Switching the Reactivity of Palladium Diimines with “Ancillary” Ligand to Select between Olefin Polymerization, Branching Regulation, or Olefin Isomerization

Article abstract of DOI:10.1002/anie.202012400

Coordinating solvents are commonly employed as ancillary ligands to stabilize late transition metal complexes and are conventionally considered to have little effect on the reaction products. Our work identifies that the presence of ancillary ligand in Pd-diimine catalyzed polymerizations of α-olefins can drastically alter reactivity. The addition of different amounts of acetonitrile allows for switching between distinct reaction modes: isomerization–polymerization with high branching (0 equiv.), regular chain-walking polymerization (1 equiv.), and alkene isomerization with no polymerization (>20 equiv.). Optimization of the isomerization reaction mode led to a general set of conditions to switch a wide variety of diimine complexes into efficient alkene isomerization catalysts, with catalyst loading as low as 0.005 mol %.

Full text of DOI:10.1002/anie.202012400

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Switching the Reactivity of Palladium Diimines with ‘Ancillary’
Ligand to Select for Olefin Polymerization, Branching Regulation,
or Olefin isomerization
Authors: Eva Harth, Glen Jones, Hatice Basbug Alhan, Lucas Karas,
and Judy Wu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202012400
Link to VoR: https://doi.org/10.1002/anie.202012400

10.1002/anie.202012400
Angewandte Chemie International Edition
COMMUNICATION
Switching the Reactivity of Palladium Diimines with ‘Ancillary’
Ligand to Select for Olefin Polymerization, Branching Regulation,
or Olefin isomerization
Glen R. Jones,[a]† Hatice E. Basbug Alhan,[a]† Lucas J. Karas,[a] Judy I. Wu,[a] and Eva Harth[a]*
[a]
Dr Glen R. Jones, Hatice E. Basbug Alhan, Lucas J. Karas, Prof Judy I. Wu, Prof Eva Harth
Department of Chemistry, Center of Excellence in Polymer Chemistry
University of Houston
3585 Cullen Blvd., Houston, Texas, USA, 77004
E-mail: harth@uh.edu
† These authors contributed equally
Supporting information for this article is given via a link at the end of the document
Abstract: Coordinating solvents are commonly employed as ancillary
ligands to stabilize late transition metal complexes and are
conventionally considered to have little effect on the reaction products.
Our work identifies that the presence of ancillary ligand in Pd-diimine
catalyzed polymerizations of α-olefins can drastically alter reactivity.
The addition of different amounts of acetonitrile allows for switching
between distinct reaction modes: isomerization-polymerization with
high branching (0 eq.), regular chain-walking polymerization (1 eq.),
and alkene isomerization with no polymerization (>20 eq.).
Optimization of the isomerization reaction mode led to a general set
of conditions to switch a wide variety of diimine complexes into
efficient alkene isomerization catalysts, with catalyst loading as low
as 0.005 mol%.
Scheme 1. Typical structures of late transition metal catalysts for olefin
polymerization showing a bidentate ligand in blue and an ancillary ligand in red.
Counterions for the cationic complexes 1 and 3 are not shown.
Acetonitrile has previously been found to have a very similar
coordination strength to ethylene[4] (Keq(-85°C) = (4.6 ± 0.9) × 10-
2) when complexed with a palladium diimine catalyst,[11] indicating
that under typical polymerization conditions where monomer is in
a large excess, a nitrile complex should have at most only a
moderate inhibition on the rate of propagation. This decrease in
turnover frequency (TOF) has also been reported for a higher α-
olefin (1-hexene).[12] Polymerizing 1-hexene in the presence of
nitriles was found to decrease the TOF of the catalyst with a larger
effect seen for more strongly donating nitriles. Similarly, using
increasing concentrations of 3,5-bis(trifluoromethyl)benzonitrile
gave decreasing molecular weights over a 3-hour reaction period.
