Tải bản đầy đủ - 0 (trang)
4 Nucleophilic Abstraction in Hydrides, Alkyls, and Acyls

4 Nucleophilic Abstraction in Hydrides, Alkyls, and Acyls

Tải bản đầy đủ - 0trang



Addition and Abstraction


Alkyls and Acyls

Alkyl groups can be exchanged between metals, typically with inversion

at carbon. This transmetalation reaction provides a route for the racemization of a metal alkyl during the early stages of an oxidative addition reaction, while there is still some of the low-valent metal left in the

reaction mixture. In Eq. 8.28, exchange of a (CR3)+ fragment between

the metals turns the Pd(0) partner into Pd(II), and the Pd(II) into

Pd(0). The stereochemical outcome of an OA can be clouded by

exchange reactions such as these.


Acyls undergo abstraction by nucleophiles in the last steps of Eqs.

8.22 and 8.23. As in the abstraction of Eq. 8.28, the reaction goes with

reduction of the metal by two units, so a Pd(II) acyl is ideal because

the Pd(0) state is easily accessible.

The recurrence of Pd(II) in this chapter is no accident—it has a very

high tendency to encourage nucleophilic attack at the ligands in its

complexes. Being on the far right-hand side of the d block, it is very

electronegative (Pauling electronegativity: 2.2), and its d orbitals are

very stable. This means that polyene-to-metal electron donation is more

important than metal-dπ-to-polyene-π* back donation, and so the

polyene is left with a net positive charge.


In common with 2e nucleophiles, 0e electrophiles, such as H+ or Me+,

can attack a ligand. Unlike nucleophiles, however, they can also attack

the M–L bond or the metal itself because, as zero electron reagents,

wherever they attack, they do not alter the electron count of the

complex. The resulting mechanistic complexity and unreliable selectivity makes electrophilic attack far less controllable and less useful than

nucleophilic attack. Polysubstitution is also more common in the electrophilic case.14

Electrophilic Addition and Abstraction


Addition to the Metal

Oxidative addition by the SN2 or ionic mechanisms involves two steps:

initial electrophilic addition to the metal (Eq. 8.29 and Sections 6.3 and

6.5), followed by substitution.





Without the second step, the reaction becomes a pure electrophilic

addition. An example is the reaction of the highly nucleophilic Co(I)

anion, [Co(dmg)2py]−, with an alkyl triflate, a reaction known to go with

inversion at carbon (Eq. 8.30). Protonation of metal complexes to give

metal hydrides is also very common (Eq. 3.28 and Eq. 3.29).

The addition of any zero-electron ligand to the metal is also an electrophilic addition: AlMe3, BF3, HgCl2, Cu+, and even η1-CO2, when it

binds via carbon, all act in this way. Each of these reagents has an empty

orbital by which it can accept a lone pair from the metal.

Addition to a Metal–Ligand Bond

Protonation reactions are common—for example, in Eq. 8.31, protonation of LnM–H can give a dihydrogen complex [LnM–(H2)]+.15 Early

metal alkyls, such as Cp2TiMe2, are readily cleaved by acid to liberate

the alkane via a transient alkane complex.


Protonation of the alkene complex shown below can occur by two

simultaneous paths: (1) direct protonation at the metal and (2) initial

protonation at the alkene followed by β elimination. Path 2 leads to

incorporation of label from DCl into the alkene ligands of the resulting

pentagonal bipyramidal hydride complex.16


Addition and Abstraction



Addition to Ligand

Simple addition to the ligand occurs in protonation of Cp2Ni, as

shown by the exo attack and lack of scrambling of the deuterium



Unlike nucleophiles, where exo attack is the rule, an endo addition

is also possible for electrophiles via attack at the metal, followed by

transfer to the endo face of the ligand, particularly favored for soft

electrophiles, for example, Hg(OAc)2. Exo-proton abstraction by OAc−

completes the sequence (Eq. 8.34).

The hard electrophile CH3CO+ gives exo attack at the ligand in Eq.

8.35. The preference for exo-proton abstraction means that an endodeuterium has to be transferred to the endo position of the other ring.

This leads to loss of the resulting exo-proton, so that all five D atoms

are retained by the complex.17

Single-Electron Transfer and Radical Reactions




Abstraction of Alkyl Groups

Electrophilic metal ions, notably Hg2+, can cleave an M–alkyl bond.

Two main pathways are seen: (i) attack at the α carbon of the alkyl with

inversion at carbon (Eq. 8.36) and (ii) attack at the metal or at the M–C

bond with retention (Eq. 8.37). The difference has been ascribed to the

greater basicity of the metal in the CpFe case. The unpredictable stereochemistry again makes the reaction less useful.







