New single-site palladium catalysts for the nonalternating copolymerization of ethylene and carbon monoxide

Article abstract of DOI:10.1021/om0490785

The synthesis and first examples of structurally characterized, single-site palladium complexes containing a phosphine sulfonate chelate (PSO) for the nonalternating co-polymerization of ethylene and carbon monoxide are reported. Extra incorporation of ethylene up to 30% has been achieved relative to the alternating polyketone structure with modest activities. As exemplarily shown, high molecular weight random copolymers have been produced with M_w ≈ 370 000, polydispersity (M_w/M_n) = 2, and melting points of 220-230°C.

Full text of DOI:10.1021/om0490785

Organometallics 2005, 24, 2755-2763
2755
New Single-Site Palladium Catalysts for the
Nonalternating Copolymerization of Ethylene and
Carbon Monoxide
Andrew K. Hearley, Ru¨diger J. Nowack, and Bernhard Rieger*
Department of Materials Science and Catalyst Design, University of Ulm,
Albert Einstein Allee 11, D-89069 Ulm, Germany
Received November 25, 2004
The synthesis and first examples of structurally characterized, single-site palladium
complexes containing a phosphine sulfonate chelate (PSO) for the nonalternating co-
polymerization of ethylene and carbon monoxide are reported. Extra incorporation of
ethylene up to 30% has been achieved relative to the alternating polyketone structure with
modest activities. As exemplarily shown, high molecular weight random copolymers have
been produced with Mw ≈ 370 000, polydispersity (Mw/Mn) ) 2, and melting points of 220-
230 °C.
Introduction
certain cationic palladium complexes (R-diimines) may
polymerize ethylene to high molecular weights.6 Elegant
mechanistic studies based on cationic palladium(II)
complexes have clearly demonstrated that the pure
alternation arises from the formation of a highly stable
five-membered cationic palladium metallacycle by an
electrostatic interaction of the oxygen of the carbonyl
moiety and the cationic palladium center “back-biting”
effect, which may be opened only by a CO molecule.7
Very recently, in contrast to the previous research,
Drent and co-workers reported three alkoxy-aryl-phos-
phine ligands bearing sulfonic acid groups that when
combined in situ with palladium acetate in a near
stoichiometric (1.1:1) ratio, produced double, triple, and
quadruple extra insertions (“misinsertions”) of ethylene
in the growing polyketone chain.8a,b
The development of new, well-defined, functional
group tolerant, late transition metal catalysts for the
synthesis of polymers comprised of a nonpolar backbone
and a controlled amount of a polar monomer is of
considerable interest to both academia and industry.1
However, metal-catalyzed copolymerization reactions of
monomers containing functional groups and olefins is
by no means a trivial task. It is clear from the literature
that many attempts have been made within this area,
but so far, very few new materials have been realized.2
The palladium-catalyzed copolymerization of ethylene
and carbon monoxide on the other hand is one of the
exceptions to this general trend. It is probably one of
the most widely studied and best understood polymer-
ization reactions.3 Both ethylene and carbon monoxide
feedstocks are cheap and readily available, and the
perfectly alternating polyketone products thus formed
possess very interesting properties.4 In over 20 years
of research, in the area of late transition metal catalyzed
polymerizations of carbon monoxide and ethylene, no
“errors” or extra insertions, i.e., no double CO or double
C2H4 insertions, have ever been reported. It is clear for
thermodynamic reasons that double CO insertions do
not occur, but why double or even multiple insertions
of ethylene do not occur is somewhat astonishing,
particularly because the same catalysts may convert
ethylene to butenes very readily.5 It is also known that
Our group has been active in the areas of new catalyst
design and polymer studies of CO-olefin (co-) and (ter-)
polymerization reactions for a number of years.9 The
idea of a random CO-C2H4 copolymer has been an
ultimate goal in polyketone research, with the promise
of new materials possessing desirable material proper-
ties and the opportunity for further functionalization.
We now report the synthesis and structural character-
(5) Drent, E. Pure Appl. Chem. 1990, 62, 661.
(6) Johnson, L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc.
1995, 117, 6414.
(7) (a) Rix, F. C.; Brookhart, M. J. Am. Chem. Soc. 1995, 117, 1137.
(b) Rix, F. C.; Brookhart, M.; White, P. S. J. Am. Chem. Soc. 1996,
118, 4746. (c) Nozaki, K.; Sato, N.; Tonomura, Y.; Yasutomi, M.;
Takaya, H.; Hiyama, T.; Matsubara, T.; Koga, N. J. Am. Chem. Soc.
1997, 119, 12779. (d) Margl, P.; Ziegler, T. J. Am. Chem. Soc. 1996,
118, 7337.
(1) (a) Britovsek, G. J. P.; Gibson, V. C.; Wass, D. F. Angew. Chem.,
Int. Ed. 1999, 38, 429. (b) Boffa, L. S.; Novak, B. M. Chem. Rev. 2000,
100, 1479. (c) Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. Rev.
2000, 100, 1169.
(8) (a) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh,
R. I. J. Chem. Soc., Chem. Commun. 2002, 964. (b) Drent, E.; Pello,
D. H. L. Eur. Pat. Appl. EP0632084 (to Shell), 1995.
(2) (a) Brubaker, M. M.; Coffman, D. D.; Hoehn, H. H. J. Am. Chem.
Soc. 1952, 74, 1509. (b) Deubel, D. V.; Ziegler, T. Organometallics 2002,
21, 1603. (c) Michalak, A.; Ziegler, T. Organometallics 2001, 20, 1521.
(d) Michalak, A.; Ziegler, T. J. Am. Chem. Soc. 2001, 123, 12266. (e)
Shultz, C. S.; DeSimone, J. M.; Brookhart, M. Organometallics 2002,
20, 16. (f) Tanaka, K.; Furo, M.; Ihara, E.; Yasuda, H. J. Polymer Sci.,
Part A. 2001, 39, 1382. (g) Desurmont, G.; Tokimitsu, T.; Yasuda, H.
Macromolecules 2000, 33, 7679.
(9) (a) Niewhof, R. P.; Marcelis, A. T. M.; Sudho¨lter, E. J. R.;
Wursche, R.; Rieger, B. Macromol. Chem. Phys. 2000, 201, 248. (b)
Wursche, R.; Rieger, B. Macromol. Chem. Phys. 2000, 201, 2869. (c)
Wursche, R.; Debardemaeker, T.; Klinga, M.; Rieger, B. Eur. J. Inorg.
Chem. 2000, 2063. (d) Huhn, W.; Hollmann, F.; Hild, S.; Kriewall, Th.;
Rieger, B. Macromol. Mater. Eng. 2000, 283, 115. (e) Mansour, A.;
Hollmann, F.; Rieger, B. Polymer 2001, 42, 93. (f) Privalko, V. P.; Pissis,
P.; Polizos, G.; Korskanov, V. V.; Privalko, E. G.; Dolgoshey, V. I.;
Kramarenko, V. Yu.; Huhn, W.; Hollmann, F.; Rieger, B. J. Macromol.
Sci.-Phys. 2002, B41 (1), 99.
(3) Drent, E.; Budzelaar, P. H. M. Chem. Rev. 1996, 96, 663, and
references therein.
