Comparison of selected zirconium and hafnium amine bis(phenolate) catalysts for 1-hexene polymerization

Article abstract of DOI:10.1021/om4006005

The kinetics of 1-hexene polymerization using a family of three zirconium and hafnium amine bis-phenolate catalysts, M[t-Bu-ON^XO]Bn₂ (where M = Zr (a) or Hf (b), and X = THF (1), pyridine (2), or NMe₂ (3)), have been inves

Full text of DOI:10.1021/om4006005

Article
pubs.acs.org/Organometallics
Comparison of Selected Zirconium and Hafnium Amine
Bis(phenolate) Catalysts for 1‑Hexene Polymerization
D. Keith Steelman,† Paul D. Pletcher,† Jeffrey M. Switzer,‡ Silei Xiong,‡ Grigori A. Medvedev,‡
W. Nicholas Delgass,‡ James M. Caruthers,*,‡ and Mahdi M. Abu-Omar*,†,‡
†Brown Laboratory, Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, Indiana 47907, United States
‡School of Chemical Engineering, Purdue University, Forney Hall of Chemical Engineering, 480 Stadium Mall Drive, West Lafayette,
Indiana 47907, United States
S
* Supporting Information
ABSTRACT: The kinetics of 1-hexene polymerization using a
family of three zirconium and hafnium amine bis-phenolate
catalysts, M[t-Bu-ONXO]Bn2 (where M = Zr (a) or Hf (b),
and X = THF (1), pyridine (2), or NMe2 (3)), have been
investigated to uncover the mechanistic effect of varying the
metal center M. A model-based approach using a diverse set of
data including monomer consumption, evolution of molecular
weight, and end-group analysis was employed to determine
each of the reaction-specific rate constants involved in a given
polymerization process. This study builds upon the mechanism of polymerization for 1a−3a, which has been previously reported
by applying the same methodology to the hafnium containing analogues, 1b−3b. It has been observed that each elementary step-
specific rate constant that involves the insertion of a monomer is reduced by an order of magnitude. As previously reported for
catalysts 1a−3a, a quantitative structure−activity relationship was uncovered between the logarithm of the monomer-
independent chain transfer rate constants and the Hf−X bond distance for catalysts 1b−3b. However, this dependence on the
pendant ligand is 2.7 times weaker for the Hf-containing analogues versus those containing Zr. These findings underscore the
importance of comprehensive kinetic modeling using a diverse set of multiresponse data, enabling the determination of robust
kinetic constants and reaction mechanisms of catalytic olefin polymerization as part of the development of structure−activity
relationships.
the effect that changing the metal center will have on the
polymerization process.
INTRODUCTION
■
Production of polyolefins is a major industrial process with a
current capacity of ca. 110 billion kg per year globally.1 Today
polyolefins are produced primarily using heterogeneous
Ziegler-Natta catalysts; however, in recent years, homogeneous
single-site catalysts, specifically metallocene-type catalysts, have
attracted attention because they offer potential control of the
various kinetic steps, which in turn can be manipulated by
“catalyst design”.2−4 While high-throughput screening has
accelerated the discovery process with group 4 coordination
complexes leading to Dow’s catalysts for olefin block
copolymer synthesis,5 the promise of directly correlating kinetic
constants to descriptors of the catalyst has not yet been
realized. A major obstacle in the way of rational catalyst design
is the lack of proper quantitative kinetic analysis of all the
relevant processes (i.e., kinetic steps) that are involved in
catalytic olefin polymerization.6,7 Nevertheless, the study of
single-site catalysts for olefin polymerization is particularly
attractive because of the potential to directly correlate the
physical properties of the resulting polymer to structural
features of the catalyst based on first principles.8 These types of
correlations enable one to draw conclusions on how a catalyst
structure may be manipulated to yield specific polymeric
architectures. One particular avenue of interest is to investigate
Of the group IV elements, the metal that has received the
most attention as a homogeneous polymerization catalyst is
zirconium. Hafnium is another group IV element that is known
to act as a homogeneous polymerization catalyst. Zirconium
and hafnium in the +4 oxidation state are remarkably similar,
having the same number of outer shell d-electrons and the same
ionic radii due to the lanthanide contraction. Many of the
analogous zirconium and hafnium complexes reported in the
literature have virtually identical crystal structures.9−11 Despite
their similarities, these two metals behave drastically differently
as polymerization catalysts. When studying β-Me elimination
chain transfer pathways in propylene oligomers, Fiorani et al.
