Download PART 3 Principles and Applications of Organometallics in Catalysis

PART 3
Principles and Applications of
Organometallics in Catalysis
1. Oxidative Addition and Reductive Elimination
These processes are of great importance in synthesis and catalysis using
organometallics.
A
oxidative addition
L nM
+
A B
reductive elimination
∆ OS=+2
∆ CN=+2
L nM
B
∆ VE=+2
18 VE
16 VE
In the OA we break the (2-electron) A—B bond and form two (2-electron) bonds to
the metal, i.e. M—A and M—B. Hence, there are some requirements for OA to
occur.
(i) A vacant coordination site must be available on the metal complex, and
(ii) the metal must have an accessible oxidation state 2 units higher.
Conversely, for RE to occur, there must be a stable oxidation state 2 units lower than
that in the starting complex. The resulting complex will necessarily have a free
coordination site which may make it reactive or perhaps unstable.
The position of the OA/RE equilibrium is determined by the overall thermodynamics
of the system, i.e. the relative stabilities of the two oxidation states, the relative bond
strengths of A—B, M—A and M—B, and entropy terms.
2. Insertion and Elimination
X
Once we have arranged
our ligands by addition
M
A B
(either oxidative or
simple) we can combine
18 V.E.
and transform them
within the coordination
sphere by the process of
insertion and its reverse
X
elimination. The two
A
M
main types of insertion
B
are shown here. In
principle, the reactions are
18 V.E.
reversible, but usually
only one direction is
observed.
1,1-migratory insertion
X
M
A
B
16 V.E.
X
1,2-migratory insertion
A
M
B
16 V.E.
1
3. Catalysis; basic principles
what catalysts do and don't do
The effect of a catalyst is to change the rate of conversion of a substrate into products,
but they do not change the position of an equilibrium. The thermodynamics of the
reaction concerned have to be favourable at the outset; catalysts can't perform the
miracle of pushing a reaction up the thermodynamic hill!
The catalyst is not consumed by the reaction (although it may be changed subtly) but
goes on to mediate the same reaction successively a number of times.
Sometimes, the thermal reaction and the catalysed reaction give different products.
This is because the catalyst has accelerated a reaction that is normally kinetically
unfavourable. Generally, we will be interested in catalysts that increase the rate of
reaction, but it should be noted that inhibitors which slow down certain reactions are
also of great commercial and academic importance (e.g. flame retardents, corrosion
inhibitors, antiknock additives in fuel, antioxidants in food).
"catalyst" or "reagent"?
In catalysis by organometallic systems, the catalyst brings together substrates within
the coordination sphere. This means that the catalyst must have a free coordination
site, or at least be able to free-up a coordination site by ligand dissociation or
isomerisation. The substrates are activated in some way by the process of coordination
and a reaction takes place. Often the most difficult step is that of decomplexation of
the transformed substrate. Indeed, although many "organic" processes can be
mediated by organometallics, relatively few are truly catalytic. This is largely because
the transformed fragment refuses to de-complex from the metal centre in order to
free-up a coordination site for the next substrate molecule. Hence the distinction
between the organometallic as a catalyst or reagent.
heterogeneous catalysis
The "vacant coordination site" is located at a phase boundary. Only some (often very
few) of the surface atoms are active catalysts.
Advantages
• The catalyst can be separated from the
reaction mixture readily (it is not in the
same phase)
• Amenable to continuous processes (bed of
catalyst in a tube).
2
Disadvantages
• Low specificity
• High reaction temperatures
• Mechanisms hard to determine
homogeneous catalysis
The catalyst and substrate are in the same phase (usually liquid but sometimes gas).
Disadvantages
• Separation of catalyst from the
reaction mixture often problematic.
• Less amenable to continuous
processes.
Advantages
• High specificity (tailoring)
• Low reaction temperatures
• Mechanisms can be studied more
easily
heterogenised homogeneous catalysts
Many attempts have been made to marry the advantages of both homo- and
heterogeneous processes, e.g. by attaching a homogeneous catalyst on a polymer
support or by performing the reaction in a two-phase liquid/liquid system using a
water soluble catalyst.
H
P
Ar3P
Rh
PAr3
Ar3P
CO
P
Ph
Ph
Ph3P
Rh
Cl
Ph
P
Ar=
Ph
Cl
Rh
Ph3P
PPh3
PPh3
SO3-Na+
A water soluble hydroformylation
catalyst (see Section 7)
Polymer supported Wilkinson's
Catalyst (see Section 4)
3
4. Catalytic Hydrogenation of Alkenes
Hydrogenation catalysts add H2 to the double bond of an alkene to give alkanes. We
will look at two types, defined according to how they activate the H2 molecule.
