Palladium(ii)-catalyzed copolymerization of styrenes with carbon monoxide: Mechanism of chain propagation and chain transfer

Article abstract of DOI:10.1039/b911392d

A mechanistic interpretation of the [(1,10-phenanthroline)Pd(CH ₃)(CH₃CN)]⁺[BArF]^- (1a) and [(2,2′-bipyridine)Pd(CH₃)(CH₃CN)] ⁺[BArF]^- (1b) (BArF = 3,5-(CF₃

Full text of DOI:10.1039/b911392d

View Article Online / Journal Homepage / Table of Contents for this issue
This paper is published as part of a Dalton Transactions themed issue on:
Metal-catalysed Polymerisation
Guest Editors: Barbara Milani and Carmen Claver
University of Trieste, Italy and Universitat Rovira i Virgili, Tarragona, Spain
Published in issue 41, 2009 of Dalton Transactions
Image reproduced with permission of Eugene Chen
Articles published in this issue include:
PERSPECTIVES:
New application for metallocene catalysts in olefin polymerization
Walter Kaminsky, Andreas Funck and Heinrich Hähnsen,
Dalton Trans., 2009, DOI: 10.1039/B910542P
Metal-catalysed olefin polymerisation into the new millennium: a perspective outlook
Vincenzo Busico, Dalton Trans., 2009, DOI: 10.1039/B911862B
HOT PAPERS:
Activation of a bis(phenoxy-amine) precatalyst for olefin polymerisation: first evidence for an outer
sphere ion pair with the methylborate counterion
Gianluca Ciancaleoni, Natascia Fraldi, Peter H. M. Budzelaar, Vincenzo Busico and Alceo
Macchioni, Dalton Trans., 2009, DOI: 10.1039/B908805A
Palladium(II)-catalyzed copolymerization of styrenes with carbon monoxide: mechanism of chain
propagation and chain transfer
Francis C. Rix, Michael J. Rachita, Mark I. Wagner, Maurice Brookhart, Barbara Milani and James
C. Barborak, Dalton Trans., 2009, DOI: 10.1039/B911392D
Visit the Dalton Transactions website for more cutting-edge organometallic and catalysis research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Palladium(II)-catalyzed copolymerization of styrenes with carbon monoxide:
mechanism of chain propagation and chain transfer†
Francis C. Rix,a Michael J. Rachita,a Mark I. Wagner,a Maurice Brookhart,*a Barbara Milanib and
James C. Barborakc
Received 11th June 2009, Accepted 7th August 2009
First published as an Advance Article on the web 3rd September 2009
DOI: 10.1039/b911392d
A mechanistic interpretation of the [(1,10-phenanthroline)Pd(CH3)(CH3CN)]+[BArF]- (1a) and
[(2,2¢-bipyridine)Pd(CH3)(CH3CN)]+[BArF]- (1b) (BArF = 3,5-(CF3)2-C6H3) catalyzed perfectly
alternating copolymerization of styrenes with CO is reported. The copolymerization in CH2Cl2 or
chlorobenzene has been found to be first order in styrene and inverse first order in CO concentrations.
The microscopic steps involved in the catalytic cycle have been studied via low temperature NMR
techniques. Palladium alkyl chelate complex [(2,2¢-bipyridine)Pd(CHArCH2C(O)CH3]+[BArF]- (5bs)
3
and [(2,2¢-bipyridine)Pd(h -CH(CH2C(O)CH3)Ar)]+[BArF]- (5bp), existing in equilibrium, were
prepared. Treatment of 5s,p with 13CO followed by 4-tert-butylstyrene at -78 oC allowed for 13C NMR
monitoring of the alternating chain growth of a series of palladium acyl carbonyl complexes. The acyl
carbonyl species, representing the catalyst resting state, is in equilibrium with a palladium acyl styrene
complex. The equilibrium constant, K4, measured between [(phen)Pd(CO)(C(O)CH3]+[BArF]- (3a) and
+
-
[(phen)Pd(C(O)CH3)-(C6H5C CH2)] [BArF] (8a), was determined to be 2.84 2.8 ¥ 10-7 at
-66 ◦C. The barrier to migratory insertion in 8a was determined (DG‡ (-66 ◦C) = 15.6 0.1 kcal mol-1).
From the experimentally determined kinetic and thermodynamic data for the copolymerization of
styrene with CO a mechanistic model has been constructed. The ability of this model to predict catalyst
turnover frequency (TOF) was used as a test of its validity. A series of para-substituted styrenes,
=
=
p-XC6H4CH CH2 (X = –OCH3, –CH3, –H, –Cl), were copolymerized with CO. A Hammett treatment
of TOF for the series showed that electron-donating groups increase the rate of copolymerization
p
(r = -0.8). The ratio of chain transfer to chain propagation was found to increase with styrene
concentration and decrease with CO concentration. Polymer end group analysis showed the presence of
a, b-enone end groups. The reactivity of model systems, coupled with a study of the effect of added
acetonitrile, support a chain transfer mechanism involving b-hydrogen transfer to monomer from a
palladium alkyl styrene intermediate.
90 ◦C, and solvent variations including alcohols, dichloromethane,
Introduction
chlorobenzene and water. The addition of co-reagents, such as
1,4-benzoquinone (BQ), or co-catalysts, such as Brønsted acids
is often advantageous. The choice of the best reaction conditions
depends both on the nature of the palladium complex as pre-
catalyst and on the nature of the alkene comonomer.
During the last fifteen years considerable interest has been focused
on the synthesis of perfectly alternating carbon monoxide/alkene
polyketones (eqn (1)). This reaction is homogeneously catalyzed
by Pd(II) complexes containing a wide array of bidentate ligands
including bidentate phosphine donors, bidentate nitrogen donors,
hybrid phosphorus-nitrogen systems and phosphino-phosphite
ligands.1–9 A range of reaction conditions have been employed:
CO pressures ranging from 1 to 40 bar, temperatures from 30 to
(1)
It is well known that diphosphine ligands are the ligands
of choice when aliphatic alkenes are the comonomers; the
o-anisole modified 1,3-bis(diphenyl phosphino) propane
aDepartment of Chemistry, University of North Carolina at Chapel Hill,
Chapel Hill, North Carolina, 27599-3290, USA. E-mail: mbrookhart@
unc.edu
R
(bdompp) was applied for the industrial production of Carilonꢀ,
bDipartimento di Scienze Chimiche, Universita` di Trieste, 34127, Trieste,
Italy
the CO/ethylene/propylene terpolymer.10 Even though CO/vinyl
arene co- and terpolymers were not exploited at a commercial level,
the corresponding copolymerization reaction has been the subject
of extensive investigations mainly aimed at understanding the
factors affecting catalyst stability, polymer molecular weight11–13
and polymer stereochemistry.14–25 In addition, it was demonstrated
that the nature of the chain termination reaction is strictly related
to the reaction media: when the copolymerization was carried
cDepartment of Chemistry, University of North Carolina at Greensboro,
Greensboro, North Carolina, 27412, USA
† Electronic supplementary information (ESI) available: Descriptions of all
kinetic experiments, tables of kinetic data and description of equilibrium
constant measurements. Details of polymerisation experiments including
NMR and GPC characterisations of copolymers, and representative 13C
NMR spectra showing 3a*, 4a* and copolymer chain growth beginning
with 3a*. See DOI: 10.1039/b911392d
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8977–8992 | 8977
©

out in methanol, nucleophilic attack of the alcohol on the
Pd–acyl bond of the growing polymer chain rapidly occurred
leading to polymers of very low molecular weight (Mw not higher
than 20 000 Da) accompanied by fast catalyst decomposition to
Pd(0);14 when the solvent was 2,2,2-trifluoroethanol, the catalysts
were remarkably stable, no alcoholysis took place and b-hydrogen
elimination was the main termination process,26 thus making
feasible the synthesis of polyketones of high molecular weight,
up to 106 Da.13 Living CO/vinyl arene copolymerization was
achieved when aprotic solvents, like CH2Cl2, were used.18,19,21
Previously, some of us reported a detailed mechanistic investiga-
tion of the key steps involved in the CO/ethylene copolymerization
catalyzed by the complex [Pd(CH3)(OEt2)(phen)][BArF] (phen =
1,10-phenanthroline; BArF = (3,5-(CF3)2-C6H3)4B-).20,27 The ki-
netic free energy barriers for the migratory insertion of carbonyl
alkyl-, ethylene acyl-, and ethylene alkyl-palladium complexes as
well as the relative binding constants for CO and ethylene were
determined by in situ, low temperature NMR studies, allowing
detection of key intermediates for the reaction and establishing
that double ethylene insertion occurs only once in ca. 106 insertion
events. The lack of double ethylene insertion events arises not
only from a higher binding affinity of CO relative to ethylene but
also from a significantly lower barrier for CO insertion into the
palladium–alkyl bond.
Fig. 1 Linear dependence of TOF on the concentration of TBS in the
1a-catalyzed copolymerization of TBS/CO. TOF was a measure of CO
uptake at 25 ◦C in CH2Cl2.
conversion, is inversely proportional to CO concentration.18 This
dependence is found to be inverse first-order in CO concentration
as shown by a drop in the rate of polymerization by a factor of
2.70 on going from 14.7 psi CO (TOF = 0.0115 h-1) to 40 psi CO
(TOF = 0.0044 h-1) (Fig. 2).
NMR studies of the CO/vinyl arene copolymerization were re-
ported by Carfagna and Nozaki. Carfagna identified key interme-
diates involved in the control of stereochemistry during the copoly-
merization process using palladium complexes containing N–N
i
bidentate ligands like Pr-DAB (iPr-DAB = 1,4-diisopropyl-1,4-
diaza-1,3-butadiene),28 bis-oxazoline,29 and a-diimines.30 Nozaki
investigated Pd-complexes based on the phosphino-phosphite
ligand BINAPHOS and demonstrated that, in contrast to the
Pd-diphosphine systems, Pd-BINAPHOS catalyzes the CO/vinyl
arene copolymerization by virtue of 1,2-regiochemistry of styrene
insertion into the Pd–acyl bond.31
We report here an in-depth mechanistic study of the catalytic
cycle of the CO/vinyl arene copolymerization catalyzed by
[Pd(CH3)(CH3CN)(N-N)][BArF] (N-N = phen 1a, 2,2¢- bipyri-
dine (bpy) 1b). The structure and reactivity of key intermediates,
a quantitative analysis of substituents effects in para-substituted
styrenes, and the mechanism of chain transfer are discussed.
Fig. 2 Inverse dependence of TOF on the concentration of CO in the
1a-catalyzed copolymerization of TBS/CO at 25 ◦C in chlorobenzene:
᭹14.7 psi CO; ꢀ 40 psi CO. kobs (14.7 psi) = 0.0115 h-1, kobs (40 psi) =
0.0044 h-1.
Results and discussion
Changes in the alkene concentration also have an effect on the
relative rates of chain propagation and chain transfer as reflected
in changes in molecular weight distribution (MWD). In particular,
increasing the TBS concentration resulted in broadening the
MWD from near living at a low TBS concentration to 1.64 at
a monomer concentration of 1.1 M (Table 1).
1. Copolymerization of 4-tert-butyl styrene with CO
Cationic palladium complexes 1a,b are excellent catalysts for the
copolymerization of 4-tert-butylstyrene (TBS) or styrene with
carbon monoxide, yielding the corresponding perfectly alternating
polyketones.18
During TBS/CO copolymerization catalyzed by 1a, the catalyst
turnover frequency (TOF) can be assessed by measuring the rate
of CO uptake under varying experimental conditions. Measuring
turnover frequency at a number of different TBS concentrations
shows that the rate of CO uptake is directly proportional to TBS
concentration (Fig. 1), suggesting that, under standard conditions
(1 atm CO at 25 ◦C) chain growth is first-order in the alkene.
Conversely, if TBS concentration is held constant and CO pressure
is increased, the rate of chain growth, as a function of monomer
Table 1 TOF and MWD as a function of [TBS] for copolymerization of
TBS and CO catalyzed by 1a in CH2Cl2 at 25 ◦C under 1 atm CO
[TBS]/M
Initial TO/min
Mn
5800
Mw/Mn
0.27
0.55
0.83
1.1
0.30
0.49
0.67
0.91
1.29
1.46
1.54
1.64
12 200
13 600
13 800
8978 | Dalton Trans., 2009, 8977–8992
This journal is
The Royal Society of Chemistry 2009
©

