Palladium-catalyzed copolymerization of ethene with acrolein dimethyl acetal: Catalyst action and deactivation

Article abstract of DOI:10.1021/ja045948y

Acrolein dimethyl acetal (ADMA) can be copolymerized with ethene using a cationic α-diimine palladium chelate catalyst to yield branched copolymers. Catalyst deactivation occurs via methanol elimination to give an inert η³-1-methoxyallyl palladium species. This process can be retarded by the addition of an aliquot of methanol to the reaction mixture, increasing overall catalyst productivity. Copyright

Full text of DOI:10.1021/ja045948y

Published on Web 09/11/2004
Palladium-Catalyzed Copolymerization of Ethene with Acrolein Dimethyl
Acetal: Catalyst Action and Deactivation
Weidong Li, Xiaochun Zhang, Auke Meetsma, and Bart Hessen*
Dutch Polymer Institute/Center for Catalytic Olefin Polymerization, Stratingh Institute for Chemistry and Chemical
Engineering, UniVersity of Groningen, Nijenborgh 4, 9747 AG Groningen, The Netherlands
Received July 7, 2004; E-mail: hessen@chem.rug.nl
Table 1. Ethene/ADMA Copolymerization with 2 as Catalyst
The incorporation of polar heteroatom-containing functionalities
into (intrinsically apolar) polyolefins by the copolymerization of
simple olefins with functional olefinic comonomers is of substantial
interest as a potential route to polyolefin materials with improved
dyeability, wettability, or permeability properties.1 A main obstacle
is the poor compatibility of Lewis acidic polymerization catalysts
with Lewis basic heteroatom-containing functionalities, leading to
a significant decrease in catalyst performance or even rapid catalyst
deactivation. The advent of (relatively soft Lewis acidic) late
transition-metal-based catalysts, in particular cationic (R-diimine)-
Pd-based systems,2 has provided encouraging results such as the
successful (if rather slow) copolymerization of ethene and methyl
acrylate to give branched copolymers.3 Nevertheless, many issues
still need to be resolved to estimate the full potential for catalytic
routes to functional polyolefins.
A recent paper described the copolymerization of ethene with
olefins containing ω-ether functionalities using cationic (R-diimine)-
Pd-catalysts.4 This was only successful when the ether functionality
was separated from the olefinic group by a spacer containing a
quaternary carbon atom. This led to the conclusion that “chain
walking” of the metal center to the ether group, followed by
â-alkoxide transfer to the metal, is the likely catalyst deactivation
pathway. Here we describe the successful copolymerization of
ethene with acrolein dimethyl acetal (ADMA) using a cationic (R-
diimine)Pd-catalyst and show that, in this case, catalyst deactivation
occurs by the formation of inert cationic (η3-allyl)palladium species
via alcohol elimination, rather than by â-alkoxide transfer to the
metal.
Reaction of (DAD)Pd(Me)Cl (1; DAD ) ArNdCMe-CMed
NAr; Ar ) 2,6-diisopropylphenyl)2a with ADMA and NaBArf4
(BArf4 ) B[3,5-(CF3)2C6H3]4) in CH2Cl2 yielded the five-membered
chelate complex [(DAD)Pd(CH2CHMeCH(OMe)2][BArf4] (2). The
complex was identified by 1H and 13C NMR spectroscopy,5
elemental analysis, and a single-crystal X-ray structure determina-
tion.6 The chelate fragment on Pd is highly disordered, as the crystal
packing in these (DAD)Pd compounds is strongly dominated by
the aryl groups of the ligand. Nevertheless, it can be seen that 1,2-
insertion of ADMA into the Pd-Me bond has taken place, and
that one of the acetal OMe groups is coordinated to the metal. The
compound is air-stable and thermally stable in solution at least up
to 50 °C. Complex 2 catalyzes the copolymerization of ethene and
ADMA at ambient temperature (results are listed in Table 1). It is
seen that addition of ADMA comonomer significantly slows down
the polymerization rate, but that copolymers with up to 2.0 mol %
(6.9 wt %) comonomer can be obtained. A catalyst deactivation
process is present, leading to full deactivation within 20 h (see below
for details).