(i) Oxidative addition Wilkinson's Catalyst [RhCl(PPh3)3] is the best known of this
or any type of hydrogenation catalyst. The activation of H2 is accomplished by its
oxidative addition to the coordinatively unsaturated, 16 VE Rh(I) centre
An outline mechanism for the hydrogenation of alkenes by Wilkinson's catalyst (L =
PPh3):H
L
L
H2
Cl
Rh
L
Cl
Rh
H
Identify the reaction
L
L
type in each step,
L
count electrons and
calculate oxidation
states in all the
-L
irrev.
intermediates.
H
H
H
Cl
L
H
H
Rh
Cl
L
Rh
H
L
L
L
L
What does the reaction below tell us about the regio- and stereoselectivity of
Wilkinson's Catalyst?
O
O
D
D
D2, [Rh(PPh3)3Cl]
O
O
(ii) Heterolytic H2 activation
4
[RuCl2(PPh3)3] is believed to activate H2 by a heterolytic process.
Cl
H2
RuCl(PPh3)3
RuCl2(PPh3)3
-HCl
RuHCl(PPh3)3
H
H
How might we accelerate this process?
Note that we have not drawn any Ru(IV) species. Why is this?
The catalytic cycle is different from that in Wilkinson's catalyst in that there is not an
OA/RE cycle. Instead, the Ru(II) centre prefers to mediate σ-bond metathesis and
insertion steps:
Identify the
reaction type in
each step,
count electrons
and calculate
oxidation states
in all the
R
intermediates.
H
H
R
RuCl(PPh3)3
HRuCl(PPh3)3
R
R
H
H
R
RuCl(PPh3)3
RuCl(PPh3)3
H
H2
H
5
H
5. Fischer-Tropsch Chemistry
The "reformation" of natural gas (see below) provides a C1 feedstock called variously
"water-gas" or "synthesis gas" or plain "syngas" which is used in the production of a
variety of chemicals. Synthesis gas can be converted to methanol (a useful feedstocksee later) or long-chain alkanes and alcohols via the Fischer Tropsch reaction.
CO + 3H2
heat
CH4 + H2O
het. cat.
CH3OH
Synthesis Gas
homogeneous F.-T.
Fischer-Tropsch
CH3(CH2)nCH3
+
CH3(CH2)nCH2OH
+
H2 O
HOCH2CH2OH
A wide variety of heterogeneous catalysts are used, but perhaps the most common is
Fe or Fe oxides. Some "models" of what is occurring at the surface of the catalyst in
this reaction have been developed in homogeneous systems, but it is hard to relate
these systems to the real heterogeneous catalyst.
See section 28-3 of Cotton & Wilkinson 5th Edition.
6. Carbonylation of Methanol
Another important industrial process involving CO is the carbonylation of methanol to
give acetic acid. The feedstocks may all be produced from coal or methane by virtue
of a combination of Syngas and Fischer-Tropsch technology. Spectroscopic evidence
for all the key intermediates has been collected. The iodomethane promoter is key to
the success of this reaction.
[Rh]
CH3 OH + CO
MeI
6
CH3COOH
The cycle (below) uses familiar processes in each step:
feedstock
product
CH3COOH
HI
CH3OH
H 2O
CH3I promoter
CH3COI
CO
CO
I
Rh
I
I
Rh
I
CO
I
Me
CO
catalyst
CO
I
I
O
CO
I
Rh
I
-
CO
Me
I
O
Rh
I
Me
CO
Monsanto Acetic Acid Process
Classify the organometallic intermediates and the reactions that interconvert them.
Why is it necessary to convert the MeOH to MeI?
7
Although rate-determining step is the oxidative addition, the reaction is usually still
performed under a high pressure (15 atm.) of CO. This presumably prevents catalyst
decomposition.
There are severe problems with corrosion of reactor vessels by HI and MeI and so
expensive alloys or glass reactors have to be used. Also, MeI is very volatile and
seriously carcinogenic, and must be continuously recycled from effluent gases.
The process can be tuned to produce acetic anhydride (BP-Monsanto Process). Many
of the uses of acetic anhydride are based on acetylation of alcohols and phenols. The
side products from such processes such as acetic acid and methyl acetate can be
recycled using a further adaptation of the Monsanto process.
