ignorance. In the chemistry of transition metals there are no “just salts”; any derivatives are coordination compounds, to which all considerations about electron counting, d-configuration, coordination saturation, etc. apply. Iron dichloride, as we are used to writing it, would turn out to be a Fe(2+) complex of type MX 2 with configuration d 6 and number of electrons 10. Not enough! Fine? After all, we have already figured out that ligands can be implicit. To make the reaction we need a solvent, and for such reactions it is most likely THF. The dissolution of the crystalline iron salt in THF occurs precisely because the donor solvent occupies free spaces, and the energy of this process compensates for the destruction of the crystal lattice. We would not be able to dissolve this “salt” in a solvent that does not provide the metal solvation services due to Lewis basicity. In this case, and in a million similar ones, solvation is simply a coordination interaction. Let us write, just for definiteness, the result of solvation in the form of the FeX 2 L 4 complex, in which two chlorine ions remain in the coordination sphere in the form of two X-ligands, although most likely they are also displaced by molecules of the donor solvent with formation of a charged complex FeL 6 2+. In this case it is not so important. Either way, we can safely assume that we have an 18-electron complex on both the left and the right.

Substitution, addition and dissociation of ligands are closely and inextricably linked

If we remember organic chemistry, then there were two mechanisms of substitution at a saturated carbon atom - SN1 and SN2. In the first, the substitution occurred in two stages: the old substituent first left, leaving a vacant orbital on the carbon atom, which was then occupied by a new substituent with a pair of electrons. The second mechanism assumed that leaving and coming were carried out simultaneously, in concert, and the process was one-stage.

In the chemistry of coordination compounds, it is quite possible to imagine something similar. But a third possibility appears, which the saturated carbon atom did not have - first we attach a new ligand, then we detach the old one. It immediately becomes clear that this third option is hardly possible if the complex already has 18 electrons and is coordination saturated. But it is quite possible if the number of electrons is 16 or less, that is, the complex is unsaturated. Let us immediately recall the obvious analogy from organic chemistry - nucleophilic substitution at an unsaturated carbon atom (in an aromatic ring or at a carbonyl carbon) also occurs first as the addition of a new nucleophile, and then the elimination of the old one.

So, if we have 18 electrons, then the substitution occurs as an abstraction-addition (fans of “smart” words use the term dissociative-associative or simply dissociative mechanism). Another way would require expanding the coordination sphere to a count of 20 electrons. This is not absolutely impossible, and such options are sometimes even considered, but it is definitely very unprofitable and every time in case of suspicion of such a path, very significant evidence is required. In most of these stories, the researchers eventually concluded that they had overlooked or missed something, and the associative mechanism was rejected. So, if the original complex has 18 electrons, then first one ligand must leave, then a new one must take its place, for example:

If we want to introduce a hapto-ligand occupying several sites into the coordination sphere, then we must first vacate them all. As a rule, this occurs only under fairly severe conditions, for example, in order to replace three carbonyls in chromium carbonyl with η 6 -benzene, the mixture is heated under pressure for many hours, releasing the released carbon monoxide from time to time. Although the diagram depicts the dissociation of three ligands with the formation of a very unsaturated complex with 12 electrons, in reality the reaction most likely occurs in stages, leaving one carbonyl at a time, and benzene entering the sphere, gradually increasing hapticity, through the stages minus CO - digapto - minus one more CO - tetrahapto - minus one more CO - hexagapto, so that less than 16 electrons are not obtained.

So, if we have a complex with 16 electrons or less, then the replacement of the ligand most likely occurs as an addition-elimination (for those who like deep-sounding words: associative-dissociative or simply associative): the new ligand first comes, then the old one leaves. Two obvious questions arise: why does the old ligand leave, because 18 electrons are very good, and why not do the opposite in this case, as in 18-electron complexes. The first question is easy to answer: each metal has its own habits, and some metals, especially late ones, with almost completely filled d-shells, prefer the 16-electron count and the corresponding structural types, and therefore throw out the extra ligand, returning to their favorite configuration. Sometimes the spatial factor also interferes with the matter; the existing ligands are large and the additional one feels like a bus passenger at rush hour. It’s easier to get off and take a walk than to suffer like this. However, you can push out another passenger, let him take a walk, and we will go. The second question is also simple - in this case, the dissociative mechanism would first have to give a 14-electron complex, and this is rarely beneficial.

Here's an example. For variety, let's replace the X-ligand with an L-ligand, and we won't get confused about oxidation states and charges. Once again: upon substitution, the oxidation state does not change, and if the X-ligand has left, then the loss must be compensated for by the charge on the metal. If we forget about this, then the oxidation number would decrease by 1, but this is incorrect.

