Transcript A density functional study on chain isomerization and co

Theoretical studies on polymerization
and co-polymerization processes
catalyzed by late transition metal
complexes
Artur Michalak and Tom Ziegler
University of Calgary; Calgary, Alberta, Canada
CC
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
O
C
C
C
N
O
C
N
C
C
Ni
C
C
C
Ni
C
C
C
C
O
O
C
C
C
C
C
C
C
O
C
C
C
C
C
O
C
C
C
C
C
C
Introduction
In the olefin polymerization catalyzed by homogeneous single-site catalysts, by modification of the ligands one can
potentially control the structure and properties of a resulting polymer. Therefore, it is important to understand the
reletionship between the catalyst structure and polymer properties. We present the results of computational studies on
the substituent effect on the factors controlling the polymer branching in the olefin polymerization catalyzed by Pdbased diimine catalysts introduced by Brookhart et al. (JACS 1995, 117,6414; 1996,118,11664; 1996,118,267,1998,
120, 888). We studied the effect of ligand modification on the relative stabilities of isomeric alkyl and olefin complexes,
as well as the regioselectivity of olefin insertion. Further, based on the results of DFT calculations, the growth and
isomerization of a polymer chain have been modeled by a stochastic approach.
The olefin polymerization catalysts based on late transition metals are not only able to polymerize and copolymerize a-olefins, but also, due to their less oxophilic character, exhibit tolerance towards compounds containing
polar functional groups. The Brookhart Pd-based catalysts are able to co-polymerize ethylene and a-olefins with
methyl acrylate. The neutral Ni-based catalysts proposed by Grubbs et al. (Organometallics 1998, 17,3149) tolerate
presence of functional monomers, and has been shown to co-polymerize ethylene with a-w-functional olefins
(Science,2000,287, 460).
In the present studies we investigated the polar monomer binding modes in the complexes involving Ni- and Pdbased catalysts with Brookhart and Grubbs ligands. The result show that while in the case of Ni-based diimine catalyst
(inactive in co-polymerization) complexes with polar molecule bound by oxygen atom are preferred, for Pd-based
Brookhart system, as well as Grubbs catalysts based on both, Ni and Pd, the C=C bound p-complexes are energetically
favoured. The difference between the former and the latter cases comes from the difference in the electrostatic
contribution to the energy of the interaction between polar monomer and the catalysts.
Further, we present the results of computational studies on co-polymerization of ethylene and methyl acrylate
catalyzed by Brookhart Pd- and Grubbs Ni-based systems. We studied the insertion of acrylate, stability of insertion
product, including chelate structures, and the chelate opening by ethylene.
Computational details
DFT calculations (ADF program) with Becke-Perdew XC functional; triple-zeta
STO basis set for Pd, double-zeta with polarization function for C,N,H,O; frozen
core: 1s for C,N,O, 1s-3d for Pd; first-order scalar relativistic correction.
2
I. Substituent effects in olefin polymerization
catalyzed by Brookhart Pd-catalyst
C
C
C
C
C
C
C
C
C
N
C
N
C
C
Pd
C
C
C
C
C
C
The summary of the calculations is presented in Scheme 2. For the generic
catalyst model (a) the complete analysis of the reaction mechanism has been
performed (Michalak; Ziegler Organometallics, 1999, 18, 3998). For the large
models (b-i) the system studied are: (i) the isomeric alkyl complexes involving nand iso-propyl groups; (ii) the ethylene and propylene p-complexes with n- and
iso-Pr; (iii) ethylene and propylene (2,1- and 1,2-) insertion TS with n-propyl.
An analysis of the relative stability of isomeric alkyl complexes is qualitatively
presented in Fig. 1. There are two factors important here: the relative stability of
alkyl radicals, and the Pd-alkyl bonding energy; they change in the opposite
directions: the branched radical is more stable, but it is more weakly bound. As a
result of decreasing bonding energy, the preference of the branched isomer
observed in a generic system is slightly enhanced for the real catalysts.