(4) Ash, C. E. J. Mater. Educ. 1994, 16, 1.
10.1021/om0490785 CCC: $30.25 © 2005 American Chemical Society
Publication on Web 04/22/2005

2756 Organometallics, Vol. 24, No. 11, 2005
Scheme 1. Synthesis of Ligands 1-4
Hearley et al.
a random copolymerization of CO and C2H4 (extra
ethylene insertions) using a palladium catalyst.8a,b As
expected, the results obtained for 2 were in agreement
with those previously reported, in both incorporation of
extra ethylene insertions (15%) and the activity (50 g
mmol-1 h-1) at 30 bar C2H4 and 10 bar CO.
Having established in our laboratories, using our
autoclave systems and equipment, that extra ethylene
insertions may be produced using ligand 2, we wanted
to address a number of further key points that we felt
were still unanswered from the previous study.8a
Therefore, 2-(diphenylphosphino)benzenesulfonic acid,8b
ligand 1, bearing no substituents was prepared and
tested under standard conditions (entry 1). In contrast
to ligands 2-4 applying the conditions stated in Table
1, both the percentage of extra ethylene insertions and
the amount of polymer produced were very low. We
assume, due to the open nature of the system, i.e., no
steric bulk in close proximity to the metal center, as
would be expected from a catalyst derived from ligand
2, that the catalyst containing ligand 1 allows â-hydride
elimination to become a more significant termination
step. This observation, however, does conflict with the
patent literature, where it is claimed using the identical
ligand 1 and Pd(OAc)2, with gas pressures CO/C2H4 (45:
40), that up to 15% extra ethylene insertions may be
produced.8b
We wanted to further establish if an alkoxy group
(electron-donating group) is a requirement in the ortho-
position of the unsulfonated phenyl ring(s) of the ligand
or would any relatively bulky group generate the same
results in both activity and copolymer composition.
Therefore, the ortho-methyl-substituted derivative 3
was prepared. The results clearly show (cf. entry 3 vs
entry 6) that although the ligand 3-derived catalyst
produces extra insertions, with almost the same activi-
ties as compared to ligand 2, the percentage of extra
ethylene insertions relative to the alternating structure
is approximately 50% less than that found for ligand 2.
Simple molecular modeling calculations clearly show
that the oxygen of the ortho-methoxy group orientates
itself directly above the palladium center (∼3.4 Å).
Other ligand derivatives, bearing substituents such as
CF3, F, and SMe2, which may potentially interact with
the metal center, are currently under study in our
laboratories.
By increasing the pressure to 65 (entry 7) and 110
bar (entry 8), with a gas blend ratio of 10:55 and 10:
100 CO/C2H4, the extra incorporation of ethylene in-
creased to 15.1 and 18.3%, respectively, but the activi-
ties decreased to 26 and 12 g mmol-1 h-1. When the
ratio of ligand 3 to Pd(OAc)2 was increased to 2:1, pure
white polymers were obtained, with higher activities.
However, the number of extra ethylene insertions
decreased significantly, for example from 8.4 (entry 12)
to 2.4% (entry 9). A detrimental effect from the addition
of a second equivalent of ligand to the copolymerization
reaction is likely, due to a shift in reaction equilibrium
toward a dimeric bis-phosphine species (vide infra),
which is capable of producing only purely alternating
copolymers (cf. Figure 1).
Table 1. Results of the in Situ CO-C2H4
Copolymerization Reactions25e
extra C2
C2H4 CO
activity (g insertion
entry
ligand
T/°C (bar) (bar) mmol-1 h-1
)
(%)d
1
1
2
2
3
3
3
3
3
110
110
110
110
110
110
110
110
30
20
30
20
20
30
55
100
30
30
30
50
10
10
10
20
10
10
10
10
10
10
10
10
<1a
25a
48c
38a
56a
54c
26b
12b
69a
49a
47c
29c
1.4
8.1
15.0
2.9
6.1
8.1
2
3
4
5
6
7
15.1
18.3
2.4
8
9
3 (2 equiv) 110
10
11
12
3
4
4
110
110
110
8.4
8.2
14.7
aIn a 100 mL (50 mL solvent) autoclave. b25 mL (10 mL solvent)
c
autoclave. 250 mL (100 mL solvent) autoclave. d Total mol % of
C2H4 relative to nonalternating copolymer, calculated from 13C{1H}
NMR spectra (measured in 1,1,3,3-hexafluoropropan-2-ol/C6D6).
e 1 ) o-H, 2 ) o-OMe, 3 ) o-Me/toluenesulfonic acid, 4 ) o-Me.
Based on Pd over experiment run time of 1 h.
ization of the first examples of neutral Pd(II) complexes
based on {SO3}-containing ligands. The new complexes
show higher activities and are able to insert larger
amounts of extra ethylene units relative to the perfectly
alternating structure as compared to the in situ co-
polymerization reactions previously reported.8a,b
Results and Discussion
In Situ Polymerizations. Ligands 1, 3, and 4 were
synthesized by addition of the appropriate chlorodi-
phenylphosphine to a tetrahydrofuran solution of di-
lithiated benzenesulfonic acid (1, 4) and toluenesulfonic
acid (3) at 0 °C, followed by acidic workup (Scheme 1).
Ligand 2 was prepared by the known literature
procedure for comparative purposes.8a The 31P NMR in
chloroform-d1 of each ligand consists of a single peak
at approximately -9 ppm relative to 80% H3PO4. No
1J(P-H) coupling is observed in the proton NMR spectra,
confirming that ligands 1, 3, and 4 are not zwitterionic,
in contrast to that found for ligand 2.
Results of the in situ copolymerization reactions of
carbon monoxide and ethylene using ligands 1-4 are
shown in Table 1.
First, we repeated the copolymerization reaction with
the ortho-methoxy-substituted ligand 2 and Pd(OAc)2
under the identical reactions conditions as described in
the literature, particularly as it is the only example of
Comparison between ligand 3 (entry 6) and ligand 4
(entry 10) clearly shows that a methyl group in the para-
position to the {SO3} group has no visible effect on

Single-Site Palladium Catalysts
Organometallics, Vol. 24, No. 11, 2005 2757
percentage of extra ethylene insertions, compared to the
experiments when only 1 equiv of ligand is used.
To account for the differences observed between the
reactions conducted in dichloromethane and those in
methanol, we propose that a palladium-acetate inter-
mediate, [P∧OPd(OAc)], is the most likely candidate as
the initiating species, particularly when very low con-
centrations of ligand and metals are used (∼10 mg) and
large volumes of solvents (100 mL) as shown in Scheme
3. Many examples of palladium-phosphine complexes
containing an acetate moiety are known in the litera-
ture.10 Both the dimeric [P∧OPdO∧P] complex and
Pd(OAc)2 are also still likely to be formed in the reactor.
The latter provides a simple explanation for the forma-
tion of Pd(0) after addition of CO and the discolored
polymeric material.
Figure 1. Molecular structure of [P∧OPdO∧P] 1a shown
Having established by NMR and in situ reactions that
complexes containing two ligands form very readily in
dichloromethane solutions, it would seem likely that
these bis-phosphine complexes are plausible derivatives
in the catalytic cycle, potentially as a resting or deac-
tivated species.
at 50% probability. Hydrogen atoms are omitted for clarity.
either activity or number of extra ethylene insertions.