observed that as a general rule zirconocene type catalysts have
increased activity over their hafnocene type catalysts; however,
for bis(Cp*)-metallocenes, hafnium has a significantly larger
activity than its zirconium analogue, making it one of the few
examples where the general rule is broken.10 Further studies by
Collins and Ferrara showed the same phenomenon with an
Received: June 22, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om4006005 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
ONTHFO Ligand (1). In a typical synthesis, an 80 mL reaction vessel
was charged with 2,4-di-tert-butylphenol (6.19 g, 30.0 mmol), 2-
(aminomethyl) tetrahydrofuran (1.55 mL, 15 mmol) and 37%
histological grade formaldehyde (6.00 mL, 80 mmol), distilled water,
and a stir bar while a maximum volume of 80 mL was maintained. The
biphasic reaction mixture was placed in a CEM microwave reactor and
allowed to warm to 100 °C over 5 min while stirring. The reaction was
allowed to stand at 100 °C for 30 min and then cooled to room
temperature. The aqueous layer was removed, and cold, dry methanol
was added to the organic phase. This mixture was shaken for 30 min,
and the resulting solid isolated by vacuum filtration. The crude ligand
product was purified by crystallization from ethanol (28% yield).
Synthesis of Zr[t-Bu-ONTHFO]Bn2 (1a). In a typical synthesis, a
100 mL flask was charged with tetrabenzylzirconium (0.557 g, 1.22
mmol), 20 mL of toluene, and a stir bar and fitted with a rubber
septum. A second 100 mL flask was charged with the t-Bu-ONTHFO
ligand (0.609 g, 1.13 mmol) and 20 mL of toluene. The two flasks
were placed under an inert atmosphere, and the ligand solution was
added to the tetrabenzylzirconium solution via a cannula. The reaction
was allowed to warm to 60 °C and stir for 2 h resulting in a bright
yellow solution. The solution was concentrated to about 10 mL and
placed into a −10 °C freezer. Yellow crystals formed within 2 days, and
the mother liquor was removed via a cannula. The crystals were dried
under vacuum (84% yield). The precatalyst was recrystallized by vapor
diffusion of pentane into a precatalyst/toluene solution to afford an
analytically pure complex.
additional note that the hafnium analogues produce polymers
with a significantly larger molecular weight, Mw.11
One specific family of nonmetallocene catalysts, first
pioneered by Kol and co-workers, that has sparked interest
utilizes an amine bis-phenolate (salan) ligand system (see
Figure 1).12,13 The reason for choosing this particular family of
Figure 1. 1-Hexene polymerization catalyzed by zirconium/hafnium
salan-type catalysts 1a−3b when combined with the activator
B(C6F5)3.
ligands as part of our detailed kinetic studies is the relative ease
of synthesis and the ability to tune the catalysts’ coordination
environment.14 Furthermore, these catalysts exhibit high
activity, comparable to metallocene catalysts, with 1-hexene in
conventional organic solvents such as toluene. This feature
enables the investigation of kinetic data in the condensed phase
thereby eliminating mass transfer limitations that are inherent
in gas phase polymerization reactions. Following up on Kol’s
earlier qualitative observations that the nature of the pendant
ligand (X) and its distance from the metal center (Zr−X)
influence chain transfer,15 we have shown a linear correlation
between the logarithm of the chain transfer rate constants,
kvinylidene and kvinylene, and the Zr−X bond distance, which was
probed by quantitative kinetic modeling of a diverse set of
multiresponse data.16,17 In this study, we continre the use of
quantitative kinetic modeling of multiresponse data for the
salan-type catalysts to elucidate the effect of changing the metal
center from Zr to Hf on the rate constants that comprise the
olefin polymerization mechanism.