7. Hydroformylation of Alkenes
This was one of the first commercial processes to use a homogeneous catalyst.
Millions of tons of aldehydes p.a. are produced in this way.
R
catalyst
R
+
H2/CO
R
CHO
CHO
branched
linear
The original process (1930s) used [Co2(CO)8] and this is still the most popular. There
are some problems with this system
- HCo(CO)4, which is the active species, is volatile and unstable
- it is also a good hydrogenation catalyst
- large amount of branched aldehydes produced (linear aldehydes are the most
valuable)
- no spectroscopic "handle" except IR \ mechanistic studies difficult.
The proposed catalytic cycle is shown overleaf.
8
H
CO
OC
Co
CO
CO
CO
O
H
C
R
CO
R
Co
H
feedstock
CO
OC
product
catalyst
O
C
R
H
H
R
CO
CO
Co
Co
H
CO
CO
CO
CO
H2
feedstock
O
R
Co
Co
OC
CO
R
CO
C
OC
CO
CO
CO
CO
Co
R
CO
CO
Hydroformylation of an Alkene Catalysed by [Co(CO4)H]
9
CO
feedstock
Most of the problems associated with the cobalt catalyst can be circumvented by use
of the Rh catalyst [Rh(PPh3)2(CO)H], which only produces linear aldehydes:
H
PPh3
Ph3P
Rh
PPh3
CO
PPh3
O
C
R
R
H
Ph3 P Rh
OC
H
PPh3
feedstock
product
catalyst
O
R
H
H
C
Rh H
R
Ph3P
OC PPh
3
PPh3
Rh
PPh3
CO
H2
R
feedstock
O
R
C
Ph3P Rh
OC
Ph3P Rh
OC
PPh3
PPh3
R
CO
Ph3P Rh
OC
PPh3
CO
feedstock
Classify the reactions in this cycle
Which step decides whether a linear or branched aldehyde is formed?
How might we increase the proportion of linear product?
10
8. Wacker Process
Complexes of alkenes often undergo nucleophilic attack to give metal alkyls which
may then rearrange to give other products.
M
M
.. Nu
Nu
This is used as the basis for production of millions of tons of aldehydes p.a. in the
Wacker Process.
It was established in the 19th century that aqueous PdCl2 oxidises ethylene to
acetaldehyde. The reaction consumes the PdCl2 (and is thus hideously expensive) and
deposits black Pd metal. The key discovery that made this reaction catalytically
feasible is that acidified Cu(II) will oxidise the Pd(0) before it has a chance to
precipitate.
As the Pd(II) is reduced by the
ethylene (mechanism later) it is
quickly re-oxidised by two moles of
Cu(II). The Cu(I) product is airsensitive and is reoxidised itself to
Cu(II). Pd(0) is not oxidised by air at
a significant rate under these
conditions.
C2 H4
PdCl2/O2 /H2 O/Cu(II)
CH3CHO
Mechanistic work on the process suggested the following rate equation:
Rate = k[PdCl4 2-][C2H4]
[Cl-]2[H+]
What can you say about the mechanism from this data?
Hints: Which species are involved, and how many moles of them?
Are they adding to or leaving the active site?
Because the reaction takes place in water, it is not feasible to determine the order of
reaction with respect to [H2O].
11
O2/H2 O/H+
Cu(II)
Cu(I)
CH3CHO
HCl
H2O
PdCl2
Pd(0)
Cl-
reductive
elimination
Cl
Cl
Me
Cl Pd
Cl
Pd
O H
H2 O
H2O
1,2 insertion
ClCl Pd
H2O
HO
Cl
H
Cl Pd
H2O
OH
H+
β-elimination
Cl-
-
Cl
Cl
Pd
H
Cl Pd
H2 O
OH
H2O
Proposed Catalytic Cycle for the Wacker Process
12
OH
9. Alkene Metathesis.
Discovered in the mid-1950s by Dupont, Standard Oil and Phillips Petroleum:
H2 C
CHR
+
+
CHR
CH2
CHR
H2C
CHR
CH2
Catalysed by heterogeneous and homogeneous systems.
