Catalytic activity of cationic diphospalladium(II) complexes in the alkene/CO copolymerization in organic solvents and water in dependence on the length of the alkyl chain at the phosphine ligands

Article abstract of DOI:10.1016/S0022-328X(00)00154-6

A series of diphos ligands CH₂(CH₂PR₂)₂ (1a-x) (a-g: R=(CH₂)_nOH, n=1, 3-8; h-k: R=(CH₂)_nCH(CH₂OH)₂, n=3-6; l-u: R=C_nH_2n+1, n=1-8, 10, 14; v-x: R=CH(CH₃)₂, (CH₂)₂CH(CH₃)₂, (CH₂)₃CH(CH₃)₂, (Scheme 1), provided with functionalities of different polarity, was prepared photochemically by hydrophosphination of the corresponding 1-alkenes with H₂P(CH₂)₃PH₂ or reaction of Grignard reagents with Cl₂P(CH₂)₃PCl₂. The water-soluble palladium complexes [(R₂P(CH₂)₃PR₂)Pd(OAc)₂] (2a-k) were obtained by reaction of Pd(OAc)₂ with the ligands 1a-k in ethanol-acetonitrile. Treatment of PdCl₂(NCC₆H₅)₂ with 1l-x afforded the dichloropalladium(II) complexes [(R₂P(CH₂)₃PR₂)PdCl₂] (3l-x). Upon chloride abstraction with AgBF₄ in dichloromethane-acetonitrile the dicationic palladium(II) complexes [(R₂P(CH₂)₃PR₂)Pd(NCCH ₃)₂][BF₄]₂ (4l-x) are formed. The structure of 4n (R=n-Pr) was investigated by an X-ray structural analysis. In particular the water-soluble complexes 2c-k proved to be highly active in the carbon monoxide/ethene copolymerization under biphasic conditions (water-toluene). In the presence of an emulsifier and methanol as activator the catalytic activity increased by a factor of about three. Also higher olefins could be successfully incorporated into the copolymerization with CO and the terpolymerization with ethene and CO. The catalytic activity of the dicationic complexes 4l-x in the propene or 1-hexene/CO copolymerization strongly depends on the length of the alkyl chain R. At 25°C a maximum is achieved in the case of 4q (R=nHex) which is five times more active than the corresponding catalyst with the dppp-ligand. This maximum is shifted to 4t (R=n-C₁₀H₂₁) if the temperature is raised to 60°C. The 1-alkene/CO copolymers are distinguished by their regioregular microstructure and their ultra high molecular weights. Compared to the sulfonated dppp-SO₃ catalyst the water-soluble complexes 2c,e,f,h are responsible for a higher 1-hexene incorporation in the terpolymerization of ethene with 1-hexene and CO.

Full text of DOI:10.1016/S0022-328X(00)00154-6

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 602 (2000) 173–187
Catalytic activity of cationic diphospalladium(II) complexes in the
alkene/CO copolymerization in organic solvents and water in
dependence on the length of the alkyl chain at the
phosphine ligands
Ekkehard Lindner a,*, Markus Schmid a, Joachim Wald a, Joachim A. Queisser b,
Michael Gepra¨gs b, Peter Wegner a, Christiane Nachtigal a
a Institut fu¨r Anorganische Chemie der Uni6ersita¨t Tu¨bingen, Auf der Morgenstelle 18, D-72076 Tu¨bingen, Germany
b BASF Aktiengesellschaft, D-67056 Ludwigshafen, Germany
Received 17 December 1999; accepted 11 February 2000
Dedicated to Professor Reinhard Schmutzler on the occasion of his 65th birthday
Abstract
A series of diphos ligands CH2(CH2PR2)2 (1a–x) (a–g: R=(CH2)nOH, n=1, 3–8; h–k: R=(CH2)nCH(CH2OH)2, n=3–6;
l–u: R=CnH2n+1, n=1–8, 10, 14; v–x: R=CH(CH3)2, (CH2)2CH(CH3)2, (CH2)3CH(CH3)2, (Scheme 1), provided with
functionalities of different polarity, was prepared photochemically by hydrophosphination of the corresponding 1-alkenes with
H2P(CH2)3PH2 or reaction of Grignard reagents with Cl2P(CH2)3PCl2. The water-soluble palladium complexes
[(R2P(CH2)3PR2)Pd(OAc)2] (2a–k) were obtained by reaction of Pd(OAc)2 with the ligands 1a–k in ethanol–acetonitrile.
Treatment of PdCl2(NCC6H5)2 with 1l–x afforded the dichloropalladium(II) complexes [(R2P(CH2)3PR2)PdCl2] (3l–x). Upon
chloride
abstraction
with
AgBF4
in
dichloromethane–acetonitrile
the
dicationic
palladium(II)
complexes
[(R2P(CH2)3PR2)Pd(NCCH3)2][BF4]2 (4l–x) are formed. The structure of 4n (R=n-Pr) was investigated by an X-ray structural
analysis. In particular the water-soluble complexes 2c–k proved to be highly active in the carbon monoxide/ethene copolymeriza-
tion under biphasic conditions (water–toluene). In the presence of an emulsifier and methanol as activator the catalytic activity
increased by a factor of about three. Also higher olefins could be successfully incorporated into the copolymerization with CO and
the terpolymerization with ethene and CO. The catalytic activity of the dicationic complexes 4l–x in the propene or 1-hexene/CO
copolymerization strongly depends on the length of the alkyl chain R. At 25°C a maximum is achieved in the case of 4q
(R=nHex) which is five times more active than the corresponding catalyst with the dppp-ligand. This maximum is shifted to 4t
(R=n-C10H21) if the temperature is raised to 60°C. The 1-alkene/CO copolymers are distinguished by their regioregular
microstructure and their ultra high molecular weights. Compared to the sulfonated dppp-SO3 catalyst the water-soluble complexes
2c,e,f,h are responsible for a higher 1-hexene incorporation in the terpolymerization of ethene with 1-hexene and CO. © 2000
Elsevier Science S.A. All rights reserved.
Keywords: Palladium complexes; Water solubility; Homogeneous catalysis; Copolymerization; Polyketones
1. Introduction
tracts considerable attention. This interest stems from
the easy and cheap access of the starting materials, a
plausible and well investigated reaction mechanism [1],
the polymers engineering properties [2], the possibility
of further derivatization of the carbonyl group [3], and
from the potential utility of the polyketones as
photodegradable materials [4]. Commonly the copoly-
merization is catalyzed by cationic palladium(II) com-
The palladium catalyzed strictly alternating copoly-
merization of alkenes with carbon monoxide still at-
* Corresponding author. Tel.: +49-7071-2972039/22; fax: +49-
7071-295306.
E-mail address: ekkehard.lindner@uni-tuebingen.de (E. Lindner)
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 2 - 3 2 8 X ( 0 0 ) 0 0 1 5 4 - 6