The dispersity of the obtained polymers was low (>1.10),
indicating that nitriles do not promote chain transfer and a living
polymerization can be maintained. No significant differences
between polymer samples apart from the lower rate in the
presence of ancillary ligands were reported, which was attributed
to an active/dormant equilibrium whereby the acetonitrile adduct
is not capable of undergoing polymerization.[12]
Late transition metal catalysts for olefin polymerization are
valued for their polar monomer tolerance and the ability to control
branching through a ‘chain-walking’ mechanism.[1] Such catalysts
are typically cationic or neutral palladium or nickel complexes
employing bidentate or multidentate ligands bound to the active
site. These ligands alter the electronic and steric environment and
can be tuned to adjust the extent of β-hydride elimination and
reinsertion, which are the key steps in branch formation through
the ‘chain-walking’ mechanism,[2] and the electrophilicity of the
active site (crucial for polar monomer tolerance.) Examples of this
tuning can lead to high molecular weight polyethylene with low
branching,[3] regular branching structure,[4] hyperbranched
polyolefins,[5] and, if hydride elimination is sufficiently favourable,
catalysts capable of alkene isomerization without the propensity
to undergo polymerization.[6]
Recent work has suggested that the absence of ancillary
ligands could be key to internal alkene polymerizability; observed
Ancillary ligands, commonly a coordinating solvent such as
acetonitrile (NCMe), are often employed to stabilize late transition
metal complexes. For example, nitriles are used with palladium
diimines in order to yield a stable precatalyst complex (1, scheme
1), where the ancillary ligand will occupy the vacant coordination
site (prior to exposure to monomer) and prevent the formation of
an unstable 14 electron complex.[7] Other examples of late
transition metal complexes employing ancillary ligands are also
shown including a Drent-type neutral phosphine-sulfonato Pd(II)
complex[1d, 8] (2), a cationic biphosphine monoxide palladium(II)
complex[9] (3), and a neutral salicylaldiminato Ni(II) complex[3, 10]
(4).
in
a
study which increased the branching of α-olefin
polymerizations with a palladium diimine catalyst through a Lewis
acid triggered isomerization of 1-alkenes to internal alkenes which
can then polymerize with high branching (isomerization-
polymerization).[13] The proposed mechanism states that the
Lewis acid undergoes a counterion swap to a complex which
favors isomerization as well as forming an adduct with the
ancillary nitrile ligand and thereby prevent competitive binding
with internal alkenes. An ancillary ligand effect has also been
seen in ethylene polymerizations with neutral salicylaldiminato
Ni(II) complexes: the temperature dependence on the degree of
branching changes according to the lability of the ancillary ligand
(1.1 to 7.3/1000 carbons). Density functional theory (DFT) studies
on a related complex suggested that the overall barrier to β-
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202012400
Angewandte Chemie International Edition
COMMUNICATION
hydride elimination is lower for an ancillary ligand adduct
compared to an ethylene adduct.[14]
Time
0 eq. NCMe[a]
46% Polymer
1 eq. NCMe[a]
20 eq.NCMe[a]
0% Polymer
Conv. [b]
16% Polymer
Herein we report that acetonitrile ancillary ligand can be used
to drastically tune the reactivity of palladium diimines to select for
distinct reaction modes. Polymerization in the absence of ancillary
ligand yields rapid isomerization of α-olefins to internal alkenes
which then polymerize to give a highly branched polymer. 1 eq. of
ancillary ligand with respect to palladium preferentially
polymerizes α-olefin, with any isomers formed polymerizing at a
much lower rate. Adding an excess of ancillary ligand (20eq. with
respect to palladium) transforms the system into an alkene
isomerization catalyst without any observable polymerization.
Optimization of the isomerization pathway in the presence of
excess acetonitrile leads to a general set of experimental
conditions which can be employed to switch a wide range of
palladium diimine polymerization precatalysts into effective
isomerization catalysts.