For the olefin p-complexes, the preference of the isomer with branched alkyl
observed in a generic system, has been reversed for the real catalysts as a result of
steric repulsion (Fig. 3a). The p-complex stabilization energies are decreased in a
‘real’ systems, as a result of decreased stability of isomer with branched alkyl and
increased stability of reference alkyl complex (Fig. 2). Also, in the ‘real’ systems,
ethene p-complexes are stabilized more strongly, than those of propene (Fig. 3b) the opposite effect has been observed for the generic catalyst model.
The steric repulsion in ‘real’ systems also affects the regioselectivity of propene
insertion (Fig. 3c): for a generic system the 2,1-insertion is strongly preferred,
while for the catalysts with the largest substituents this is again reversed: the 1,2insertion TS become lower in energy.
The complete results are presented in the recent article (Michalak, Ziegler
Organometallics, 2000, 19, 1850).
C
C
C
C
C
C
C
C
C
C
C
C
N
C
C
N
C
C
Pd
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
N
C
C
C
C
C
N
C
C
Pd
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
N
C
C
C
C
N
C
C
Pd
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
N
N
C
C
Pd
C
C
C
C
C
C
C
C
C
C
3
Propene polymerization:
N
P
N= -NAr-CR-CR-NAr-
Pd CH
2
N
CH-CH 3
1
1,2 ins.
C CH
3
Pd
C H2
H
N
N
isom.
C CH
Pd
3
C H2
N
(3)
4
...
H
...
1,2 ins.
...
H
N
Pd
N
3
2,1 ins.
...
P
P
2
2,1 ins.
= C nH2n+1
(2)
P
N
1,2 ins.
P
2,1 ins.
(1)
P
N-
1,2 ins.
...
CH
C H
CH3
2,1 ins.
...
isom.
(4)
CH2
C H
Pd
CH2
N
5
N
1,2 ins.
...
• two possible insertion paths (1) and (2); each introduces a methyl branch;
H
2,1 ins.
...
Scheme 1.
• chain staightening isomerization reaction (4) removes a branch;
• there are alkyl complexes and olefin p-complexes with primary, secondary,
and tertiary carbon atom attached to the metal, present in the catalytic cycle.
4
Substituent effects in olefin polymerization catalyzed by Brookhart Pd-catalyst
- summary of calculations
R1
R2
R1
CH
N
N
P d CH
2
N
CH2
Pd
R1
H
N
CH
Pd
CH
N
R2
3 (a-i :) R1=C2H5 , R2=H
C2 H4
R1
R2
CH
N
Pd
CH
CH2 CH3
N
C3 H6
1 (a-i :) R1=CH3 , R2 =H
1'(a-i ): R1=H , R2=CH3
CH2
CH2
N
2 (a-i :) R1=C2H5 , R2=H
2'(a-i ): R1=CH3 , R2 =CH3
R2
CH
R1
N
R2
CH
P d CH
2
N
CH-CH3
1,2
=
Ar - N
Na:
b:
c:
d:
e:
f:
g:
R
N - Ar
R = H ; Ar = H
R = H; Ar = -C
6H 5
R = H ; Ar = 2,6-C
6H 3Me2
i
R = H ; Ar = 2,6-C
6H 3 ( P r)2
R = M e; Ar = H
R = Me ; Ar = 2,6-C
6H 3Me2
i
R = M e ; Ar = 2,6-C
6H 3( P r)2
5 (a-i ): R1=C2H5 , R2=H
2,1
4 (a-i :) R1=C2H5 , R2=H
4'(a-i ): R1=CH3 , R2 =CH3
R
N-
R1
N
Pd
N
R2
CH
CH2
CH
CH3
Ar - N
N - Ar
h : (R2 = An); A r = H
i
i : (R2 = An) ; A r = 2,6-C
6 H3 ( P r)2
6 (a-i ): R1=C2H5 , R2=H
alkyl complexes
(n-, iso-Pr)
ethene, propene
complexes (n-, iso-Pr)
ethene, propene (2,1- and 1,2-)
insertion TS (n-Pr)
Scheme 2.