174
E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
Scheme 1.
plexes, provided with bidentate phosphorus and nitro-
gen ligands [5] and weakly coordinating anions [6]. In
particular diphosphines with a hydrocarbon backbone
consisting of three methylene groups belong to the most
active catalysts [5,7]. Late transition metal complexes
are suitable for the copolymerization of polar and
non-polar monomers. A great variety of olefins [8] and
imines [9] were employed and even carbon monoxide
was replaced by other polar building blocks [10].
Steric and electronic properties of the catalyst exert
strong influence in metal mediated polymerizations. In
the polymerization of olefins with b-diiminenickel(II)
and palladium(II) complexes bulky substituents are
needed to suppress chain transfer and termination reac-
tions [11]. Gibson et al. showed that cobalt(III) and
iron(II) complexes with steric demanding tridentate ni-
trogen donors are suitable for the polymerization of
ethylene, while less bulky ligands are responsible for
oligomers [12]. Moreover the microstructure of the
resulting polymers can be controlled by the catalyst. In
the copolymerization of propene with carbon monoxide
the regio- and stereoregularity of the polyketones is
dependent on the kind of the substituent attached to
the phosphorus donor. With [(dppp)Pd(NCMe)2][BF4]2
(dppp=1,3-bis(diphenylphosphino)propane)
regioir-
regular materials are obtained, whereas alkyl sub-
stituents lead to almost perfectly head-to-tail connected
units [13]. Recently optically active polyketones were
prepared by using chiral palladium(II) complexes
[13b,14]. Finally the ligand offers the possibility to
adapt the solubility of the catalyst for different media
[15]. The present work is devoted to an efficient ligand
synthesis, which offers an easy access to diphos ligands
with phosphorus attached alkyl substituents of different

E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
175
chain length. These alkyl groups contain terminal polar
or non-polar functions. Palladium(II) complexes with
these ligands were applied in the copolymerization of
carbon monoxide with ethene and different a-olefins in
water, organic solvents, and in a biphasic system with
and without emulsifiers. By variation of the kind of
surfactants, the electronic and steric factors of the
catalysts and the reaction temperature, the impact on
the catalytic activity and the molecular weights of the
polyketones were studied and optimized.
If (PhCN)2PdCl2 is subjected to a CH2Cl2 solution of
the diphosphines 1l–x, the colorless, air-stable palladi-
um(II) complexes 3l–x are formed (Scheme 1), which
were purified by recrystallization from different sol-
vents. Chloride abstraction to give the dicationic
bis(acetonitrile)palladium(II) complexes 4l–x succeeded
with two equivalents of AgBF4 in CH2Cl2/CH3CN. In
the presence of moist air, acetonitrile is slowly ex-
changed for water. Whereas 4l–x are soluble in polar
or even less polar organic media, 4u, which is provided
with tetradecyl substituents, is also soluble in non-polar
media.
2. Results and discussion
2.2. Crystal structure of 4n
2.1. Synthesis of the diphos ligands 1a–x and
diphospalladium(II) complexes 2a–k, 3l–x, and 4l–x
The dicationic diphospalladium(II) complex 4n crys-
tallizes in the monoclinic space group P21/n with four
formula units per unit cell [20]. The bond angles
P1ꢀPdꢀP2 (91.42(3)°) and N1ꢀPdꢀN2 (87.2(1)°) differ
slightly from an ideal square-planar geometry (see
Table 1). Remarkably the six-membered ring, formed
by the diphos ligand and the metal atom, is arranged in
a chair conformation (Fig. 1). The Pd center and the C2
With some modifications the diphos ligands 1b–k,n–
u,w,x were prepared according to a procedure described
by Maier [16]. An excess of the corresponding olefin
was photochemically hydrophosphinated with the dipri-
mary phosphine H2P(CH2)3PH2 (Scheme 1). This con-
venient synthesis is nearly quantitative and simplifies
the purification of the products. Ligand 1a was only
mentioned in a patent and is available by reaction of
four equivalents of formaldehyde with H2P(CH2)3PH2
[17]. Finally the ligands 1l,m,v were made accessible by
treatment of Cl2P(CH2)3PCl2 [18] with the Grignard
reagents RMgX. The tetrachlorodiphosphine was ob-
tained in high yields from H2P(CH2)3PH2 and
trichloromethyl chloroformate. All phosphines are col-
orless, air-sensitive and oily (1b–t,v,w) or solid (1a,u,x)
products. Whereas the hydroxyalkyl functionalized spe-
cies 1a–k are soluble in alcohols and with the exception
of 1f,g, also in water, 1l–x are soluble in all common
organic solvents. Their characterization was carried out
by means of MS, IR, and NMR spectroscopy (see
Section 4).
,
,
atom are located 0.463(2) A below and 0.744(4) A
above the plane that is formed by the atoms P1, P2, C1,
and C2. The coordination of acetonitrile to the palla-
dium atom causes longer NꢀC (N1ꢀC16=1.137(4) and
,
N2ꢀC18=1.138(4) A) and CꢀC distances (C16ꢀC17=
,
1.459(5) and C18ꢀC19=1.456(4) A) compared to non-
coordinated acetonitrile (NꢁC=1.117(7) and
CꢀCH3=1.430(7) A) [21].
,
2.3. Copolymerization of ethene with carbon monoxide
The copolymerization of olefins with carbon monox-
ide is controlled by different factors, which originate
from chemical and physical effects. Steric and electronic
properties of the catalysts and the substrate molecules,
but also the solubility of the starting materials in the
corresponding solvent, and their transport over the
gas–liquid boundary have to be taken into account.
The efficient synthesis of the ligands 1a–x (Scheme 1)
offers an easy access to a variety of diphospalladium(II)
complexes 2a–k and 4l–x with adjustable steric and
electronic properties. Two series of palladium com-
plexes with hydroxyalkyl and alkyl substituents at-
tached to the phosphorus donors were employed in the
copolymerization in dichloromethane with small
amounts of methanol as activator (4l–u, Fig. 2), [22]
methanol (2a–k, 4l–t, Fig. 3), and water (2a–k, Fig. 4)
as solvents. In order to eliminate the coordinating
carboxylate anions, an excess of HBF4 was added to
the catalyst solutions of 2a–k. In dichloromethane and
methanol a polymerization temperature of 60°C proved
to be optimal, while in water the highest reaction rates
were observed at 80°C. Generally an increased catalytic
Upon reaction of palladium(II) acetate with the
phosphines 1a–k in mixtures of ethanol or chloroform
with acetonitrile at room temperature (r.t.) bright yel-
low solutions of 2a–k were obtained (Scheme 1). After
removal of the solvents, these palladium(II) complexes
slowly decompose [19]. Spectroscopic data of 2a–k are
summarized in Section 4 [20].
Table 1
,
Selected interatomic distances (A) and angles (°) for 4n
Bond lengths
PdꢀN(1)
PdꢀN(2)
PdꢀP(1)
2.103(3)
2.125(3)
2.2507(8)
PdꢀP(2)
C(16)ꢀC(17)
C(18)ꢀC(19)
2.2501(8)
1.459(5)
1.456(4)
Bond angles
N(1)ꢀPdꢀN(2)
N(1)ꢀPdꢀP(1)
87.18(11)
90.05(8)
P(1)ꢀPdꢀP(2)
N(2)ꢀPdꢀP(2)
91.42(3)
90.98(8)

