Inhibitory role of carbon monoxide in palladium(II)-catalyzed nonalternating ethene/carbon monoxide copolymerizations and the synthesis of polyethene-block-poly(ethene-alt-carbon monoxide)

Article abstract of DOI:10.1021/om700523m

We report the first well-defined palladium-based system for the living homopolymerization of ethene, as well as the living copolymerization of ethene with carbon monoxide. We demonstrate this by the synthesis of polyethene-block-poly-(ethene-alt-carbon mo

Full text of DOI:10.1021/om700523m

3
636
Organometallics 2007, 26, 3636-3638
Inhibitory Role of Carbon Monoxide in Palladium(II)-Catalyzed
Nonalternating Ethene/Carbon Monoxide Copolymerizations and the
Synthesis of Polyethene-block-poly(ethene-alt-carbon monoxide)
†
†
,†
‡
David K. Newsham, Sachin Borkar, Ayusman Sen,* David M. Conner, and
Brian L. Goodall‡
Department of Chemistry, The PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802, and
Emerging Technology Department, Rohm and Haas Company, Spring House, PennsylVania 19477
ReceiVed May 25, 2007
Summary: We report the first well-defined palladium-based
system for the liVing homopolymerization of ethene, as well as
the liVing copolymerization of ethene with carbon monoxide.
We demonstrate this by the synthesis of polyethene-block-poly-
(ethene-alt-carbon monoxide). In addition, it has been possible
to monitor chain growth by sequential insertions of carbon
monoxide and ethene into palladium-carbon bonds. The
mechanistic studies haVe also allowed us to pinpoint the hitherto
not well-understood reason for the general failure to obtain
alkene/carbon monoxide copolymers with low carbon monoxide
content.
The palladium-catalyzed alternating copolymerization of
ethene with carbon monoxide (CO) has been widely studied.
1
Figure 1. Effect of carbon monoxide pressure on polymer yield
and composition. Curves merely guide the eye. Reaction condi-
The processing of the resultant polyketone (alt-PK) is prob-
lematic due to its insolubility in common solvents and its very
high melting point (∼265 °C), both of which result from high
crystallinity brought about by dipolar interactions between
carbonyl groups present at 50 mol % in the copolymer. The
problem of processibility can be circumvented by lowering the
CO content in the copolymer by producing nonalternating
copolymers. It has been suggested that as low as 5-10 mol %
CO incorporation into polyethene would be sufficient to impart
such desirable properties as adhesion, paintability, and hardness
tions: Pd
2
(dba)
3
, 10 µmol; (P∼SO
3 2 4
H), 20 µmol; C H , 300 psi;
CH Cl , 20 mL, 75 °C, 3 h.
2
2
incorporation from 0 to 50 mol % cannot be synthesized using
this system by simply raising the alkene:CO feed ratio. Herein,
we address this question and demonstrate that it is possible to
produce nonalt-PK with as low as 10 mol % CO, albeit in low
yields. Further, using a well-defined palladium complex with
this ligand, we demonstrate the first catalytic system for the
living polymerization of both PE and alt-PK, thereby allowing
the synthesis of the block copolymer: polyethene-block-poly-
1
without sacrificing the processibility associated with polyethene.
However, traditional catalysts for the production of alkene/CO
(ethene-alt-carbon monoxide).
copolymers do not yield materials with <50 mol % CO even
1
at high alkene:CO feed ratios. There are several reasons for
the strict alternation observed with conventional catalysts: CO
binds more strongly to the active palladium species and inserts
more readily into the Pd-alkyl bond than ethene does; however,
successive insertions of CO (i.e., insertion of CO into a Pd-
acyl bond) are thermodynamically disfavored.
Recently, a system that generates nonalternating ethene/CO
copolymers (nonalt-PK) with as little as 35 mol % CO has been
-
reported, and it employs the anionic phosphine ligand (P∼SO3 )
shown below.2 Additionally, the same system polymerizes
ethene to linear polyethene (PE) unlike conventional phosphine-
or imine-ligated palladium systems. A key unanswered mecha-
nistic question is why copolymers with the full range of CO
,3
The addition of small amounts of CO to the monomer feed
-
during ethene homopolymerization using the original (P∼SO3 )-
based system caused a dramatic reduction in activity despite
the system’s ability to produce PE, nonalt-PK, and alt-PK.
However, it was possible to produce nonalt-PK incorporating
as little as 10 mol % CO by adding only 5 psi (35 kPa) CO
into an autoclave with 300 psi (2070 kPa) ethene (see Figure
*
Corresponding author. E-mail: asen@psu.edu.
The Pennsylvania State University.
Rohm and Haas Company.
†
‡
(1) (a) Catalytic Synthesis of Alkene-Carbon Monoxide Copolymers and
1
). This lowered the melting point of the copolymer to 118 °C,
Cooligomers; Sen, A., Ed.; Catalysis by Metal Complexes 27; Kluwer
Academic: Dordrecht, 2003. (b) Mul, W. P.; Oosterbeck, H.; Betel, G. A.;
Kramer, G.-J.; Drent, E. Angew. Chem., Int. Ed. 2000, 39, 1848-1851.
as determined by DSC. The activity of the system could be
improved by increasing the CO pressure but not without
restoring the near-alternating character of the resulting copoly-
mer. The relationship between CO pressure, yield, and composi-
tion is consistent with the mechanism originally proposed for
(2) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh, R. I.
Chem. Commun. 2002, 964-965.
3) Hearley, A. K.; Nowack, R. J.; Rieger, B. Organometallics 2005,
4, 2755-2763.
(
2
1
0.1021/om700523m CCC: $37.00 © 2007 American Chemical Society
Publication on Web 06/22/2007

