Density Functional Study of Neutral
Salicylaldiminato Ni(II) Complexes
as Olefin Polymerization Catalysts
Mary Chan, Liqun Deng and Tom Ziegler
Department of Chemistry, University of Calgary
Calgary, Alberta Canada T2N 1N4
The recent discovery of the ability of salicylaldiminato Ni(II) complexes to promote
ethene polymerization creates the potential for the new class of olefin polymerization
catalyst.1 The skeleton structure of the catalyst is shown in Figure 1. The major advantage of
this type of catalysts is that they produce a neutral active center and thereby avoids the ion-
pairing problems encountered with the homogenous single-site catalysts in current use. The
influence of the substituents on the catalyst backbone have been studied and the activity of
various substituted catalysts are summarized in Table 1.
Table 1: Reactivity of the Catalyst
i-pr R X Catalyst*
O N H H 26.7
R t-Bu H 46.7
H Ph H 81.3
9-Phenanthrenyl H 93.3
9-Anthracenyl H 98.7
X H OMe 13.3
H NO2 253.3
Figure 1: Skeleton Structure of Catalyst
* kg of PE/mol of Ni
Introduction: reactivity of catalyst 3
The nickel complexes are inactive as polymerization catalysts without the presence of
phosphine scavenger such as Ni(COD)2 and B(C6F5)3, and activity appears to be independent
of scavenger type.
One of the two areas of structural modification is at the 3 position of the salicylaldiminato ring
(labeled as R in Figure 1). The experimental data show that bulky substituents at this position
enhances catalyst activity.
The second area of structural modification is at the 5 position of the salicylaldiminato ring
(labeled as X in Figure 1). It was found that electron donating groups at this position decrease
the catalyst activity while catalysts with an electron withdrawing group significantly increases
For compounds where R=H an induction period was observed, ranging 8 min for X=H to 20
min for X=NO2. No induction period was observed when R is substituted with a bulky
organic group. It was speculated that this may be due to slow olefin insertion into the Ni–Ph
bond or slow abstraction of the phosphine from the nickel.
Objectives of Theoretical Study: mechanism 4
To investigate the fundamental reactions of activation, propagation and termination in
polymerization process as shown in Figure 2 for the salicylaldiminato nickel catalyst.
Generation of the Active Species:
P(CH 3 )3 R
O P O O
Ni O Ni Ni P
C2H4 Ni P N N
H N H H
Termination via -Hydrogen Transfer:
P P P
O H O O
Ni Ni H Ni H
N N N
H H H
Termination via -Hydrogen Elemination:
N N H
Figure 2: Fundamental Reactions of the Polymerization Process
Objectives of Theoretical Study: substitution 5
To determine the effects of the substituent X by comparing the energy of the model systems 1,
2 and 3 shown in Figure 3.
To determine the effects of the bulky organic groups by comparing the energy of the model
systems 3, 4 and 5 shown in Figure 3.
To determine the effects of changing the donor atom from nitrogen to phosphorus by
comparing the energy of the model systems 5 and 6 shown in Figure 3.
(CH3) 3P P
(CH3) 3P P Ni
(CH3) 3P P
Ni H O N H
O N O N
System (1) System (2) System (3)
(CH3) 3P P (CH3) 3P ipr (CH3) 3P ipr
O N O P
NO2 NO2 NO2
System (4) System (5) System (6)
Figure 3: Structures of Catalyst Systems to be Studies
Computational Details 6
The density functional theory calculations were carried out using the Amsterdam
Density Functional (ADF) program version 126.96.36.199 Geometry optimizations were carried out
by augmenting the local exchange-correlation potential of Vosko et al.3 with Becke’s
nonlocal exchange corrections4 and Perdew’s nonlocal correlation corrections.5 The frozen-
core approximation was used to treat the core orbtials for all atoms. A Slater type triple-zeta
basis set was used to describe the valence orbitals for the nickel whereas a double-zeta basis
set was used for the non-metals. A single-zeta polarization function was also included for all
Catalyst systems 4, 5 and 6 were investigated by a combined DFT and molecular
mechanics approach using the QM/MM implementation in the ADF program.6 It
incorporates a modified Amber957 force field which includes Rappé’s universal force field
van de Waals parameters for nickel.8 The partition scheme developed by Morokuma and
Maseras was used to couple the QM and the MM regions.9 The MM regions were defined as
the bulky 2,6-diisopropylphenyl or 9-anthracenyl groups when they are present.