5
a)
Relative energy
n-P r
Fig. 1
n-P r
is o-P r
0
n-P r
is o-P r
alkyl
radical s
Substituent effects in olefin polymerization
catalyzed by Brookhart Pd-catalyst
is o-P r
'real'
complexes
Electronic preference
(generic system)
generic
complexes
a)
Alkyl bonding energy
b)
R'
R'
R
N
Pd
R'
R'
CH3
R
N
Pd
R
N
n-P r, is o-P r
0
R'
is o-P r
CH3
is o-P r
R'
R'
CH2
R'
CH 3
alkyl
radical s
iso-propyl a lkyl
n-P r
'real'
complexes
n-P r
generic
complexes
b)
CH3 R
Fig. 2
n-P r
is o-P r
n-P r
n-propyl a lkyl
R'
R'
N
Relative energy
R
N
Steric preference
(real systems)
Pd
R'
R
N
Pd
R'
R'
is o-P r
prope ne p-complex
alkyl complexes
+ free olefin
En -Pr >>Eis o-Pr
'real'
complexes
c)
R'
R'
R
N
olefin
p-compl exes
generic
complexes
R'
R
N
R
N
R'
0
is o-Pr
En -Pr >
~ E
R'
C
R'
e the ne p-complex
R
N
R
N
Pd
C
C
C
CH 3
2 ,1 - ins. TS
R'
R'
R'
R'
Pd
C
R
N
C
R'
CH3
1 ,2 - ins. TS
Fig. 3
6
II. Simulations of polymer growth and isomerization
The results of our studies on the ethylene and propylene
polymerization catalyzed by Brookhart complexes are used as input
data for stochastic simulations of polymer growth and isomerization.
Based on the assumption that relative probabilities of the events
possible in the catalytic cycle are equal to the relative rates of the
elementary reactions, these simulations allow one to investigate an
effect of the temperature and the olefin pressure on the polymer
structure. It is known from experimental studies (Guan Z; et al.
Science 1999, 283,2059) that in the case of Brookhart catalyst the
change in the olefin pressure does not affect the total number of
branches, but strongly influence the microstructure of resulting
polymer: from mostly linear polymers obtained for high pressure to
hyper-branched structures for low pressure values.
In our simulations we build one polymer chain at a time, starting
from a olefin insertion into the Pd-methyl group. The structure of the
polymer is remembered, and at every step the stochastic choice of the
next event is made, on the basis of relative probabilities, calculated
from the energetics of elementary reactions. For example, if the
primary carbon (a) [Fig4.] is attached to the metal, then the olefin
uptake followed by insertion, as well as one isomerization reaction
(leading to a secondary carbon being linked to Pd) are possible; if the
primary atom (b) is attached to the metal, a choice is made between
olefin capture/insertion and an isomerization leading to a secondary
carbon; if the secondary carbon (c) is attached to the metal, then a
capture/insertion event is possible, or two equivalent isomerization
events; from carbon (d), besides a capture/isomerization, two
inequivalent isomerisation reaaction are taken into account; etc. In the
propylene case, the two insertion paths (1,2- and 2,1-) are considered.
Energetics of elementary reactions
Relative probabilities of the
elementary reactions
Choice of a path
relative probabilities = relative rates:
pi ri
i pi  1

pj rj
e.g. isomerization vs. isomerization:
piso.1 riso.1 kiso.1
G1, 2


 exp(
)
piso.2 riso.2 kiso.2
kT
isomerization vs. insertion:
piso .1
riso .1
kiso .1


p ins . 1, 2 rins .1, 2 k ins .1, 2 Kco mpl. pole fin
etc.
(c)
(a)
(d)
(e)
(b)
Fig.4.
7
Simulations of polymer growth and isomerization
As a result of a set of simulations, the polymer structures are
obtained, the average number of branches is calculated, and analysis
of the branches is performed: number and length of primary branches,
secondary ones, etc. (Fig. 5). If the termination reactions were taken
into account, the molecular weights could be also obtained. Here,
however, since we focused on the polymer structure, we assumed no
termination, and each simulation was performed until the chain
reached a length of 1000 carbon atoms.