It may be concluded that, both electronically and
sterically, there is as expected no appreciable difference
observed between benzene or toluenesulfonic acid moi-
eties. Therefore, similar groups in these positions may
be used to further aid catalyst solubility.
In summary, sterically bulky substituents and/or
R-groups, providing electron density in close proximity
to the metal center such as OR (where R ) Me, Et, iPr)
in the ortho-position of the unsulfonated phenyl rings
of the phosphine ligand, combined with high pressures
(high ethylene to carbon monoxide ratios) and temper-
atures, give suitable conditions for the formation of
extra ethylene insertions. Stoichiometric metal-to-ligand
ratios are also required, as 2 equiv of ligand to the metal
center reduces the number of extra ethylene insertions
relative to the alternating copolymer.
Neutral Palladium Complexes. Having established
the prerequisites of the catalyst system by in situ
reactions, we were very interested to ascertain the
structural nature of the palladium species formed
during the copolymerization reactions, particularly as
it is believed that a monoligated {P∧OPd} neutral
complex is most likely responsible for the incorporation
of extra ethylene insertions via a neutral metallacycle,
vide infra. First, NMR studies were carried out in
methanol-d4 at room temperature, to reproduce stand-
ard autoclave-loading conditions. However, owing to the
poor solubility of Pd(OAc)2 in pure methanol, further
NMR reactions had to be performed using a mixture of
CD3OD/CD2Cl2 (1:2). Ligands 1-4 react cleanly to give
complexes [P∧OPdO∧P] 1a-4a, Pd(OAc)2, and acetic
acid in less than 20 min, as shown in Scheme 2.
Complementary IR autoclave reactions were also
performed in both dichloromethane and methanol using
a 250 mL Bu¨chi (ATR-IR) autoclave). Once again it is
clear that the reaction in dichloromethane produces the
dimeric palladium complex [P∧OPdO∧P], Pd(OAc)2, and
acetic acid as the only detectable species, as shown by
comparison of the IR spectra to pure samples. Carbon
monoxide (νCO ) 2134 cm-1) was introduced into the
reaction vessel, and almost immediately a new CO band
appeared at ca. 1943 cm-1, which was tentatively
assigned to a palladium-carbonyl complex.
Therefore, pure [P∧OPdO∧P] complexes 1a-4a were
synthesized by reacting 2 equiv of the appropriate ligand
with 1 equiv of Pd(OAc)2 in dichloromethane at room
temperature. In all cases the yields were quantitative.
It must be mentioned that in general the complexes once
formed are not particularly soluble in any common
organic solvents. However, suitable single crystals for
X-ray analysis were grown from dilute solutions of
ligands 1-4 and Pd(OAc)2 in an NMR tube. A repre-
sentative molecular structure of complex 1a is il-
lustrated in Figure 1.11
Selected bond lengths and angles of 1a are depicted
in Table 2.
Molecule 1a is centrosymmetric, with each P∧O ligand
located in a cis-orientation around the metal center,
thus forming two puckered six-membered rings with
each P and O atom in a distorted square planar
configuration (O(1)-Pd(1)-P(1) angle of 85.08(7)°). The
oxygen O(3) atom of the {SO3} moiety takes up a
position above the Pd(1) atom. The second oxygen O(3#)
of the opposite {SO3} moiety takes up a position below
the plane of the Pd(1) atom. A π-interaction between
the two unsulfonated phenyl rings, from the two differ-
ent phosphine ligands, is also observed, with an average
Carom- - -Carom distance of 3.66 Å.
It is apparent from the literature that there are very
few examples of complexes containing a direct {M-O-
SO2} bond. To the best of our knowledge the structures
reported in this study are the first examples of pal-
ladium complexes containing a phosphorus-sulfonate
chelate. It is well known that sulfonate {O2S-O-}
groups are regarded as very weak coordination ligands.
The O(1)-Pd(1) bond length is 2.100(3) Å, which is
comparable to the other two known palladium struc-
tures that contain a {OTs-} ligand (Pd-O ) 2.151 and
2.123 Å). In both cases the {OTs-} group is used as a
(10) (a) Del Piero, G.; Cesari, M. Acta Crystallogr., Sect. B: Crys-
tallor. Cryst. Chem. 1979, 35, 2411. (b) Louie, J.; Hartwig, J. F. Angew.
Chem., Int. Ed. Engl. 1996, 35, 2359. (c) Bianchini, C.; Lee, H. M.;
Meli, A.; Oberhauser, W.; Peruzzini, M.; Vizza, F. Organometallics
2002, 21, 16.
(11) Analogous crystal structures of 2a and 3b will be reported
elsewhere.
As observed in the in situ copolymerization reaction
(entry 9), 2 equiv of the ligand lead to a reduced

2758 Organometallics, Vol. 24, No. 11, 2005
Hearley et al.
Scheme 2
Scheme 3
Table 2. Selected Bond Distances (Å) and Angles (deg) of Complex 1a
C(6)-S(1) ) 1.778
O(1)-Pd(1) ) 2.100
C(7)-P(1) ) 1.825
O(2)-S(1) ) 1.433
O(1)-S(1) ) 1.500
P(1)-Pd(1) ) 2.2458
C(6)-C(1)-P(1) ) 125.3
S(1)-O(1)-Pd(1) ) 111.01
O(1)-Pd(1)-P(1) ) 85.08
C(2)-C(1)-P(1) ) 116.4
C(7)-P(1)-C(1) ) 105.10
O(1)-S(1)-C(6) ) 104.00
C(1)-C(6)-S(1) ) 122.4
C(1)-P(1)-Pd(1) ) 112.86
Table 3. Results of the [P∧OPdO∧P] CO-C2H4
ethylene incorporation was observed. By raising the
temperature to 150 °C (entry 9) and 200 °C (entry 10)
again, extra incorporation of ethylene was also detected,
however, at the expense of activity. It must be noted
that even at 200 °C very little decomposition of the
catalyst was apparent.
Copolymerization Reactionsg
extra C2
C2H4
CO
activity (g
insertion
entry complex T/°C (bar) (bar) mmol-1 h-1
)
(%)d
1
2
3
4
5
6
7
8
9
10
1a
2a
2a
2a
2a
3a
3a
3a
3a
3a
110
110
150
110
110
110
110
110
150
200
30
30
30
30
30
30
60
110
30
60
10
10
10
10
10
10
20
10
10
10
2.5a
<1
1.2
4.1
<1e
1.2f
2.1
3.3
4.9
4.3
4.9
106a
We propose that the [P∧OPdO∧P] complex in metha-
nol, upon addition of monomer(s), most likely forms two
species, as shown in Scheme 4. Cleavage of the Pd-O
bonds by protonation and/or carbonylation of the com-
plex would lead to a dicationic cis-(bis-phosphine) pal-
ladium(II) dicarbonyl complex (a). It has been shown
that the carbonyl ligand may undergo nucleophilic
attack by an alcohol (methanol) to form a monocationic
ester complex (b).13 No example of a cationic Pd(II)
complex has ever been reported in the literature that
is able to introduce extra ethylene insertions in a CO-
C2H4 copolymer.