5
mins
16% 1-hex
32% 2-hex
6% 3-hex
75% 1-hex
6% 2-hex
3% 3-hex
82% 1-hex
13% 2-hex
5% 3-hex
Remaining
monomer
Conv.[b]
85% Polymer
52% Polymer
0% Polymer
1
hour
0% 1-hex
14% 2-hex
1% 3-hex
25% 1-hex
19% 2-hex
4% 3-hex
27% 1-hex
60% 2-hex
13% 3-hex
Remaining
monomer
Conv.[b]
97% Polymer
71% Polymer
0% Polymer
3
0% 1-hex
2% 2-hex
1% 3-hex
1% 1-hex
23% 2-hex
5% 3-hex
5% 1-hex
75% 2-hex
20% 3-hex
hours
Remaining
monomer
Branching[c]
128
84
N/A
[a] Room temperature, 2mL of 1-hex in 6mL of dichloromethane with 0.2mol%
palladium diimine catalyst, 0.2 mol% NaBArF4. Equivalencies of NCMe are
calculated with respect to Pd catalyst. [b] Conversion calculated by 1H NMR in
CDCl3. [c] Branching is given per 1000 carbon atoms and was determined by
quantitative 13C NMR.
To test this hypothesis, we conducted a series of experiments with
varying amounts of acetonitrile (table 1), analysing the extent of
polymerization and isomerization in each case.
Polymerization of 1-hexene from the Pd methyl chloride 5,
activated with NaBArF in the absence of acetonitrile proceeds
4
with significant isomerization, scheme 2, table 1. Sampling after 5
minutes revealed that the reaction had achieved 46% conversion
to polymer. Analysis of the sample shows that only 16% is the
starting material, 1-hexene, with the rest having been isomerized
to internal alkenes (32% 2-hexene and 6% 3-hexene). A
conventional understanding of Pd-diimine catalysts would
suggest that the internal alkenes generated in the isomerization
side reaction are too weakly coordinating to polymerize at an
appreciable rate.[12, 18] However, after 1 hour no 1-hexene remains
in the reaction (85% polymer) and after 3 hours a significant
amount of the internal alkenes have successfully been
incorporated (97% polymer, Mn = 48600 Da, Đ = 1.18, S10.5,
table S3), suggesting that the NCMe present in a typical
polymerization significantly inhibits internal alkene coordination.
Branching analysis of the precipitated polymer showed 128
branches per 1000 carbon atoms, with carbon NMR showing
significant amounts of ethyl and propyl branches indicating that 2-
hexene and 3-hexene have been polymerized (S10.4, table S2,
figure S4). Overall, monomer isomerization appears to be favored
over polymerization, giving isomers which then polymerize to give
a highly branched polymer. In-situ activation of a Pd-diimine
Scheme 2. Switching reactivity of a palladium diimine by activation with varying
amounts of NCMe. 0 eq. results in rapid isomerization of 1-hexene and
polymerization of the resultant internal hexenes.
1 eq. of NCMe gives
polymerization with very low rates of internal alkene polymerization. 20 eq. of
NCMe gives rapid isomerization and no measurable polymerization.
Isomerization is
polymerizing α-olefins with Pd-diimine catalysts, occurring
through β-hydride elimination and subsequent reinsertion.[15]
a
well-known side reaction when
A
typical reaction with catalyst 1 will yield around 71% polymer and
28% internal alkenes in 3 hours (1% 1-alkene left), with branching
analysis by 13C NMR (S10.3, figure S3) showing methyl and butyl
branching (S11.1, scheme S4), indicative of chain-walking
polymerization of 1-hexene.[5, 16] Allowing the reaction to proceed
for an additional 18 hours shows a slight decrease in the internal
olefin signals via 1H NMR and the appearance of ethyl and propyl
branches (characteristic of chain-walking polymerization of 2 and
3-hexenes) (S11.2, scheme S5; S11.3, scheme S6) by
quantitative 13C NMR (S10.3, table S1, figure S3). [13, 17]
precursor with NaBArF for 1-hexene polymerization has
4
Previous studies into the effect of ancillary ligands on Pd-
diimine catalyzed reactions showed only an inhibition on the rate
of polymerization.[12] However, the removal of ancillary ligand
through reaction with a Lewis acid has been implicated as being
crucial to increasing the polymerizability of internal alkenes.[13] We
reasoned that the presence and concentration of a labile ligating
species could play a crucial role in the reactivity of Pd-diimines.