In Figure 6 we show the examples of polymer structures obtained
from propene polymerization under different monomer pressure, for
the catalyst with unsubstituted phenyl rings. For high pressure (panel
a), the structure of the polymer is mostly linear, with relatively long
primary branches, and with small number of the higher-order
branches. With decrease in the pressure (panels b and c), the number
of atoms in the main chain decreases, while the number of the higher
order-branches increases, leading to a hyper-branched polymer
structure.
In Table 1, we listed some data obtained for propylene
polymerization with ‘real’ Brookhart catalyst. Here we can observe a
similar effect of the change in the olefin pressure: while the total
number of branches remains constant, the structure of the polymer is
strongly affected. Again, with decrease in the pressure, the number
and length od higher-order branches increases.
Fig.5.
main chain
primary branch
secondary branch
tertiary branch
etc.
Results:
- Polymer chain;
- Total No. of branches;
- Classification of branches:
no. of branches of a given type,
and their length;
- Molecular weight;
8
Simulations of polymer growth and isomerization - results
a)
b)
c)
Table 1:
Effect of olefin pressure change Brookhart system: R=Me; Ar= Ph(i-Pr)2
p=1
p=0.1
p=0.01
Av. Total No. of Branches
238
236
236
1o branches -av. Length
longest
1.59
26
1.66
30
2.11
36
2o branches -av. Length
longest
2.33
12
2.29
16
2.15
14
3o branches -av. Length
longest
0.27
4
0.51
6
1.38
7
4o branches -av. Length
longest
0.00
2
0.01
4
0.11
4
5o branches -av. Length
longest
0.00
0
0.01
1
0.00
1
Figure 6:
Effect of olefin pressure on polymer structure;
a) p=1, b) p=0.1; c) p=0.01.
9
III. Binding mode of polar monomers
The first step in polymerization processes involves a formation of the catalyst-monomer complex; in a-olefins
polymerization the monomer is bound by the double C=C bond. In order to incorporate polar monomers into
polymer chain in random co-polymerization process, it is required that its insertion follows the same reaction
mechanism, i.e. it involves formation of the corresponding p-complex. Therefore, it seem s to be important, that the
stabilization energy of the p-complex is larger than that of complexes in which the monomer is bound by polar
group.
In this studies we compare the binding modes of methyl acrylate and vinyl acetate for Ni- and Pd-based catalysts
with Brookhart and Grubbs ligands. The cationic Brookhart Pd-based systems have been show to co-polymerize
olefins with methyl acrylate, while the Ni-based complexes are inactive in co-polymerization process. The neutral
Grubbs catalysts tolerate the presence of polar molecules in the polymerization of ethylene and are able to copolymerize olefins with w-functionalized compounds.
The stabilization energies of the polar monomer complexes are listed in Table 2; the systems with polar molecule
bound by C=C bond or by O atom were considered. The results for methyl acrylate and vinyl acetate are
qualitatively similar. Namely, in all the systems but Ni-based Brookhart catalysts, the p-complexes are more stable
than corresponding O-bound ones. In Table 3, the contributions to the bonding energy are listed for methyl acrylate
complexes with generic catalyst models. It can be observed that the main difference between the Ni-Brookhart
catalyst and all the remaining systems is an increased electrostatic contribution in O-bound complex in comparison
to the p complex case. Thus, it can be concluded that the use of neutral catalyst seems to be promising for copolymerization purposes. In the case of Gruubs ligands, the preference of the p complexes is enhanced for both, Niand Pd-based systems.
A comparison of Brookhart Ni- and Pd-based catalysts also suggests that in a search for random copolymerization catalyst the systems in which the O-bound complexes are preferred can be excluded. Here, the use of
computational studies can be very helpful.
10
Binding mode of polar monomers
Monomer2 C=C
complex
Catalyst
Ligand1-Metal
A) generic catalyst
Brookhart - Ni
Brookhart - Ni
Table 2.