Communications
Organometallics, Vol. 26, No. 15, 2007 3637
Scheme 1. General Catalytic Cycle for Ethene/Carbon
Monoxide Copolymerization1
,4
the classical alternating copolymerization system (see Scheme
1,4
1
). As shown, a chelate resting state (A) exists in the catalytic
cycle and, unlike CO, the more weakly coordinating ethene is
less able to disrupt the chelate (A f D), a requirement for
sequential ethene insertions; note that this issue does not arise
in ethene homopolymerization. However, CO binding is strong
enough to disrupt the chelate, leading to further insertions (A
f B f C f A). Thus, with conventional alternating copo-
lymerization catalysts, CO is required for the alkene insertion
step to proceed.
We sought to look for a stable chelate similar to A in Scheme
for the present system in order to confirm that it proceeds via
Figure 2. ORTEP structures of cis-(P∼SO
3
)Pd(Me)(Py) (1) and
)Pd(COMe)(Py) (2). Hydrogen atoms are omitted for
1
cis-(P∼SO
3
the same general mechanism as bis-phosphine-ligated systems.
The active species in the original catalytic system reported by
Drent and Pugh (prepared in situ from Pd(OAc)2 and P∼SO3H)
clarity.
2
. The Pd-carbon bond in both 1 and 2 is found cis to the
2
was not well-defined. Therefore, we used a model compound
coordinated phosphine moiety, presumably because of the
stronger trans effect of phosphorus compared to oxygen. The
insertion of CO into the Pd-C bond of 1 was found to be
with a methyl group and pyridine occupying two coordination
sites, cis-(P∼SO3)Pd(Me)(Py), 1. This species was synthesized
1
31
and characterized by H and P NMR and IR spectroscopy and
single-crystal X-ray diffraction as described in the Supporting
Information. The ORTEP structure is shown in Figure 2. All
reactions were carried out in dichloromethane to maintain
solubility of the products. When 1 in CD2Cl2 was exposed to
reversible, as shown by the following experiment. First,
13
(
P∼SO3)Pd( COMe)(Py) was formed by the insertion of
13
13
CO, as confirmed by a strong carbonyl C NMR signal at
27 ppm. The solvent and excess CO were removed under
13
2
vacuum, and the product was redissolved in CD2Cl2 before being
1
00 psi (690 kPa) ethene at 25 °C, PE was produced. The ethene
12
13
exposed to 50 psi (345 kPa) of regular CO. After 18 h,
C
polymerization has living characteristics under these conditions,
13
NMR analysis showed a lack of C enrichment in the Pd
1
and the chain growth was monitored by H NMR spectroscopy.
13
complex, indicating slow decarbonylation of CO followed by
The growth of the PE peak at 1.3 ppm was not accompanied
by any signals in the vinyl region that would indicate chain
termination by â-hydrogen abstraction. Also, CH2 groups R and
â to the Pd center were seen respectively as multiplets at 0.8
and 0.6 ppm, along with the methyl end group at 0.9 ppm.
When 1 was exposed to a mixture of 50 psi (345 kPa) CO
and 50 psi (345 kPa) ethene in CD2Cl2 at 25 °C, alt-PK was
formed. As with ethene homopolymerization, it was possible
to observe stepwise insertion of CO (or 13CO for C NMR
experiments) and ethene into the growing, methyl-terminated
co-oligomer. To accomplish this, 1 was first exposed to 15 psi
12
5
rapid insertion of CO as predicted by Ziegler.
When a solution of 2 in CD2Cl2 was exposed to 100 psi (690
kPa) ethene at 0 °C, slow insertion of ethene into the Pd-acyl
bond occurred to give a mixture of (P∼SO3)Pd(CH2CH2COMe),
3
, and unreacted 2 but not PE. NMR analysis showed a new
1
terminal methyl H NMR signal at 1.9 ppm, and the CH2 groups
R and â to the metal appeared as a doublet of triplets at 1.3
ppm and as a broad multiplet at 2.1 ppm, respectively. Also, a
new P signal appeared at 25 ppm and a broad C carbonyl
signal appeared at 216 ppm. IR analysis showed a new carbonyl
13
31
13
-1
absorption at 1643 cm , which is consistent with the formation
(
105 kPa) CO, causing the immediate formation of cis-(P∼SO3)-
6,7
of a five-membered chelate like structure A, Scheme 1.
1
Pd(COMe)(Py), 2. The H NMR signal for Pd-CH3 at 0.2 ppm
was replaced by the Pd-COCH3 signal at 1.8 ppm. Also, the
1
Low-temperature H NMR analysis suggests that the five-
membered chelate 3 is likely in equilibrium with the open-chain
pyridine-coordinated analogue, 3‚Py, and that the former is
favored at low temperatures. At -80 °C, a doublet due to the
3
1
1
3
P signal at 23 ppm was replaced by a signal at 11 ppm and a
C carbonyl signal appeared at 227 ppm. IR analysis showed
-
1
a carbonyl absorption at 1695 cm . Single-crystal X-ray
diffraction data were also collected as described in the Sup-
porting Information. The ORTEP structure is shown in Figure
(
5) Haras, A.; Michalak, A.; Rieger, B.; Ziegler, T. J. Am. Chem. Soc.
005, 127, 8765-8774.
6) Agostinho, M.; Braunstein, P. Chem. Commun. 2007, 58-60.
(7) Reddy, K. R.; Chen, C.-L.; Liu, Y.-H.; Peng, S.-M.; Chen, J.-T.;
Liu, S.-T. Organometallics 1999, 18, 2574-2576.
2
(
(4) Rix, F. C.; Brookhart, M.; White, P. S. J. Am. Chem. Soc. 1996,
1
18, 4746-4764.