176
E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
Fig. 1. ORTEP plot of 4n. Thermal ellipsoids are drawn at the 20% probability level.
in this medium when the number of hydroxy function-
alities is doubled, as in 2h–k. In addition, the increased
polarity and modified steric bulk of 2h–k exert a posi-
tive impact on the catalytic activity.
The lower efficiency of the catalysts 2a–f and 2h–k
in water mainly reflects the poor solubility of ethene in
the aqueous phase and the possibility for water to
coordinate to the metal center as a ligand, which was
demonstrated recently [24]. Crucial steps in the mecha-
nism of the ethene/CO copolymerization are the easy
performance was established upon replacing the
P(CH3)2 or P(CH2OH)2 groups in 4l or 2a for longer
alkyl or hydroxyalkyl chains. In the rhodium catalyzed
hydrogenation of olefins a similar effect has been re-
ported [23]. If the catalysis is carried out in
dichloromethane the catalytic performance between
R=Me (4l) and longer alkyl chains does not change
significantly. Nevertheless, the well known complex
[(dppp)Pd(NCMe)2][BF4]2 [(dppp)Pd] [5b] is still supe-
rior to 4a–u (Fig. 2). However, if polar solvents like
methanol (Fig. 3) or water (Fig. 4) are employed
shorter alkyl groups induce less catalytic activity. With
exception of 2a, which decomposes rapidly under the
selected copolymerization conditions, and 2b being less
active than the corresponding non-functionalized com-
plex 4n, the two complex series 2a–g and 4l–t show a
similar catalytic performance in methanol. In this sol-
vent, the (dppp)Pd complex remains the most active
catalyst. However, the corresponding water soluble
(dppp-SO3)Pd complex with four meta positioned sul-
fonate groups exhibited lower catalytic activity (see
Table 2) in water. Recently reported results by Sheldon
and co-workers could not be reproduced under the
applied reaction conditions [15c]. Complex 2g, bearing
long P((CH2)8OH)2 groups is insoluble in water. How-
ever, catalysts with longer alkyl chains can be applied
Fig. 2. Catalytic activity in the ethene/CO copolymerization in
CH2Cl2. aActivity in (g polymer)×(g Pd×h)−1
.