ADMA
(mmol)
yield
(g)
ADMA
(mol %)c
3
conditionsa
productb
Mw
(
×
10-
)
PDI
A
A
A
A
B
B
B
B
0
4.2
8.4
17.4
4.2
8.4
19.1
26.1
3.50
1.21
0.64
0.44
1.60
1.50
1.20
0.52
13.0
4.5
2.4
1.6
1.2
1.1
0.9
0.4
186
105
52
31
50
32
16
13
1.5
1.7
2.0
2.3
1.7
1.7
1.7
1.6
0
0.29
0.32
0.66
0.72
1.30
1.67
2.02
a Reaction conditions A: 13.4 µmol 2, 15 mL CH2Cl2, 5 bar ethene,
ambient temperature, 20 h run time. B: 67 µmol 2, 3 mL CH2Cl2, otherwise
1
as A. b Productivity kg(PE) mol-1 h-1
.
c Determined by H NMR.
or a secondary carbon (in 1:1 to 1:0.8 ratio for lowest to highest
comonomer content) revealed by the presence of either a triplet or
a doublet resonance for the acetal CH proton, indicating, respec-
tively, comonomer incorporation on a chain/branch end or within
a chain/branch. Increasing the comonomer concentration leads not
only to increased comonomer incorporation but also to a lowering
of the polymerization rate and of the polymer molecular weight.
Upon workup of the ethene/ADMA copolymerization reaction
mixtures, orange crystals of a palladium species were isolated. NMR
spectroscopy,8 elemental analysis, and a single-crystal structure
determination9 revealed the formation of the 1-methoxyallyl
complex [(DAD)Pd(η3-1-MeOCHCHCH2)][BArf4] (3, Scheme 1).
Complex 3 itself is not active in ethene homopolymerization, and
its recovery in >80% yield after recrystallization indicates that its
formation is the main catalyst deactivation process.
Unlike the copolymerizations with ADMA described above, the
copolymerization of ethene with allyl ethyl ether (AEE), using 2
as an introduced catalyst species, could not be achieved. Instead,
rapid catalyst deactivation is observed, and the cationic allyl
complex [(DAD)Pd(η3-CH2CHCH2)][BArf4] (4)10 was recovered in
high yield. In the reaction mixture, the formation of ethanol was
observed by GC/MS analysis. As the reaction of [(DAD)PdMe-
(OEt2)][BArf4]2a with 1 equiv of AEE resulted in clean formation
of the stable five-membered chelate [(DAD)Pd(CH2CHMeCH2-
OEt)][BArf4] (5),11 the catalyst decomposition via ethanol elimina-
tion and formation of the cationic palladium allyl species apparently
takes place only in the presence of excess comonomer.
A possible route to the formation of these allylic species is shown
in Scheme 2. An equilibrium between the Pd-alkyl species and
the hydride-olefin complex (implied in chain-transfer processes
for these catalysts12), followed by olefin exchange by an allylic
ether comonomer, can lead either to 1,2-insertion of the comonomer
(reforming a chelate) or to elimination of a molecule of alcohol,
producing an η3-allylic species. This type of reaction was recently
implied in the catalytic substitution of allyl alcohol with amines to
give allylic amines, catalyzed by cationic palladium bis(phosphin-
The branched PE-co-ADMA copolymers contain two main types
of acetal groups, as seen by 1H and 13C NMR spectroscopy.7 These
correspond to CH(OMe)2 groups attached to either a primary carbon
9
12246
J. AM. CHEM. SOC. 2004, 126, 12246-12247
10.1021/ja045948y CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1
of methanol to the reaction mixture. The identification of catalyst
deactivation pathways thus provides information that can lead to
approaches that improve the catalytic production of functionalized
polyolefins.
Acknowledgment. We thank J. Vorenkamp for polymer GPC
analysis.
Supporting Information Available: Text giving full experimental
and characterization data and crystallographic data for 2-5, as well as
positional and thermal parameters and bond distances and angles (PDF,
CIF). This material is available free of charge via the Internet at http://
pubs.acs.org.