previously been reported in the literature with similar ethyl and
propyl branches found in the resultant polymer, however the
authors attribute this branching to ligand electronic effects and do
not report a crude NMR.[19] Given our results, it seems likely that
these branches are also due to isomerization of 1-hexene to
internal alkenes followed by polymerization. Repeating the
reaction of 5 activated with NaBArF with the addition of 1
4
equivalent of acetonitrile (table 1, scheme 2) gives a drastically
different result. Activation in the presence of acetonitrile gives the
Table 1. Reaction of 1-hexene with palladium diimine catalyst (5), with varying
concentration of ancillary ligand, NCMe.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202012400
Angewandte Chemie International Edition
COMMUNICATION
Scheme 3. Left: Proposed mechanistic details of the switch in reactivity of palladium diimines in the presence of varying amounts of ancillary ligand, NCMe. Right:
Calculated free energies of alkene displacement with acetonitrile from Pd-alkyl and Pd-hydride complexes computed at the PCM-B3LYP-D3/6-31+G(d)-
LANL2DZ(Pd) level, in implicit dichloromethane solvent
.
well-known isolable precatalyst 1, and polymerization results
agree with this, showing a typical polymerization of 1-hexene as
previously reported in various publications.[12, 20] After 5 minutes
16% conversion to polymer is seen, with the remaining monomer
being ~75% 1-hex and 9% isomers. The polymerization of 1-
hexene proceeds to 71% (Mn = 34200, Đ = 1.06, S10.5, table S3)
with only 1% 1-hex remaining after 3 hours, the rest having been
converted to isomers. Branching analysis of this polymer shows
~84 branches per 1000 carbons (S10.4, table S2, figure S4)., and
a branching distribution of methyl and butyl branches, consistent
ancillary ligand concentration.[12] However, the decreased rate of
polymerization is only one effect of the increased ancillary ligand.
Despite inhibiting polymerization, isomerization still occurs. For
example, addition of 4 eq. of acetonitrile with respect to Pd
catalyst 5 gives just 9% polymer in three hours, with 54% of the
remaining monomer isomerized to internal alkenes (S10.6, table
S4). Further increasing acetonitrile to 20 eq. gives no observable
polymerization in 3 hours, hence transforming the reaction into an
alkene isomerization (95% conversion, table 1). Under these
conditions the ratio of product formed is 77:23 2-hexene-3-
hexene with an E:Z ratio of 3:1 for the 2-hexene. This result is
significant as previous attempts to utilize Pd-diimines as
isomerization catalysts necessitated bulky ‘sandwich’ ligands to
increase steric hindrance at the reactive center to result in weaker
binding affinity of internal olefins and an external hydride source
to form a Pd-H species.[6a] A similar polymerization inhibition can
be seen when more strongly binding ancillary ligands are used.
Addition of 1 eq. of dimethyl sulfide or pyridine exhibits no
polymerization over 3 hours but still shows some isomerization
(28% and 7%, respectively). Increasing the amounts of these
ligands to 20 eq. showed no polymerization or isomerization for
pyridine, and 9% isomerization for dimethyl sulfide.
Mechanistically we propose that higher concentrations of
acetonitrile will favor isomerization over polymerization through
displacement of coordinated alkenes (K5, K6, K7, K8), as shown in
scheme 3. The agostic complex 6 can either undergo β hydride
elimination, coordination of an alkene, or coordination of
acetonitrile. High concentrations of acetonitrile would shift the
equilibrium K4 to the right, slowing polymerization and
isomerization which is in agreement with the data in table 1
showing fast isomerization and polymerization when no ancillary
with 1,2 and 2,1 insertion and chain-walking of 1-hexene only.[16b,
21]
These results show that the presence of an ancillary ligand
has a large effect on Pd-diimine catalysts, enabling a switch in
reactivity between an isomerization-polymerization (0 eq., where
monomer is isomerized to internal alkenes which can then
polymerize) and a regular 1-hexene polymerization (1 eq., 1-
hexene is polymerized preferentially over internal alkenes.)