NMR Scale Polymerization of 1-Hexene. The procedure for
NMR scale polymerization is based on literature.17 For a typical
polymerization, Zr[t-Bu-ONTHFO]Bn2 (1) (6.1 mg, 0.0075 mmol) was
dissolved in 0.5 mL of toluene in a small vial and sealed with a screw-
cap septum. The vial containing the precatalyst solution was pierced
with a 1 mL syringe. The vial and syringe were placed in a N2 bag and
allowed to equilibrate to 25 °C. Tris(pentafluorophenyl)boron (4.3
mg, 0.0084 mmol), 1-hexene (0.1265 g, 1.50 mmol), and diphenyl-
methane (9.5 mg 0.056 mmol) were added to a 2 mL volumetric flask
and diluted to the mark with toluene-d8. This solution was placed in an
NMR tube and sealed with a septum. The monomer/activator solution
was placed in the spectrometer and allowed to equilibrate to 25 °C
using a VT controller. A measurement was taken to determine the
initial concentration of monomer relative to the internal standard. The
NMR tube was removed from the spectrometer, and the catalyst
precursor solution was added to the activator/monomer solution by
piercing the septum while the syringe remained in the N2 bag. The
reaction mixture was shaken for ca. 30 s and placed back into the
spectrometer. Spectra were acquired at predetermined time intervals
until the reaction reached completion. Each sample was prepared for
GPC analysis by evaporation over mild heat before dissolution in
hexanes and filtration through an alumina plug to remove the
quenched catalyst. Evaporation of solvent yielded clear, colorless
EXPERIMENTAL PROCEDURE
■
General Procedure. All manipulations were performed under dry
inert atmosphere in a glovebox or at a vacuum manifold using air-
sensitive techniques under N2 or Ar atmosphere. Toluene and pentane
were distilled over activated alumina and a copper catalyst using a
solvent purification system (Anhydrous Technologies) and degassed
through freeze−pump−thaw cycles. Both solvents were stored over
activated molecular sieves. Tetrabenzylzirconium was purchased from
STREM and used as received. The monomer 1-hexene was purchased
from Aldrich and purified by distillation over a small amount of
dimethyl bis(cyclopentadienyl)zirconium and stored over molecular
sieves. Tris(pentafluorophenyl)boron was purchased from STREM
and purified by sublimation. Diphenylmethane was purchased from
Aldrich and stored over molecular sieves. CH3OD was purchased from
Cambridge Isotopes and used as received. Toluene-d8 was used as
received and stored over molecular sieves. 1H and 2H NMR
experiments were performed on a Varian INOVA600 MHz or Bruker
DRX500 MHz spectrometer.
1
poly(1-hexene). The array of H spectra was collected on an INOVA
600 MHz spectrometer and analyzed using MestReNova.
Batch Polymerization of 1-Hexene. The procedure for Manual
Quench is based on literature.18 For a typical polymerization, Zr[t-Bu-
ONTHFO]Bn2 (0.073 g, 0.090 mmol) was dissolved in 5.0 mL of
toluene in a small vial that was sealed with a screw-cap septum. The
vial containing the precatalyst solution was pierced with a 10 mL
syringe. The vial and syringe were placed in a N2 bag and allowed to
equilibrate to 25 °C. Tris(pentafluorophenyl)boron (0.053 g, 0.099
mmol), and 1-hexene (1.575 g, 18.71 mmol) were added to a 25 mL
flask and diluted to the mark with toluene. This solution was diluted to
26 mL with 1 mL of toluene, and 1 mL of the resulting solution was
removed for quantification of the initial monomer concentration
through NMR analysis. The flask was sealed with a septum and moved
from a N2 filled glovebox to a vacuum manifold and placed under
argon. The monomer/activator solution was allowed to equilibrate to
25 °C using a temperature-controlled silicone oil bath. The catalyst
precursor solution was added to the activator/monomer solution by
piercing the septum while the syringe remained in the N2 bag. The
resulting yellow solution was allowed to stir while aliquots were
removed at selected times, and each was injected into a 10 mL
The ligands and precatalysts (1a−3b) were prepared following
modified literature procedures.12,13 We describe herein the details for
one representative procedure and provide the others in the Supporting
Information.