The observation that chain growth in TBS/CO copolymer-
ization is proportional to TBS concentration and inversely
proportional to CO concentration is consistent with the micro-
scopic steps of the catalytic cycle observed for the 1a-catalyzed
CO/ethylene copolymerization.20,27 Under polymerization con-
ditions, pre-catalyst 1a undergoes ligand exchange with CO to
form the Pd-methyl-carbonyl derivative (i), from which methyl
migration occurs, generating the Pd-acyl-carbonyl intermediate
(ii) (Scheme 1). This species, analogous to the corresponding
derivative where the acyl fragment belongs to the growing polymer
chain, represents the catalyst resting state. No acyl migration to
coordinated CO occurs, since this reaction is thermodynamically
unfavourable.32,33
Scheme 2 Preparation of acetonitrile free pre-catalyst 5b and its
activation.
[Pd(CH3)2(bpy)] 2b. Protonation of 2b with [H(OEt2)2][BArF], at
-30 ◦C then carbonylation yielded the Pd-acyl-carbonyl species
[Pd(C(O)CH3)(CO)(bpy)]+ 3b. (Complex 3b can also be generated
by reaction of [Pd(CH3)(CO)(bpy)][BArF] 4b,20 with carbon
monoxide in cold CD2Cl2). Subsequent treatment of the cold
mixture with TBS under a N2 purge, to displace CO, gave 5b.
Complex 5b exists in solution as a rapidly equilibrated mixture
of the chelated species [(bpy)Pd(CHArCH2C(O)CH3]+ 5bs and
3
the p-benzyl complex [Pd(bpy)(h -CH(CH2C(O)CH3)Ar]+ 5bp
(Scheme 2). Static H NMR spectra, obtained at ca. -100 ◦C,
1
indicated a 5bs; 5bp ratio of ca. 3 : 1. The regiochemistry of TBS
insertion was studied through the preparation of a 13C labelled
version of 5b, 5b*, through use of 99% 13CO. The labelled carbonyl
1
carbon displayed a Jcc of ca. 40 Hz to both the CH3 and CH2
carbons in 5bs and 5bp. Such a large coupling to the methylene
carbon is indicative of exclusive 2,1 TBS insertion.
Scheme 1 General proposed mechanism for the Pd-catalyzed CO/vinyl
arene copolymerization.
Low temperature 13C NMR spectroscopy of the 13C-labelled
carbonyl region was used to observe chain growth. An NMR
◦
sample prepared from 5b* in CD2Cl2 was cooled to -78 C then
For propagation to proceed, the Pd-acyl-carbonyl species
(ii) has to be in equilibrium with the Pd-acyl-alkene intermediate
(iii), from which the migratory insertion of the olefin into the
Pd–acyl bond occurs yielding the Pd-alkyl intermediate (iv). The
latter can either bind CO, followed by regeneration of the resting
state and completion of the catalytic cycle, or undergo b-hydrogen
elimination as a chain transfer reaction leading to termination of
chain growth.
The turnover frequency of a catalytic system following the
mechanism depicted in Scheme 1 is given by the expression
TOF = Keqkins[TBS]/[CO]. This expression takes into account
the pre-equilibrium between the resting state and the Pd-acyl-
alkene species, as well as the rate of alkene insertion in the
Pd-acyl complex. The two subsequent reactions, involving CO co-
ordination and CO migratory insertion in the Pd-alkyl derivative,
should be fast and are excluded from the TOF expression. This
TOF expression is in accord with the observed rate dependence of
polymerization on TBS and CO concentrations.
purged with 13CO. Returning the sample to the probe allowed
observation of the carbonyl acyl complex 6b*, as evidenced
by the three characteristic carbonyl bands at 172.2, 218.7 and
207.5 ppm for the Pd–CO, the a-acyl carbonyl and –CH2C(O)CH3,
respectively, to be observed (Scheme 3). The sample was removed,
exposed to 6 equiv. of TBS at -78 ◦C and then returned to the cold
probe. After 40 min, the first species detected upon insertion, at the
expense of 6b*, is 7b*. After 2 h at -80 ◦C, 8b* became detectable
in a mixture of 6b* and 7b*. Raising the probe to -60 ◦C, to
hasten the polymerization, allowed 8b* to grow as monitored by
13C NMR spectroscopy.
Chemical shift assignments of the –C(O)CH3 carbonyls were
made after monitoring the reactions of unlabelled 5b with 13CO
and TBS by 13C NMR spectroscopy. Unlabelled 5b was treated
with 13CO and ca. 10 equiv. of TBS at -78 ◦C. The first two
insertion products, analogous to 6b* and 7b* but with unlabelled
end groups, confirmed the assignments for 13C(O)CH3 resonances
in 6b*, 7b* and allowed identification of the internal carbonyl
resonances of 8b*. Further insertions occurred between -60 and
20 ◦C, which finally led to an envelope of bands in the 207–208 ppm
range continually increasing in intensity relative to the Pd–CO
and Pd–CO(R) bands. These experiments clearly indicate that the
carbonyl acyl form of the catalyst is the catalyst resting state under
these conditions.
2. In situ NMR studies of copolymerization
(a) Observation of the Pd-acyl-carbonyl species as the catalyst
resting state. The in situ NMR studies were performed starting
from complex 5b (Scheme 2) to avoid the presence of acetoni-
trile. Complex 5b was prepared in two steps from the dialkyl
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8977–8992 | 8979
©

Scheme 3 13C NMR spectroscopic monitoring of sequential chain growth in the copolymerization of TBS and CO using 5b*.
Scheme 4 13C NMR monitoring of sequential chain growth in the copolymerization of TBS and CO using 4a*.
A similar in situ 13C NMR analysis of the copolymeriza-
tion of 13CO and TBS has been conducted using labelled
[Pd(CH3)(13CO)(phen)]+ 4a* as the pre-catalyst (Scheme 4). Com-
plex 4a* was prepared via protonation of [Pd(CH3)2(phen)] 2a
with [H(OEt)2]+[BArF]- in CH2Cl2 followed by a 13CO purge at
-78 ◦C.20 Precipitation with hexanes gave colourless, analytically
pure 4a*, which was used to monitor polymer chain growth.
Exposing 4a* to 13CO at -30 ◦C in an NMR tube and
then cooling to -100 ◦C generated the palladium-acetyl-CO
complex 3a*. When 3a* is treated with 14 equiv. of TBS at
-30 ◦C, followed by periodic temperature quenching of the
growing polymer chain, the sequential formation of the acyl
carbonyl resting state(s) 6a* and 7a* was monitored by NMR
at -100 ◦C. After several insertions were observed, the sample
was warmed to 20 ◦C in order to facilitate polymerization.
Returning the sample to the cold probe resulted in observation
of a broad envelope of bands at ca. 207 ppm as well as a new
carbonyl resonance at 197 ppm. A similar resonance at 197.8 was
previously observed in the 13C NMR spectrum of a high molecular
weight TBS/13CO copolymer. This minor resonance was iden-
tified as an a,b-unsaturated enone capped polymer chain end
a similar approach was taken in order to observe this reactive
intermediate in the CO/styrene copolymerization (Scheme 5).
=
(4-(CH3)3C-C6H4CH CHC*(O)-poly). Consiglio has also iden-
Scheme 5 Reaction scheme of 4a with styrene.
tified an enone end group in the copolymerization of 13C-labelled
styrene with CO,14 and the analogous end group was recognized
in the CO/styrene and CO/4-Me-styrene polyketones synthesized
with Pd-phenanthroline based systems in trifluoroethanol.26
A CD2Cl2 solution of 4a was treated with 10 equiv. of styrene
at -78 ◦C, and placed in a pre-cooled (-66 ◦C) NMR probe. The
+
=
formation of [Pd(C(O)CH3)(C6H5CH CH2)(phen)] (8a) and its
subsequent rearrangement, via acyl migration to styrene, to the
(b) Spectroscopic observations of a palladium-acyl-alkene com-
plex. Although the catalyst resting state for TBS/CO copolymer-
ization, the acyl carbonyl complex, was clearly observed via low
temperature NMR spectroscopy, the propagating palladium-acyl-
alkene species was not detected. In analogy to the study performed
on the migratory insertion reactions of ethylene acyl complexes,20
equilibrated mixture of 5as and p-benzyl 5ap was observed. The
+
=
two other intermediates, [Pd(CH3)(C6H5CH CH2)(phen)] (9a)
and [Pd(C(O)CH3)(CO)(phen)]+ (3a), each making up less than 4%
of the total composition of the reaction mixture, were also present
during this experiment. The time course of the composition of the
reaction is shown in Fig. 3.
8980 | Dalton Trans., 2009, 8977–8992
This journal is
The Royal Society of Chemistry 2009
©