It is often difficult to differentiate between a true electrophilic abstraction or addition, a one-step process in which a pair of electrons is

implicated (Eq. 8.36 and Eq. 37), from a two-step process involving


Addition and Abstraction

single-electron transfer (SET) from the metal to E+ going via the

radical intermediate, E• (Eq. 8.38 and Eq. 8.39).18 First row metals

prefer 1e to 2e OS changes (Co(I), (II), (III) versus Ir(I), (III), (V)),

and are therefore more likely to give radical pathways. For example,

halogens, X2, give electrophilic cleavage of M–R to form RX. One

common mechanism involves SET oxidation of the metal, which

increases the electrophilic character of the alkyl and generates halide

ion, so that, paradoxically, it is nucleophilic abstraction of the alkyl

group by halide ion that leads to the final products. Co(III) alkyls are

known to behave in this way, and the intermediate Co(IV) species,

formed via Ce(IV) oxidation, are stable enough to be detected by

EPR at −50°C (Eq. 8.38, R  =  n-hexyl). Addition of Cl- leads to the

nucleophilic abstraction of the alkyl with inversion.



Nucleophiles can also give SET reactions, for example, [Cp*Mo­

(CO)3(PMe3)]+ reacts with LiAlH4 to give paramagnetic [Cp*Mo(CO)3­

(PMe3)], observed by EPR. Loss of CO, easy in this 19e species, leads

to Cp*Mo(CO)2(PMe3), which abstracts H•, probably from the THF

solvent, to give the final product, Cp*MoH(CO)2(PMe3).

Radical traps, such as galvinoxyl, TEMPO, and DPPH (Q•), are

sometimes used as a test for the presence of radicals, R•, in solution;

in such a case, the adduct Q–R is expected as product. Unfortunately,

this procedure can be misleading in organometallic chemistry because

typical Q• abstract H from some palladium hydrides at rates competitive with those of typical organometallic reactions; [PdHCl(PPh3)2]

reacts in this way but [PdH(PEt3)3]BPh4 is stable.19

Radical abstraction from a ligand is also possible. For example, in

Eq. 8.40, an alkyl radical abstracts a hydrogen atom from a coordinated

water.20 In fact, the process is better seen as a concerted H+ transfer

from the water and an e− transfer from the Ti(III) center.




• Nucleophilic addition is more predictable (Section 8.3) than the

other pathways considered.

• Nucleophilic attack at a ligand is favored by weak and electrophilic attack by strong back bonding to that ligand.


  1.  B. Jacques, J. P. Tranchier, F. Rose-Munch, E. Rose, G. R. Stephenson, and

C. Guyard-Duhayon, Organometallics, 23, 184, 2004.

  2.  S. E. Gibson and H. Ibrahim, Chem. Commun., 2002, 2465.

  3.  M. A. Todd, M. L. Grachan, M. Sabat, W. H. Myers, and W. D. Harman,

Organometallics, 25, 3948, 2006.

  4. Y. H. Kim, T. H. Kim, N. Y. Kim, E. S. Cho, B. Y. Lee, D. M. Shin, and Y. K.

Chung, Organometallics, 22, 1503, 2003.

  5.  J. Barluenga, K. Muniz, M. Tomas, A. Ballesteros, and S. Garcia-Granda,

Organometallics, 22, 1756, 2003.

  6.  L. C. Song, L. X. Wang, G. J. Jia, Q. L. Li, and J. B. Ming, Organometallics,

31, 5081, 2012.

  7. A. G. Tskhovrebov, K. V. Luzyanin, F. M. Dogushin, M. F. C. G. da Silva,

A. J. L. Pombeiro, and V. Y. Kukushkin, Organometallics, 30, 3362, 2011.

  8.  S. G. Davies, M. L. H. Green, and D. M. P. Mingos, Tetrahedron, 34, 3047,


  9. D. Harakat, J. Muzart, and J. Le Bras, RSC Advances, 2, 3094, 2012 and

references cited; V. Imandi, S. Kunnikuruvan, and N. N. Nair, Chem. Eur.

J., 19, 4724, 2013.

10.  J. M. Takacs and X. T. Jiang, Curr. Org. Chem., 7, 369, 2003.

11.  A. Comas-Vives, A. Stirling, A. Lledos, and G. Ujaque, Chem. Eur. J., 16,

8738, 2010.