Results: generation of the active catalyst 7
The activation of the salicylaldiminato nickel complexes requires the removal the phosphine
ligand to produce a coordinately unsaturated site. This process is modeled by the taking the
enthalpy change of the dissociation of the phosphine ligand from the square planar precursor. The
dissociation enthalpies are reported in Table 2. The first 3 entries show that the electronic nature of
the substituent X at the 5 position of the salicylaldiminato ring has a small influence on the enthalpy
of dissociation. Entries 5 and 6 show that changing the electronic nature of the donor atom have a
larger effect of 2.7 kcal/mol on the
Table 2: Phosphine Dissociation Enthalpy Changes dissociation enthalpy. A comparison
∆H of entries 3 and 4 shows that the 2,6-
System R X R* (kcal/mol)
diisopropylphenyl group on the
imino ligand has little effect on the
1 H H N–H 27.3
dissociation enthalpy. Finally,
2 H OMe N–H 27.5
systems 4 and 5 shows that the
3 H NO2 N–H 28.2
anthracenyl group on the 3 position
4 H NO2 N–Iph 27.9
decreases the dissociation enthalpy
5 An NO2 N–Iph 23.7
by 4.2 kcal/mol.
6 An NO2 P–Iph 21.0
An = 9-anthracenyl
Iph = 2,6-diisopropylphenyl
Results: chain propagation energies 8
Chain propagation is assumed to follow the Cossée-Arlman mechanism.10 The insertion is
initiated by the coordination of the olefin to the metal center followed by insertion into the carbon-
metal bond. Two geometrical isomers are possible for each of the p-complex, the transition state,
and the insertion product: one where the alkyl chain is trans to the nitrogen and one where is it
trans to the oxygen. The complexation energies and insertion barriers for both isomers are
reported in Table 3. The insertion barriers as well as the complexation energy from the trans to N
isomers are relatively insensitive to changes in the catalyst structure. The anthracenyl group
decrease the olefin complexation energy for the trans to O isomer by 5 kcal/mol.
Table 3: Olefin Complextion Energy and Insertion Barriers
Catalyst Olefin Complexation (kcal/mol) Insertion Barrier (kcal/mol)
Alkyl trans to N Alkyl trans to O Alkyl trans to N Alkyl trans to O
1 -18.2 -17.1 15.3 25.0
2 -18.3 -17.4 15.5 25.0
3 -17.1 -16.7 14.1 24.0
4 -16.1 -16.1 14.0 24.3
5 -17.1 -11.1 14.0 23.3
6 -16.8 -10.8 15.6 25.2
Results: chain propagation mechanism 9
The insertion barriers in Table 3 indicate that there is a significant difference between the two
isomers. The geometrical arrangement around the nickel changes after each insertion and
therefore, insertion cannot proceed through the energetically more favorable pathway at all times
due to the formation of the undesired isomer. The barrier of cis/trans isomerization from the p-
complex was determined to be 11.4 kcal/mol. This suggests that the lowest energy pathway for
the insertion process follows the sequence outlined in Figure 4 of 1) complexation with the olefin,
2) cis/trans isomerization, and 3) insertion to extend the polymer chain.
Ni + C2 H4
Figure 4: Lowest Energy Pathway for the Insertion Process
Results: termination mechanism 10
hydride + propylene The reaction profiles for the -
hydrogen transfer (BHT) as well as
the -hydrogen elimination (BHE)
BHE chain termination mechanism for
BHE system 4 appears in Figure 5. The
transition hydride propene
state BHE mechanism involves the transfer
of a hydrogen to the metal. The
propene ethene barrier to form the nickel hydride is
12.5 kcal/mol complex
12.5 kcal/mol, but dissociation of
active catalyst polymer requires another 38.0
kcal/mol. The BHT mechanism
29.9 kcal/mol involves the transfer of the hydrogen
-16.1 kcal/mol to a coordinated monomer. Although
the BHT termination barrier is high at
29.9 kcal/mol, the relative energies
propene required for the various steps show
p-complex ethyl complex that BHT mechanism is still the
Figure 5: Potential Energy Profile for Termination Reactions preferred pathway.