89a
5a
105a
102a
87b
47b
53a
75c
aIn a 100 mL (50 mL solvent) autoclave. b25 mL (10 mL solvent)
autoclave. c 250 mL (100 mL solvent) autoclave. d Total mol % of
C2H4 relative to purely alternating copolymer, calculated from
13C{1H} NMR spectra (measured in 1,1,3,3-hexafluoropropan-2-
e
f
ol/C6D6). In dichloromethane. Addition of 1 equiv Pd(OAc)2. g 1a
) o-H, 2a ) o-OMe, 3a ) o-Me. Based on Pd over experiment run
time of 1 h.
Alternatively, one of the P∧O ligands may split off
completely, particularly at higher pressures and tem-
peratures, to yield a monoligated {Pd(O∧P)} complex,
which in turn would be able to insert extra ethylene,
as depicted in Scheme 4. It is well known from the
literature that tertiary phosphines, such as PPh3, are
able to dissociate in solution and migrate between
metal centers.14 Having established that a monomeric
{Pd(O∧P)} complex is responsible for the extra C2H4
insertions, we searched for an alternative preactivated
palladium precursor.15
Monosubstituted Palladium Complexes. Very
recently the synthesis and structural characterization
of a preactivated cationic palladium 1,3-bis(diphenylphos-
phino)propane (dppp) complex, bearing a dicyclopenta-
dienylethoxy ligand, was reported by our group.16
Complexes 2b and 3b were synthesized using the
similar methodology, by first converting the appropriate
ligands 2 and 3 to the corresponding sodium salts using
Na2CO3. The dimeric bis[µ2-chloro-1-η2,5η1-6-ethoxy-exo-
5,6-dihydrodicyclopentadienepalladium(II)] complex ([Cp-
OEtPdCl]2) was added at -20 °C in dichloromethane,
weakly coordinating (leaving) group, for example,
[dpppPd(H2O)(OTs)][OTs].12 The bond between the oxy-
gen of the {SO3} group and the palladium center in the
new complexes reported here may be therefore best
described as a phosphine-chelate assisted coordination.
Results of the copolymerization reaction of ethylene
and carbon monoxide using complexes 1a, 2a, and 3a
are shown in Table 3.
As previously observed for ligand 1, complex 1a (entry
1) does not produce any significant amounts of polymer
under the conditions used. Entries 2 and 6 were more
encouraging; both complexes 2a and 3a produced pure
white polymers with good activities (cf. in situ). How-
ever, as expected in both cases, very little extra ethylene
was incorporated relative to the purely alternating CO-
C2H4 copolymer. In an attempt to increase the number
of extra ethylene insertions, 1 equiv of Pd(OAc)2 was
added (entry 5) to complex 2a under identical reactions
conditions. The polymer formed was almost black in
color due to reduction of the Pd(OAc)2 to Pd(0). Analysis
revealed no increase in either activity or percentage of
extra misinsertions compared to entry 2. Upon increas-
ing the relative pressures (entries 7 and 8), extra
(13) Mu¨cke, A.; Rieger, B. Macromolecules 2002, 35, 2865.
(14) (a) Evans, D.; Yagupsky, G.; Wilkinson, G. J. Chem. Soc. A
1968, 2660. (b) O’Connor, C.; Wilkinson, G. J. Chem. Soc. A 1968, 2665.
(15) A stoichiometric reaction between [Pd(COD)ClMe] and ligand
3 in dichloromethane yielded [P∧OPdO∧P] as the only phosphine-
containing product.
(12) (a) Benetollo, F.; Bertani, R.; Bombieri, G.; Toniolo, L. Inorg.
Chim. Acta 1995, 5, 233. (b) Tulloch, A. A. D.; Winston, S.; Danapou-
lous, A. A.; Eastham, G.; Hursthouse, M. B. J. Chem. Soc., Dalton
Trans. 2003, 699.
(16) Meinhard, D.; Hollmann, F.; Huhn, W.; Thewalt, U.; Klinga,
M.; Rieger, B. Organometallics 2004, 23, 5637.

Single-Site Palladium Catalysts
Organometallics, Vol. 24, No. 11, 2005 2759
Scheme 4. Proposed Activation of [P∧OPdO∧P] in Methanol
Scheme 5. Synthesis of Complexes 2b and 3b
Table 4. Selected Bond Distances (Å) and Angles (deg) of Complex 2b
C(25)-S(1) ) 1.765
C(20)-P(1) ) 1.847
P(1)-C(26) ) 1.829
O(5)-Pd(1) ) 2.189
O(5)-S(1) ) 1.485
P(1)-Pd(1) ) 2.302
Pd(1)-C(4)A ) 2.30
Pd(1)-O(5)-S(1) ) 118.4
C25(1)-S(1)-O(5) ) 104.6
Pd(1)-C(10)A ) 1.94
C(20)-P(1)-C(13) ) 105.2
C(20)-P(1)-C(26) ) 108.3
P(1)-Pd(1)-O(5) ) 93.2
C(20)-P(1)-Pd(1) ) 110.2
O(6)-S(1)-O(5) ) 111.6
to yield the neutral complexes 2b and 3b as pale yellow
solids, as depicted in Scheme 5. Each complex was
structurally characterized using standard 1D and 2D
NMR measurements. Full analyses of complexes 2b and
3b are described in the Supporting Information.
Suitable single crystals for X-ray analysis of 2b were
grown from dichloromethane/pentane. A crystal struc-
ture of complex 2b is shown in Figure 2. Selected bond
lengths and angles of 2b are depicted in Table 4.
The structure of 2b clearly shows, as Drent and co-
workers correctly proposed, that the active catalytic
species is based on a {P∧O} chelate.8a The distance
between the O(2) of the o-OMe group and the Pd(1)
center is 3.35 Å. Such a close proximity of an electron-
donating group near the metal center is likely to have
a significant influence on the catalysis. The use of a
dicyclopentadienylalkoxy ligand provides a simple but
highly effective method of generating a preactivated (σ
and π bond) catalyst.
The polymerization reactions with complexes 2b and
3b are shown in Table 5. In general, all copolymeriza-
tion reactions were performed in dichloromethane with
2 wt % B(C6F5)3 as activator. The use of boranes, such
as B(C6F5)3, as phosphine scavengers with neutral
salicylaldiminato Ni(II) complexes18 and promoters/
cocatalysts in combination with cationic palladium bis-
phosphine complexes has been described in the litera-
(17) It is clear from the structural refinement of complex 2b that
the dicyclopentadienyl ligand exists in two chemically equivalent,
diastereomeric forms and each coordinates to the free Pd site in two
possible orientations. As a consequence, superimposed electron density
is observed in this area of the unit cell, which cannot be resolved or at
least it cannot be freely refined. However, it is possible to model an
overlay of both diastereoisomers sharing the same sites and to
constrain the atomic positions, such that the atoms do not shift to
chemically meaningless positions during refinement cycles.
Figure 2. Molecular structure of 2b diastereoisomer (A)
shown at 30% probability. Hydrogen atoms are omitted for
clarity.17
(18) Wang, C.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs, R.
H.; Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3149.

2760 Organometallics, Vol. 24, No. 11, 2005
Hearley et al.