analogous ethylene case: Krel = 34 4.20 In both systems, alkene
binding cis to acetyl is favoured relative to methyl, presumably
for steric reasons. The increase in scatter observed here is likely
due to measuring decreasing amounts of 9a and 3a. The drift is
possibly the result of a systematic error such as a fixed amount
of impurity coincident with the resonances for 4a or 8a. The
significance of the drift appears minor and the rapid equilibria
hypothesis is consistent with the good fit obtained between the
data and the mechanism in Scheme 5.
The ratio K4/K3 yields K4 after determining K3. Previous studies
indirectly measured the binding affinities of styrene relative to
3,5-(CF3)2-benzonitrile (KA = 0.125
0.01), and benzonitrile
relative to CO (KB = (2.87 1.35) ¥ 10-6).34 To determine K3, the
binding affinity of 3,5-(CF3)2-benzonitrile relative to benzonitrile,
KC, at -66 ◦C was measured: KC = (2.2 0.02) ¥ 10-2. K3 at
Fig.
from the reaction of [Pd(CH3)(CO) (phen)]+ 4a with styrene:
[Pd(CH3)(CO)(phen)]+ 4a; [Pd(C(O)CH3)(styrene)(phen)]+ 8a;
[Pd(h -CHPhCH2C(O)CH3)(phen)]+ 5ap, [Pd(CHPhCH2C(O)CH3)-
(phen)]+ 5as; ꢀ [Pd(CH3)(styrene)(phen)]+ 9a; ꢂ [Pd(C(O)CH3)(CO)-
(phen)]+ 3a. Inset is ratio of (K4/K3 = [8a][4a]/[9a][3a]) over the course of
the experiment.
3 Experimental and calculated composition vs. time data
᭹
◦
-66 C is given by KAKBKC = (7.89 3.76) ¥ 10-9 (DG◦ = 7.7
ꢀ
ꢁ
3
0.2 kcal mol-1). K4 was then calculated from (K4/K3)KAKBKC,
K4 = (1.18 0.61) ¥ 10-6 (DG◦ = 5.68 0.24 kcal mol-1) (Scheme 6).
Composition vs. time data were simulated numerically with the
program Gear–Git.35 The mechanism in Scheme 5 was used as
the basis for calculating compositions iteratively, with changing
migratory insertion rate constants, until the “best fit” with
experimental data was achieved. To simplify the curve fitting,
several constraints were imposed upon the calculations. Rate
constants for the two equilibria were given values much larger than
the insertion rate constants while maintaining the necessary ratios
dictated by K3 and K4. Initial compositions were held at values
obtained from simple exponential or biexponential fits of the data.
The iterative fitting procedure then gave k1 = 2.32 ¥ 10-4 s-1 (DG‡ =
15.4 kcal mol-1) and k2 = 1.47 ¥ 10-4 s-1 (DG‡ = 15.6 kcal mol-1).
The k1 is in good agreement with the value determined separately
from saturation kinetics experiments: k1(sat.) = 2.4 0.3 ¥ 10-4 s-1.
The data obtained in this study concerning styrene binding
affinities and insertion barriers combined with the previously
obtained data for ethylene allows a detailed comparison of the
relative insertion abilities of these monomers in phenanthroline
palladium(II) systems. A free energy diagram constructed from this
data is shown in Scheme 7. Facile interconversion of the alkene
complexes is justified by the rapid associative ligand exchange
observed previously at low temperatures. The relative insertion
abilities of the two alkenes are determined by the energy differences
between the transition states in accord with the Curtin–Hammett
principle. Thus, while there is a lower barrier to insertion for the
styrene acyl complex relative to the ethylene acyl complex, this
difference is offset by ethylene’s larger binding affinity. The net
result is a DDG‡ of 1.2 kcal mol-1 favouring ethylene insertion.
A similar diagram can be constructed for the methyl migratory
insertion reaction. Migratory insertion from the styrene methyl
complex is more favourable than the ethylene analogue, but is
again offset by a larger ethylene binding affinity whose sum is a
DDG‡ of 1.45 kcal mol-1 favouring ethylene insertion.
Consecutive migratory insertion reactions convert 4a into
5as, 5ap via 8a. The first migratory insertion reaction of
[Pd(CH3)(CO)(phen)]+ has been previously studied using different
trapping agents: CO, ethylene and methyl acrylate. The evidence
indicates that the insertion occurs via reversible methyl migration
to CO followed by ligand trapping.20 The limiting rate constant
obtained in these previous studies will serve to test the numerical
simulation of the data in Fig. 3. The second migratory inser-
tion reaction converting [Pd(C(O)CH3)(styrene)(phen)]+ 8a into
the isomers 5as and 5ap, is analogous to the previously ob-
served first-order conversion of [Pd(C(O)CH3)(ethylene)(phen)]+
to [Pd(CH2CH2C(O)CH3)(phen)]+.
From Fig. 3, it is apparent that the fractions of 9a and 3a,
while changing, are roughly the same, while the ratios 4a/9a and
8a/3a are not, throughout the experiment. Previous studies of
ligand substitution on complex 4a and analogues suggest that the
presence of 9a is due to styrene reacting with 4a to yield 9a and
free CO. A large driving force for CO binding insures that the
concentration of free CO is very small. Even so, the liberated CO
can react with either 8a to form 3a or with 9a to reform 4a. In this
manner, CO and styrene are equilibrated throughout the available
sites. The roughly equivalent amounts of 9a and 3a suggest that
the amount of CO in solution, which reacted with 5 or escaped to
the head-space is very small. Thus, CO acts to couple the equilibria
represented by K3 and K4 and as the amount of free CO available
decreases, with formation of 8a (and thus 5a), the ratios 4a/9a and
8a/3a must change.
The fact that the populations of 9a and 3a change reveals that the
equilibria are being attained at least on a similar time-scale as the
insertion reactions. If either equilibrium were established slowly,
then the fractions of 9a and 3a would be fixed. If equilibria K3
and K4 are fast relative to the insertions then, while the minuscule
[CO] and moderate [styrene] may change, the relative equilibrium
constant (K4/K3 = [8a][4a]/[9a][3a]) will be constant. A plot of
K4/K3 vs. time is shown as an inset in Fig. 3. The ratio K4/K3 is
roughly constant at 150 30, while displaying a slight upward drift
with an increase in scatter as the experiment progresses. A similar
ordering of the relative equilibrium constants was observed in the
3. Comparison of calculated and experimental turnover
frequencies for styrene/CO copolymerization
From the experimentally determined kinetic and thermodynamic
data for the copolymerization of styrene and CO catalyzed by
1a, the TOF can be predicted using the expression TOF(calc)
Keqkins[styrene]/[CO]. Keq in the TOF expression is K4 (Scheme 5)
=
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8977–8992 | 8981
©

Scheme 6 Equilibria studied to determine K3 and K4 values.
extrapolated to 25 ◦◦C ((2.29 2.1) ¥ 10-5), kins is k2 (Scheme 5)
increase in DG‡ upon changing the migrating center from acetyl
to the first repeat unit. An increase in the transition state energy
corresponds to a reduction in the rate of styrene insertion and
therefore a smaller value for kins than k2 predicts. If these two
correction factors are used to re-estimate TOF, Keq is (1/5)K4
(4.58 ¥ 10-6) and kins is 9.7 s-1, then the calculated TOF becomes
1.9 ¥ 10-2 s-1. The corrections indeed drop the estimated TOF
into close approximation of the observed TOF. The ability to
approximate TOF through the use of both thermodynamic and
kinetic data strongly supports the proposed mechanism for the
propagation of styrene/CO copolymerization.
extrapolated to 25 C (23 4 s-1), [styrene] is the initial concen-
tration of styrene in a given experiment, and [CO] is the estimated
concentration of CO in the solution of styrene and CH2Cl2 at
◦
25 C (6.5 ¥ 10-3 M).36‡ Using a styrene concentration of 2.9 M
gives a calculated TOF of 0.23 s-1. The experimentally determined
TOF at this concentration is 0.011 s-1. The calculated TOF is seen
to overestimate the experimental TOF by roughly a factor of 20.
This over approximation is not surprising when the models used to
determine K4 and k2 are considered. For instance, the equilibrium
constant is a measure of the displacement of CO by styrene in
a palladium-acetyl complex (Scheme 5). Such a model does not
take into account the steric bulk of the growing polymer, which
will have an effect of decreasing K4. During the copolymerization
of CO with ethylene it was found that Keq for ethylene binding to
a palladium-acetyl-CO complex was 5 times greater than ethylene
binding after just one ethylene insertion.20 It was also noted in
the ethylene/CO copolymerization that there is a 0.5 kcal mol-1
4. Substituent effects on rate of chain growth
Having a detailed understanding of the microscopic steps in the
catalytic cycle, it was of interest to examine the effects of aryl
substituents on the rate of copolymerization. A series of para-
=
substituted styrenes, p-XC6H4CH CH2 (X = –OCH3, –CH3, –H,
–Cl) were copolymerized with CO (1 atm) at 25 ◦C. The TOF
was determined by measuring the rate of CO uptake over the first
two hours of polymerization.
‡ Bryndza’s calculation of [CO] compares favourably with that reported for
CHCl3 (8 ¥ 10-3 M), a,b ClCH2CH2Cl (6 ¥ 10-3 M)b. Since [CO] in toluene
is similar (7 ¥ 10-3 M)b to that in CH2Cl2 the value of 6.5 ¥ 10-3 M is a
reasonable value for the solution of styrene and CH2Cl2. (a) W. F. L i n ke,
Solubilities of Inorganic and Metal Organic Compounds, 4th edn, vol 1,
Van Nostrand, Princeton, NJ, 1958. (b) W. J. Knebel and Angelici, Inorg.
Chem.,1974, 13, 632.
Electron-rich styrenes (OCH3, tBu) were found to increase
copolymerization rates, while electron-deficient styrenes (H, Cl)
p
inhibited copolymerization. A Hammett plot of log(TOF) vs. s
p
gave a r exp of -0.8 for the observed turnover frequency (Fig. 4).
8982 | Dalton Trans., 2009, 8977–8992
This journal is
The Royal Society of Chemistry 2009
©

Scheme 7 Curtin–Hammett plots for migratory insertion reactions in [Pd]R(alkene)+ complexes: alkene = ethylene, styrene; [Pd] = (phen)Pd; R = Me,
Ac (Ac = acyl).
As shown above, the TOF = kpoly[styrene]/[CO], where kpoly
=
Keqkins (see Scheme 1). Thus, substituent effects on kpoly will result
from effects on both Keq and kins values and the r based on kx
poly
values will be the sum of r(Keq) and r(kins) values. Models for both
(2)
Keq and kins processes are available.20,34
The system which best models kins is shown in eqn (2) and
involves the migratory insertion of styrene methyl complex 10a to
yield p-benzyl complexes 11a. Rix, et. al. showed that a Hammett
p
p
plot of logkx vs. s values gave a r of +1.1 0.1 for these
ins
insertions.34 This positive r value indicates that the rate of insertion
The relative binding affinities of substituted styrenes to the
(phen)Pd(CH3)+ moiety have been previously determined and
increases as the monomer becomes more electron-deficient.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8977–8992 | 8983
©

5. Mechanism of chain transfer
(a) b-Hydrogen elimination vs. monomer-assisted b-hydrogen
abstraction leading to enone end-capped polymers. Although a
reasonable understanding of the microscopic steps involved in
the chain propagation of TBS/CO copolymerization exists, very
little is understood about the mechanism of chain transfer in
the aprotic solvent CH2Cl2. Three key observations have been
made that must be included in any proposed chain transfer
mechanism: (1) the ratio of chain transfer to chain propagation
increases with increasing TBS concentration; (2) the ratio of
chain transfer to chain propagation decreases with increasing
carbon monoxide pressure; and (3) polymer end group analysis has
shown the presence of a,b-enone end groups. Two distinct chain
transfer mechanisms, each of which account for the experimental
observations and share a number of common intermediates, have
been proposed (Scheme 8). The most straightforward mechanism,
which is consistent with recent advances from this laboratory,34,37
is to invoke a chain transfer mechanism dependent on b-hydrogen
elimination from 12 followed by associative displacement of
the resulting enone 13 (pathway A, Scheme 8). Associative
substitution of the enone in 13 with TBS results in free polymer
bearing an enone end group and in the Pd-hydride complex 14.
Complex 14 will undergo hydride migration to TBS followed
by CO coordination to generate palladium alkyl complex 15.
Complex 15 will insert CO leading to the growth of a new
polymer chain through the catalyst resting state 16. An alternative
mechanism for chain transfer, again starting from the palladium
alkyl complex 12, can be envisioned, however. Complex 12, rather
than undergoing b-hydrogen elimination, can initially coordinate
TBS forming palladium alkyl olefin complex 18 (pathway B,
Scheme 8). Complex 18 can then undergo monomer-assisted
b-hydrogen abstraction generating palladium alkyl enone 19.
Compound 19 is funnelled back into the propagating cycle after
enone displacement with CO (chain transfer) forming the alkyl
carbonyl species 15.
p
Fig. 4 Hammett plot of log (TOF) vs. s for the rate of copolymerization
of a series of substituted styrenes with CO using 1a in CH2Cl2 at 25 ◦C.
represent an excellent model for the effect of substituents on
Keq(poly) (Scheme 1).34 From observation of a series of pairwise
equilibria shown in eqn (3), the relative binding affinities of
substituted styrenes could be determined.
(3)
Binding affinities were shown to decrease in the order of
OCH3 > CH3 > H > Cl > CF3. In order to manipulate the KAX data
for a Hammett treatment, KBX was found from the ratio KBX
AX/KAH (eqn (4)). Plotting the logKBX vs. s gave a Hammett
plot with a slope of r = -2.2 0.1. The negative slope indicates
electron-rich styrenes bind more readily to the cationic palladium
fragment.
=
p
K
p
The differences between the two types of chain transfer
mechanisms outlined in Scheme 8 are subtle. For example, both
mechanisms predict a decrease in the ratio of chain transfer to
chain propagation rates with increasing CO concentration. As
CO pressure increases, the rate of trapping of {12p, 12, 12s} will
increase and its lifetime will be decreased. Each trapping event
will result in a chain extension (propagation). Thus, in either
mechanism, increasing CO pressure will result in an increase in
the ratio of propagation to chain transfer and a narrowing of
MWD as observed experimentally (see above). Both A and B
pathways can account for increasing MWD with increasing TBS
concentration as this increases the ratio of chain transfer to chain
propagation. In the case of pathway A if we assume {12p, 12,
12s} is in rapid equilibrium with 13 (with equilibrium favouring
{12p, 12, 12s}) prior to formation of 14, then the rate of chain
transfer will be proportional to TBS concentration. In pathway
B, TBS concentration dependence would be clearly observed for
chain transfer if conversion of {12p, 12, 12s} to 18 is either rapidly
reversible favouring {12p, 12, 12s} or if conversion of 18 to 19 is
rapid relative to formation of 18 and thus trapping {12p, 12, 12s}
by TBS to form 18 is rate-determining.
(4)
Based on the above two model systems, the predicted r value
for the copolymerization reaction is -2.2 (Keq) + 1.1 (kins) = -1.1,
which is remarkably close to the experimentally determined value
of -0.8. Clearly, substituent effects on the Keq and kins terms oppose
one another, but the dominant effect is on the Keq term. Thus,
the copolymerization is accelerated when the aryl substituents are
electron-donating.
This finding is in agreement with experimental data reported
in the literature on the CO/vinyl arene copolymerization pro-
moted both by dicationic bischelated palladium complexes12
[Pd(N-N)2][PF6]2 and monocationic monochelated derivatives
[Pd(CH3)(CH3CN)(N-N)][PF6] (N-N = phen and its substituted
derivatives).13 For all the tested catalysts, 4-Me-styrene was
always found to be a more reactive alkene than styrene: higher
productivities and polymers of higher molecular weight were
obtained for the substituted vinyl arene with respect to styrene.
Both chain transfer mechanisms depicted in Scheme 8 have been
previously proposed to explain chain transfer in early transition
8984 | Dalton Trans., 2009, 8977–8992
This journal is
The Royal Society of Chemistry 2009
©