Scheme 2
References
(1) See: Boffa, L. S.; Novak, B. M. Chem. ReV. 2000, 100, 1479 and
references cited therein.
(2) (a) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995,
117, 6414. (b) Brookhart, M. S.; Johnson, L. K.; Killian, C. M.; Arthur,
S. D.; Feldman, J.; McCord, E. F.; McLain, S. J.; Kreutzer, K. A.; Bennett,
A. M. A.; Coughlin, E. B.; Ittel, S. D.; Parthasarathy, A.; Tempel, D. J.
(DuPont). WO 96/23010, 1996. (c) Ittel, S. D.; Johnson, L. K.; Brookhart,
M. Chem. ReV. 2000, 100, 1169 and references cited.
(3) (a) Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc. 1996,
118, 267. (b) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J.
Am. Chem. Soc. 1998, 120, 888.
(4) Chen, G.; Ma, X. S.; Guan, Z. J. Am. Chem. Soc. 125, 2003, 6697.
(5) Selected NMR data for 2. 1H (500 MHz, CD2Cl2, 25 °C): δ 4.12 (d, J )
5.5 Hz, CH(OMe)2), 3.12 (s, 3H, OMe), 2.87 (m, 2H, PdCH2), 2.84 (s,
3H, OMe), 1.3 (m, PdCH2CHMe), 0.74 (d, 3H, J ) 7.0 Hz, PdCH2CHMe).
13C (126 MHz, CD2Cl2, 25 °C): δ 119.48 (d, JCH ) 172 Hz, CH(OMe)2),
60.51 (q, JCH ) 147 Hz, OMe), 56.55 (q, JCH ) 145 Hz, OMe), 38.80 (t,
JCH ) 140 Hz, PdCH2), 38.75 (d, JCH ) 112 Hz, PdCH2CHMe), 14.65
(q, JCH ) 131 Hz, PdCH2CHMe).
(6) Crystal data for 2: [C34H53N2O2Pd][C32H12BF24], Mr ) 1491.45, mono-
clinic, P21/n, a ) 12.8098(2) Å, b ) 28.259(2) Å, c ) 19.154(1) Å, â )
103.450(1)°, V ) 6743.3(7) Å3, Z ) 4, Dc ) 1.469 g cm-3, T ) 100(1)
K, µ(Mo KR) ) 0.71073 Å, wR(F2) ) 0.2107 for 11 901 unique
reflections, 950 parameters and 44 restraints, R(F) ) 0.0726 for 8207
reflections with Fo g 4.0σ(Fo). A disorder model with two alternative
conformations (SOF of the major fraction ) 0.507(8)) and with bond
restraints was used to model the CH2CH(Me)CH(OMe)2 fragment.
(7) 1H NMR (500 MHz, CDCl3, 25 °C): δ 4.33 (t, J ) 5.7 Hz, CH2CH(OMe)2),
4.00 (d, J ) 6.7 Hz, CHCH(OMe)2), 3.33 and 3.29 (s, OMe). 13C NMR
(126 Hz, CDCl3, 25 °C): δ 108.9 and 104.6 (CH(OMe)2), 54.0 and 52.5
(OMe).
idene)cyclobutene hydride complexes.13 This process appears to
be facile for AEE, thus preventing the formation of copolymer,
whereas for the acetal comonomer ADMA, the allyl formation is
sufficiently retarded to allow the formation of a significant amount
of copolymer before catalyst deactivation.
If this deactivation mechanism is correct, it is conceivable that
the catalyst lifetime may be improved by the addition of alcohol
to the reaction mixture. A series of experiments using conditions
A (defined in Table 1) with 4.2 mmol of ADMA and 37 equiv/Pd
of methanol (0.5 mmol) showed that this does lead to improved
overall productivity. Run times of 6, 12, 20, and 40 h gave polymer
yields of 0.90, 1.21, 1.72, and 2.31 g, respectively. In the absence
of methanol, a maximum polymer yield of 1.2 g was reached after
12 h, after which no additional polymer was formed. Although
methanol addition improves overall productivity, it is clear that
gradual catalyst deactivation still takes place. The addition of
substantially larger amounts of methanol (500 equiv/Pd) again led
to a decrease in catalyst productivity, possibly due to the occurrence
of an alternative catalyst deactivation process under these conditions.