E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
177
the nature of the olefin [8] and the catalysts influence
the microstructure of the resulting polymers [13]. With
chiral diphospalladium(II) complexes optically active
polyketones are accessible, since a-olefins are prochiral
[14]. This chapter describes the impact of the chain
length of the phosphorus attached alkyl substituents in
4l–x on the 1-alkene/CO copolymerization.
While [(dppp)Pd(NCMe)2][BF4]2 [(dppp)Pd] is gener-
ally superior to 2a–k and 4l–x in the synthesis of
ethene/CO copolymers, the latter complexes are suc-
cessful catalysts if ethene is replaced for 1-alkenes.
However, 2a–k and 4l–x are not active if styrene is
applied. The copolymerization of carbon monoxide
with propene and 1-hexene, respectively (Fig. 5, Table
4), was carried out with complexes 4l–x in
dichloromethane in the presence of small amounts of
methanol as activator [22]. For both olefins a distinct
maximum in the catalytic efficiency in dependence on
the length of the alkyl chain was found. At 25°C, the
most efficient catalyst in this sequence is 4q, which is
about five times more active than (dppp)Pd (Fig. 5,
Table 4, runs 1, 7, 25, 29, 35). Unlike the ethene/CO
copolymerization the catalytic activity decreases if the
bulky diphospalladium(II) complexes 4t,u are applied
(Fig. 5, Table 4, runs 9, 10, 38, 39). If the temperature
is raised to 60°C not only the catalytic efficiency is
clearly improved, also its optimum is shifted to com-
plexes with longer alkyl chains (4s,t, Fig. 5, runs 23,
24). A further amelioration was achieved by the intro-
duction of branched alkyl groups, although the employ-
Fig. 3. Catalytic activity in the ethene/CO copolymerization in
methanol. aActivity in (g polymer)×(g Pd×h)−1
.
i
ment of Pr functions (4v) results in a deterioration of
the catalysis. In particular at 60°C the activity is higher
by a factor of 1.5 in the case of complexes 4w,x (Table
4, runs 27, 28). iHex (4w) and neoHex (4x) substituents
represent a compromise between steric shielding and
flexibility. Steric and electronic properties of the cata-
lysts also control the molecular weights and the mi-
crostructure, (e.g. the regioregularity) of the copolymers
(Table 4). In agreement with the catalytic efficiency also
the molecular weights are increased on going from
shorter to longer alkyl chains (4l“4t). Ultra high
molecular weights of about Mw=420 000 g mol−1 are
obtained with 4s and 4t at 25°C. As expected low
molecular weights are obtained at 60°C. Generally the
molecular weights are lower with an increasing chain
length of the 1-alkene (Table 4). However, it should be
mentioned that the molecular weights of the 1-hexene
and 1-tetradecene/CO copolymers are the highest ever
reported (Mw=210 000 and 120 000 g mol−1, respec-
tively, Table 4) [8b]. DSC measurements of the polyke-
tones obtained with propene and CO show weak glass
transitions of −5 to 10°C and little distinct melting
points between 130 and 150°C. In contrast to the elastic
material obtained with the (dppp)Pd complex [2a], they
represent non-elastic thermoplastics. At 60°C brittle
products are formed. 1-Hexene and 1-tetradecene/CO
Fig. 4. Catalytic activity in the ethene/CO copolymerization in water.
aActivity in (g polymer)×(g Pd×h)−1
.
availability of empty coordination sites and the follow-
ing p/s rearrangement of the olefin [1]. The perfor-
mance of aqueous phase catalysis with the
diphospalladium(II) complexes 2a–f and 2h–k is con-
siderably improved by addition of a small amount of
methanol, ten equivalents of HBF4, an anionic surfac-
tant, and a second organic phase (Table 3). An
emulsifier (potassium dodecylsulfate) reduces the sur-
face tension of the aqueous phase and accelerates the
transport of ethene and CO over the gas–liquid
boundary [25]. Optimal results were obtained if the
above-mentioned improvements were combined.
2.4. Copolymerization of 1-alkenes with carbon
monoxide
Academic and in particular industrial research was
mainly focused on the examination of catalysts for the
copolymerization of ethene with carbon monoxide, al-
though 1-alkene/CO copolymers have different interest-
ing features. Their physical properties are dependent on

178
E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
Table 2
Selected catalytic and analytical data of ethene–CO copolymerizations a
b
c
e
f
Run Compound
Solvent
TP (°C)
tP (h)
Activity d
Solution viscosity pred (ml g−1
)
Mw (kg mol−1
)
D
g
g
g
g
g
g
g
1
2
3
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
(dppp)Pd
4l
4p
4q
4r
4s
4u
(dppp)Pd
4l
4n
4p
4q
4r
2a
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
80
80
80
80
80
80
80
80
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2904
1945
2601
2581
2379
2577
2481
1899
913
1479
1770
1608
1585
80
n.d.
n.d.
86.64
82.34
113.96
n.d.
119.01
n.d.
n.d.
316.92
314.41
554.73
n.d.
n.d.
n.d.
n.d.
342.24
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
78.79
n.d.
261.1
81.9
n.d.
n.d.
n.d.
119.6
92.8
1013.4
251.5
n.d.
2.5
4.5
n.d.
n.d.
n.d.
2.5
3.6
2.5
1.6
n.d.
n.d.
n.d.
4.8
4.5
3.2
3.0
1.4
1.8
3.3
3.9
2.9
4.1
3.8
2.7
5.1
3.1
n.d.
n.d.
241.2
275.2
171.5
263.7
246.5
316.0
10.1
MeOH h
MeOH h
MeOH h
MeOH h
MeOH h
H2O h
2b
2c
2e
2f
112
1366
1648
1400
365
326
431
487
493
640
627
(dppp-SO3)Pd
2b
2c
2d
2f
2h
2i
H2O h
94.6
89.8
H2O h
H2O h
108.8
82.9
128.5
159.5
106.2
H2O h
H2O h
H2O h
2j
H2O h
545
n.d.
a The copolymerizations were carried out with 10−5 mol of the palladium(II) complexes 2a–k and 4l–m in 30 ml of the corresponding solvent
with 30 bar of ethene and carbon monoxide each.
b Polymerization temperature.
c Polymerization time.
d Activity in g polymer×(g Pd×h)−1
.
e Molecular weight determined by GPC.
f D=Mw/Mn.
g 30 ml of CH2Cl2 and 2 ml of MeOH.
h With 10−4 mol of HBF4.
Table 3
Aqueous phase copolymerization of ethene with CO. Effect of the emulsifier, methanol and toluene on the catalytic activity a
d
e
Run
Water (ml)
Methanol (ml)
Emulsifier b (mg)
Toluene (ml)
Activity c
Mw (kg mol−1
)
D
1
2
3
4
5
6
30
30
30
30
30
30
0
2
0
0
0
2
0
0
300
0
300
300
0
0
0
20
20
20
567
883
776
1024
1564
1672
43.8
44.0
40.6
51.3
56.3
46.5
2.4
5.5
2.1
2.4
2.6
2.2
a Copolymerization with 2e and ten equivalents of HBF4 at 80°C, 30 bar of ethene and CO each, for 2 h.
b Potassium dodecylsulfate.
c Activity in g polymer×(g Pd×h)−1
.
d Molecular weight determined by GPC.
e D=Mw/Mn.
copolymers show no glass transitions, but the material
obtained with 1-tetradecene as comonomer melts
sharply at 42°C. Inverse gated 13C-NMR spectra were
employed to determine the head-to-tail ratio of the
1-alkene/CO building blocks in the polyketones [13]. In
the case of propene as a comonomer the head-to-tail
ratio is steadily improved in the sequence 4l to 4t. The
polyketones produced by the complexes 4p and 4t are
almost perfectly regioregular (Table 4, runs 2–6, 10,
15–20, 24). If 1-hexene and 1-tetradecene are used as