Synthesis of 6,6′-((((Tetrahydrofuran-2-yl)methyl)-
azanediyl)bis(methylene))bis(2,4-di-tert-butyl-phenol), t-Bu-
B
dx.doi.org/10.1021/om4006005 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
volumetric flask containing 1 mL of deuteromethanol. A 1 mL aliquot
from the quenched solutions was removed, and a 0.5 mL solution of
toluene-d spiked with diphenylmethane as an internal standard for
quantification of 1-hexene consumption (via 1H NMR on Varian
Inova600). Each sample was prepared for GPC analysis by evaporation
over mild heat before dissolution in hexanes and filtration through an
alumina plug to remove the quenched catalyst. Evaporation of solvent
yielded clear, colorless poly(1-hexene).
As a result, a minimal set of elementary steps is determined
that can fit the multiresponse data. For the zirconium-based
systems 1a, 2a, and 3a, such a minimal set turned out to include
initiation, propagation via normal insertion, 2,1-misinsertion,
recovery from misinsertion, and chain transfer16 resulting in the
formation of vinylidene and vinylene species (see Scheme 1).
Scheme 1. Elementary Kinetic Steps Used in Fitting the Data
In the case of vinyl end group analysis, a 1 mL aliquot was worked
up as described above. The resulting polymer was dissolved in CDCl3
and diluted to the mark in a 2 mL volumetric flask. Diphenylmethane
was used as an internal standard, and the method of standard additions
a
for Catalysts 1−5
1
was used in quantification of the end groups by H NMR. All end-
group analysis measurements were taken on a Bruker DRX500
spectrometer at 25 °C.
2
In the case of H analysis for active-site counting, the remaining
quenched reaction solution (8 mL) was worked up as described above.
The resulting polymer was dissolved in CH2Cl2, and diluted to the
mark in a 2 mL volumetric flask. Benzene-d6 was used as an internal
standard, and the method of standard additions was used in
quantification of active sites by 2H NMR. All active site measurements
were taken on a Bruker DRX500 spectrometer at 25 °C.
Gel Permeation Chromatography (GPC) Analysis. The
procedure used to analyze polymer samples using GPC methods
was taken from Novstrup et al.,6 and it is summarized below. Poly(1-
hexene) samples were added to THF at room temperature and
allowed to dissolve for 4 h. Solutions were then passed through a 0.2
μm filter to remove any particulate matter. The GPC analysis was
performed on a Waters GPCV 2000 for systems 1a and 3a, and on a
Viscotek TDAmax for systems 1b, 2a, 2b, and 3b. On the Waters
GPCV 2000, samples were injected through a 101.3 μL injection loop
and passed through two Polymer Laboratories PLGel 5 μm Mixed-C
columns in series in a 45 °C oven at a flow rate of 1.0 mL min−1. On
the Viscotek TDAmax, samples were injected through a 200 μL
injection loop and passed through three Viscotek T6000M 10 μm
General Mixed Org columns in series in a 35 °C oven at a flow rate of
1.0 mL min−1. The analysis made use of the differential RI detector
and a viscometer. Molecular weights were assigned by way of a
universal calibration curve created with polystyrene standards ranging
from 580 g mol−1 to 3 114 000 g mol−1. The calibration was verified
through the analysis of a broad standard, SRM 706a, provided by the
National Institute of Standards and Technology.
RESULTS
■
The complete kinetic analysis for the zirconium-based systems
1a, 2a, and 3a has been reported in previous publications.16,17
Here we present the experimental data and a complete kinetic
analysis for 1-hexene polymerization by hafnium-based
analogues 1b, 2b, and 3b. For each system, we followed our
previously developed kinetic modeling method6,16,17 based on
the analysis of multiresponse data that includes (1) monomer
consumption, (2) MWD, (3) active site counts, and (4) vinyl
end group counts as measured by 1H NMR. We determine the
active site count at any point in the course of the reaction as the
number measured by quenching with methanol-d4 and
a
The ordinary differential equations (ODEs) that describe the mass-
action kinetics associated with this mechanism are provided in the
Supporting Information.