And one more strange thing. A metal-pyridine bond was formed due to the lone pair on nitrogen. In organic chemistry, in this case we would definitely show a plus on the pyridine nitrogen (for example, upon protonation or formation of a quaternary salt), but we never do this in coordination chemistry with either pyridine or any other L-ligands. This is terribly annoying for everyone who is accustomed to the strict and unambiguous system of drawing structures in organic chemistry, but you will have to get used to it, it is not so difficult.

But there is no exact analogue of SN2 in the chemistry of coordination compounds; there is a distant one, but it is relatively rare and we do not really need it.

Stable and labile ligands

We could not talk about the mechanisms of ligand substitution at all if not for one extremely important circumstance that we will use a lot: ligand substitution, be it associative or dissociative, necessarily presupposes the dissociation of the old ligand. And it is very important for us to know which ligands leave easily and which leave poorly, preferring to remain in the coordination sphere of the metal.

As we will soon see, in any reaction some of the ligands remain in the coordination sphere and do not change. Such ligands are usually called spectator ligands (if you don’t want such simple, “unscientific” words, use the English word spectator in the local transcription spectator, spectator ligand, but, I beg you, not spectator - this is unbearable!). And some directly participate in the reaction, turning into reaction products. Such ligands are called actors (not actors!), that is, active ones. It is quite clear that ligand-actors need to be easily introduced and removed into the coordination sphere of the metal, otherwise the reaction will simply get stuck. But it is better to leave spectator ligands in the coordination sphere for many reasons, but at least for such a banal one as the need to avoid unnecessary fuss around the metal. It is better that only ligand actors and in the required quantities can participate in the desired process. If there are more available coordination sites than necessary, extra ligand actors may sit on them, and even those that will participate in side reactions, reducing the yield of the target product and selectivity. In addition, spectator ligands almost always perform many important functions, for example, they ensure the solubility of complexes, stabilize the correct valence state of the metal, especially if it is not quite ordinary, help individual stages, provide stereoselectivity, etc. We won’t decipher it yet, because we will discuss all this in detail when we get to specific reactions.

It turns out that some of the ligands in the coordination sphere should be tightly bound and not prone to dissociation and replacement by other ligands. Such ligands are usually called coordinationally stable . Or simply stable, if it is clear from the context that we are talking about the strength of the bonds of the ligands, and not about their own thermodynamic stability, which does not concern us at all.

And ligands that easily and willingly enter and leave, and are always ready to give way to others, are called coordination labile , or simply labile, and here, fortunately, there are no ambiguities.

Cyclobutadiene as a ligand

This is probably the most striking example of the fact that in the coordination sphere a very unstable molecule can become an excellent ligand, and by definition, coordination stable, if only because if it dares to leave the warm and cozy sphere outside, nothing good will await it (at the cost of the output will be precisely the energy of anti-aromatic destabilization).

Cyclobutadiene and its derivatives are the best known examples of anti-aromaticity. These molecules exist only at low temperatures, and in a highly distorted form - in order to get as far as possible from antiaromaticity, the cycle is distorted into an elongated rectangle, removing delocalization and maximally weakening the conjugation of double bonds (this is otherwise called the Jahn-Teller effect of the 2nd kind: degenerate system, and cyclobutadiene square is a degenerate biradical, remember the Frost circle - it is distorted and reduces symmetry to remove the degeneracy).

But in complexes, cyclobutadiene and substituted cyclobutadienes are excellent tetrahapto ligands, and the geometry of such ligands is exactly a square, with identical bond lengths. How and why this happens is a separate story, and not nearly as obvious as it is often made out to be.

Coordination labile ligands

You need to understand that there is no reinforced concrete fence with barbed wire and security towers between the areas of labile and stable ligands. Firstly, it depends on the metal, and LMKO works well in this context. For example, late transition metals prefer soft ligands, while early transition metals prefer hard ones. Let's say, iodide holds very tightly to the d 8 atoms of palladium or platinum, but rarely enters the coordination sphere of titanium or zirconium in the d 0 configuration at all. But in many metal complexes with less pronounced features, iodide manifests itself as a completely labile ligand, easily giving way to others.