3638 Organometallics, Vol. 26, No. 15, 2007
Communications
ortho-hydrogens of the free pyridine is observed at 8.7 ppm.
Additionally, the doublet of triplets at 1.3 ppm and the multiplet
at 2.1 ppm from the CH2 groups R and â to the metal,
respectively, split into unique proton signals at 0.80, 1.01, and
species was shifted to 22 ppm, indicating that it underwent
ethene insertion to form a Pd-alkyl species (similar to 3).
Finally, the living nature of the ethene homopolymerization
and ethene/CO copolymerization allows the formation of PE
and alt-PK blocks in a single chain. Block cooligomers were
synthesized to demonstrate this, while maintaining the solubility
of the products in CD2Cl2. First, 1 in CD2Cl2 was exposed to
100 psi (690 kPa) ethene at 25 °C for 2 h to grow a PE block
with an average of four ethene units. The solvent and excess
ethene was removed under vacuum, and the redissolved system
was then exposed to a mixture of 50 psi (345 kPa) ethene and
50 psi (345 kPa) CO for 9 h to grow a second alt-PK block.
The PE and alt-PK blocks were identified by their characteristic
1
.55 ppm with the fourth obscured by the methyl signal,
suggesting the presence of a rigid chelate ring structure.
In contrast to 1 and 2, the chelated species 3 was unreactive
toward ethene alone at 100 psi (690 kPa) and 25 °C, consistent
with the notion that the alkene is too weakly binding to easily
disrupt the chelate (or displace the coordinated pyridine) and
give consecutive ethene insertions under these conditions. Under
the more forcing conditions used to generate the data for Figure
1
, ethene insertion does occur slowly, resulting in nonalt-PK.
1
However, unlike alt-PK formation, the yield of nonalt-PK is
low because of the sluggishness of this step. We hypothesize
that the step A f D of Scheme 1 does proceed in the present
neutral system, albeit slowly, because the carbonyl group in
species A binds less strongly to the metal unlike in the traditional
cationic palladium systems used for alternating ethene/CO
copolymerizations.
H NMR chemical shifts at 1.3 and 2.7 ppm, respectively. The
connectivity between the two blocks was observed as two CH2
signals at 1.5 and 2.4 ppm. The assignment of the connecting
1
CH2 groups was established by comparing with the H NMR
shifts of 3-hexanone. The protons labeled a and b in Et-C(O)-
b
a
C H2-C H2-Me are observed at 1.6 and 2.4 ppm, respectively.
Compound 3, when redissolved in CD2Cl2 and exposed to
Acknowledgment. The research was funded by the National
Science Foundation and Rohm & Haas Company. X-ray
crystallographic analysis was performed by Dr. Hemant
Yennawar at The Pennsylvania State University.
5
0 psi (345 kPa) CO at 25 °C, formed (P∼SO3)Pd(COCH2-
CH2COMe), 4, which corresponds to structure C in Scheme 1
1
(
or its open-chain pyridine-coordinated analogue). The H NMR
signals for the CH2 groups R and â to the metal in 4 were shifted
to 1.9 and 2.5 ppm, respectively. The 31P signal at 25 ppm was
Supporting Information Available: Experimental procedures,
characterization data for all new compounds, X-ray crystal structure
coordinates, and files in CIF format. This material is available free
of charge via the Internet at http://pubs.acs.org.
replaced by a peak at 12 ppm, and the 13C carbonyl signal at
2
16 ppm was replaced by two carbonyl signals at 207 and 228
ppm. Compound 4, in turn, was exposed to 100 psi (690 kPa)
ethene at 0 °C for 18 h. The 31P NMR signal for the Pd-acyl
OM700523M