Scheme 8 Possible mechanisms describing chain transfer within the propagating catalytic cycle of TBS/CO copolymerization (GP = growing polymer).
metal catalyzed Ziegler–Natta polymerizations.38 More recently,
theoretical calculations have been performed that assess the
energetic feasibility of these two pathways in late transition metal
catalyzed olefin polymerizations. Calculations by Morokuma
and co-workers have suggested that the b-hydrogen transfer to
monomer process for chain termination in palladium diimine
catalyzed ethylene polymerization would require a very large ac-
tivation energy.39 b-Hydrogen elimination followed by associative
displacement appeared to be the most energetically feasible chain
transfer pathway. Conversely, Ziegler and co-workers calculated
that b-hydrogen elimination from a nickel diimine catalyst would
be energetically unfavourable.40,41 Their calculations indicated that
chain transfer in these systems occurred by the b-hydrogen transfer
Scheme 9 Preparation of CO free palladium-alkyl 20a.
to monomer mechanism. Several experiments described below
were carried out in an attempt to discriminate between these two
mechanisms in TBS/CO copolymerization catalyzed by 1a.
A mechanistic scheme, which corresponds to the chain transfer
step involving associative displacement (pathway A, Scheme 8), is
shown in Scheme 10. Treating 21 as a steady-state intermediate,
the kinetic expression for disappearance of 20a is -d20a/dt =
k3k4[TBS][20a]/(k-3 + k4[TBS]). A kinetic scheme corresponding
to pathway B (Scheme 8) is shown in Scheme 10. Assuming
24 never reaches significant concentrations (see below), the
kinetic expression for disappearance of 20a is -d20a/dt =
k5Keq[20a][TBS]. For mechanistic Scheme 10 path A, there is the
possibility that at high TBS concentration k4[TBS] will be greater
than k-3 and thus conversion will become independent of TBS
(b) Model studies aimed at probing the mechanism of chain
transfer. Complex 20as,p, a model for intermediate {12p, 12,
12s}, can be prepared by treatment of complex 1a with CO
followed by TBS as shown in Scheme 9. Treatment of 20as,p with
TBS results in formation of the free enone 22 and the p-benzyl
complex 23 (see Scheme 10); this reaction corresponds directly to
the chain transfer step. In order to gain additional information
concerning this reaction a kinetic study was undertaken.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8977–8992 | 8985
©

Scheme 10 Mechanism for generation of b-acetyl-4-tert-butylstyrene 22 from palladium-alkyl 20a.
concentration and controlled only by rate of formation of 21
chain transfer is dependent on associative displacement after
b-hydrogen elimination, the addition of an external ligand capable
of displacing the enone should increase the rate of chain transfer.
If, on the other hand, chain transfer is independent of associative
displacement, as is the case in the monomer assisted b-hydrogen
abstraction mechanism, then an external ligand will have little
effect on chain transfer. In order to test this hypothesis a series
of TBS/CO copolymerizations catalyzed by 1a were performed
in which acetonitrile was added in varying amounts (eqn (5)).
Acetonitrile was chosen as a competing ligand on both kinetic as
well as thermodynamic grounds. Thermodynamically, acetonitrile
has been shown to form much more stable complexes with
(phen)Pd+ fragments than styrene. For example, the Keq for ligand
exchange in the palladium st◦yrene complex 9a with acetonitrile
was found to be ca. 160 at 25 C.42
(rate = k3[20a]). Therefore, kinetic experiments spanning a range
of TBS concentrations may exhibit classic saturation behaviour.
In contrast, Scheme 10 path B will always exhibit first-order
TBS dependence (provided 24 does not build up at high TBS
concentrations) with no saturation behaviour possible.
Kinetic measurements were carried out using 0.01 M solutions
of 20as,p in CD2Cl2 with added TBS in varying concentrations
between 0.07 M (7 equiv.) and 1.2 M (130 equiv.). Changes in
the concentration of 20as,p and the liberated enone, 22, were
1
both monitored by integration of their respective H NMR acyl
methyl signals. Even at the highest concentration of TBS, no other
palladium species besides 20a and 23 could be detected. As can
be seen from Fig. 5 a plot of kobs vs. [TBS] is linear over the full
range of TBS concentrations indicating the reaction is first-order
in TBS at all concentrations. There is no evidence of saturation
behaviour. Thus, the kinetic behaviour of this model is consistent
with either of the two possible mechanisms shown in Scheme 10.
If Scheme 10A does apply, then the linearity of Fig. 5 indicates
that k-3 >> k4[TBS] even at 1.1 M TBS.
(5)
Since acetonitrile is also a smaller ligand which binds in an
1
2
h rather than h mode, it should be kinetically very effective in
displacing enone in an associative ligand displacement reaction.
In a typical experiment, 0.1 mmol of 1a in CH2Cl2 were treated
with 55 mmol of TBS (1.1 M) and a varying mol% of CH3CN
with respect to TBS. The homogenous reaction mixture was
stirred under 1 atm of CO at 25 ◦C for 20 h before the white
polymer was precipitated from the reaction medium through the
addition of methanol. Table 2 summarizes results of the effect of
varying the mol% acetonitrile in solution on the MWD of the
resultant polymer. These data show that there is no significant
relationship between MWD and mol% acetonitrile. In the absence
of acetonitrile the copolymer produced has a MWD of 1.71. In
the presence of acetonitrile, MWD varies from 1.71 at 10 mol% to
1.45 at 100 mol% additive. The lack of broadening of the MWD
Fig. 5 Plot of kobs vs. [TBS] for the formation of enone 22 from 20a.
A more direct way to discriminate between a b-hydrogen
elimination chain transfer mechanism and a monomer assisted
b-hydrogen abstraction mechanism is through the study of the
effects of added ligands on copolymerization. If the rate of
8986 | Dalton Trans., 2009, 8977–8992
This journal is
The Royal Society of Chemistry 2009
©

Table 2 Effects of added acetonitrile on MWD and on the rate of chain
BArF as a counterion in these systems allowed for extensive
1H NMR studies to be undertaken to probe the mechanism of
copolymerization. A series of in situ polymerization monitored
by NMR spectroscopy definitively showed the catalyst resting
state is a palladium acyl carbonyl complex (ii) (Scheme 1).
Chain propagation from (ii) was shown to be first-order in
styrene concentration and inverse first order in CO concentration
suggesting that TOF was dependent on a pre-equilibrium between
(ii) and palladium acyl styrene complex, (iii). The equilibrium
constant for the substitution of CO in the catalyst resting state
by styrene was determined experimentally from palladium acyl
carbonyl complex 3a to be 2.8 ¥ 10-7 at -66 ◦C. The TOF is also
dependent on the rate of insertion of styrene into the palladium-
acyl bond in (iii). This rate was determined by monitoring the
appearance and subsequent disappearance of palladium acyl
styrene complex 8a. The rate of formation of 8a from palladium
methyl carbonyl complex 4a was 2.3 ¥ 10-4 s-1 and the rate of
styrene insertion into the palladium–acetyl bond was determined
to be 1.5 ¥ 10-4 s-1 (DG‡ = 15.6 0.1 kcal mol-1 at -66 ◦C).
The thermodynamic as well as kinetic data collected in the model
studies was used to calculate an expected TOF for styrene/CO
copolymerization from TOF(calc) = Keqkins[styrene]/[CO] of 1.9 ¥
10-2 s-1. The experimentally determined TOF was found to be 1.1 ¥
10-2 s-1, which is in good agreement with the calculated value.
The dependence of TOF on both the pre-equilibrium term
Keq and the rate of styrene insertion (kins) was experimentally
established through a Hammett treatment of the copolymeriza-
tion. Styrenes substituted in the para position showed a linear
propagation in a non-living copolymerization of TBS at 1.1 M and CO at
1 atm using 1a as catalyst in CH2Cl2 at 23 ◦
C
CH3CN (mol%)
Initial TO/min
Mn
Mw/Mn
0
10
25
50
75
0.68
0.52
0.42
0.34
0.25
0.21
14 600
18 700
16 600
15 300
13 300
14 600
1.71
1.71
1.63
1.57
1.59
1.45
100
Table 3 Effects of added acetonitrile on polymer MWD in a living
copolymerization of TBS at 0.27M and CO at 1 atm using 1a as catalyst
in CH2Cl2 at 23 ◦
C
CH3CN (mol%)
Mn
Mw/Mn
0
10
25
50
8600
7900
7200
6900
1.17
1.15
1.13
1.14
indicates the added ligand does not increase the rate of chain
transfer. To determine the effect acetonitrile had on the rate of
polymerization, CO uptake was measured over the first two hours
of the reaction (Table 2). Initial TOF dropped considerably with
just a few mol% of acetonitrile and dropped by as much as 70%
for a 1 : 1 mixture of TBS and acetonitrile. This rate suppression is
most likely an effect of acetonitrile competing with CO for binding
to the acyl complex and thus reducing both the concentration of
the acyl CO complex (ii) and of the palladium acyl olefin complex
(iii) (Scheme 1).
The series of experiments described above were performed at
a high TBS concentration in which non-living polymerization
occurs. Another set of experiments were performed at lower
TBS concentrations in which near-living polymerization results
(Table 3). These conditions focus directly on the ability of
acetonitrile to trigger chain transfer via enhanced associative
substitution. Under the reaction conditions, TBS concentration
was held constant at 0.27 M while the concentration of acetonitrile
was progressively increased. The MWD of the resulting polymers
precipitated after 2 h ranges only from 1.17 to 1.13 over a 50%
increase in additive concentration.
p
correlation between s and both Keq and kins. Electron donating
p
groups were found to increase Keq (r = -2.2) while decreasing
p
p
kins (r = 1.1) predicting a r for copolymerization of -1.1. The
p
experimentally determined r value for the copolymerization of
CO with a series of substituted styrenes was found to be -0.8
supporting the validity of the combined rate expression.
The mechanism of chain transfer in TBS/CO copolymerization
was also probed experimentally. Chain transfer was found to
increase with TBS concentration and decrease with CO pressure.
Two mechanisms consistent with these observations, b-hydrogen
elimination/associative substitution and monomer assisted
b-hydrogen abstraction, were envisioned to both be possible from
1
an alkyl intermediate. Data from in situ H NMR kinetic exper-
iments on formation of enone from model complex 20as,p were
consistent with either mechanism. However the monomer assisted
b-hydrogen abstraction mechanism was supported through the
study of additive effects on the rate of chain transfer. Addition of
acetonitrile, which was shown to efficiently compete with styrene
for binding at the metal center, had no effect on the rate of chain
transfer. The inability of acetonitrile to trigger chain transfer
supports a mechanism in which styrene assisted b-hydrogen
abstraction is favoured over b-hydrogen elimination/associative
substitution.
Once again, acetonitrile was ineffective at increasing the rate of
chain transfer. The independence of the rate of chain transfer on
the concentration of acetonitrile supports a chain transfer mech-
anism which relies on monomer assisted b-hydrogen abstraction
(pathway B, Scheme 8).§
Summary
Palladium complexes 1a and 1b efficiently copolymerize styrenes
and CO to high molecular weight polyketones. The use of
§ Similar to the arguments made for the possible effects of added CH3CN,
one could propose that CO should also function as an excellent ligand for
displacement of enone from 21 in the pathway A mechanism, Scheme 8.
Therefore (provided formation of 13 is not rate-limiting) both trapping
and chain transfer should show a first-order dependence on [CO] and thus
no effect on MWD should result from changes in CO concentration. Using
this analysis, the simple observation that MWD narrows with increasing
[CO] itself supports pathway B for chain transfer.
Experimental
General comments
All reactions, except where indicated, were carried out in flame
dried glassware under a dry, oxygen-free argon atmosphere using
standard Schlenk and drybox techniques. Non-deuterated solvents
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8977–8992 | 8987
©