Mechanism proposed by Chauvin in 1970 involves a metallocyclobutane complex:
R'2
C
M CR2 + R'2C CR'2
M
M
CR'2
Schrock
alkylidene
CR'2 + R'2C CR2
C
R2
"
"
from e.g. WCl6 + ZnMe2
"
"
Me
W
W
CH2
Me
Homogeneous Alkene Metathesis Catalysts
• Transient alkylidene complexes:
Cl
Ti
AlMe2
-AlClMe2
CH2
Tebbe's Reagent
13
Ti
CH2
• Stable alkylidene complexes:
P
CH3
F3C
O
F3C
Cl
N
Ru
Mo
H3 C
F3C
O
CHPh
Cl
P
CHCMe2Ph
CF3
Grubbs Catalyst
Schrock Catalyst
The two catalysts above are particularly important in Ring Opening Metathesis
Polymerisation:R
R
R
L nM
L nM
+
L nM
norbornene
metallocyclobutane
+
L nM
n
polymer
R
14
10. Alkene Polymerisation; Ziegler-Natta Catalysis (see MPC
Organometallic Option)
Ziegler and Natta shared the Nobel Prize in 1963 for their earlier work on the
heterogeneous polymerisation of ethene and propene. The catalysts used are similar to
those active as metathesis catalysts (early TM in high oxidation state, see Section 9),
the earliest and most popular is a mixture of TiCl3 and the alkylating agent AlEt2Cl.
Homogeneous analogues of the Ziegler-Natta system have been very widely
studied. The parent complex of several generations of catalyst, zirconocene dichloride,
was popularised by Sinn and Kaminsky in 1976.
Zr
Cl
excess MAO
very high activity for olefin polymerisation
Cl
MAO= [MeAlO]n
AlMe 3 + H2O!
Me
Me
probable structure of MAO
Al
O
Al
Me
O Al
Me
n
Me
Development of the mechanism of this process has a long history and is still going on
today (see later).
Cossee-Arlman Mechanism (1964)
The polymer chain grows by successive insertion of alkene into the M–C
bond at the catalyst centre.
M
CH2
P
+
H2C
CHR
M
CH2 CHR
P
CH 2
This rather simple approach has some problems:
(i) Why does the polymer not
chain-terminate by a
β-elimination step?
α
β
P
M
H
(ii) Insertion of alkenes into M—C bonds was (and still is) rare.
15
P
M
H
Green-Rooney Mechanism (1978)
Green suggested that both of the above problems could be solved by the polymer
chain performing an α-elimination to give a alkylidene hydride. The reason why the
chain does not β-eliminate is because it α-eliminates! This alkylidene could react with
the incoming alkene in a metathesis-like step and then reductively eliminate the
intermediate alkyl hydride to give the expected polymer.
H
M
H
α
β
P
M
α
P
β
C 2H4
β
H
H
M
M
P
α
P
Holes in this mechanism were obvious from the outset:
(i) The metallacycle could just as easily eliminate the other M—C bond to give the
unobserved ethyl-branched polymer (draw this out!!).
(ii) The mechanism could not apply to a d0 or d1 catalyst because the α-elimination
requires ∆OS = +2.
16
Hence, the modified Green-Rooney mechanism was developed; the alkylidene hydride
was replaced by an agostic interaction (see earlier!)
H
M
H
α
C2H4
P
M
α
P
H
M
olefin
insertion
P
The idea behind this was two-fold. First, the formation of the agostic bond would
swing the growing polymer chain around so as to not only present the M—C bond to
the incoming alkene, but also open up the coordination site. Secondly, it will also
weaken the M—C bond (remember, the agostic is a three-centre two-electron bond).
Most researchers still broadly agree with the modified Green-Rooney mechanism.
Recent Developments
ACTIVE
SPECIES
CH2
Zr+
Zr+ CH3
H
14e alkyl cation
Ethene coordination and insertion
‡
Zr+
Zr+
Zr+
CH2
CH2
H
H
17
CH2
H
Generation of Alkyl Cations
Involves the use of strong Lewis Acids to abstract an alkyl group (as R- ) from the
metal centre:
Catalyst System
Cl
Zr
Me
MAO
Zr
Cl
MAO
Zr
Me
alkylation
+
CH3
[Me-MAO]-
alkyl abstraction
Model Systems
Me
Zr
[CPh3]+X-
Zr
Me
+
CH3
+ Ph3CMe
XX= [B(C6F5)4]-
how do you make this?
Me
Zr
B(C6F5)3
Zr+ CH3
Me
[MeB(C6F5)3]-
18
CATALYSIS PROBLEM CLASS
Answer ALL the questions in the catalysis handout!!
19