2
performing H NMR measurement of the concentration of
chains with deuterated end groups. The sites that have
undergone 1,2-insertion are defined as primary sites, and the
sites that have undergone 2,1-misinsertion are defined as
secondary sites. Within this analysis, each system is studied
independently, and no a priori assumptions are made with
respect to the elementary steps. As explained in detail in the
Supporting Information, the analysis procedure begins with the
most basic mechanism, i.e., initiation and propagation, and
fitting is attempted to the entire data set; only after a simple
mechanism is shown to fail, a new elementary step, e.g., chain
transfer, is added, and the fitting is attempted again.
Also it is noted that the catalyst participation may not be 100%
of the nominal precatalyst amount, and it may vary from system
to system and experiment to experiment. By catalyst
participation, here we mean the fraction of precatalyst that
can be activated and initiated once the reactant species are
combined. This is separate from time-dependent deactivation.
For the hafnium-based systems 1b, 2b, and 3b, the results of
the kinetic analysis are here presented. We chose the system 2b
to illustrate the quality of kinetic fitting. The similar figures for
C
dx.doi.org/10.1021/om4006005 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
systems 1b and 3b are in the Supporting Information. The
main conclusion is that the kinetic mechanism for hafnium-
based systems is essentially the same as for zirconium
analogues.
Hf−Pyridine Catalyst 2b. The experimental data along
with the kinetic modeling fits are presented in Figure 2.
The specific features of this system are as follows: (1)
Catalyst participation is nearly 100%. (2) In the case of the
batch scale experiments, significant catalyst deactivation is
observed as evidenced by bending of the monomer
consumption curve in Figure 2C and the steep decline in
primary active site counts over the course of the reaction in
Figure 2E. In the case of the NMR scale experiments, the
deactivation either does not occur or is much less significant.
For that reason, deactivation is not considered as part of the
catalytic reactions. (3) The amount of chain transfer is
relatively high as evidenced by the significant vinylidene
concentration in Figure 2F and the fact that the MWD does
not change much after 30% conversion of the monomer. The
vinylidene formation is via a monomer-independent reaction as
evidenced by the upward curvature in the vinylidene
concentration versus monomer conversion plot (Figure 2F).
(4) The vinylene end group concentration is much lower than
that of vinylidene (Figure 2F), where the vinylene formation is
via monomer-dependent reaction as evidenced by the linear
accumulation in Figure 2F.
Hf−THF Catalyst 1b. The experimental data along with the
kinetic modeling fits are presented in the Supporting
Information. The specific features of this system are as follows:
(1) Catalyst participation is approximately 50%. (2) Faster
chain transfer rate and slower propagation rate compared to its
zirconium analogue result in a much higher chain transfer
frequency (i.e., the measured vinyl terminated groups are 100
times higher at the end of the reaction). However, the chain
transfer rate of this catalyst remains the lowest compared to
catalyst 2b and 3b. (3) Fewer secondary sites are formed,
meaning there is less dormancy as compared to the zirconium
analogue. The vinylene count is quite small, indicating that the
actual chain transfer rate from secondary sites is negligible.
Hf−NMe2 Catalyst 3b. The experimental data along with
the kinetic modeling fits are presented in the Supporting
Information. The specific features of this system are as follows:
(1) Catalyst participation is approximately 40%. (2) There is a
decline in active catalyst sites over the course of the reaction,
although it is not as steep as in systems 1b and 2b. (3) No
secondary catalyst sites were measured, although a small
amount of vinylene end groups was detected. This peculiar
behavior was also observed for the EBIZrMe2/B(C6F5)3
catalyst.6,7 Vinylene is typically expected to form following
chain transfer of secondary sites. It is likely in this system that
secondary sites do form, but they rapidly undergo either chain
transfer or monomer-dependent recovery. Since no secondary
Figure 2. Multiresponse data set with fits for catalyst 2b. NMR-scale
experiments: (A) Monomer consumption. Data, symbols; fits, lines.