Other things being equal:

• L-ligands are usually more labile than X-ligands;
• the lability of X-ligands is determined by the hardness/softness and nature of the metal;
• “implicit” ligands are very labile: solvents and bridges in dimers and clusters, so much so that their presence in the coordination sphere is often completely neglected and structures without them are drawn with a formally unsaturated coordination sphere;
• Dihapto ligands, for example alkenes and alkynes, behave like typical L-ligands: they are usually quite labile;
• ligands with greater hapticity are rarely labile, but if a polyhapto ligand can change the mode of binding to mono-hapto, it becomes more labile, for example, η 3 -allyls behave this way;
• chelate ligands forming 5- and 6-membered chelate rings are stable, and chelates with fewer or more ring atoms are labile, at least at one center (the chelate ring opens and the ligand remains hanging as a simple one). This is how acetate behaves, for example;

Coordinatively stable ligands

Let's repeat it all again, only on the other side

In the coordination sphere of metals, the following are generally preserved (coordinationally stable):

• 5- and 6-membered chelators;
• polyhapto-ligands: in order to knock cyclopentadienyls or benzene (arenes) out of the coordination sphere, you have to use all sorts of special techniques - they just don’t come out, often withstanding even prolonged heating;
• ligands bound to a metal with a high proportion of π-donor effect (back-donation);
• soft ligands for late transition metals;
• “last” ligand in the coordination sphere.

The last condition looks strange, but imagine a complex that has many different ligands, among which there are no absolutely stable ones (no chelators or polyhapto-ligands). Then, in reactions, the ligands will change, relatively speaking, in order of relative lability. The least labile and the last to remain. This trick occurs, for example, when we use palladium phosphine complexes. Phosphines are relatively stable ligands, but when there are many of them, and the metal is rich in electrons (d 8, d 10), they give way, one after another, to actor ligands. But the last phosphine ligand usually remains in the coordination sphere, and this is very good from the point of view of the reactions in which these complexes participate. We will return to this important issue later. Here is a fairly typical example when only one, “last” phosphine remains from the initial coordination sphere of the palladium phosphine complex in the Heck reaction. This example brings us very close to the most important concept in the reactions of transition metal complexes - the concept of ligand control. We'll discuss it later.

Remetallation

When replacing some ligands with others, it is important not to overdo the reactivity of the incoming ligand. When we are dealing with reactions of organic molecules, it is important for us to deliver exactly one molecule of each reactant into the coordination sphere. If two molecules enter instead of one, there is a high probability of side reactions involving two identical ligands. A loss of reactivity is also possible due to saturation of the coordination sphere and the impossibility of introducing into it other ligands necessary for the expected process. This problem especially often arises when strong anionic nucleophiles, for example, carbanions, are introduced into the coordination sphere. To avoid this, less reactive derivatives are used, in which, instead of the alkali metal cation, which determines the high ionicity of the bond, less electropositive metals and metalloids (zinc, tin, boron, silicon, etc.) are used, forming covalent bonds with the nucleophilic part . Reactions of such derivatives with transition metal derivatives produce ligand substitution products, in principle just as if the nucleophile were in anionic form, but due to reduced nucleophilicity with less complications and no side reactions.

Such ligand substitution reactions are usually called transmetallation to emphasize the obvious fact that the nucleophile seems to change metals - more electropositive to less electropositive. This name, therefore, contains an element of unpleasant schizophrenia - we seemed to have already agreed that we would look at all reactions from the point of view of a transition metal, but suddenly we lost it again and look at this reaction and only this reaction from the point of view of a nucleophile. You will have to be patient, this is how the terminology has developed and is accepted. In fact, this word goes back to the early chemistry of organometallic compounds and to the fact that the action of lithium or organomagnesium compounds on halides of various metals and metalloids is one of the main methods for the synthesis of all organometallic compounds, primarily intransition ones, and the reaction that we are now considering in chemistry of coordination compounds of transition metals is simply a generalization of the ancient method of organometallic chemistry from which it all grew.

How does transmetallation occur?

Remetallation is both similar to conventional substitution and not similar. It looks like - if we consider a non-transition organometallic reagent to be simply a carbanion with a counterion, then the carbon-non-transition metal bond is ionic. But this idea seems to be true only for the most electropositive metals - magnesium. But already for zinc and tin this idea is very far from the truth.

Therefore, two σ bonds and four atoms at their ends enter into the reaction. As a result, two new σ bonds are formed and four atoms bond to each other in a different order. Most likely, all this occurs simultaneously in a four-member transition state, and the reaction itself has a concerted character, like so many other reactions of transition metals. The abundance of electrons and orbitals for literally all tastes and all types of symmetries makes transition metals capable of simultaneously maintaining bonds in transition states with several atoms.

In the case of remetallation, we obtain a special case of a very general process, which is simply called σ-bond metathesis. Do not confuse them only with true metathesis of olefins and acetylenes, which are full-fledged catalytic reactions with their own mechanisms. In this case we are talking about the mechanism of transmetallation or another process in which something similar occurs.