The improvement is apparently obtained by retarding the allyl
formation (rather than by reactivation of the allyl species once it is
formed), as addition of methanol to CH2Cl2 solutions of complex
3 did not induce ethene polymerization activity.
(8) Selected NMR data for 3. 1H (500 MHz, CD2Cl2, 25 °C): δ 5.66 (d, J )
9.6 Hz, CHOMe), 5.27 (m, CH2CHCHOMe), 2.95 (m, CHH′CHCHOMe),
2.62 (s, 3H, OMe), 2.44 (dd, J ) 12.8 and 2.2 Hz, CHH′CHCHOMe).
13C (126 MHz, CD2Cl2, 25 °C): δ 120.33 (d, JCH ) 168 Hz, CHOMe),
98.67 (d, JCH ) 149 Hz, CH2CHCHOMe), 60.18 (q, JCH ) 141 Hz, OMe),
54.50 (t, JCH ) 156 Hz, CH2CHCHOMe).
(9) Crystal data for 3: [C32H47N2OPd][C32H12BF24], Mr ) 1445.37, mono-
clinic, C2/c, a ) 21.9494(9) Å, b ) 12.8691(5) Å, c ) 25.207(1) Å, â )
115.522(1)°, V ) 6425.4(4) Å3, Z ) 4, Dc ) 1.494 g cm-3, T ) 100(1)
K, µ(Mo KR) ) 0.71073 Å, wR(F2) ) 0.1874 for 7900 unique reflections,
597 parameters and 55 restraints, R(F) ) 0.0675 for 6294 reflections with
Fo g 4.0σ(Fo). The cation is disordered over a crystallographic inversion
center.
(10) Selected NMR data for 4. 1H (500 MHz, CD2Cl2, 25 °C): δ 5.64 (m,
allyl CH), 3.35 (d, 2H, J ) 7.0 Hz, CHH), 3.04 (d, 2H, J ) 12.8 Hz,
CHH). 13C (126 MHz, CD2Cl2, 25 °C): δ 121.03 (d, J ) 170 Hz, allyl
CH), 65.78 (t, J ) 185 Hz, CH2).
1
(11) Selected NMR data for 5. H (500 MHz, CD2Cl2, 25 °C): δ 3.53 (dd, J
) 7.7 and 6.0 Hz, CHH′OEt), 3.14 (t, J ) 7.6 Hz, CHH′OEt), 2.85 (m,
2H, OCH2Me), 1.67 (m, 2H, PdCH2), 1.29 (m, PdCH2CHMe), 0.82 (t, J
) 7.0 Hz, OCH2Me), 0.79 (d, J ) 7.0 Hz, PdCH2CHMe). 13C (126 MHz,
CD2Cl2, 25 °C): δ 83.69 (t, J ) 145 Hz, CH2OEt), 72.84 (t, J ) 146 Hz,
OCH2Me), 47.77 (t, J ) 150 Hz, PdCH2), 37.45 (d, J ) 161 Hz,
PdCH2CHMe), 14.62 (q, J ) 127 Hz, PdCH2CHMe), 13.69 (q, J ) 127
Hz, OCH2Me).
In conclusion, we have successfully copolymerized ethene with
acrolein dimethyl acetal using a palladium catalyst to give a
branched polyethene copolymer with pendant acetal groups. Allyl
formation via methanol elimination was found to be the dominant
catalyst deactivation pathway. The lifetime of the catalyst (and with
it overall copolymer production) could be improved by the addition
(12) See: Tempel, D. J.; Johnson, L. K.; Huff, R. L.; White, P. S.; Brookhart,
M. J. Am. Chem. Soc. 2000, 122, 6686 and references cited therein.
(13) Ozawa, F.; Ishiyama, T.; Yamamoto, S.; Kawagishi, S.; Murakami, H.;
Yoshifuji, M. Organometallics 2004, 23, 1698.
JA045948Y
9
J. AM. CHEM. SOC. VOL. 126, NO. 39, 2004 12247