Scheme 6. Proposed Mechanism for the
Incorporation of Extra C2H4 Insertions via a
Neutral Metal Chelate
Table 5. Results of the {P∧OPdCp-OEt} CO-C2H4
Copolymerization Reactionse
extra C2
slave factor
entry complex T/°C C2H4 CO mmol-1 h-1
activitya (g
insertion
)
(%)d
1
2b
2b
2b
2b
2b
2b
3b
3b
110
110
110
50
150
110
110
110
100
100
100
100
100
100
100
100
40
5
292
89
126
<5
87
134
67
40
3.1
2
26.4
3b
4
5
15.1
5
alternating
3.3
5
5
6c
7
5
29.6
40
5
2.2
8
7.5
a Based on Pd over experiment run time of 2 h in a 250 mL
autoclave in dichloromethane with 2% B(C6F5)3 as activator at 65
bar. b No activator. c In methanol. d Total mol % of C2H4 relative
to purely alternating copolymer, calculated from 13C{1H} NMR
spectra (measured in 1,1,3,3-hexafluoropropan-2-ol/C6D6). e 2b )
o-OMe, 3b ) o-Me.
ture.19 It has been shown that such additives provide
increased catalyst stability with respect to reduction and
also provide means of regenerating the catalytically
active species.
analysis of each sample showed that the four samples
have an identical chemical structure; that is, each
sample retains the same ratio of nonalternating vs
alternating regions. Also, further inspection of the peaks
in the 13C NMR spectra, corresponding to double vs
triple extra-ethylene insertions, further supports our
findings. To further confirm the NMR and fractionation
studies, we turned to gel permeation chromatography
(GPC) for further insight. The average molecular weight
of the crude polymer Mw ) 370 000 with a polydispersity
Mw/Mn ) 2. The molecular weight is high and consistent
with the 1H NMR results (HFIP-d2/benzene-d6), where
no end groups could be detected. The results show that
one type of polymer is most likely produced in the
reaction, consisting of random blocks of both alternating
and nonalternating parts.
Mechanistic Proposal. To account for the polymers
observed, we propose the following mechanism for the
extra incorporation of ethylene into the growing CO-
C2H4 copolymer, as depicted in Scheme 6. As previously
mentioned by Drent and co-workers,8a a stereoelectronic
destabilization of the neutral chelate enables ethylene
to compete with carbon monoxide and to open the five-
membered metallacycle. It is known for cationic pal-
ladium complexes, under typical polymerization condi-
tions, that a CO insertion is both rapid and reversible.21
Therefore we assume the neutral chelate provides
precedence for the decarbonylation reaction over the
formation of a CO inserted six-membered chelate. In
turn, increasing the pressure and employing high gas
blend ratios CO/C2H4 (1:20) statistically provides higher
probability for an ethylene compared to a carbon
monoxide insertion. The resulting seven-membered
chelate is believed to open more easily, and further
ethylene units can be incorporated. A higher energy
barrier for the rate-determining ethylene insertion step
may account for the low activities found for these
systems compared to the cationic palladium catalysts.
We propose that the polymerization is able to proceed
via two intertwined pathways (Scheme 6), where the
same active species may switch between the production
of alternating and nonalternating blocks in the same
Entry 1 corresponds to the complex 2b with a C2H4/
CO ratio of 2.5:1 and an overall pressure of 65 bar. The
reactivity is very high (approximately 300 g mmol-1 h-1
)
compared to the in situ reactions. However, the percent-
age of extra ethylene relative to the purely alternating
polyketone as expected is low (3.1%). As previously
observed in the in situ polymerizations, the ratio of gas
blend plays a significant role for extra ethylene incor-
poration into the copolymers. The result demonstrates
that the catalyst alone is not able to incorporate large
quantities of extra ethylene into the growing polymer.
Further copolymerization reactions using complex 2b
were performed using a gas blend ratio of C2H4/CO )
20:1 in the presence of B(C6F5)3 (entry 2) or pure
dichloromethane (entry 3). The reaction performed in
methanol (entry 6) at 110 °C and 65 bar showed high
activity (134 g mmol-1 h-1) and the highest number of
extra ethylene insertions relative to the purely alternat-
ing polymer structure (30%). Other bulkier groups such
as biphenyl or phenoxy, which are currently under
investigation in our laboratories, may provide a chance
to insert even more ethylene relative to the alternating
copolymer.20
Polymer Properties. The 13C NMR (methylene
region) of the copolymer (Table 5, entry 2) is in ac-
cordance with that previously found in the literature8a
(Supporting Information). Analyses of the polymer(s) by
differential scanning calorimetry (DSC) showed that the
melting points of the materials are reduced from ca.
260-270 °C (perfectly alternating) to 220 °C (entry 2),
which indicates a significant decrease in the interactions
between the polymer chains. One major reason for the
polymer analysis is to establish if a single polymer or
reactor blend is produced during the CO-C2H4 co-
polymerization reaction. Therefore, fractionation of the
polymeric material was carried out using benzene/
hexafluoro-2-propanol (HFIP). Approximately 2 g of the
polymer was dissolved in HFIP and precipitated into
four equal portions, by the addition of benzene. 13C NMR
(19) Barlow, G. K.; Boyle, J. D.; Cooley, N. A.; Ghaffar, T.; Wass, D.
F. Organometallics 2000, 19, 1470.
(20) Schmid, M.; Eberhardt, R.; Klinga, M.; Leskela, M.; Rieger, B.
Organometallics 2001, 20, 2321.
(21) Brumbaugh, J. S.; Sen, A. J. Am. Chem. Soc. 1988, 110, 803.

Single-Site Palladium Catalysts
Organometallics, Vol. 24, No. 11, 2005 2761
in THF (60 mL) was added n-BuLi (58 mL, 92.8 mmol, 1.6 M
in hexane) at 0 °C. After stirring for 6 h at room temperature,
a solution of bis(phenyl)chlorophosphine (10.24 g, 46.4 mmol)
in THF (30 mL) was added dropwise at 20 °C and the reaction
was stirred overnight. After this time, NH4Cl (2.48 g, 46.4
mmol) in water (50 mL) was added and the organic solvent
removed in vacuo. The aqueous solution was washed with
diethyl ether (2 × 50 mL) and the organic phase acidified with
HCl (37% in H2O). After extraction with dichloromethane (2
× 50 mL) the crude product was dried over MgSO4. The solvent
was removed in vacuo and recrystallized from dichloromethane/
methanol at -20 °C. The resulting white crystals were filtered
polymer chain. The insertion of a C2H4 molecule, directly
followed by CO, leads to random blocks of alternating
copolymer. Alternatively, an extra insertion of C2H4
yields nonalternating blocks. The pressure, tempera-
ture, and gas blend ratio contribute significantly to the
polymer composition, thus providing a method to tune
material properties.
Conclusion
In conclusion, we have been able to demonstrate the
first structurally characterized, very stable (under the
conditions used), neutral, single-site catalysts contain-
ing the first example of a phosphorus-sulfonate chelate
for the nonalternating copolymerization of ethylene and
carbon monoxide. Up to 30% extra ethylene has been
incorporated into the CO-C2H4 copolymers. As mea-
sured in one case (Table 5, entry 2) the copolymer
produced from these new catalysts has high molecular
weight with narrow polydispersity, indicating one type
of polymer comprised of both alternating and nonalter-
nating parts is formed.