were distilled under a nitrogen atmosphere from a drying agent
immediately prior to use: CH2Cl2 from P4O10; hexanes and
diethyl ether from sodium benzophenone ketyl. Acetonitrile was
deoxygenated and dried via passage over a column of activated
alumina.43 CD2Cl2 and CDCl2F were dried over CaH2, submitted
to 5 freeze pump thaw cycles and then vacuum transferred into
glass Schlenk tubes fitted with Kontes high vacuum Teflon plugs,
and then stored under Ar. CP grade CO was purchased from
National Welders Supply and used as received. 4-Tert-butylstyrene
was generously donated by Deltech and stored under Ar over
molecular sieves. Styrene, 4-chlorostyrene, and 4-vinylanisole were
used as received from Aldrich.
complete, the reaction mixture was stirred for an additional 5 min;
then excess acetonitrile was removed at reduced pressure. The
product obtained in this way appeared as an air-stable, colourless
or pale-yellow glass, which could be converted to a solid by
tituration with toluene. An analytical sample was obtained by
repeated washing with toluene and drying for 24 h in vacuo.
Found: C 47.26, H 2.38, N 3.35. C47H26BF24N3Pd requires C
46.81, H 2.17, N 3.48. (1,10-Phenanthroline)Pd(CH3)2 (31.6 mg,
0.1 mmol) was dissolved in dry acetonitrile (10 mL) in a Schlenk
tube under a N2 atmosphere, and while this solution was stirred,
H+(OEt2)2Ar’4B- (110 mg, 1.09 ¥ 10-4 mol) in acetonitrile (10 mL)
was added over a 10 min period. After addition was complete,
the reaction mixture was stirred for an additional 5 min then
excess acetonitrile was removed at reduced pressure. The product
obtained in this way appeared as an air-stable, colourless or
pale-yellow glass, which could be converted to a solid by titration
with toluene. An analytical sample was obtained by repeated
washing with toluene and drying for 24 h in vacuo. Anal. calcd
for C47H26BF24N3Pd: C 46.81, H 2.17, N 3.48. Found: C 47.26, H
2.38, N 3.35. IR (CD2Cl2) nCN = 2326 cm-1. 1H NMR (250 MHz,
CD2Cl2, 20 ◦C) d 8.89 (1 H, dd, Ha), 8.81 (1 H dd, Ha¢), 8.62
(2 H, m), 8.03 (2 H, AB q,), 7.92 (2 H, oct,), 7.74 (8 H, br t,
Ar¢-Ho), 7.56 (4 H, s, Ar¢-Ho), 2.52 (3 H, s, CH3CN), 1.28 (3 H, s,
IR spectra were recorded on a Mattson-Polaris FT-IR Spec-
trophotometer. 1H NMR spectra were recorded on Varian XL400,
Bruker AMX300, Bruker WM250 at 400, 300, 250 MHz, respec-
tively. 13C NMR were recorded on a Varian XL400 and Bruker
1
AMX300 at 100 and 75 MHz, respectively. H and 13C chemical
shifts were referenced to residual 1H signals and to the 13C signals
of the deuterated solvents, respectively. In the 1H NMR data, s, br s,
d, dd, br d, br dd, br t, m, q and oct refer to singlet, broad singlet,
doublet, double doublet, broad doublet, broad double doublet,
broad triplet, multiplet, quartet, octet.
NMR probe temperatures were measured using an anhydrous
methanol sample except for temperatures below -95 ◦C, which
were determined using a thermocouple.
Atom labelling schemes for the phenanthroline, bipyridine and
BArF counterion resonances are as follows:
Pd-CH3). 13C NMR (100 MHz, CD2Cl2, 20 ◦C) d 162.1 (q, JCB
50 Hz, Ar¢-Ci), 149.6 (d, JCH = 179 Hz, Ca or Ca¢), 148.8 (d, JCH
=
=
183 Hz, Ca¢ or Ca), 148.3 (s, Ce or Ce¢), 144.4 (s, Ce¢ or Ce), 140.2
(d, JCH = 167 Hz, Cc or Cc¢), 139.5 (d, JCH = 167 Hz, Cc¢ or Cc),
135.2 (d, JCH = 159 Hz, Ar¢-Co), 131.2 (s, Cf or Cf¢), 130.6 (s, Cf¢
2
or Cf), 129.3 (q, JCF = 31 Hz, Ar¢-Cm), 128.1 (d, JCH = 167 Hz,
Cd or Cd¢), 127.9 (d, JCH = 169 Hz, Cd¢ or Cd), 126.2 (d, JCH
=
169 Hz, Cb or Cb¢), 125.7 (d, JCH = 171 Hz, Cb¢ or Cb), 125.0 (q,
1JCF = 272 Hz, CF3), 122.6 (q, 2JCH = 10 Hz, CH3CN), 117.9 (d,
JCH = 164 Hz, Ar¢-Cp), 3.9 (q, JCH = 138 Hz, CH3), 3.4 (q, JCH
=
136 Hz, CH3).
The 1H NMR resonances were assigned into groups of a, b, c, d, f
or a¢, b¢, c¢, d¢, f¢ according to their characteristic coupling patterns.
The 13C NMR resonances were assigned in pairs such as Ca or Ca¢
and Ca¢ or Ca, based on their chemical shifts and 1JCH. The relation
between the 1H and 13C assignments and the stereochemistry with
respect to the ligands X,Y has not been ascertained.
Alternative procedure for the preparation of 1a: a flame
dried Schlenk flask was charged with 1.97 g (5.85 mmol) of
[Pd(CH3)(Cl)(phen)], prepared similarly as [Pd(CH3)Cl(bpy)],48,51
and suspended in 50 mL of CH2Cl2. The suspension was then
transferred by cannula to a solution of 5.93 g (1.1 equiv.) of
[Na][BArF] and CH3CN (3 mL) in 40 mL CH2Cl2. After stirring
for 3 h at room temperature, a yellow cloudy suspension (NaCl)
remains. The solution was filtered through Celite and a yellow
glass remained following solvent removal in vacuo. The glass was
washed three times with hexane to give a cream coloured foam
isolated in 87% yield (4.70 g). The 1H and 13C NMR spectra were
identical to 1a prepared above.
Errors reported for rate constants determined by kinetics are the
standard deviations of several runs. Errors in DG‡, DH‡ and DS‡
were calculated based on the derivations by Binsch and Girolami,
respectively, and also incorporated a 1◦ error in temperature.44,45
p
Values for s were taken from March’s text.46
C, H, N analyses were performed by Oneida Research Labora-
tories of Whitesboro, NY.
[H+(OEt2)2][BarF],47 CDCl2F,48 [Pd(CH3)2(TMEDA)],49 [Pd-
(CH3)2(phen)],34 [Pd(CH3)2(bpy)],49 [Pd(CH3)Cl(COD)],50 [Pd-
(CH3)Cl(bpy)],51 [Pd(CH3)(CO)(bpy)][BArF],20 [Pd(CH3)(CO)-
(phen)][BArF],20 [Pd(C(O)CH3)(CO)(phen)][BArF],20 were pre-
pared according to literature procedures.
[Pd(CH3)(NCCH3)(bpy)][BArF] (1b). Complex 1b was pre-
pared in the same manner as its phenanthroline analog 1a
from solutions of [Pd(CH3)2(bpy)] (29.2 mg, 0.1 mmol) in dry
acetonitrile (10 mL) and [H+(OEt2)2][BArF] (110 mg 1.1 ¥
10-4 mol) in acetonitrile (10 mL). Found: C 46.31, H 2.16, N 3.12.
C45H26BF24N3Pd requires C 45.73, H 2.22, N 3.56. IR (CD2Cl2)
nCN = 2328 cm-1. 1H NMR (400 MHz, CD2Cl2, 20 ◦C) d 8.58 (1
H, d, J = 5.5 Hz, Ha or Ha¢), 8.52 (1 H, d, J = 4.5 Hz, Ha¢ or Ha),
8.06 (4 H, m, Hf, Hf¢, Hc, Hc¢), 7.92 (8 H, s, Ar¢-Ho), 7.69 (4 H, s,
Ar¢-Hp), 7.56 (2 H, m, Hb, Hb¢), 2.44 (3 H, s, CH3CN), 1.14 (3 H, s,
I. Preparations
[Pd(CH3)(NCCH3)(phen)][BArF]
(1a). [Pd(CH3)2(phen)]
(31.6 mg, 0.1 mmol) was dissolved in dry acetonitrile (10 mL) in a
Schlenk tube under a N2 atmosphere, and while this solution was
stirred, [H+(OEt2)2][BArF] (110 mg, 1.09 ¥ 10-4 mol) in acetonitrile
(10 mL) was added over a 10 min period. After addition was
Pd-CH3). 13C NMR (100 MHz, CD2Cl2, 20 ◦C) d 162.5 (q, JCB
50 Hz, Ar¢-Ci), 157.8 (s, Ce or Ce¢), 153.2 (s, Ce¢ or Ce), 149.7 (d,
=
8988 | Dalton Trans., 2009, 8977–8992
This journal is
The Royal Society of Chemistry 2009
©