12.  G. Dong, P. Teo, Z. K. Wickens, and R. H. Grubbs, Science, 333, 1609, 2011.

13.  T. Tachinami, T. Nishimura, R. Ushimaru, R. Noyori, and H. Naka, J. Am.

Chem. Soc., 135, 50, 2013.

14.  G. R. Clark, P. M. Johns, W. R. Roper, T. Sohnel, and L. J. Wright, Organometallics, 30, 129, 2011.


Addition and Abstraction

15.  A. Miedaner, J. W. Raebiger, C. J. Curtis, S. M. Miller, and D. L. DuBois,

Organometallics, 23, 2670, 2004.

16.  R. A. Henderson, Angew. Chem. -Int. Ed. Engl., 35, 947, 1996.

17. M. J. Mayor-Lopez, J. Weber, B. Mannfors, and A. F. Cunningham, Organometallics, 17, 4983, 1998.

18. K. M. Smith, Organometallics, 24, 778, 2005.

19.  A. C. Albeniz, P. Espinet, R. Lopez-Fernandez, and A. Sen, J. Am. Chem.

Soc., 124, 11278, 2002.

20.  J. Jin and M. Newcomb, J. Org. Chem., 73, 7901, 2008.


8.1. Where would a hydride ion attack each of the following?

8.2. Predict the outcome of the reaction of CpFe(PPh3)(CO)Me with

each of the following: HCl, Cl2, HgCl2, and HBF4/THF.

8.3. Explain the outcome of the reaction shown below:



8.4. [CpCo(dppe)(CO)]2+ (A) reacts with 1° alcohols, ROH, to give

[CpCo(dppe)(COOR)]+, a reaction known for very few CO complexes. The ν(CO) frequency for A is 2100 cm−1, extremely high

for a CO complex. Br− does not usually displace CO from a carbonyl complex, but it does so with A. Why is A so reactive?

8.5. Nucleophilic addition of MeO− to free PhCl is negligibly slow

under conditions for which the reaction with (η6-C6H5Cl)Cr(CO)3

is fast. What product would you expect, and why is the reaction

accelerated by coordination?

8.6. Given a stereochemically defined starting material (either erythro

or threo), what stereochemistry would you expect for the products of the following electrophilic abstraction reaction:





Let us say that for a related 16e complex LnM(CHDCHDCMe3)

gave precisely the same products, but of opposite stereochemistries. What mechanism would you suspect for the reaction?

8.7.You are trying to make a methane complex LnM(η1-H−CH3)+

(8.17), by protonation of a methyl complex LnMMe with an acid

HA. Identify three things that might go wrong and suggest ways

to guard against each.

8.8. (cod)PtCl2 reacts with MeOH/NaOAc to give a species

[{C8H12(OMe)}PtCl]2. This in turn reacts with PR3 to give

1-methoxycyclooctadiene (8.18) and PtHCl(PR3)2. How do you

think this might go?

8.9. [CpFe(CO)(PPh3)(MeC≡CMe)]+ reacts with (i) LiMe2Cu (a

source of Me−) and (ii) I2 to give compound 8.19; explain this

reaction. What product do you think might be formed from


8.10.  Equation 8.24 and Equation 8.26 involve substrates with different

oxidation states of carbon in the substrate hydrocarbon, one

starts from ethylene, the other from acetylene. Explain how both

reactions can give the same product, CH3CHO, when both are

hydration reactions where we do not expect the oxidation state

of carbon to change.



The catalysis of organic reactions1 is one of the most important applications of organometallic chemistry and has been a significant factor in

the rapid development of the field as a whole. Organometallic catalysts

now have numerous applications in the pharmaceutical,2 fine chemical,

and commodity chemical industries and are beginning to contribute to

the rising topics of energy and green chemistry. By bringing about a

reaction at lower temperature, a catalyst can save energy input and, by

improving selectivity, can minimize product separation problems and

waste formation. With growing regulatory pressure to market drugs in

enantiopure form, asymmetric catalysis has come to the fore as a practical way to make such products on a large scale from racemic or achiral



A catalytic cycle consists of a set of reactions that occurs only in the

presence of a catalyst and that leads to product formation from reactants, or substrates. The catalyst can mediate an indefinite series of

The Organometallic Chemistry of the Transition Metals, Sixth Edition.

Robert H. Crabtree.

© 2014 John Wiley & Sons, Inc. Published 2014 by John Wiley & Sons, Inc.