E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
179
to considerable degradation and cross-linking at tem-
peratures above their melting point of 260–270°C, CO/
ethene–copolymers exhibit a very narrow window of
melt processability [27]. The statistical incorporation of
minor amounts of 1-alkenes (e.g. propene, 1-butene,
1-hexene) as termonomers into the CO/ethene copoly-
mer structure leads to easier processable polyketones
with lowered crystallinity and melting points. Consider-
able work has already been done in the area of CO/
ethene/1-alkene terpolymers and Shell was the first
company to commercialize a CO/ethene/propene ter-
polymer with a CO/propene content of 5–8 mol%
under the trade name Carilon® [1a,8b,14c,26,28].
comonomers even the catalysts with shorter alkyl
chains lead to polyketones with a high regioregularity
(Table 4, runs 30–34, 40).
It is remarkable that the water-soluble complexes
2a–k, in particular those which are provided with the
described long-chain alkyldiphosphine ligands, are also
active in the copolymerization of 1-alkenes with carbon
monoxide. At 60°C an unexpected productive biphasic
catalytic system consisting of 1-hexene, an aqueous
solution of 2d, ten equivalents of HBF4, 2 ml of
methanol, and potassium dodecyl sulfate, was found
(Table 5, run 2). If the anionic surfactant is replaced for
neutral (polyethylene glycol) or cationic (benzalkonium
tetrafluoroborate) emulsifiers the catalytic efficiency
strongly decreases (Table 5, runs 3, 4). The anionic
emulsifier affords the best results since it is able to
stabilize the positively charged complex at the interface.
However, stable emulsions are formed, which prevent
the re-use of the aqueous phase. Compared to 1-hexene
and 1-decene the catalytic activity in the case of
propene as monomer remains low even if toluene is
employed as a second phase (see Table 5). This obser-
vation is attributed to the poor solubility of propene at
60°C (Table 5, runs 2, 6, 7). Furthermore, under bipha-
sic conditions functionalized olefins like 6-hexenol and
10-undecenol are also successfully copolymerized with
CO (Table 5, runs 8, 9).
The remarkable preference of the diphospalladi-
um(II) complexes 2a–k and 4l–x compared to
(dppp)Pd for a-olefins as comonomers can be proved if
these catalysts are employed in terpolymerizations of
ethene, 1-alkenes and carbon monoxide.
Semicrystalline polyketones form a new class of poly-
meric materials with properties in the range of typical
engineering thermoplastics such as polyamides,
polyesters or polyacetals [1a,26]. Co- and terpolymers
of CO, ethene, and 1-alkenes fall within this class. Due
As part of these studies on the performance of the
water-soluble diphospalladium(II) complexes 2a–k as
catalysts in polyketone synthesis, aqueous phase ter-
polymerization reactions of CO and ethene with
propene, 1-hexene, and N-vinyl formamide were inves-
tigated. In all cases, the polyketones precipitated from
the reaction solutions as bright white powders and were
completely insoluble in common organic solvents.
In aqueous phase terpolymerization reactions of CO,
ethene and 1-hexene, the highest catalyst activities were
achieved with the hydroxyhexyl substituted diphospal-
ladium(II) complex 2e. Running the polymerization at
a pressure of 80 bar and a temperature of 90°C with a
tenfold excess of para-toluenesulfonic acid (relative to
the catalyst) and a minor amount of methanol as
activators proved optimal in regard of catalytic activity
and molecular weights of the terpolymers. Further-
more, by adding an anionic emulsifier such as sodium
dodecyl sulfate or a mixture of sodium alkylsulfonates
(K30®), we observed a significant increase in catalyst
activity and found the terpolymers to be of higher
molecular weight and 1-hexene contents (see Table 6).
CO/ethene/1-hexene-terpolymers of exceptionally high
1-hexene incorporation resulted from the use of 2f as
catalyst. Generally, the polydispersities Mw/Mn were in
the range of 3.7–5.5. However, in some cases (Table 6,
runs 6, 8, and 9) the molecular weight distribution
(MWD) curves showed bimodal distributions with a
small number of polymer chains at very high molecular
weights of \5×105 g mol−1. This was strongly pro-
nounced in the case of the catalysts 2f and 2h. The
broad molecular weight distributions seem to indicate
that there is more than one type of catalytically active
catalyst species or more than one initiation–termina-
tion mechanism operating. As was the case for CO/
olefin copolymerizations, catalyst activities for
(dppp-SO3)Pd (both in the form of the acetate complex
and the combination of the sulfonated dppp-ligand and
[Pd(CH3CN)2((Tosylate)2], (dppp-SO3)PdTos2) were sig-
nificantly lower in CO/ethene/1-alkene terpolymeriza-
tions. Furthermore, these catalysts led to terpolymers of
lower molecular weight and 1-hexene incorporation
when compared to the catalysts stated above.
Fig. 5. Catalytic activity in the 1-alkene/CO copolymerization in
CH2Cl2. aActivity in (g polymer)×(g Pd×h)−1
polymerization. cPropene/CO copolymerization.
.
b1-Hexene/CO