JCH = 184 Hz, Ca or Ca¢), 148.8 (d, JCH = 184 Hz, Ca or Ca¢), 140.9
(d, JCH = 169 Hz, Cc or Cc¢), 140.7 (d, JCH = 168 Hz, Cc¢ or Cc),
135.5 (d, JCH = 159 Hz, Ar¢-Cp), 129.7 (q, 2JCF = 31.6 Hz, Ar¢-Cm),
127.8 (d, JCH = 170 Hz, Cf or Cf¢), 127.5 (d, JCH = 170 Hz, Cf¢ or
Cf), 125.3 (q, 1JCF = 272 Hz, CF3), 123.7 (d, JCH = 167 Hz, Cb or
2
Cb¢), 120.8 (d, JCH = 166 Hz, Cb¢ or Cb), 122.6 (q, JCH = 10 Hz,
H3C-CN), 118.2 (d, JCH = 164 Hz, Ar¢-Cp), 3.8 (q, JCH = 136 Hz,
Pd-CH3 or CH3CN), 3.5 (q, JCH = 138 Hz, CH3CN or Pd-CH3).
1H NMR (400 MHz, CD2Cl2, 20 ◦C) d 8.40 (1 H, d, J = 4.8 Hz,
Ha or Ha¢), 8.31 (1 H, d, J = 4.8 Hz, Ha¢ or Ha), 8.06 (4H, m,
Hf, Hf¢, Hc, Hc¢), 7.72 (8 H, s, Ar¢-Ho), 7.55 (4 H, s, Ar¢-Hp), 7.54
(2 H, m, Hb, Hb¢), 7.50 (2 H, d, J = 8 Hz, Hb or Hg), 7.12 (2 H, d,
J = 8 Hz, Hg or Hb), 3.94 (1 H, dd, J = 7.4 Hz, J = 5.0 Hz, CHx),
3.08 (1 H, AB dd, J = 19.8 Hz, J = 7.4 Hz, CHAHB), 2.96 (1 H,
AB dd, J = 19.8 Hz, J = 5.0 Hz, CHAHB), 2.44 (3 H, s, CH3),
1.34 (9 H, s, C(CH3)3). 1H NMR (400 MHz, CDCl2F, -101 ◦C) d
6.91 (1 H, d, J = 7.2 Hz, p-Hb¢), 6.64 (1 H, d, J = 6.4 Hz, p-Hb),
3.8 (1 H, broad resonance (W1/2 = 20 Hz, p-H), 3.60 (3 H, d, J =
7.2 Hz, s-CH), 3.51 (3 H, dd, J = 19 Hz, J = 7 Hz, s-CH2), 3.16
(1 H, broad, J = 19.5 Hz, p-CH2), 2.91 (1 H, dd, J = 19.5 Hz,
J = 10 Hz, p-CH2), 2.85 (3 H, d, J = 19 Hz, s-CH2), 2.62 (9 H, s,
s-CH3), 2.33 (3 H, s, p-CH3), 1.36 (9 H, s, p-C(CH3)3), 1.24 (27 H, s,
s-C(CH3)3). The remaining aromatic resonances form a complex
series of multiplets in the range of 8.8 to 7.1 ppm. The resonances
were referenced to residual methylene chloride at 5.32 ppm.
An equilibrium constant was determined to be ca. 3 to 1 from the
ratio of 5bs to 5bp resonances in the -101 ◦C 1H NMR spectrum.
13C NMR (100 MHz, CD2Cl2, -80 ◦C) d 238.4 (s, s-CO), 204.5
(s, p-CO), 161.5 (q, 1JCB = 50 Hz, Ar¢-Ci), 155.6 (s, s-Ce or s-Ce¢),
154.8 (s, p-Ce or p-Ce¢ or p-Cd), 153.9 (s, p-Ce¢ or Ce or p-Ce), 152.8
(Ce or p-Ce or p-Ce¢), 151.9 (d, JCH = 186 Hz, p-Ca or p-Ca¢), 151.3
(s, s-Ce or s-Ce¢), 150.6 (d, JCH = 187 Hz, s-Ca or s-Ca¢), 148.4
[Pd(C(O)CH3)(CO)(bpy)][BArF] (3b). An NMR tube was
charged with [Pd(CH3)(CO)(bpy)][BArF] (4b) (ca. 100 mg, 8.5 ¥
10-5 mol) and CD2Cl2 (0.9 mL) then purged with CO at -78 ◦C for
ca. 15 min. The conversion to 3b was observed spectroscopically,
some precipitation was observed.
IR (CH2Cl2) nCO = 2128 (PdCO), 1746 (PdC(O)CH3) cm-1. 1H
NMR (400 MHz, CD2Cl2, -80 ◦C) d 8.33 (1 H, d, J = 5.1 Hz, Ha¢
or Ha), 8.15 (1 H, d, J = 5.3 Hz, Ha or Ha¢) 7.8 (4 H, m, Hf, Hf¢,
Hc, Hc¢), 7.78 (8 H, s, Ar¢-Ho), 7.57 (1 H, m, Hb¢ or Hb) 7.51 (4 H, s,
Ar¢-Hp), 7.46 (1 H, m, Hb or Hb¢), 2.77 (3 H, s, CH3). 13C NMR
(100 MHz, CD2Cl2, -80 ◦C) d 217.4 (s, PdC(O)CH3), 172.5 (s,
PdCO), 161.5 (q, 1JCB = 50 Hz, Ar¢-Ci), 154.3 (s, Ce or Ce¢), 151.6
(s, Ce¢ or Ce), 150.7 (d, JCH = 183 Hz, Ca or Ca¢), 150.3 (d, JCH
=
191 Hz, Ca¢ or Ca), 141.9 (d, JCH = 170 Hz, Cc or Cc¢), 141.2 (d,
JCH = 169 Hz, Cc¢ or Cc), 134.3 (d, JCH = 158 Hz, Ar¢-Co), 128.3
(q, 2JCF = 35 Hz, Ar¢-Cm), 127.8 (d, JCH = 185 Hz, Cf or Cf¢), 127.7
(d, JCH = 181 Hz, Cf¢ or Cf), 124.0 (q, 1JCF = 273 Hz, CF3), 123.1
(d, JCH = 168 Hz, Cb or Cb¢), 122.8 (d, JCH = 167 Hz, Cb¢ or Cb),
117.2 (d, JCH = 165 Hz, Ar¢-Cp), 40.6 (q, JCH = 132 Hz, CH3).
[Pd(13C(O)CH3)(13CO)(bpy)][BArF] (3b*). An NMR tube
charged with [Pd(CH3)(13CO)(bpy)][BArF] in CD2Cl2 was briefly
purged with 13CO (99%) at -78 ◦C to give 3b*. The spectroscopic
13
data was identical to 3b except for: IR (CH2Cl2) n
=
CO
(d, JCH = 185 Hz, s-Ca¢ or s-Ca), 148.4 (s, s-Ca), 147.7 (d, JCH
=
2079 (Pd13CO), 1705 (Pd13C(O)CH3) cm-1. H NMR (400 MHz,
1
183 Hz, p-Ca¢ or p-Ca), 143.4 (s, s-Ca), 139.9 (d, JCH = 170 Hz,
◦
2
1
CD2Cl2, -80 C) d 2.77 (3 H, d, JCH = 5.9 Hz, CH3). 13C{ H}
s-Cc and s-Cc¢), 134.3 (d, JCH = 158 Hz, Ar¢-Co), 130.4 (d, JCH
=
NMR (100 MHz, CD2Cl2, -80 ◦C) d 40.6 (dd 1JCC = 30 Hz, 3JCC
=
157 Hz, p-Cg or p-Cg¢), 129.8 (d, JCH = 153 Hz, p-Cg¢ or p-Cg),
128.4 (q, 2JCF = 31 Hz, Ar¢-Cp), 127.5 (d, JCH = 174 Hz, p-Cf or
11 Hz, CH3).
p-Cf¢), 127.2 (d, JCH = 168 Hz, s-Cf or s-Cf¢), 126.9 (d, JCH
=
[Pd(g3-CH(CH2C(O)CH3)C6H4-p-C4H9)(bpy)][BArF]
[Pd(CH(CH2C(O)CH3)C6H4-p-C4H9)(bpy)][BArF]
(5bp),
(5br).
169 Hz, s-Cf¢ or s-Cf), 126.4 (d, JCH = 169 Hz, s-Cb), 125.0 (d,
1
JCH = 178 Hz, s-Cg), 124.1 (q, JCF = 272 Hz, CF3), 122.6 (d,
[H+(OEt2)2][BArF] (1.00 g, 9.84 ¥ 10-4 mol) in CH2Cl2 (15 mL)
was slowly syringed onto a bright yellow slurry of [Pd(CH3)2(bpy)]
(2b) (288 mg, 9.84 ¥ 10-4 mol) in CH2Cl2 (30 mL) giving a yellow-
orange solution at -30 ◦C. The temperature was maintained at
-30 ◦C while the solution was stirred for 1 h, then purged with
CO for 45 min. The presence of 3b (nCO = 2128, 1746 cm-1) was
verified by IR spectroscopy. After cooling the solution to -78 ◦C,
it was purged with N2 for 1 h to remove dissolved CO, then 1
equiv. of 4-tert-butylstyrene solution (9.84 ¥ 10-4 mol in 4.6 mL
CH2Cl2) was added. The N2 purge was continued as the reaction
was warmed to -30 ◦C for 1 h and then to 0 ◦C for 1 h. The
small amount of black precipitate that formed upon warming
was removed by filtration through Celite. The filtrate was reduced
JCH = 166 Hz, s-Cb or s-Cb¢), 122.1 (d, JCH = 169 Hz, p-Cb or
p-Cb¢), 121.8 (d, JCH = 168 Hz, s-Cb¢ or s-Cb), 117.2 (d, JCH
=
165 Hz, Ar¢-Cp), 114.7 (s, p-Ca), 107.9 (d, JCH = 165 Hz, p-Ca¢),
104.2 (d, JCH = 172 Hz, p-Ca), 56.7 (t, JCH = 127 Hz, s-CH2), 52.0
(d, JCH = 160 Hz, p-C), 44.0 (d, JCH = 141 Hz, s-CH), 42.9 (t,
JCH = 125 Hz, p-CH2), 35.5 (s, p-C(CH3)3), 34.2 (s, s-C(CH3)3),
30.2 (q, JCH = 125 Hz, s-C(CH3)3), 29.9 (q, JCH = 127 Hz, p-CH3),
29.6 (q, JCH = 122 Hz, p-C(CH3)3), 28.2 (q, JCH = 130 Hz, s-CH3).
The remaining resonances for (p-Cf¢ or p-Cf) and (p-Cb¢ or p-Cb)
may be hidden by d 126.9 (s-Cf¢ and s-Cf) and d 122.6 (s-Cb or
s-Cb¢), respectively.
[Pd(g3-CH(CH213C(O)CH3)C6H4-p-C4H9)(bpy)][BArF] (5bp*)
and [Pd(CH(CH213C(O)CH3)C6H4-p-C4H9)(bpy)][BArF] (5br*).
Complex 5b* was prepared similarly to 5b, but on a smaller scale.
◦
under vacuum at 0 C to a bright yellow glass. Triturating with
hexane, then drying in vacuo gave a brittle yellow glass which was
crushed to a yellow powder (910 mg, 6.85 ¥ 10-4 mol, 70% yield).
Found: C 50.83, H 2.71, N 1.97. C56H41F24O1N2Pd requires C
51.14, H 3.14, N 2.13. IR (CH2Cl2) nCO = 1719 (5bp), 1617 (sh,
5bs) cm-1. Labelling scheme:
◦
1
The 13C{ H} NMR spectrum was identical to 5b at -80 C except
for 1JCC couplings.
IR (CH2Cl2, 25 C) n
◦
= 1680 (5bp*), 1582 (5bs*) cm-1.
13
CO
13C{ H} NMR (100 MHz, CD2Cl2, -80 ◦C) d 56.7 (dd,
1
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8977–8992 | 8989
©