Ocomplex
E(C=C) -E (O)
C
C
MA
VA
-17.1
-17.1
-21.1
-17.7
+4.0
+0.6
N
N
C
Pd
C
Brookhart - Pd
Brookhart - Pd
MA
VA
-20.7
-20.1
-17.3
-15.0
-3.4
-5.1
Grubbs - Ni
Grubbs - Ni
MA
VA
-17.7
-16.9
-10.1
-9,7
-7.6
-7.2
Grubbs - Pd
Grubbs - Pd
MA
VA
-24.3
-21.7
-10.2
-9.6
-14.1
-12.1
C
C
O
C
C
O
C=C complex
C
C
C
N
B) ‘Real’ Catalyst
Brookhart - Ni
Brookhart - Pd
MA
MA
-10.1
-13.6
-13.1
-10.6
+3.0
-3.0
N
Pd
C
C
C
O
C
Grubbs - Ni
MA
-12.8
-6.5
-6.3
C
O
C
C
1Brookhart: -N(Ar)-C(R)-C(R)-N(Ar)-; generic: Ar=H, R=H; real: Ar=Ph(i-Pr)2, R=Met
Grubbs:
generic: R1=R2=X=H; real R1= Ph(i-Pr)2; R2=X=H
2
MA=methyl acrylate;
VA=vinyl acetate
O- complex
11
Binding mode of methyl acrylate - contributions to the bonding energy
Eb = Eb,dist. + Eg = [Esteric + Eorb] + Eg = [ (Eel + EPauli ) + Eorb] + Eg ;
Eb,dist. - interaction energy of distorted reactants;
Eg
- geometry distortion term;
Esteric - total steric interaction;
Eorb - orbital interaction;
Eel - electrostatic interaction;
Epauli - Pauli repulsion.
Table 3.
Catalyst/binding mode
Eel
Epauli
Esteric
Eorb
Eb,dist
Generic catalyst
Brookhart - Ni / C=CBrookhart - Ni / O-
-96.5
-62.2
115.1
63.0
23.7
-0.8
-69.2
-40.8
-45.5
-41.6
Brookhart - Pd /C=CBrookhart - Pd /O-
-102.1
-45.11
120.6
50.3
21.0
0.7
-60.4
-32.1
-39.4
-31.3
Grubbs - Ni /C=CGrubbs - Ni /O-
-110.6
-50.7
141.4
58.5
35.0
7.63
-82.8
-30.0
-47.8
-22.4
Grubbs - Pd /C=CGrubbs - Pd /O-
-116.9
-36.1
144.7
47.2
30.6
6.77
-70.4
-24.4
-39.7
-17.6
12
Binding mode of polar monomers - ‘real’ systems
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
O
C
Pd
O
C
C
C
C
C
C
C
C
C
N
C
Ni
C
C
C
N
C
C
C
N
N
C
C
C
C
C
C
C
C
C
CC
C
C
C
C
C O
C
C
C
C
C
O
Brookhart, Ni - O-complex
CC
C
C
Brookhart, Pd - p-complex
C
C
C
C
C
C
O
C
C
O
C
C
C
N
C
C
Ni
C
C
C
C
C
C
C
C
O
C
C
Grubbs, Ni - p-complex
13
IV. Co-polymerization of ethylene and methyl acrylate
The elementary reactions in olefin-acrylate co-polymerization occurring after insertion of acrylate are shown in
Scheme 3. In the present studies we investigated the 1,2- and 2,1-insertion of acrylate, stability of the insertion
products: g-and b-agostic complexes, 4-, 5-, and 6-member chelates, as well as chelete opening by the ethylene and
acrylate molecules. Since the 1,2-insertion bariier is higher by 4.5 kcal/mol for Brookhart catalyst, we present here
only the reaction paths following the 2,1-insertion.
The energetics of the elementary reactions for Brookhart catalyst is shown in Figure 7. The acrylate p-complex
stabilization energy lies between the numbers for ethylene and propylene, obtained from the same generic model for
the catalyst. One can expect, however, that for the real catalysts with a bulky substituents this trend will be reversed,
as for ethylene and propylene (see Fig.3). In agreement with experimental results, the chelates are more stable the
the agostic insertion products. The stability of the chelates increases from 4- to 6-member ring. The latter has been
found experimentally to be a resting state in the process. In the ethylene p-complexes (Fig. 8) the chelating bond is
still present; the Pd-O bond is broken in the TS geometries. The ethylene insertion is easiest in the case of 4-member
chelate; the insertion barriers increase for 5- and 6-member structures.