(B) MWDs at the end. {Blue, Red, Green}, [C]0 = {3.0, 3.0, 6.0} mM
and [M]0 = {0.30, 0.60, 0.60} M. Data, solid; fits, dashed. Batch scale
experiments ([C]0 = 3.0 mM, [C]0 = 0.60 M): (C−F). (C) Monomer
consumption. Data, symbols; fit, line. (D) MWDs at (solid) 1694 s,
(dashed) 4352 s, (dotted) 10963 s. Data, black; fits, magenta. (E)
Active site counts. Primary, filled circles (data)/solid line (fit);
secondary, open circles (data)/dashed line (fit). (F) End group
analysis. Filled circles (data)/solid line (fit), vinylidene; open circles
(data)/dashed line (fit), vinylene. In (A), black circles same as in (C)
for comparison.
a
Table 1. Rate Constants for 1-Hexene Polymerization with the M[t-Bu-ONXO]Bn2/B(C6F5)3 Catalysts 1a−3b
X
Zr-THF (1a)
2.37
0.08 (+0.02/−0.01) 0.04 (+0.02/−0.01)
Hf-THF (1b)
Zr-Pyr (2a)
Hf-Pyr (2b)
Zr-NMe2 (3a)
2.59
0.16 (+0.04/−0.02) 0.04 (+0.01/−0)
Hf-NMe2 (3b)
M−X/Å
2.33
2.51
2.47
2.56
ki/M−1 s−1
>0.05
0.0017 (+0.0002/−
0.0001)
kp/M−1 s−1
8.0 (+0.8/−0.2)
0.53 (+0.06/−0.06)
1.8 (+0.2/−0.1)
0.20 (+0/−0.02)
0.00028 (+0.00002/−0) 0.055 (+0.007/−
11 (+1/−1)
0.95 (+0.07/−0.09)
b
k
mis/M−1 s−1
rec/M−1 s−1
0.054 (+0.026/−
0.0081 (+0.0002/−
0.031 (+0.004/−
0.0012 (+0.0003/−
0.003)
0.001)
0.005)
0.004)
0)
k
0.047 (+0.021/−
0.06 (+0.004/−0.005) 0.028 (+0.004/−
0.0002 (+0/−0.0002)
3.8 (+0.3/−0.2)
0.04 (+0.03/−0.02) N/A
0.002)
0.005)
kvinylidene (10−3)/ 0.14 (+0.014/
0.84 (+0.02/−0.04)
0.27 (+0.07/−0.06)
2.4 (+0.1/−0.1)
12.2 (+0.8/−0.6)
5.5 (+0.2/−0.2)
s−1
−0.02)
b
kvinylene (10−3)/
0.051 (+0.002/−
0.65 (+0.06/−0.05) 2nd order
8.72 (+0.07/−0.04)
s−1
0.003)
a
b
In toluene at 25 °C. See Figure 1 for precatalyst structures and Scheme 1 for reactions steps. Errors are in parentheses. The misinsertion reaction
in the system 3b mechanism is followed immediately by monomer-independent β-H elimination to form vinylene.
D
dx.doi.org/10.1021/om4006005 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
sites are observed even late in the reaction when monomer
concentration is low, a fast monomer-independent chain
transfer event is more probable.
DISCUSSION
■
In this study, the complete set of kinetic rate constants for three
zirconium amine bis-phenolate catalyst systems and three
hafnium analogues have been presented. For each system, a rich
data set including MWD has been collected and successfully
fitted by comprehensive kinetic modeling. With one possible
exception, the mechanism of 1-hexene polymerization for these
catalysts (1a−3b) consists of the following elementary reaction
steps: initiation, normal propagation, misinsertion, recovery,
and chain transfer. For system 3b, there is not enough
information to include or exclude a recovery reaction.
Figure 3. Log(kvinylidene) vs M−X bond length.