1
and dried to give the desired ligand. Yield ) 9.7 g (61%). H
NMR (CDCl3) δ: 8.9-7.2 (m, 14H, arom). 31P NMR (CDCl3) δ:
4.4 (s); MS: found 341.0, calc 341.4. Anal. Found: C, 65.64,
H, 5.53. Calc: C, 65.61, H, 5.51.
Ligand 3. 2-{Bis(2-methylphenyl)phosphino}toluene-
sulfonic Acid. To a solution of dry toluenesulfonic acid (4.34
g, 25.2 mmol) in THF (60 mL) was added n-BuLi (31.54 mL,
50.4 mmol, 1.6 M in hexane) at 0 °C. After stirring for 6 h at
room temperature, a solution of 2-{bis(2-methylphenyl)}-
chlorophosphine (6.12 g, 25.2 mmol) in THF (30 mL) was added
dropwise at 20 °C and the reaction was stirred overnight. After
this time, NH4Cl (1.34 g, 25.2 mmol) in water (50 mL) was
added and the organic solvent removed in vacuo. The aqueous
solution was washed with diethyl ether (2 × 50 mL) and the
organic phase acidified with HCl (37% in H2O). After extraction
with dichloromethane (2 × 50 mL) the crude product was dried
over MgSO4. The solvent was removed in vacuo and the crude
product recrystallized from dichloromethane/ethanol at -20
°C. The resulting white crystals were filtered and dried to give
Experimental Section
General Considerations. All reactions were carried out
under a dry argon or nitrogen atmosphere using standard
Schlenk line techniques or in a glovebox. Solvents were
obtained from commercial sources and dried (as appropriate,
pentane, hexane, tetrahydrofuran over sodium (ketyl radical),
dichloromethane over CaH2, and methanol over magnesium
turnings) and deoxygenated prior to use. Deuterated solvents
were dried over the appropriate drying agent (chloroform-d1,
dichloromethane-d2 over CaH2, benzene-d6 over Na, and
methanol-d4 over 4 Å molecular sieves). The NMR solvents
were deoxygenated using three freeze-thaw-pump cycles and
transferred using a high-vacuum line. Commercial reagents
were used as supplied, unless otherwise stated. Toluene and
benzene sulfonic acid were dried by Dean-Stark distillation
from benzene, prior to use. 2-{Bis(2-methylphenyl)chlorophos-
phine and 2-{bis(2-methoxyphenyl)phosphino}benzenesulfonic
acid (ligand 2) were prepared according to the literature
methods.8a,22 1H and 13C NMR spectra (chemical shifts relative
to residual solvent) and 31P{1H} (chemical shifts relative to
80% H3PO4) were recorded on either a Bruker 400 MHz or
Bruker 500 MHz spectrometer at ambient temperature unless
otherwise stated.
N,N-Diethylaminodichlorophosphine [Cl2PNEt2]. A 1
L Schlenk flask was charged with diethyl ether (400 mL) and
PCl3 90.6 g (0.66 mol) and cooled to -20 °C in an ice/NaCl
bath. Diethylamine (86.3 g, 1.18 mol) in 100 mL of diethyl
ether was added dropwise over a period of 1 h. The reaction
mixture was allowed to warm to room temperature and further
stirred for 1 h. The diethyl ether was removed by distillation
under argon under normal pressure (oil bath 70 °C). The crude
product was distilled in vacuo (oil pump), with an oil bath
temperature of 190 °C, from the inorganic salts. The pure
product was obtained by fractional distillation through a
Vigreux column (head temperature 30.5 °C). Yield ) 80.4 g,
colorless liquid (70%). 1H NMR (CDCl3) δ: 3.26-3.35 (sext,
4H, J ≈ 7 Hz); 1.15 (t, 6H, J ) 7.07 Hz), 13C{1H} NMR (CDCl3)
δ: 13.9 (d, 2C, J ) 5.12 Hz); 41.4 (d, 2C, J ) 22.69 Hz), 31P
NMR (CDCl3) δ: 163.26 (s). Caution: the solid residue is
pyrophoric and should be destroyed carefully, by the slow
addition of water.
1
the desired ligand. Yield ) 5.2 g (54%). H NMR (CDCl3) δ:
2.29 (3H, s, CH3-C6H3SO3H), 2.41 (6H, s, 2CH3), 6.8-8.3 (11H,
m, arom). 31P NMR (CDCl3) δ: -8.75 (br, s). MS: found 384.4,
calc 384.4. Anal. Found: C, 65.64, H, 5.53. Calc: C, 65.61, H,
5.51.
Ligand 4. 2-{Bis(2-methylphenyl)phosphino}benzene-
sulfonic Acid. To a solution of dry benzenesulfonic acid (7.17
g, 45.3 mmol) in THF (60 mL) was added n-HexLi (36.26 mL,
90.6 mmol, 2.5 M in hexane) at 0 °C. After stirring for 18 h at
room temperature, a solution of 2-{bis(2-methylphenyl)chloro-
phosphine (9.83 g, 40.7 mmol) in THF (40 mL) was added
dropwise at 20 °C and the reaction stirred overnight. After
this time, NH4Cl (2.42 g, 45.3 mmol) in water (50 mL) was
added and the organic solvent removed in vacuo. The aqueous
solution was washed with dichloromethane (2 × 50 mL) and
the organic phase acidified with HCl (37% in H2O). After
extraction with dichloromethane (2 × 50 mL) the crude product
was dried over MgSO4. The solvent was further removed in
vacuo and the crude product recrystallized from dichloro-
methane/ether at -20 °C. The resulting white crystals were
filtered and dried to give the desired ligand. Yield ) 11.1 g
1
(66%). H NMR (CDCl3) δ: 2.39 (6H, s, 2CH3), 7.0-8.4 (11H,
m, arom). 31P NMR (CDCl3) δ: -8.43 (s). MS: found 370.4,
calc 370.4. Anal. Found: C, 64.43, H, 5.10. Calc: C, 64.85, H,
5.17.
Complex 1a. Ligand 1 (0.9 g, 2.88 mmol) was added to a
Schlenk flask containing Pd(OAc)2 (0.35 g, 1.44 mml). Di-
chloromethane (30 mL) was added and the solution stirred or
24 h. During this time the desired complex precipitated from
the solution as a bright yellow powder. The solvent was
removed by cannula, and the complex was washed with
dichloromethane (3 × 20 mL). Yield ) 1.03 g (98%). Anal.
Found: C, 57.70, H, 4.63. Calc: C, 57.76, H, 4.62. Suitable
crystals for single-crystal X-ray analysis were grown from a
dichloromethane solution, containing 2 equiv of ligand 1 and
1 equiv of Pd(OAc)2 in dichloromethane under argon in an
NMR tube.
Ligand 1. 2-(Diphenylphosphino)benzenesulfonic Acid.
To a solution of dry benzenesulfonic acid (7.34 g, 46.4 mmol)
Complex 2a. Ligand 2 (1 g, 2.48 mmol) was added to a
Schlenk flask containing [Pd(OAc)2] (0.28 g, 1.2 mmol). Di-
chloromethane (30 mL) was added and the solution stirred for
(22) Dang, T. P. J. Organomet. Chem. 1975, 91, 105.