180
E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
Table 4
Copolymerization of carbon monoxide with 1-alkenes in dichloromethane
a
b
c
f
g
Run
Compound
Olefin
TP (°C)
tP (h)
Activity d
TON e (cycles)
Mw (kg mol−1
)
D
H–T h (%)
1
2
3
4
5
6
7
8
(dppp)Pd
4l
4m
4n
4o
4p
4q
4r
4s
4t
4u
4v
4w
4x
(dppp)Pd
4l
4m
4n
4o
4p
4q
4r
4s
4t
4u
4v
4w
4x
(dppp)Pd
4l
4m
4n
4o
4p
4q
4r
4s
4t
4u
4q
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
Propene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Tetradecene
25
25
25
25
25
25
25
25
25
25
25
25
25
25
60
60
60
60
60
60
60
60
60
60
60
60
60
60
25
25
25
25
25
25
25
25
25
25
25
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
4
4
4
4
4
4
4
4
4
45
22
1642
803
108.2
156.2
135.0
213.6
227.0
303.0
235.6
321.1
414.9
418.0
330.4
38.7
318.4
273.9
8.5
22.7
24.8
23.0
23.5
30.4
34.5
44.7
32.4
30.8
26.7
3.4
2.6
1.8
2.8
2.2
2.8
2.3
1.9
1.5
1.9
2.2
2.4
1.7
1.8
4.1
3.7
2.3
3.8
2.7
3.6
3.3
3.5
3.1
3.2
2.9
3.5
3.1
3.7
4.9
4.0
3.7
2.7
2.6
3.5
2.9
3.5
3.8
3.1
2.6
2.9
66
75
95
107
125
160
181
199
173
159
132
101
65
247
180
476
125
441
551
689
785
872
890
916
959
641
189
1300
1421
32
3904
4561
5838
6604
7261
6312
5801
4816
3685
2371
9012
6568
2896
761
2683
3353
4192
4776
5306
5415
5573
5835
3900
1150
7910
8646
486
98
ꢀ100
ꢀ100
n.d. i
n.d.
n.d.
ꢀ100
n.d.
n.d.
n.d.
n.d.
64
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
75
92
97
ꢀ100
ꢀ100
n.d.
n.d.
n.d.
ꢀ100
n.d.
n.d.
n.d.
n.d.
75
4
4
4
4
5.4
38.0
42.5
33.6
4
16
16
16
16
16
16
16
16
16
16
16
48
23
85
87
96
101
124
116
111
97
350
85.0
88
1292
1323
1459
1535
1885
1763
1687
1475
1444
1642
124.2
112.2
209.9
155.8
162.8
192.8
142.5
111.8
169.4
118.9
ꢀ100
ꢀ100
ꢀ100
ꢀ100
n.d.
n.d.
n.d.
n.d.
n.d.
ꢀ100
95
72
a Propene: 100 ml of CH2Cl2, 2 ml of MeOH, saturated with propene at r.t., 51 bar of CO; 1-hexene (1-tetradecene): 70 ml of CH2Cl2, 30 ml
of 1-hexene (1-tetradecene), 2 ml of MeOH, 60 bar of CO.
b Polymerization temperature.
c Polymerization time.
d Activity in g polymer×(g Pd×h)−1
.
e Turnover number.
f Molecular weight determined by GPC.
g D=Mw/Mn.
h H–T: content of head-to-tail connected units.
i n.d.=not determined.
very small particles (diameters of 10–30 mm) whereas
the particle sizes of the materials made accessible in
runs 4 and 6 were in the range of 50–120 mm. In
conclusion, the ‘emulsion process’ led to larger and
more compact polymer particles which is a very
profitable property for further processing.
Interestingly, performing the polyketone synthesis in
the presence of an anionic emulsifier also led to poly-
mer powder samples of increased bulk densities and
mean particle sizes. For example, the polyketones ob-
tained in runs 2, 4, and 6 (Table 6) had bulk densities
of 0.073, 0.164, and 0.200 g ml−1, respectively, and
scanning electron microscopy studies revealed that the
polyketone from run 2 was predominantly made up of
The addition of the anionic emulsifier leads to a
stable emulsion of the 1-alkene in water by finely