The values of the rate constants are shown in Table 1
including error bounds, which were determined using the
methodology for determining error bounds discussed in the
previous paper.16 Examining the summarized kinetic data in
Table 1, the following conclusions emerge: (1) The monomer-
dependent rate constants ki, kp, kmis, and krec are slower for the
Hf systems than for the Zr systems. In particular, the
propagation rate is 1 order of magnitude slower in all the
hafnium-based systems. (2) kvinylidene, which is monomer-
independent chain transfer, is not uniformly slower for Hf
versus Zr. It depends on the pendant of the ligand. For
example, for the THF pendant (1a and b), kvinylidene for Hf is
larger than that for Zr, and the rate constants are comparable
for both metals in the case of the pyr pendant (2a and b). (3)
Vinylene formation does not behave consistently across all
pendants with Hf as it does for Zr. For Hf−Pyr it appears
second order; for Hf−NMe2 it is apparently fast (consistent
with fast kvinylidene). We do not currently have an explanation for
this behavior. (4) Each hafnium complex exhibits less
secondary site formation than its zirconium analogue.
A possible reason for the reduction in the rate of all
elementary steps that require the insertion of a monomer is due
to the larger metal−carbon bond enthalpy of the hafnium
systems as compared with the analogous zirconium systems.19
In our previous paper we pointed out a linear correlation
between the logarithm of the rate of monomer-independent
chain transfer and the bond distance between the zirconium
and the pendant group observed in the precatalyst.16 A similar
linear relationship appears to be holding for the monomer-
independent chain transfer rate for the hafnium-based systems
as shown in Figure 3. However, the hafnium-based system
exhibits a much weaker dependence on the bond length, as the
slope of this correlation is 2.7 times smaller. In our previous
study,16 we speculated that this increase in bond distance allows
for more steric freedom to accommodate the β-hydride agostic
interaction necessary for chain transfer to occur. Since the
effective size of the hafnium metal center is generally believed
to be similar to that of zirconium, it is unclear why this
correlation is weaker in hafnium-based systems. However, it is
likely that the exact reason lies with the intrinsic properties of
the metal center and how these properties control the β-
hydrogen transfer reactions.
reaction steps were determined for each system. The
mechanism includes initiation, normal propagation, misinser-
tion, recovery, and chain transfer. In conjunction with the
previous study of zirconium analogues, this report allows for
the first quantitative comparison between similarly ligated
hafnium and zirconium-based olefin polymerization catalysts.
The most important findings are as follows: the 1 order of
magnitude decrease in kp for the hafnium catalysts; an overall
decrease in all monomer-dependent reaction steps; and the
correlation between the logarithm of monomer-independent
chain transfer and the hafnium pendant ligand (Hf−X) bond
distance. The last observation is similar to the one previously
reported for zirconium systems, but in case of the hafnium
catalysts the dependence is 2.7 times weaker. However, it is also
interesting that there does not appear to be such a correlation
that can be drawn for the propagation rate constant. Subsquent
studies are ongoing to ascertain the dependence of kp on the
steric and electronic nature of the pendant.
ASSOCIATED CONTENT
* Supporting Information
■
S
The synthesis of all ligands and precatalysts, as well as a
complete set of experimental procedures for each system and
kinetic modeling material, and crystal data (CIF). These
materials are available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: mabuomar@purdue.edu; caruther@purdue.edu.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support was provided by the U.S. Department of
Energy by Grant No. DE-FG02-03ER15466. This research was
supported in part by the National Science Foundation through
TeraGrid resources provided by Purdue University under Grant
No. TG-CTS070034N. Computing resources were also
provided by Information Technology at Purdue.
CONCLUSIONS
■
REFERENCES
A comprehensive kinetic study of three catalytic systems based
on hafnium amine bis-phenolate complexes has been
completed, and the relevant rate constants and elementary
■
(1) World Polyolefins Analysis, Chemical Market Associates, Inc.
(CMAI), 2005; www.cmaiglobal.com.
E
dx.doi.org/10.1021/om4006005 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(2) (a) Chen, E. Y-X.; Marks, T. J. Chem. Rev. 2000, 100, 1391−1434.
(b) Li, H.; Marks, T. J. Proc. Natl. Acad. Sci. U. S. A. 2006, 103,
15295−15302.
(3) Manz, T. A.; Phomphrai, K.; Medvedev, G. A.; Krishnamurthy, B.
B.; Sharma, S.; Haq, J.; Novstrup, K. A.; Thomson, K. T.; Delgass, W.