Catalytic Cycles


cycles and is thus only needed in substoichiometric amount relative to

reactants or products. The catalyst loading is typically from 1 ppm to

1% relative to reactants, meaning that the number of cycles initiated

by each molecule of catalyst runs from 106 to 102, respectively. Catalysis

can be useful either by speeding up a reaction or modifying its selectivity, or both. The same reactants can give quite different products

depending on the catalyst: ethylene oxidation with O2, for example, can

give the epoxide or acetaldehyde.

Of interest here are soluble complexes, or homogeneous catalysts,

as opposed to insoluble materials, or heterogeneous catalysts, so

named because the catalyst and substrates are in the same phase

only in the first case. Some reactions, such as hydrogenation, are

amenable to both types of catalysis, but others are currently limited

to one or the other, for example, O2 oxidation of ethylene to the

epoxide over a heterogeneous Ag catalyst or Wacker air oxidation

of ethylene to acetaldehyde with homogeneous Pd(II) catalysts.

Homogeneous catalysis extends far beyond organometallics to cover

acid or base catalysis, organocatalysis, and coordination catalysis,

such as H2O2 decomposition by Fe2+. Electrocatalysis is also a rising


Catalytic mechanisms are easier to study in homogeneous cases,

where powerful methods such as NMR can assign structures and

follow reaction kinetics. Homogeneous catalysts are at a disadvantage,

however, in being difficult to separate from the product. Sometimes,

this requires special techniques, but in polymer synthesis, the catalyst

still remains in the final product. Homogeneous catalysts are also heterogenized by covalently grafting onto solid supports to aid separation.

Although now technically heterogeneous, the catalyst often retains the

characteristic reactivity of the homogeneous form. We can distinguish

between homogeneous or heterogenized homogeneous catalysts that

have a single type of active site, or a small number of them (homotopic),

from metal and metal oxide surfaces that can have a cocktail of sites

(heterotopic). The first case tends to give higher selectivity than the

second. Homogeneous catalysts are also amenable to tuning by change

of ligand.


A homogeneous precursor can give rise to a homogeneous catalyst;

however, it can also decompose to give catalytically active solid material. A particularly dangerous form of decomposition gives rise to suspended nanoparticles, of typical diameter 10–1000 Å. These can mislead

by masquerading as homogeneous catalysts. Many early “homogeneous”


Homogeneous Catalysis

catalysts, formed by reduction of metal salts in polar solvents, may well

have given active nanoparticles, and even today, ambiguities can easily

arise.4 One might think an asymmetric catalyst has to be homogeneous,

but in one recent case, an impressive level of asymmetric induction

(90% e.e.) was achieved by modification of a nanoparticle surface with

an asymmetric “ligand.”5 Two catalytic reactions not normally seen for

true homogeneous catalysts can be considered a “red flag”: nitrobenzene reduction and arene hydrogenation. Careful work in homogeneous catalysis should include tests for heterogeneity; sometimes, both

types even occur together.6 The possibility that the true catalyst is very

different from the complex originally introduced into the reaction

mixture has led authors to term the original complex the catalyst precursor (or precatalyst).


Before trying to find a catalyst for a given reaction, we need to check

that the reaction itself has favorable thermodynamics, as is the

case for the alkene isomerization of Eq. 9.1, for example. If a reaction is disfavored, as in splitting H2O to H2 and O2, then no catalyst,

however efficient, can bring it about without energy input. To get

round this problem, we might couple an unfavorable reaction to a

strongly favorable process or provide energy in the form of photons,

as in photosynthesis, or a voltage, as in electrolysis. In the absence

of these effects, the catalyst only increases the rate but does not

change the position of equilibrium, decided by the thermodynamics

of substrates, S, versus products, P. In the energy diagram of Figure

9.1a, for example, S is slightly less stable than P, so the reaction

favors P. For 9.1 → 9.2, the additional conjugation in 9.2 is sufficient

to make the reaction favorable. Normally, the substrate binds to

form a substrate–catalyst complex, M.S (Fig. 9.1). Stronger M–S

binding might be thought to be better, but this is not always so. If

binding is too strong, M.S will be too stable, and the activation

energy to get to “M.TS” may be just as large as it was in going from

S to TS in the uncatalyzed reaction, so no rate acceleration would

be achieved. Nor can S bind too weakly because it would then be

excluded from the metal and fail to be activated. The product P,

initially formed as M.P, must be the least strongly bound of all so

that S can displace P to give back M.S and start a new cycle. Many

of these ideas also apply to enzymes.7


Tài liệu bạn tìm kiếm đã sẵn sàng tải về

4 Nucleophilic Abstraction in Hydrides, Alkyls, and Acyls

Tải bản đầy đủ ngay(0 tr)