2762 Organometallics, Vol. 24, No. 11, 2005
Hearley et al.
24 h. During this time the desired complex precipitates from
solution as a bright yellow powder. The solvent was removed
by cannula, and the complex was washed with dichloro-
methane (3 × 20 mL). Yield ) 1.1 g (98).
(arom), 134.5 (arom), 131.7 (arom), 131.6 (arom), 21.3 (CH3),
147.1 (aromq), 131.9 (arom), 132.1 (arom), 131.7 (arom), 140.1
(arom), 126.4 (aromq), 23.3 (aromq). 31P NMR (CDCl3) δ: 12.0
(s). Anal. Found: C, 61.40, H, 5.91. Calc: C, 61.28, H, 5.81.
In Situ Copolymerizations. In a typical reaction, 0.04
mmol (∼10 mg) of Pd(OAc)2, 0.06 mmol of ligand 1-4, and
10, 50, or 100 mL of MeOH were transferred to a 25 mL (Parr),
100 mL (Roth), or 250 mL (Bu¨chi Limbo) autoclave. The
autoclave was quickly closed and purged with argon and then
charged with ethylene until a pressure of 30 bar was reached.
Subsequently, CO was introduced to reach the pressure of 40
bar (C2H4/CO ) 30:10). After the solution was completely
saturated with gas (15 min), the 25 mL or 100 mL autoclave
was heated using a preheated silicon oil bath (110 °C) for 1 h.
The 250 mL autoclave was heated and temperature main-
tained using its own electrical heating block and water-cooling
system for 2 h. After this time the cooled contents were filtered
and washed with methanol (3 × 50 mL) and dried in vacuo.
Copolymerization Using 1a-4a. In a typical reaction,
0.012 mmol (∼10 mg) of complex 1a-4a and 50 or 100 mL of
MeOH were transferred to a 100 mL (Parr) or 250 mL (Bu¨chi,
Limbo) autoclave. The autoclave was quickly closed and purged
with argon and then charged with ethylene until a pressure
of 30 bar was reached. Subsequently, CO was introduced to
reach a pressure of 40 bar (C2H4/CO ) 30:10). After the
solution was completely saturated with gas (15 min), the 100
mL autoclave was heated using a preheated silicon oil bath
(110 °C) for 1 h. The 250 mL autoclave was heated and
temperature maintained using its own electrical heating block
and water-cooling system for 2 h. After this time the cooled
contents were filtered and washed with methanol (3 × 50 mL)
and dried in vacuo.
Copolymerizations Using 2b-3b, at High Tempera-
ture and Pressures. In a typical reaction, 0.015 mmol (∼10
mg) of complex 2b-3b was dissolved in 10 mL of dichloro-
methane and added to a gas buret under argon. A 250 mL
(Bu¨chi, Limbo) autoclave was charged with the desired solvent
(CH2Cl2 or MeOH) and activator (if required). The autoclave
was heated to the desired temperature, and the desired
pressure (10 bar below actual pressure) of gas was added using
an 800 mL reservoir/gas mixing chamber. The carbon monoxide/
ethylene gas feed rates were controlled by a remote personal
computer (PC). Once the equilibrium of the system had been
reached, the contents of the gas buret was added (using the
required pressure). Polymerization reactions were run for 2
h. After this time the cooled contents were filtered and washed
with methanol (3 × 50 mL) and dried in vacuo.
Note: Complex 2a is insoluble in all common organic
solvents at room temperature. Anal. Found: C, 52.49, H, 4.01.
Calc: C, 52.84, H, 3.99.
Complex 3a. A Schlenk tube (100 mL) was charged with
ligand 3 (0.583 g, 1.52 mmol) and [Pd(OAc)2] (0.171 g, 0.76
mmol). Dichloromethane (40 mL) was added and the solution
stirred for 24 h at room temperature. The desired complex
precipitated during the reaction. The solvent was removed
with a cannula and the complex washed with pentane (3 × 20
mL) to afford a bright yellow powder in almost quantitative
1
yield (0.65 g). H NMR (CDCl3) δ: 2.60 (6H, s, CH3), 2.87 (3,
s, CH3), 2.97 (6H, s, CH3), 6.8-8.3 (m, 22H, arom). 31P NMR
(CDCl3) δ: -3.44 and -4.02. Anal. Found: C, 57. 70, H, 4.63.
Calc: C, 57.76, H, 4.62.
Complex 4a. An NMR tube was charged with ligand 4 (0.02
g, 0.054 mmol) and [Pd(OAc)2] (0.012 g, 0.027 mmol). Deuter-
ated dichloromethane (0.5 L) was added. After 30 min at room
temperature the solvent and volatiles were removed in high
vacuum and the complex washed with pentane (3 × 2 mL) to
afford a bright yellow microcrystalline powder. Yield ) 0.042
1
g (92%). H NMR (CD2Cl2) δ: 2.58 (3H, s, CH3), 2.80 (3H, s,
CH3), 2.87 (6H, s, CH3), 2.99 (3H, s, CH3, 6.8-8.3 (m, 24H,
arom). 31P NMR (CDCl3) δ: -3.1 and -3.8 (JP-P 109 Hz). Anal.
Found: C, 56.80, H, 4.18. Calc: C, 56.84, H, 4.29.
Complex 2b. Ligand 2 (2.00 g, 4.97 mmol) was added to a
Schlenk flask containing Na2CO3 (0.63 g, 0.59 mmol). Di-
chloromethane (30 mL) was added and the solution stirred for
4 h. After this time, the desired white sodium salt precipitated.
[Cp-OEtPdCl]2 (1.58 g, 2.48 mmol) was added with stirring at
-20 °C, and the solution was allowed to further stir overnight.
The complex was filtered and washed with diethyl ether (3 ×
20 mL) to yield complex 2b as a pure pale yellow solid. Yield
) 1.49 g (88%). 1H NMR (CDCl3) δ: 0.78 (3H, t, CH3), 2.74
and 3.15 (2H, m, CH2-O), 3.36 (1H, d, O-CH), 2.69 (1H, m,
CH), 2.60 and 2.32 (2H, m, CH2), 2.26 (1H, d, CH), 3.74 (1H,
s, CH), 2.93 (1H, m, CH), 1.65 (1H, m, CH), 1.04 (1H, d, CH2)
and 1.33 (1H, d, CH2), 6.37 (1H, m, dCH), 6.37 (1H, m, dCH),
7.83 (1H, m, arom), 7.39 (1H, m, arom), 7.48 (1H, m, arom),
8.10 (1H, m, arom), 3.58 (3H, s, CH3) and 3.60 (3H, s, CH3),
6.89 (2H, m, arom), 7.07 (2H, m, arom), 6.91 (2H, m, arom),
7.24 (2H, m, arom). 13C{1H} NMR (CDCl3) δ: 15.4 (CH3), 63.3
(CH2), 80.4 (CHCp), 39.1 (CHCp), 31.6 (CH2), 41.4 (CHCp), 51.2
(CHPd), 53.7(CHCp), 55.0 (CHCp), 35.8 (CH2), 129.7 (dCH), 134.6
(dCH), 133.6 (aromP), 148.2 (aromS), 120.8 (arom), 128.2
(arom), 125.4 (arom), 137.9 (arom), 57.3 (CH3), 160.6 (aromt),
111.1 (arom), 133.3 (arom), 127.6 (arom), 137.5 (arom), 115.3
(aromp). 31P NMR (CDCl3) δ: 12.1 (s). Anal. Found: C, 56.26,
H, 5.21. Calc: C, 56.18, H, 5.01.