Furthermore, what is thought of as typical reactivity of Pd-diimines
with α-olefins (polymerization with isomerization as a side
reaction) is shown to be largely a consequence of the ‘ancillary’
ligand added to give a stable isolable catalyst. The difference in
the polymeric products between reactions with 0 eq. vs. 1 eq. of
ancillary ligand can be seen in the thermal properties of the
polymer; the higher branching imparted by internal alkene
polymerization gives a highly amorphous polymer (Tg = -70°C),
whereas exclusive 1-alkene polymerization allowed for ‘chain-
straightening’ to yield linear segments with some crystallinity (Tm
= 6°C).
Further increasing the amount of acetonitrile gives a lower
rate of polymerization, in agreement with a previous study into
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202012400
Angewandte Chemie International Edition
COMMUNICATION
generates anionic [AlCl4]- capable of salt metathesis with the
[BArF ]- anion. To test if an [AlCl4]- counterion complex gives a
4
Scheme 4. Activation of 5 with NaBArF or AlCl3. The tetrachloroaluminate
4
counterion gives a higher rate of isomerization: quantitative conversion in just 5
minutes (0.2 mol% catalyst loading) with no observable polymer peaks by 1H
NMR.
Scheme 5. Isomerization of α-olefin with a variety of Pd complexes including
sterically hindered and unhindered Pd-diimines and Pd(COD)MeCl.
ligand is used, and slower isomerization when 20 eq. are added.
However, displacement from the alkyl palladium complexes 8 and
9 (yielding complex 7) is expected to be more favourable than
from hydride complexes 10 and 12 (yielding 11) to due steric
factors. Therefore, polymerization would be inhibited more than
isomerization when [NCMe] is high, accounting for the
observation that isomerization is the sole pathway for reactions
employing 20 eq. of ancillary ligand. Computed Gibbs free
higher rate of isomerization we conducted an experiment where
precatalyst 5 was activated with the Lewis acid, AlCl3, to directly
obtain an [AlCl4]- counterion derivative, 13, of complex 1 (scheme
4).
Addition of 1-hexene and 20 eq. of acetonitrile to this complex
gave a sharp increase in the rate of isomerization compared to
the reaction with complex 1, with quantitative conversion to
internal alkenes observed in just 5 minutes at 0.2 mol% catalyst
loading. The higher isomerization rate of complex 13 implicates
energies,
performed
at
PCM-B3LYP-D3/6-31+G(d)-
LANL2DZ(Pd) in implicit dichloromethane solvent (scheme 3,
right), suggest that it is easier to displace alkenes from alkyl rather
than hydride complexes.
counterion ion exchange from [BArF ]- to [AlCl4]- as a key
4
mechanistic step in AlCl3 triggered branching regulation.[13]
The quantitative conversion observed with complex 13 in 5
minutes suggests that the complex is highly active and that much
lower catalyst concentrations could be employed. To test this we
reduced the catalyst loading, with entries 1-3 in table 2 showing
that quantitative isomerization can be obtained in 18 hours with
catalyst loadings as low as 0.005 mol%, corresponding to a as
well as providing further evidence of the dual action of AlCl3 in
branching regulation (removing NCMe and forming a new
1 equivalent of ancillary ligand is enough to greatly inhibit
internal alkene polymerization, meaning that only 1-hexene can
efficiently coordinate and insert to yield a polymer, hence the
characteristic methyl and butyl branching distribution (see
supporting information for an explanation of branch formation
from 1-alkenes vs. internal alkenes S11.1, scheme S4; S11.2,
scheme S5; S11.3, scheme S6). Any internal alkene that is
generated will polymerize but at a greatly reduced rate. In the
absence of ancillary ligand any isomer generated is more readily
polymerized. We postulate that this would drive the isomerization
equilibrium to form more isomer, hence the branching structure of
the polymer contains higher branching overall because internal
alkenes cannot form linear segments, instead forming a higher
number of ethyl and propyl branches (S10.4, table S2, figure S4).
The higher rate of polymerization observed in the case of 0 eq. vs
1 eq. can again be explained through the active/dormant
equilibrium (K4) in scheme 3.
Table 2. Isomerization including catalyst scope and loading.
Cat.
Loading
(mol%)
Entry
Cat.