Results: termination barriers 11
The BHT termination barriers do not show the same isomeric dependence as the insertion
barriers. The termination barriers form the two isomers of system 3 are 23.9 and 23.7 kcal/mol and
therefore, only the barrier form the more stable p-complex was compared. The results are summarized
in Table 4. The first 3 entries show that the electronic nature of the substituent X at the 5 position have
relatively small effects on the termination barrier. The electron releasing group appears to increase the
barrier while the electron withdrawing group shows a decrease. Entries 3 and 4 shows that the bulky
2,6-diisopropylphenyl group increases
Table 4: BHT Termination Barriers
the termination barrier drastically by 6.0
System R X R* (kcal/mol) kcal/mol. The effect have been
rationalized by the destabilization of the
1 H H N–H 25.7 transition state by repulsive steric
2 H OMe N–H 26.2 interactions.11 Entries 5 and 6 show that
3 H NO2 N–H 23.9 changing the donor atom from nitrogen
4 H NO2 N–Iph 29.9 to phosphorus decreases the termination
5 An NO2 N–Iph 29.3 barrier by 3.3 kcal/mol. Closer analysis
6 An NO2 P–Iph 26.0 of the geometries show that this may be
due to steric rather than electronic
An = 9-anthracenyl
Iph = 2,6-diisopropylphenyl
Influence of Substitution on Catalyst Activity 12
Electron withdrawing substituents at the 5 position of the salicylaldiminato ring alters the
energies in favor of polymerization by decreasing the phosphine dissociation energy and the
insertion barrier. Electron releasing groups at this position would hinder polymerization as
they increase the phosphine dissociation energy and insertion barrier. The magnitude of the
changes in energy caused by this substituent is relatively small.
The effects of varying the electronic nature of the donor atom from nitrogen to phosphorus is
not easily predictable due to the opposing trends observed. The phosphorus analog has a
lower phosphine dissociation energy, but at the same time shows a lower termination barrier.
The most significant effect of the 2,6-diisopropylphenyl group on the imino ligand is to
increase the termination barrier by 6.0 kcal/mol. Catalyst with this substituent should show
marked increase in activity.
The function of the 9-anthracenyl group on the 3 position is to decrease the phosphine
dissociation energy by 4.2 kcal/mol. This is due to the destabilization of the precursor by
steric repulsion between this substituent and the phosphine ligand. Therefore, bulky
substituents at this position should enhance catalyst activity.
The polymerization mechanism followed by neutral salicylaldiminato nickel complex have
been determined. They are activated by the dissociation of the phosphine ligand followed by
the coordination of a monomer to the metal center. The p-complexes thus form undergoes cis
to trans isomerization before insertion. Termination of the polymer occurs via the -hydrogen
The electronic nature of the substituent X on the 5 position of the salicylaldiminato ring have
relatively small effects on the energies of all fundamental polymerization reactions and is
expected have little influence on catalyst activity. Therefore, the large increase in activity of
the nitro substituted catalyst observed experimentally cannot be attributed to electronic factors
The calculated insertion barriers are relatively insensitive to electronic or steric changes to the
catalyst backbone. Therefore, the induction period observed is due to slow dissociation of the
phosphine from the nickel rather than slow insertion.
The steric effects of adding bulky groups on the 3 position of the salicylaldiminato ring and
the imino nitrogen were large in comparison to electronic effects. The calculated barriers and
enthalpy changes suggest that the addition of bulky substituents at these positions should
significantly enhance catalyst activity.
This investigation was supported by the Natural Science and Engineering Research
Council of Canada (NSERC) and by Novacor Research and Technology (NRTC) of Calgary,
Alberta, Canada. We wish to thank Dr. A. Michalak for helpful discussions.
1. Wang, C.; Friedrich, S.; Younkin, T.R.; Li, R.T.; Grubbs, R.H.; Bansleben, D.A.; Day, M.W.
Organometallics 1998, 17, 3149.
2. (a) Baerends, E.J.; Ellis, D.E.; Ros, P. Chem. Phys. 1973, 2, 41. (b) Baerends, E.J.; Ros, P. Chem. Phys.
1973, 2, 52.
3. Vosko, S.H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200.
4. Becke, A. Phys. Rev. A 1988, 38, 3098.
5. Perdew, J. P. Phys. Rev. B 1986, 34, 7406.
6. Woo, T.K.; Cavallo, L.; Ziegler, T. Theor. Chem. Act. 1998, 100, 307.
7. Cornell, W.D,; Cieplk, P.; Bayly, C.I.; Gould, I.R.; Merz, K.M.Jr.; Ferguson, D.M.; Spellmeyer, D.C.;
Fox, T.; Caldwell, J.; Koolman, P.A. J. Am. Chem. Soc. 1995, 117, 5179.
8. Rappé, A.K.; Casewit, C.J.; Colwell, K.S.; Goddard, W.A. III; Skiff, W.M. J. Am. Chem. Soc. 1992, 114,
9. Maseras, F.; Morokuma, K. J. Comput. Chem. 1995, 16, 1170.
10. (a) Cossée, P. J. Catal. 1964, 3, 80. (b) Arlman, E. J. J. Catal. 1964, 3, 89.
11. Deng, L.; Woo, T.K.; Cavallo, L.; Margl, P.M. Ziegler, T. J. Am. Chem. Soc. 1997, 119, 6177.