NMR Analyses of Copolymers. 13C{1H} NMR copolymer
analyses were measured in hexafluoro-2-propanol/benzene-d6
(5 mm NMR tubes; volume of solvent, 0.5 mL) on a Bruker
500 MHz spectrometer (chemical shifts relative to benzene-
d6). 13C{1H} NMR spectra were simulated (using ACD/Chem-
Sketch, version 4.04) to compare calculated and observed
chemical shifts of the nonalternating copolymers, showing good
agreement. Further comparisons of spectra were made to the
literature.8a
Polymer Fractionation. Approximately 2 g of the polymer
(Table 3, entry 2) was fractionated into five approximately
equal parts using hexafluoro-2-propanol/benzene. First, pure
polyethylene was removed by filtration from hexafluoro-2-
propanol. To the hexafluoro-2-propanol solution, benzene was
added and the precipitate filtered, washed with methanol, and
dried in a vacuum oven at 80 °C. The same process was
repeated three further times. Each sample was analyzed using
13C NMR in hexafluoro-2-propanol/benzene. Fraction A is pure
polyethylene with Mw ) 3500 and Mw/Mn ) 2. Fractions B, C,
D, and E contain identical peaks (structure) with the same
extra ethylene incorporation (25-26%) as observed in the
crude polymer.
Complex 3b. Ligand 3 (2 g, 5.2 mmol) was added to a
Schlenk flask containing Na2CO3 (0.66 g, 0.62 mmol). Di-
chloromethane (30 mL) was added and the solution stirred for
4 h. After this time, the desired white sodium salt precipitated.
[Cp-OEtPdCl]2 (1.66 g, 2.6 mmol) was added with stirring at
-20 °C, and the solution was allowed to stir overnight. The
complex was filtered and washed with diethyl ether (3 × 20
mL) to yield complex 3b as a pure pale yellow solid. Yield )
1
1.54 g (89%). H NMR (CDCl3) δ: 0.89 (6H, t, CH3), 2.68 and
2.93 (2H, m, CH2), 2.83 (1H, s, CH), 1.97 (1H, m, CH), 1.02
and 1.38 (2H, d CH2), 2.11 (1H, m, CH), 3.47 (1H, m, CH),
2.60 (1H, m, CH), 1.84 (1H, m, CH), 1.66 and 1.69 (2H, d, CH2),
6.44 (1H, s, CH), 6.44 (1H, s, CH), 7.49 (1H, m, arom), 7.41
(1H, m, arom), 8.11 (1H, m, arom), 2.21 (6H, s, arom), 7.21
(2H, m, arom), 7.31 (2H, d, arom), 6.82 (2H, d, arom), 7.35
(2H, d, arom), 2.3 (3H, m, arom). 13C{1H} NMR (CDCl3) δ: 15.5
(CH3), 63.3 (CH2), 81.0 (CHCp), 39.1 (CHCp), 32.4 (CH2Cp), 41.1
(CHCp), 54.9 (CHPd), 58.1 (CHCp), 53.6 (CHCp), 35.7 (CH2Cp),
126.3 (CHCp), 129.4 (CHCp), 121.9 (CHp), 142.3 (CHS), 131.4
Gel Permeation Chromatography (GPC). The polymer
(Table 5, entry 2) was dissolved in hexafluoro-2-propanol

Single-Site Palladium Catalysts
Organometallics, Vol. 24, No. 11, 2005 2763
Table 6. Crystallographic Data for 1a and 2b
ments (samples A and B) were made, Mw(A) ) 367 800 and
Mw(B) ) 369 900. Mw/Mn(A) ) 2.02 and Mw/Mn(B) ) 2.04. The
GPC trace(s) and data are shown in the Supporting Informa-
tion).
1a
2b
empirical formula
formula mass
habit
cryst size [mm]
cryst syst
space group
a [Å]
C
36H28O6P2PdS2
C32H35O6PPdS‚CH2Cl2
769.96
789.05
X-ray Crystallography. Data of 1a and 2b were collected
with a STOE-IPDS diffractometer using graphite-monochro-
mated Mo KR radiation (Table 6). The structures were solved
by heavy-atom methods (SHELXS-86)23 and refined by full-
matrix least squares against F2 (SHEXLS-97).24 Hydrogen
atoms were included in the refinement, in calculated positions
using a riding model. For complex 1a, all other non-hydrogen
atoms were refined with anisotropic displacement parameters.
Complex 2b is disordered in the dicyclopentadienylethoxy
region, due to the presence of two diastereoisomers. Therefore,
the cyclopentadienyl moiety of each diastereoisomer was
refined isotropically using PART with DANG and DFIX
constraints.
yellow plate
0.24 × 0.17 × 0.12
monoclinic
C2/c
pale yellow plate
0.29 × 0.24 × 0.15
triclinic
P1h
9.1853 (7)
17.5867(10)
20.2249(19)
91.040(11)
3266.6(4)
4
9.0884(11)
10.2010(12)
19.395(3)
86.427(15)
1690.4(4)
2
b [Å]
c [Å]
â [deg]
V [Å3]
Z
F
calcd [g/cm-3
]
1.604
1.513
abs coeff [mm-1
F(000)
]
0.841
0.858
1600
788
temp [K]
abs corr
223(2)
223(2)
semiempirical
0.9058, 0.8236
0.707173
none
Tmin, Tmax
λ [Å], Mo KR
scan type
0, 0
Acknowledgment. We would like to thank Bayer
AG for Funding (A.K.H. and R.J.N.). All would like to
express thanks to Mr. Ulrich Ziegler for assistance with
polymer NMR measurements and Mr. B. Mu¨ller for
X-ray data collection.
0.707173
image plate/rotation image plate/rotation
2.01-25.99
2θ range [deg]
2.10-23.99
10741, 4972
382
no. of reflns measd 3188, 3188
total, unique no. of 214
params
R1(F2 > 2σ(F2)
wR2(F2 > 2σ(F2)
GooF, S
0.0360
0.0826
Supporting Information Available: Crystal structure
data of 1a and 2b, NMR analyses of complex 2b and 3b, a
selected GPC trace, and 13C NMR data from the copolymer
Table 5, entry 2, are available free of charge via the Internet
at http://pubs.acs.org.
0.0763
0.2445
0.987
0.973
res electron dens
0.380, -0.622
2.430, -0.582
[e-/Å3]
(HFIP)/0.05 M potassium trifluoroacetate (KTFAC) to yield a
concentration of 3 g/L. The sample was allowed to stand at
room temperature for 5 h. Then a 50 µL sample at room
temperature was taken, filtered (pore size ) 1 µm), and
injected into the GPC. Calibration was preformed using poly-
(methyl methacrylate) (PMMA) as standard. Two measure-
OM0490785
(23) Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467.
(24) Sheldrick, G. M. SHELXL-97; University of Go¨ttingen: Go¨t-
tingen, Germany, 1997.
(25) Polymers were often gray in color, due to catalyst decomposition.