Alkene
Product[a]
Yield[b]
1
2[c]
3[c]
4
5
5
1-hexene
1-hexene
2-hexene
2-hexene
77
77
0.2
0.01
0.005
0.2
Reacting 1-hexene with precursor 5 in the absence of ancillary
ligand yields fast isomerization to internal alkenes which can be
polymerized to give a highly branched polymer. This system is
similar to the previously reported system using a Lewis acid, AlCl3,
to alter the reactivity of palladium diimine catalysts to isomerize
then polymerize internal olefins.[13] However, the reaction without
aluminium chloride shows slightly lower branching (128 vs 150
branches/1000C). The reaction with 0 eq. of NCMe is slower to
isomerize than an AlCl3 reaction, meaning that some 1-hexene
could be incorporated in the early stages of the reaction therefore
lowering the branching. Coordination disproportionation, a well-
studied reaction of AlCl3 with NCMe in dilute solutions, yields
cationic [AlCl2(NCMe)4]+ and [AlCl4]- (S10.7, scheme S3).[22] The
reaction of AlCl3 with NCMe can be observed visually, where AlCl3
coagulates in dilute solutions of NCMe (S10.7, figure S11). An
5
1-hexene
2-hexene
81
14
15
16
17
5
1-hexene
2-hexene
76
5
1-hexene
2-hexene
75
0.2
6
1-hexene
2-hexene
76
0.2
7
1-hexene
2-hexene
80
0.2
8[d][e]
9[d][e]
10
1-decene
2-decene
62.5
53
0.2
5
1-octadecene
Allyl benzene
2-octadecene
Methylstyrene
0.2
5
95
0.2
4-pentene-
OTBS
3-pentene-
OTBS
11[e]
5
90
0.2
1
upfield shift of the NCMe protons can also be seen in H NMR,
[a] Reactions carried out at toom temperature; 2mL of monomer in 6mL of
dichloromethane with Pd-diimine catalyst, 4eq AlCl3, 20 eq of NCMe with
respect to Pd catalyst, 3 hours standard reaction time. Product shown is the
major component found in the reaction. [b] Yield calculated from 1H NMR. [c] 18
suggesting complexation. This process serves to sequester
NCMe (giving conditions similar to 0 eq. NCMe, table 1), but also
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202012400
Angewandte Chemie International Edition
COMMUNICATION
hour reaction time, 20 mL monomer in 60 mL of dichloromethane used.[d]
Reactions in chlorobenzene. [e] 4 eq. NaBArF4 used as activator.
Acknowledgements
The authors thank the Robert A. Welch Foundation for generous
support of this research (#H-E-0041) through the Center of
Excellence in Polymer Chemistry. JW thanks the National
Institute of General Medical Sciences of the National Institute of
Health (R35GM133548), and the Alfred P. Sloan Foundation (FG-
2020-12811) for grant support. We acknowledge the use of the
Sabine cluster and support from the Research Computing Data
Core at the University of Houston.
counterion via
a
disproportionation reaction), the fast
isomerization seen from the [AlCl4]- counterion complex also
shows potential as a general method for transforming palladium
diimine catalysts into highly efficient alkene isomerization
catalysts without the need for specialist ligand design or an
external hydride source to form the Pd-H species. To investigate
this further we utilized four other palladium diimine complexes
with different ligand structures (scheme 5) and subjected them to
conditions under which polymerization should be inhibited and
isomerization enhanced (activation with AlCl3 to form the [AlCl4]-
counterion in the presence of excess ancillary ligand (20 eq. of
NCMe with respect to Pd cat.)). Complex 15 contains methyl
substituents, meaning there is very little steric hindrance around
Keywords: branching • chain-walking • olefins • isomerization •
polymers
the palladium center, whereas compound 14 contains
a
References
naphthalene backbone and large aryl substituents,[23] complex 16
is the highly robust phenyl substituted catalyst first reported by
Chen for polymerization of ethylene at elevated temperatures.[24]
The precursor complexes were first activated with AlCl3 prior to
addition of ancillary ligand to prevent coordination
disproportionation. Results show that all catalysts, despite the
differences in sterics and electronics, fully isomerize 1-hexene to
internal alkenes with no detectable polymerization (entries 4-7,
table 2). Isomerization under similar conditions was also achieved
for chloro(1,5-cyclooctadiene)methylpalladiium(II), 17, the
precursor used to synthesize the Pd-diimines used in this study.