E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
181
Table 5
Aqueous biphasic copolymerization of carbon monoxide with 1-alkenes
a
c
d
g
h
Run
Olefin
Surfactant b
TP (°C)
tP (h)
Activity e
TON f (cycles)
Mw (kg mol−1
)
D
H–T i (%)
1
2
3
4
5
6
7
8
9
1-Hexene
1-Hexene
1-Hexene
1-Hexene
1-Hexene
Propene j
1-Decene
5-Hexenol
10-Undecenol
–
60
60
25
60
60
60
60
60
60
4
4
4
4
4
4
4
8
8
45
1746
193
53
170
6636
733
201
182
5959
2012
15564
5547
15.2
24.1
3.7
19.0
25.0
44.0
12.1
29.5
14.3
72.5
21
A
A
N
C
A
A
A
A
ꢀ100
2.5
3.3
3.1
5.5
3.6
4.9
4.4
48
980
794
2340
1186
ꢀ100
ꢀ100
a The copolymerizations were carried out in 30 ml of an aqueous solution of 2f (10−5 mol) with ten equivalents of HBF4, 2 ml of methanol,
20 ml of the corresponding olefin and 60 bar of CO.
b A: anionic surfactant potassium dodecylsulfate; N: neutral surfactant polyethylene glycol-600; C: cationic surfactant benzalkonium te-
trafluoroborate.
c Polymerization temperature.
d Polymerization time.
e Activity in g polymer×(g Pd×h)−1
.
f Turnover number.
g Molecular weight determined by GPC.
h D=Mw/Mn.
i H–T: content of head-to-tail connected units.
j The organic phase consists of toluene that was saturated with propene at r.t.
3. Conclusion
distributing the latter through micelle formation.
Apart from the stabilization of the positively charged
catalyst complex, this seems to lead to a higher 1-
alkene concentration at the catalytically active site. It
may well be, that the polymerization proceeds in a
manner comparable to radically induced emulsion
polymerizations, in which a water soluble initiator
starts the chain growth of the polymer in the aqueous
phase. This is then incorporated into the micelles
where the polymerization proceeds to the final
product.
The efficient synthesis of the ligands 1a–x (Scheme
1) offers an easy access to a variety of diphospalladi-
um(II) complexes 2a–k and 4l–x with tunable polar-
ity. In particular, the water-soluble hydroxyalkyl
functionalized complexes 2a–k represent an alterna-
tive to sulfonated aryldiphosphines, which are incon-
venient to synthesize. In the copolymerization of
olefins with carbon monoxide and 2a–k as catalyst
precursors, organic solvents can be replaced for the
cheap, non-toxic and non-flammable reaction medium
water. A broad spectrum of alkenes from ethene to
10-undecenol is successfully copolymerized with CO
and the above-mentioned water-soluble complexes
2a–k. In particular with higher olefins, a remarkable
catalytic activity was detected under biphasic poly-
merization conditions with an anionic surfactant. The
high a-olefin chemoselectivity of the alkyldiphospalla-
dium(II) complexes 2a–k and 4l–x compared to
(dppp)Pd was demonstrated in terpolymerization reac-
tions of a-olefins with ethene and carbon monoxide.
This feature facilitates the regulation of the physical
properties (e.g. the melting point) of these polyke-
tones. In dependence on the chain-length of the phos-
phorus attached substituents in 4l–x (i) the
regioregularity and (ii) molecular weights of the
propene/CO copolymers are controlled and (iii) an
optimum in the catalytic activity is formed, (iv) that
is shifted to longer alkyl chains when the polymeriza-
Under similar reaction conditions, the use of
propene and the water-soluble N-vinyl formamide as
termonomers and complex 2e as catalyst, led to the
corresponding terpolymers with high incorporation of
the 1-alkenes. For example, terpolymers with CO/1-
alkene contents of 9.7 mol% CO/propene and 6.2
mol% CO/N-vinyl formamide were formed at catalyst
activities of 1.16 and 0.54 kg (polyketone) g−1 (Pd)
h
−1, respectively.
The melting points of a number of terpolymer sam-
ples were measured by DSC. Melting points Tm were
detected at 241, 200, and 225°C for samples of CO/
ethene/1-hexene, CO/ethene/propene, and CO/ethene/
N-vinyl formamide terpolymers with 3.3, 9.7, and 6.2
mol% of CO/1-alkene units, respectively. This docu-
ments the considerable lowering of the melting point
of CO/ethene copolymers by the incorporation of 1-
alkenes.

Table 6
Results of ethene/CO copolymerizations and ethene/1-hexene/CO terpolymerizations in water
a
e
f
Run
Catalyst
Emulsifier b
type (g)
Termonomer
1-hexene (g)
Activity c
1-Hexene incorp. d
(mol% 1-hexene/CO)
Mw
D
Solution viscosity pred
(kg mol−1
)
(ml g−1
)
/
1
2
3
4
5
6
7
8
2e
2e
2e
2e
2e
2e
2c
2f
–
–
–
20
–
20
–
20
20
20
20
20
20
20
1420
1230
1710
2320
1530
2560
1940
1980
1500
320
–
2.3
–
3.3
–
3.2
3.9
5.3
2.6
0.6
0.9
0.4
51.0
50.0
99.0
91.0
97.0
96.0
n.d. f
148.0
163.0
45.8
n.d.
4.5
4.4
4.5
3.7
4.5
5.5
63
60
(A), 0.30
(A), 0.30
(D), 0.75
(D), 0.75
(D), 0.75
(D), 0.75
(D), 0.75
(D), 0.75
–
127
112
124
120
65
161
174
60
n.d.
10.5 (bimodal)
10.0 (bimodal)
2.8
n.d.
n.d.
9
2h
10
11
12
(dppp-SO3)Pd
(dppp-SO3)PdTos2
(dppp-SO3)PdTos2
600
980
59
56
(D), 0.75
n.d.
a All polymerizations were run in 100 ml of degassed water with 0.02 mmol of the corresponding catalyst in a 300 ml autoclave at 90°C with 40 bar of CO and ethene each for a reaction time
of 3 h. In all cases 0.2 mmol of para-toluene sulfonic acid and 3 ml of methanol were used as activators.
602
(
b The emulsifier (A) was sodium dodecylsulfate, emulsifier (D) was K30® (mixture of sodium alkylsulfonates, 40 wt.% in water, product of Bayer AG).
c Activity in g polymer×(g Pd×h)−1
.
d Determined by NMR spectroscopy.
173
e Determined by gel-permeation chromatography (GPC) against PMMA standard samples.
f n.d.=not determined.
187