N.; Caruthers, J. M.; Abu-Omar, M. M. J. Am. Chem. Soc. 2007, 129,
3776−3777.
(4) (a) Krauledat, H.; Brintzinger, H. H. Angew. Chem., Int. Ed. 1990,
29, 1412−1413. (b) Piers, W. E.; Bercaw, J. E. J. Am. Chem. Soc. 1990,
112, 9406−9707. (c) Coates, G. W.; Waymouth, R. M. J. Am. Chem.
Soc. 1991, 113, 6270−6271.
(5) Arriola, D. J.; Carnahan, E. M.; Hustad, P. D.; Kuhlman, R. L.;
Wenzel, T. T. Science 2006, 312, 714−719.
(6) Novstrup, K. A.; Travia, N. E.; Medvedev, G. A.; Stanciu, C.;
Switzer, J. M.; Thomson, K. T.; Delgass, W. N.; Abu-Omar, M. M.;
Caruthers, J. M. J. Am. Chem. Soc. 2010, 132, 558−566. Caruthers, J.
M.; Abu-Omar, M. M. J. Am. Chem. Soc. 2007, 129, 3776−3777.
(7) Landis, C. R.; Rosaaen, K. A.; Sillars, D. R. J. Am. Chem. Soc.
2003, 125, 1710−1711.
(8) (a) Angermund, K.; Fink, G.; Jensen, V. R.; Kleinschmidt, R.
Chem. Rev. 2000, 100, 1457−1470. (b) Bochmann, M. J. Organomet.
Chem. 2004, 689, 3982−3998. (c) Mohring, P. C.; Coville, N. J. Coord.
Chem. Rev. 2006, 250, 18−35. (d) Wang, B. Coord. Chem. Rev. 2006,
250, 242−258.
(9) (a) Gauthier, W. J.; Corrigan, J. F.; Taylor, N. J.; Collins, S.
Macromolecules 1995, 28, 3771−3778. (b) Razavi, A.; Ferrara, J. J.
Organomet. Chem. 1992, 435, 299−310.
(10) Resconi, L.; Piemontesi, F.; Franciscono, G.; Abist, L.; Fioranit,
T. J. Am. Chem. Soc. 1992, 114, 1025−1032.
(11) Abbas, R.; Ferrara, J. J. Organomet. Chem. 1992, 435, 299−310.
(12) Tshuva, E. Y.; Goldberg, I.; Kol, M.; Goldschmidt, Z.
Organometallics 2001, 20, 3017−3028.
(13) Tshuva, E. Y.; Groysman, S.; Goldberg, I.; Kol, M.;
Goldschmidt, Z. Organometallics 2002, 21, 662−670.
(14) Kerton, F. M.; Holloway, S.; Power, A.; Soper, R. G.; Sheridan,
K.; Lynam, J. M.; Whitwood, A. C.; Willans, C. E. Can. J. Chem. 2008,
86, 435−443.
(15) Groysman, S.; Goldberg, I.; Kol, M.; Genizi, E.; Goldschmidt, Z.
Organometallics 2003, 22, 3013−3015.
(16) Steelman, D. K.; Xiong, S.; Pletcher, P. D.; Smith, E.; Switzer, J.
M.; Medvedev, G. A.; Delgass, W. N.; Caruthers, J. M.; Abu-Omar, M.
M. J. Am. Chem. Soc. 2013, 135, 6280−6288.
(17) Switzer, J. M.; Travia, N. E.; Steelman, D. K.; Medvedev, G. A.;
Thomson, K. T.; Delgass, W. N.; Abu-Omar, M. M.; Caruthers, J. M.
Macromolecules 2012, 45, 4978−4988.
(18) Liu, Z.; Somsook, E.; Landis, C. R. J. Am. Chem. Soc. 2001, 123,
2915−2916.
(19) Kissounko, D. A.; Zhang, Y.; Harney, M. B.; Sita, L. R. Adv.
Synth. Catal. 2005, 347, 426−432.
F
dx.doi.org/10.1021/om4006005 | Organometallics XXXX, XXX, XXX−XXX