The general conditions for alkene isomerization are also shown to
give 2-alkenes as the major product for reactions with higher
olefins; 1-decene and 1-octadecene (entries 8 and 9, table 2).
Allyl benzene is converted to β-methylstyrene in good yield (entry
10, table 2). Isomerization of a sillyl ether protected alcohol (tert-
[1]
aB. P. Carrow, K. Nozaki, Macromolecules 2014, 47,
2541-2555; bF. Wang, C. Chen, Polymer Chemistry 2019,
10, 2354-2369; cA. Keyes, H. E. Basbug Alhan, E.
Ordonez, U. Ha, D. B. Beezer, H. Dau, Y. S. Liu, E.
Tsogtgerel, G. R. Jones, E. Harth, Angewandte Chemie
International Edition 2019, 58, 12370-12391; dS. D. Ittel,
L. K. Johnson, M. Brookhart, Chemical Reviews 2000,
100, 1169-1204; eL. Guo, W. Liu, C. Chen, Materials
Chemistry Frontiers 2017, 1, 2487-2494.
[2]
[3]
[4]
[5]
[6]
S. Biswas, Comments on Inorganic Chemistry 2015, 35,
300-330.
P. Kenyon, M. Wꢀrner, S. Mecking, Journal of the
American Chemical Society 2018, 140, 6685-6689.
Y. Zhang, C. Wang, S. Mecking, Z. Jian, Angewandte
Chemie International Edition 2020.
L. K. Johnson, C. M. Killian, M. Brookhart, Journal of the
American Chemical Society 1995, 117, 6414-6415.
aA. L. Kocen, K. Klimovica, M. Brookhart, O. Daugulis,
Organometallics 2017, 36, 787-790; bA. L. Kocen, M.
Brookhart, O. Daugulis, Chemical Communications 2017,
53, 10010-10013.
butyldimethyl(pent-4-en-1-yloxy)silane
was
selectively
isomerized to tert-Butyldimethyl(pent-3-en-1-yloxy)silane (entry
11, table 2) although required NaBArF4 as the activating agent.
[7]
[8]
D. H. Camacho, Z. Guan, Chemical communications
2010, 46, 7879-7893.
Conclusion
aE. Drent, R. van Dijk, R. van Ginkel, B. van Oort, R. I.
Pugh, Chemical communications 2002, 744-745; bS. Ito,
M. Kanazawa, K. Munakata, J.-i. Kuroda, Y. Okumura, K.
Nozaki, Journal of the American Chemical Society 2011,
133, 1232-1235.
We have shown that ancillary ligands commonly believed to
be benign actually have a profound effect on the reactivity of Pd-
diimine catalysts, revealing that the reactivity associated with
nitrile adduct Brookhart catalysts can be largely attributed to the
presence of this ligand. Varying the amount of ancillary ligand
added was found to switch reactions between fast isomerization
and polymerization, polymerization of α-olefins only, or exclusive
isomerization by simply varying the amount of ligand added. The
presence of NCMe disfavors polymerization of internal alkenes
through competition for open coordination sites. Computed Gibbs
free energies for key mechanistic steps show that acetonitrile
displacement is favored for Pd-alkyl complexes over Pd-hydride
species, meaning that high [NCMe] will drive the reaction to form
isomers in the absence of polymerization. Optimization and
investigation into counterion effects on the alkene isomerization
pathway also lead to the discovery that isomerization can be
further favored by employing an [AlCl4]- counterion, yielding
efficient isomerization even with catalyst loading as low as 0.005
mol%. Furthermore, activation with AlCl3 and 20 equivalents of
NCMe is shown to yield isomerization (free of polymerization) for
a wide range of both sterically hindered and unhindered Pd-
diimines and Pd(COD)MeCl, without an external hydride source.
[9]
aB. P. Carrow, K. Nozaki, Journal of the American
Chemical Society 2012, 134, 8802-8805; bI. Brassat, W.
Keim, S. Killat, M. Möthrath, P. Mastrorilli, C. F. Nobile, G.