1
1JCC = 43 Hz, 5bs*-CH2), 42.9 (d, JCC = 36 Hz, 5bp*-CH2),
tube was charged with 5b (ca. 100 mg, 7.5 ¥ 10-5mol) and CD2Cl2
(0.9 mL) then purged with CO at -78 ◦C for ca. 15 min. The
conversion of 5b to 6b was observed spectroscopically.
1
26.2 (d, JCC = 40 Hz, 5bs*-CH3). The missing resonance for
d 29.9 (5bp*-CH3) is split under the broad neighbouring peaks
d 30.2 (5bs*-(C(CH3)3)) and d 29.6 (5bp*-(C(CH3)3)).
IR (CH2Cl2) nCO = 2126 (PdCO), 1742 (PdC(O)–), 1715
(–C(O)CH3) cm-1. 1H NMR (400 MHz, CD2Cl2, -80 ◦C) d 8.45 (1
H, br s, Ha or Ha¢), 8.22 (1 H, d, J = 4.92 Hz, Ha¢ or Ha), 8.65 (4 H,
m, Hf, Hf¢, Hc Hc¢), 7.73 (8 H, s, Hb), 7.57 (2 H, br s, Hb, Hb¢), 7.36
(1 H, d, J = 8.0 Hz, Hb or Hg), 7.25 (1 H, d, JCH = 7.8 Hz, Hg or
Hb), 4.77 (1 H, d, J = 11.3 Hz, CH2), 3.70 (1 H, dd, J = 18.4 Hz,
J = 11.3 Hz, CH2), 2.80 (1 H, d, J = 18.5, CH), 2.24 (3 H, s,
CH3). 13C NMR (100 MHz, CD2Cl2, -80 ◦C) d 218.7 (s, Pd-C(O)-
CH(Ar)–), 207.5 (s, –CH2C(O)CH3), 172.2 (s, Pd-CO), 161.5 (q,
JCB = 50 Hz, Ar¢-Ci), 154.1 (s, Ce or Ce¢), 152.4 (s, Ar¢-Cp), 152.1
[Pd(CH(C6H5)CH2C(O)CH3)(phen)][BarF] (5ar) and [Pd(g3-
CH(CH2C(O)CH3)C6H5)(phen)][BArF] (5ap). Styrene (17.4 mL,
1 equiv.) was added to a CH2Cl2 (1 mL) solution of 4a (181 mg,
1.52 ¥ 10-4 mol) at 20 ◦C. The colour turned intense yellow. After
stirring briefly, the solvent was removed in vacuo to yield 5as,p as
a yellow glass (169 mg, 81%).
IR (CH2Cl2) nCO = 1717 (p), 1614 (s) cm-1. 1H NMR (400 MHz,
CD2Cl2, 20 ◦C) d 8.9 (1 H, br d, Ha or Ha¢), 8.54 (2 H, w, Hc and
Hc¢), 8.4 (1 H, br d, Ha¢ or Ha), 7.99 (1 H, d, Jd,d¢ = 8.8 Hz, Hd or
Hd¢), 7.95 (1 H, d, Jd,d¢ = 8.8 Hz, Hd¢ or Hd), 7.89 (1 H, dd, J=
8.2 Hz, J = 5.2 Hz, Hb or Hb¢), 7.7 (9 H, m, Ar¢-Ho + Hb¢ or Hb),
7.55 (4 H, s, Ar¢-Hp), 7.5 (1 H, m, p-H-C6H4), 7.4 (4 H, m, C6H4),
4.10 (1 H, dd, JAM = 7 Hz, JBM = 5 Hz, CHMAr), 3.30 (1 H, dd,
JAB = 20 Hz, JAM = 7 Hz, CHAHB), 3.11 (1 H, dd, JAB = 20 Hz,
JBM = 5.0 Hz, CHAHB), 2.58 (3 H, s, CH3).
(d, JCH = 193 Hz, Ca or Ca¢), 151.9 (s, Ce¢ or Ce), 150.1 (d, JCH
=
188 Hz, Ca¢ or Ca), 141.3 (d, JCH = 168 Hz, Cc or Cc¢), 141.0 (d,
JCH = 169 Hz, Cc¢ or Cc), 129.4 (d, JCH = 145 Hz, Cb or Cg), 128.8
(s, Ar¢-Ci), 128.3 (q, 2JCF = 31 Hz), 127.4 (d, JCH = 182 Hz, Cf or
Cf¢), 127.3 (d, JCH = 182 Hz, Cf¢ or Cf), 126.7 (d, JCH = 157 Hz, Cg
or Cb), 124.0 (q, 1JCF = 273 Hz, CF3), 122.7 (d, JCH = 167 Hz, Cb
or Cb¢), 122.6 (d, JCH = 167 Hz, Cb¢ or Cb), 117.2 (d, JCH = 156 Hz,
Ar¢-Cp), 62.1 (d, JCH = 137 Hz, CH), 46.1 (t, JCH = 129 Hz, CH2),
Low temperature 1H NMR revealed 5as–5ap = 11 : 1.
◦
1
5as: H NMR (400 MHz, CD2Cl2, -90 C) d 8.92 (1 H, br d,
phen), 8.63 (1 H, br d, phen), 8.39 (2 H, m, phen), 7.8 (12 H, m,
Hb and Hb¢ + Hd and Hd¢ + Ar¢-Ho), 7.3 (2 H, m, Ph), 7.2 (3 H,
m, Ph), 3.88 (1 H, br d, J = ca. 7.5 Hz, CHAr), 3.55 (br dd, J =
21 Hz, J = ca. 7.5 Hz, CHH), 2.94 (br d, J = 21 Hz, CHH), 2.63
(3 H, s, CH3).
34.2 (s, C(CH3)3), 30.3 (q, JCH = 126 Hz, C(CH3)3), 29.4 (q, JCH
=
128 Hz, C(O)CH3).
1
(a) 13C{ H} NMR monitoring of chain growth:
5b* → 6b* → 7b* → 8b*.
Complex 5b*, prepared as described from [Pd(CH3)2(bpy)]
(14.4 mg, 4.92 ¥ 10-5 mol), was dissolved in CD2Cl2 (0.9 mL), then
syringed into a stoppered NMR tube and cooled to -78 ◦C. The
sample was briefly purged with 13CO (99%), then placed in a cold
NMR probe (-80 ◦C). The conversion of 5b* to 6b* and some
[Pd(CH3)(13CO)(bpy)][BArF] to 3b*, respectively, was observed
◦
1
5ap: H NMR (400 MHz, CD2Cl2, -90 C) d 9.03 (m, phen),
8.4 (m, phen), (remainder aromatics are obscured), 7.03 (1 H, d,
ortho H of Bz), 6.66 (1 H, d, ortho H of Bz), 4.5 (1 H, m, CHAr),
3.2 (d, J = ca. 20 Hz), 2.32 (3 H, s, CH3). The missing resonances
of 5ap are obscured by those of the major isomer 5as.
1
by 13C{ H} NMR spectroscopy. The sample was then removed
Warming the sample to -81 ◦C caused the methyl resonances of
the two isomers to broaden significantly. The rate of conversion of
the minor isomer 5ap into the major, 5as, was determined from
k = p(W - Wo), where W = 12 Hz, Wo = 3 Hz, k = 36 Hz,
from the NMR probe and placed in a dry ice/isopropanol bath.
4-t-Butyl styrene (10 mL, 5.49 ¥ 10-5 mol, ca. 6 equiv., as
determined spectroscopically) was syringed into the cold sample
which was shaken rapidly and returned to the cold NMR probe
(-80 ◦C). The resonances for 7b* began to appear 40 min after the
addition of olefin. After 2 h all the carbonyl resonances for 7b*
and all but one for 8b* could be identified. The resonances of 3b*
slowly decreased in intensity.
DG‡
= 9.7 0.1 kcal mol-1. Knowing Keq = 11 : 1, DG◦ =
min→maj
0.9 kcal mol-1, gives DG‡
= 10.6 0.1 kcal mol-1.
maj→min
Trapping with Styrene. A 5 mm NMR tube was charged with
4a (10 mg, 8 ¥ 10-6 mol) and CD2Cl2 (0.7 mL). Styrene (9 mL,
10 equiv.) was added to the sample at -78 ◦C. The sample was
briefly shaken to dissolve the frozen styrene on the inside of
the tube. The sample was then placed in a pre-cooled NMR
Labelling scheme:
probe (-66 ◦C) where the conversion of 4a to the intermediate
[Pd(C(O)CH3)(C6H5CH CH2)(phen)] (8a) en route to 5as,p was
+
=
observed. The characterization of 8a was limited because free and
coordinated styrene exchange on the NMR timescale. The methyl
resonances of 4a (d = 1.5), 8a (d = 2.15) and 5as,p (d = 2.65) (time
averaged) and the carbonyl acyl 3a (d = 2.9) and the styrene methyl
9a (d = 0.9) were monitored with time. 8a: 1H NMR (300 MHz,
◦
CD2Cl2, -66 C), d 8.6 (2 H, br d, J ~ 8.5 Hz, Hc and Hc¢), 8.4
◦
1
7b*: 13C{ H} NMR (100 MHz, CD2Cl2, -80 C) d 219.2 (7-C2),
(2 H, br d, J ~ 5 Hz, Ha and Ha¢), 7.97 (2 H, s, Hd and Hd¢), 7.9 (2
H, dd, J ~ 8.5 Hz, J ~ 5 Hz, Hb and Hb¢), 2.15 (3 H, s, CH3).
207.6 (7-C4), 207.3 (7-C3), 172.3 (7-C1).
8b*: 13C{ H} NMR (100 MHz, CD2Cl2, -80 ◦C) d 218.0 (8-
1
C2), 207.3, 207.1 (8-C3, C5 or C6), 172.40 (8-C1). Some organic
carbonyl resonances are believed to be coincident with other
similar resonances. After 2.5 h at -80 ◦C, all of 3b* had
been converted to 6b*. The temperature was then raised to
-60 ◦C where distinct bands of resonances were observed for the
II. Polymerizations
In situ copolymerizations of p-t-butylstyrene with 13CO. In situ
copolymerizations of p-t-butylstyrene with 13CO. [Pd(C(O)CH(p-
C4H9-C6H4)CH2C(O)-CH3)(CO)(bpy)][BArF] (6b). An NMR
8990 | Dalton Trans., 2009, 8977–8992
This journal is
The Royal Society of Chemistry 2009
©