In Figures 9 and 10 we present the corresponding results for the Grubbs Ni-based catalyst. Here, the process is
more complicated, due to the asymmetry of the catalyst. There exist two paths starting from two isomeric alkyl
complexes, denoted as C or T (corresponding to the Pd-C bond being in cis- or trans- position with respect to
nitrogen atom). As in the case of ethylene (Chan, M; Ziegler, T., in press), the path starting from the more stable
alkyl complex (C) leads to the less stable acrylate complex, leading to the higher insertion barrier. In comparison to
the Brookhart catalyst, the energetic hierarchy of the chelates is different. Here, the 4-member ring is the most stable.
This comes mainly from the size of the metal: the Ni orbitals are more ‘compact’. Due to less oxophilic character of
the catalyst, the chelate structures are higher in energy with respect to the agostic products than in the Brookhart
catalyst case, and the ethylene p-complexes are stabilized more strongly. Therefore, it seems that in the case of
Grubbs catalyst the co-polymerization should be easier, leading to larger incorporation of the polar monomer and the
polymers characterized by larger molecular weights. However, we have not obtained the ethylene insertion barriers
yet; neither we studied the termination reactions.
14
Co-polymerization of ethylene and methyl acrylate
• acrylate or olefin p-complexes can be formed from alkyl species;
• after olefin insertion the process proceeds as in the olefin polymerization case;
• after acrylate insertion various chelate structures can be formed;
• chelates can be opened by either olefin or acrylate molecule;
Scheme 3.
P
N
N
H
Pd
N
C COOCH
3
C H2
2,1 ins.
1,2 ins.
H
P
C COOCH
Pd
3
C H2
N
O COCH 3
CH
Pd
CH
N
2
P
N
O
N
1,2 ins.
...
2,1 ins.
...
COOCH3
COCH 3
Pd
CH
N
CH2
Pd
N
...
CH
COCH 3
C H2
2,1 ins.
...
...
COCH 3
C H2
CH C H2
Pd
N
P'
5-memb. chelate
1,2 ins.
O
N
P
4-memb. chelate
1,2 ins.
b-agostic
O
N
CH
C H
N
g-agostic
P
5 memb. chelate
Pd
CH
N
H
N
CH
Pd
COOCH3
g-agostic
b-agostic
N
P
H
P
2,1 ins.
...
6-memb. chelate
1,2 ins.
...
2,1 ins.
...
15
Acrylate 2,1-insertion path - Brookhart, Pd-based catalyst
kcal/mol
0
alkyl agostic
+acrylate
N
N
insertion TS
C
C
O
O
N
C
N
Pd
C
C
Pd
O
C
C
-5
C O
+19.4
C
C
C
C
C
C
C
-10
C
C
C
-18.5
-20.7
C
C
N
-15
N
Pd
C
C
C
C
N
N
O
g-agostic
C
C
O
O
C
Pd
C
C
O
C
-20
C
C
C
acrylate 
p complex
-25
-5.3
b-agostic
C
C
C
C
C
N
N
C
Pd
C
C
O
-8.5
-30
C
C
N
N
C
-35
C
O
C
C
C
C
Pd
N
N
C
C
O
-6.1
C
Pd
C
C
C
-40
C
O
C
C
O
O
C
Fig. 7.
4-memb.
chelate
C
5-memb.
6-memb.
chelate
chelate
C
C
C
-1.1
16
Chelate opening - Brookhart, Pd-based catalyst
C
C
C
O
C
C
N
N
O
O
O
C
C
C
C
C
dP
C
O
N
C
C
N
O
C
C
C
C
C
Pd
N
N
C
dP
C
C
C
C
C
C
C
C
C
N
C
C
N
C
C
C
C
C
C
Pd
C
C
C
C
O
C
C
kcal/mol
C
O
C
C
C
C
20
O
C
N
O
N
dP
C
C
C
C
15
C
C
+25.0
C
10
4-memb.