P. Suranna, Journal of Molecular Catalysis A: Chemical
2000, 157, 41-58.
[10]
aI. Göttker-Schnetmann, B. Korthals, S. Mecking, Journal
of the American Chemical Society 2006, 128, 7708-7709;
bN. A. Kolhatkar, A. M. Monfette, S. Lin, M. J. Miri, Journal
of Polymer Science Part A: Polymer Chemistry 2012, 50,
986-995.
[11]
S. Mecking, L. K. Johnson, L. Wang, M. Brookhart,
Journal of the American Chemical Society 1998, 120, 888-
899.
[12]
[13]
[14]
[15]
A. C. Gottfried, M. Brookhart, Macromolecules 2003, 36,
3085-3100.
H. E. Basbug Alhan, G. R. Jones, E. Harth, Angewandte
Chemie International Edition 2020, 59, 4743-4749.
P. Kenyon, L. Falivene, L. Caporaso, S. Mecking, ACS
Catalysis 2019, 9, 11552-11556.
D. J. Tempel, L. K. Johnson, R. L. Huff, P. S. White, M.
Brookhart, Journal of the American Chemical Society
2000, 122, 6686-6700.
[16]
aL. Guo, S. Dai, X. Sui, C. Chen, ACS Catalysis 2016, 6,
428-441; bE. F. McCord, S. J. McLain, L. T. J. Nelson, S.
D. Ittel, D. Tempel, C. M. Killian, L. K. Johnson, M.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202012400
Angewandte Chemie International Edition
COMMUNICATION
Brookhart, Macromolecules 2007, 40, 410-420; cJ. Merna,
J. Cihlar, M. Kucera, A. Deffieux, H. Cramail, European
Polymer Journal 2005, 41, 303-312.
[17]
[18]
aF. Wang, R. Tanaka, Z. Cai, Y. Nakayama, T. Shiono,
Polymer 2017, 127, 88; bJ. D. Azoulay, G. C. Bazan, G. B.
Galland, Macromolecules 2010, 43, 2794-2800.
S. J. McLain, J. Feldman, E. F. McCord, K. H. Gardner, M.
F. Teasley, E. B. Coughlin, K. J. Sweetman, L. K.
Johnson, M. Brookhart, Macromolecules 1998, 31, 6705-
6707.
[19]
[20]
X. Sui, C. Hong, W. Pang, C. Chen, Materials Chemistry
Frontiers 2017, 1, 967-972.
A. Keyes, H. E. Basbug Alhan, U. Ha, Y.-S. Liu, S. K.
Smith, T. S. Teets, D. B. Beezer, E. Harth,
Macromolecules 2018, 51, 7224-7232.
[21]
[22]
C. M. Killian, D. J. Tempel, L. K. Johnson, M. Brookhart,
Journal of the American Chemical Society 1996, 118,
11664-11665.
aW. Libus, D. Puchalska, The Journal of Physical
Chemistry 1967, 71, 3549-3554; bM. Dalibart, J.
Derouault, P. Granger, S. Chapelle, Inorganic Chemistry
1982, 21, 1040-1046; cC. Schmulbach, Journal of
Inorganic and Nuclear Chemistry 1964, 26, 745-749.
A. Keyes, H. Dau, H. E. B. Alhan, U. Ha, E. Ordonez, G.
R. Jones, Y.-S. Liu, E. Tsogtgerel, B. Loftin, Z. Wen,
Polymer Chemistry 2019, 10, 3040-3047.
[23]
[24]
S. Dai, X. Sui, C. Chen, Angewandte Chemie 2015, 127,
10086-10091.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202012400
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents
The ancillary ligand often added to stabilize late transition metal catalysts for olefin polymerization actually has a profound effect on the
reactivity of such complexes. We demonstrate that Pd-diimine catalysts can be switched between 3 reactions modes simply by varying
the amount of ancillary ligand added to give either: isomerization-polymerization (0 eq.), regular polymerization (1 eq.) or alkene
isomerization (>20 eq.)
Institute and/or researcher Twitter usernames: @HarthEva @glenrjones92
7
This article is protected by copyright. All rights reserved.