Pd-CO, PdC(O)R and organic carbonyls at slightly different
chemical shifts, which are presumably temperature dependent.
(280 mg, 83% yield). 1H data for 20as,p at room temperature are
given below.
13C{ H} NMR (100 MHz, CD2Cl2, -60 ◦C) d 218.7 (7-C2),
20a: 1H NMR (400 MHz, CD2Cl2, +20 ◦C): d 8.7 (1 H, m, Ha
or Ha¢), 8.55 (2 H, d, J = 8 Hz, Hc, Hc¢), 7.99 (2 H, d, J = 2 Hz,
Hd, Hd¢), 7.87 (1 H, dd, J = 8 Hz, J = 5 Hz, Hb or Hb¢), 7.82 (1 H,
dd, J = 8 Hz, J = 4 Hz, Hb or Hb¢), 7.72 (8 H, s, Ho), 7.54 (4 H, s,
Hp), 7.53 (2 H, d, J = 8 Hz, Ho or Hm), 7.20 (2 H, d, J = 8 Hz, Ho
or Hm), 4.15 (1 H, overlapping dd, J = 6 Hz, J = 6 Hz, CH), 3.10
(2 H, dd, J = 6 Hz, J = 6 Hz, CH2), 2.50 (3 H, s, C(O)CH3), 1.36
(9 H, s, C(CH3)3).
1
218.3 (6-C2), 217.6 (8-C2), 207.7 (6-C3 + 7-C4), 207.4 (8-C5 or 8-C6
+ 7-C3 + 8-C3), 207.2 (8-C6 or 8-C5), 172.7 (8-C1), 172.6 (7-C1),
172.5 (6-C1).
1
(b) 13C{ H} NMR monitoring of chain growth:
5b → 6b*u → 7b*u → 8b*u → 5b*u
The “u” after the compound number denotes a non-13C labelled
acetyl end group. 5b was prepared on the same scale as 5b*. The
procedure was similar to the previous one, except that 5b was
used instead of 5b* and p-t-butyl styrene (30 mL, 1.65 ¥ 10-4 mol,
ca. 10 equiv. as determined spectroscopically) was used. 6b*u
Acknowledgements
1
exhibited two carbonyl resonances: 13C{ H} NMR (100 MHz,
We are grateful to the Department of Energy (FG05-94ER14459),
the National Science Foundation (CHE-0615704), and DuPont for
financial support of this work. F. C. R was supported in part by a
Department of Education Fellowship and J. C. B. was supported
by an NSF-ROA grant.
CD2Cl2, -80 ◦C) d 218.7 (6-C2), 172.2 (6-C1). 7b*u exhibited three
◦
1
carbonyl resonances: 13C{ H} NMR (100 MHz, CD2Cl2, -80 C)
d 219.2 (7-C2), 207.6 (7-C4), 172.3 (7-C1). After 45 min at -80 ◦C,
the temperature was raised to -60 ◦C. 8b*u exhibited four carbonyl
◦
resonances: 13C{ H} NMR (100 MHz, CD2Cl2, -60 C) d 217.5
(8-C2), 207.5 (8-C6 or 8-C5), 207.2 (8-C5 or 8-C6), 172.7 (8-C1).
After 2 h 45 min at -60 ◦C, the temperature was raised to -40 ◦C
to induce more rapid chain growth. Extensive chain growth was
1
References
1 A. Sen and T.-W. Lai, J. Am. Chem. Soc., 1982, 104, 3520.
2 T.-W. Lai and A. Sen, Organometallics, 1984, 3, 866.
3 E. Drent, J. A. M. van Broekhoven and M. J. Doyle, J. Organomet.
Chem., 1991, 417, 235.
4 E. Drent and P. H. M. Budzelaar, Chem. Rev., 1996, 96, 663.
5 C. Bianchini and A. Meli, Coord. Chem. Rev., 2002, 225, 35.
6 K. Nakano, N. Kosaka, T. Hiyama and K. Nozaki, Dalton Trans.,
2003, 4039.
7 J. Durand and B. Milani, Coord. Chem. Rev., 2006, 250, 542.
8 E. J. Garc´ıa Sua´rez, C. Godard, A. Ruiz and C. Claver, Eur. J. Inorg.
Chem., 2007, 2582.
9 T. M. J. Anselment, S. I. Vagin and B. Rieger, Dalton Trans., 2008,
4537.
1
verified by the appearance of the spectrum after 50 min: 13C{ H}
◦
NMR (100 MHz, CD2Cl2, -40 C) d 217.8 (weak, C2), 207–208
(br envelope, Cn), 172.8 (weak C1), 172.7 (weak C1¢).
◦
1
The temperature was raised to 20 C: 13C{ H} NMR (100 MHz,
CD2Cl2, 20 ◦C) d 207–208 (broad envelope).
(c) [Pd(CH3)(13CO)(phen)][BArF] (4a*) + 13CO and TBS.
A
5 mm NMR tube was charged with 4a* (25 mg, 2 ¥ 10-5 mol) a◦nd
CD2Cl2 (0.7 mL). 13CO was purged through the solution at -78 C
for ca. 20 s and the sample placed in a pre-cooled (-100 ◦C) NMR
probe. The sample was warmed to -30 ◦C to increase the rate of
the reaction; complex 3a* formed and the sample was returned to
-100 ◦C. The terminal CO of the carbonyl complex averaged with
the free CO at higher temperatures in this sample. p-t-butyl styrene
(50 mL, 2.7 ¥ 10-4 mol, 14 equiv.) was added to the sample at -78 ◦C
and the sample placed in the (-30 ◦C) NMR probe. The reaction
was followed over the course of ca. 48 h, interrupted by one 5 h
period where the sample was stored at -78 ◦C. Periodically the
sample was cooled to -100 ◦C to obtain spectra of the “quenched”
polymerizing system as 6a*, 7a*, etc.
10 N. Alperwicz, Chem. Week, 1995, 127.
11 B. Milani, A. Anzilutti, L. Vicentini, A. Sessanta, o Santi, E. Zan-
grando, S. Geremia and G. Mestroni, Organometallics, 1997, 16, 5064.
12 A. Scarel, B. Milani, E. Zangrando, M. Stener, S. Furlan, G. Fronzoni,
G. Mestroni, S. Gladiali, C. Carfagna and L. Mosca, Organometallics,
2004, 23, 5593.
13 Je`rome Durand, Ennio Zangrando, Mauro Stener, Giovanna Fronzoni,
Carla Carfagna, Barbara Binotti, Paul C. J. Kamer, Christian Muller,
Maria Caporali, Piet W. N. M. van Leeuwen, Dieter Vogt and B. Milani,
Chem.–Eur. J., 2006.
14 M. Barsacchi, G. Consiglio, L. Medici, G. Petrucci and U. W. Suter,
Angew. Chem., Int. Ed. Engl., 1991, 30, 989.
15 M. Barsacchi, A. Batistini, G. Consiglio and U. W. Suter, Macro-
molecules, 1992, 25, 3604.
16 A. Aeby, A. Gsponer and G. Consiglio, J. Am. Chem. Soc., 1998, 120,
11000.
17 Giambattista Consiglio and B. Milani, in ‘Stereochemical aspects of
cooligomerization and copolymerization’, ed. A. Sen, Dordrecht, 2003.
18 M. Brookhart, F. C. Rix, J. M. DeSimone and J. C. Barborak, J. Am.
Chem. Soc., 1992, 114, 5894.
◦
1
6a*: 13C{ H} NMR (100 MHz, CD2Cl2, -100 C) d 218.8 (6-C2),
207.8 (6-C3), 172.2 (6-C1).
◦
7a*: 13C{ H} NMR (100 MHz, CD2Cl2, -100 C) d 219.5 (7-C2),
1
207.6 (7-C4), 207.4 (7-C3), 172.2 (7-C1).
[Pd(CH(p-t-Bu-C6H4)CH2C(O)CH3)(phen)][BArF] (20ar) and
[Pd(g3-CH(CH2C(O)CH3)p-t-Bu-C6H4)(phen)][BArF] (20bp).
A
19 M. Brookhart, M. I. Wagner, G. G. A. Balavoine and H. A. Haddou,
J. Am. Chem. Soc., 1994, 116, 3641.
flame dried Schlenk flask was charged with 300 mg (0.249 mM)
of 1a, which is then dissolved in 15 mL CH2Cl2 under argon. The
solution was purged with CO at room temperature for 15 min
followed by an argon purge for 25 min at -80 ◦C. One equiv.
20 F. C. Rix, M. Brookhart and P. S. White, J. Am. Chem. Soc., 1996, 118,
4746.
21 M. Brookhart and M. I. Wagner, J. Am. Chem. Soc., 1996, 118, 7219.
22 K. Nozaki, N. Sato and H. Takaya, J. Am. Chem. Soc., 1995, 117, 9911.
23 S. Bartolini, C. Carfagna and A. Musco, Macromol. Rapid Commun.,
1995, 16, 9.
24 M. T. Reetz, G. Aderlein and K. Angermund, J. Am. Chem. Soc., 2000,
122, 996.
25 A. Bastero, C. Claver, A. Ruiz, S. Castillon, E. Daura, C. Bo and E.
Zangrando, Chem.–Eur. J., 2004, 10, 3747.
26 B. Milani, G. Corso, G. Mestroni, C. Carfagna, M. Formica and R.
Seraglia, Organometallics, 2000, 19, 3435.
27 F. C. Rix and M. Brookhart, J. Am. Chem. Soc., 1995, 117, 1137.
◦
of TBS (45.1 mL) was syringed into the Schlenk flask at -80 C
and the reaction was removed from the ice bath and stirred at
room temperature for 1 h. Solvent was removed in vacuo leaving a
yellow glass. The glass was then re-dissolved in 10 mL of toluene
and solvent was removed in vacuo once again to insure that no
free CH3CN was present. The yellow glass, which remained, was
washed three times with hexane, leaving a yellow orange powder
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 8977–8992 | 8991
©

28 C. Carfagna, G. Gatti, D. Martini and C. Pettinari, Organometallics,
2001, 20, 2175.
29 B. Binotti, C. Carfagna, G. Gatti, D. Martini, L. Mosca and C.
Pettinari, Organometallics, 2003, 22, 1115.
30 B. Binotti, G. Bellachioma, G. Cardaci, C. Carfagna, C. Zuccaccia and
A. Macchioni, Chem.–Eur. J., 2007, 13, 1570.
31 K. Nozaki, H. Komaki, Y. Kawashima, T. Hiyama and T. Matsubara,
J. Am. Chem. Soc., 2001, 123, 534.
41 L. Deng, T. K. Woo, L. Cavallo, P. Margl and T. Ziegler, J. Am. Chem.
Soc., 1997, 119, 6177.
42 F. C. Rix, ‘Mechanistic studies of palladium(II) mediated carbon-
carbon bond formation’, University of North Carolina, Chapel Hill,
1994.
43 A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosea and F. J.
Timmers, Organometallics, 1996, 15, 1518.
44 G. Binsch, ‘Dynamic Nuclear Magnetic Resonance Spectroscopy’, ed.
L. Jackman, and F. A. Cotton, Academic Press, 1975.
45 P. M. Morse, M. D. Spencer, S. R. Wilson and G. S. Girolami,
Organometallics, 1994, 13, 1646.
46 J. March, ‘Advanced Organic Chemistry’, John Wiley & Sons, 1985.
47 M. Brookhart, B. Grant and A. F. Volpe, Jr., Organometallics, 1992,
11, 3930.
48 J. S. Siegel and F. A. Anet, J. Org. Chem., 1988, 53, 2629.
49 W. de Graaf, J. Boersma, W. J. J. Smeets, A. L. Spek and G. van Koten,
Organometallics, 1989, 8, 2907.
32 J.-T. Chen and A. Sen, J. Am. Chem. Soc., 1984, 106, 1506.
33 A. Sen, J.-T. Chen, W. M. Vetter and R. R. Whittle, J. Am. Chem. Soc.,
1987, 109, 148.
34 F. C. Rix, M. Brookhart and P. S. White, J. Am. Chem. Soc., 1996, 118,
2436.
35 F. Weigert, and R. McKinney, in ‘Gear-Git Software’.
36 H. E. Bryndza, Organometallics, 1985, 4, 1686.
37 L. K. Johnson, C. M. Killian and M. Brookhart, J. Am. Chem. Soc.,
1995, 117, 6414.
38 H. Sinn and W. Kaminsky, Adv. Organomet. Chem., 1980, 18, 99.
39 D. G. Musaev, M. Svensson, K. Morokuma, S. Stromberg, K.
Zetterberg and P. E. M. Siegbahn, Organometallics, 1997, 16, 1933.
40 L. Fan, A. Krzywicki, A. Somogyvari and T. Ziegler, Inorg. Chem.,
1996, 35, 4003.
50 R. E. Ru¨lke, J. M. Ernsting, A. L. Spek, C. J. Elsevier, P. W. N. M. Van
Leeuwen and K. Vrieze, Inorg. Chem., 1993, 32, 5769.
51 R. E. Ru¨lke, J. G. P. Delis, A. M. Groot, C. J. Elsevier, P. W. N. M. Van
Leeuwen, K. Vrieze, K. Goubitz and H. Schenk, J. Organomet. Chem.,
1996, 508, 109.
8992 | Dalton Trans., 2009, 8977–8992
This journal is
The Royal Society of Chemistry 2009
©