C
O
O
+23.0
C
C
C
N
N
5
Pd
C
-10.1
5-memb.
C
0
C
C
6-memb.
-5
-10
Fig. 8.
C
+30.4
-9.1
C
-7.8
ch elates
C
ethen e insertion TS
ethen e p complexes
17
Acrylate 2,1-insertion paths - Grubbs Ni-based catalyst
C
C
C
C
C
C
C
kcal/mol
N
O
Ni
0
-5
T
C
-16.0
C
C
+20.5
T
-15
C
C
-18.7
C
-22.0
-10
O
C
O
C
C
+15.7
C
T
C
C
C
C
-22.2
-20
-25
-9.3
T
T
C
-30
Fig. 9.
N
T
-5.1
-8.8
T
T +4.7
C
alkyl agostic
insertion TS
+acrylate
acrylate
g-agostic
p complex
C
O
-4.2
-35
C
C
b-agostic
C +0.2 C
Ni
+2.4
O
C
C
C
+1.9 C
O
C
C
C
C
5-memb.
4-memb. chelate
6-memb. chelate
chelate
18
Chelate opening - Grubbs Ni-based catalyst
6-memb.,cis
6-memb.,trans
C
5-memb.,cis
C
O
5-memb.,trans
O
C
C
C
C
O
O
C
C
C
C
C
C
C
C
C
C
N
Ni
C
O
C
C
O
C
C
N
C
N
C
Ni
O
C
Ni
C
O
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
O
C
O
C
C
O
O
C
C
N
C
Ni
C
C
C
C
C
C
C
C
C
C
C
C
C
C
kcal/mol
C
C
C
C
10
C
-9.9; 6-memb. trans
C
C
C
C
5
C
C
C
C
C
C
C
C
-8.4; 5-memb. trans
O
N
O
N
C
0
C
C
C
Ni
O
O
O
O
C
Ni
C
C
-3.4; 5-memb.trans
C
C
C
C
C
C
-5
-7.7; 4-memb. trans
-5.6; 6-memb. cis
C
C
C
C
C
-10
Fig. 10.
C
-12.0; 4-memb. cis
chelates
+ ethene
ethene p complexes
4-memb.,cis
4-memb.,trans
19
Conclusions
• A increase in the substituents size in Brookhart diimine catalyst strongly affects the factors
controlling branching of the polymers: the preference of the branched alkyl complexes is enhanced,
the olefin complexes with linear alkyl become more stable, and the regioselectivity of insertion is
reversed, leading to the preference of the 1,2-insertion path for larger catalysts.
• Stochastic modeling of the polymer growth and isomerization based on the energetics of the
elementary reactions from DFT calculations allows one to qualitatively investigate the effect of the
temperature and the olefin pressure on the structures of resulting polymers; the general trends
obtained from a simplified model are in agreement with experimental observations.
• A comparison of the binding mode of polar monomers for the Ni- and Pd-based complexes
(inactive and active co-polymerization catalysts) with diimine ligands reveals that the preference of
the O-bound complex in Ni case is reversed in Pd-based system. Further, the origin of this
difference has mainly electrostatic character. Thus, use of the neutral catalysts in co-polymerization
processes seems to be promising. Indeed, in the case of Grubbs ligand, the p-complex is strongly
preferred already in Ni-system; this preference is enhanced for Pd catalyst
• Studies on the elementary reactions in olefin-acrylate co-polymerization provide the information
about the stability of the reaction intermediates. The energetic order of the chelates in Brookhart
and Grubbs systems is opposite. As a result of less oxophilic character of the Grubbs complex,
formation of the ethylene complex after acrylate insertion is easier than in the Pd-diimine case.
Acknowledgements
This work has been supported by the Novacor Research and Technology Corporation ,as well as by the NSERC.
A.M. acknowledges the University of Calgary Postdoctoral Fellowhip. Important part of the calculations has been
performed with the UofC MACI cluster.
20