E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
183
tion temperature is raised. The system operates as if a
catalytic pocket is formed by the ligand framework, in
which the substrate molecules nicely fit.
T), respectively (TMS, carbonyl resonance of glycine
(l=176.03) as the second standard); 31P, 121.49 MHz
(85% H3PO4, NH4H2PO4 (l=0.8) as second standard).
Chemical shifts are reported in ppm. Mass spectra were
acquired on a Finnigan MAT 711A modified by AMD
‘Meß- und Datensysteme’ (8 kV, 303 K) and reported as
mass/charge (m/z). IR spectra were recorded on a
Bruker IFS 48 FTIR spectrometer. DSC measurements
were performed on a Mettler-Toledo TA8000 or on a
DuPont 2000 instrument (20 K min−1). Glass transi-
tions refer to the onset temperatures of the second
heating cycle. Molecular weights were determined by
means of gel permeation chromatography (GPC), using
a set-up consisting of a Perkin–Elmer Series 10 HPLC
pump, a Perkin–Elmer LC 90 UV detector, a PSS SDV
linear XL column (eluent CHCl3) with a pore size of 10
mm or a PSS PFG linear XL column (eluent 1,1,1,3,3,3-
hexafluoroisopropanol containing 0.05 weight percent-
age of potassium trifluoroacetate) with a pore size of 7
mm. The molecular weights of the terpolymers were
measured on Shodex hexafluoroisopropanol columns.
All molecular weights refer to narrow distributed poly-
methyl methacrylate standards. Reduced solution vis-
cosities pred were measured from 0.5 weight percentage
solutions of the polyketones in a mixture of ortho-
dichlorobenzene and phenol (1:1 mixture) at 25°C in a
Ubbelohde capillary viscosimeter (Schott, Typ 53720/
II). For scanning electron microscopy studies, powder
samples were attached to a copper foil, treated with
carbon vapor and sputtered with gold. The samples were
then measured on a Hitachi S4000FE instrument at 15
kV with a SE detector. Catalyst activities are given as
mean values calculated from the polymer yield (g), the
reaction time (h) and the amount of catalyst used (g
[Pd]).
4. Experimental
4.1. General comments
All experiments were carried out under an atmosphere
of argon, if not stated otherwise. Dichloromethane was
distilled from calcium hydride, diethyl ether, THF, and
toluene from sodium–benzophenone, n-hexane from
LiAlH4, methanol from magnesium, and acetonitrile
from P4O10. PdCl2 was a gift from Degussa AG. 1-
Butene, 1-pentene, 1-hexene, 1-heptene, 1-octene, 1-
decene,
1-tetradecene,
4-methyl-1-pentene,
and
3,3-dimethyl-1-butene, methylmagnesium chloride (3 M
in THF), and ethylmagnesium bromide (1 M in THF)
were purchased from Aldrich. The a,v-hydroxyalkenes
used as starting materials are accessible by reacting
sodium diethylmalonate with the appropriate alkenyl
tosylate followed by hydrolysis and decarboxylation
steps [29]. The emulsifiers sodium and potassium dode-
cyl sulfate, polyethylene glycol-600 and benzalkonium
chloride were purchased from Fluka. The emulsifier
K30® was purchased from Bayer AG. Benzalkonium
tetrafluoroborate was obtained from benzalkonium
chloride with NaBF4 in acetone. The diprimary phos-
phine 1,3-diphosphinopropane, [16] dppp, [30] dppp-
SO3, [15c] and 1a [17] were synthesized according to
literature methods. Propene, ethene, and carbon monox-
ide were gifts from BASF Aktiengesellschaft. Elemental
analyses were carried out on a Carlo Erba analyzer,
model 1106 and Elementar, model Vario EL; Cl analyses
were carried out according to Dirschel and Erne [31] and
1
Scho¨niger [32]. The high resolution H-, 13C{1H}-, and
4.2. General procedure for the preparation of the diphos-
phines 1b–k,n–u,w,x
31P{1H}-NMR spectra were recorded on a Bruker DRX
250 spectrometer at 296 K. Frequencies and standards
are as follows: 13C{1H}-NMR 62.90 MHz, 31P{1H}-
NMR 101.25 MHz. Chemical shifts in the 1H, and
13C{1H} spectra were measured relative to partially
deuterated solvent peaks and to deuterated solvent
peaks, respectively, which are reported relative to TMS.
31P{1H} spectra were measured relative to 85% H3PO4.
In a quartz Schlenk tube 1,3-diphosphinopropane and
a 10% excess of the corresponding olefin were magneti-
cally stirred and the mixture was irradiated with the
ultraviolet light of a mercury high pressure lamp at
20°C. After 10 h, excess olefin was removed under
reduced pressure. Further purification was not neces-
sary. Analytical data of selected diphosphine ligands are
presented in Sections 4.3, 4.4 and 4.5. Full data sets for
all ligands are summarized in Refs. [20,33].
1
The terpolymer microstructures were analyzed by H
(250 MHz)- and 13C (62.8 MHz)-NMR spectroscopy on
a
Bruker instrument DPX 250 in hexafluoroiso-
propanol/C6D6 as solvent with TMS as external stan-
dard. The CP/MAS solid-state NMR spectra were
recorded on a Bruker DSX 200 and a Bruker ASX 300
multinuclear spectrometer equipped with wide bore
magnets (field strengths 4.7 and 7.05 T). Magic angle
spinning was applied up to 10 kHz (4 mm ZrO2 rotors)
and 3–3.5 kHz (7 mm ZrO2 rotors). Frequencies and
standards: 13C, 50.288 MHz (4.7 T), 75.432 MHz (7.05
Caution: The phosphines 1n–p are pyrophoric in
contact with cellulose.
4.3. 1,3-Bis[di(6-hydroxyhexyl)phosphino]propane (1e)
1,3-Diphosphinopropane (600 mg, 5.55 mmol) and
6-hexenol (2.45 g, 24.4 mmol) were reacted to give 1e.
Yield: 2.79 g (99%). MS (FD, 30°C): m/z 509.6 [M++

184
E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
H]. Anal. Calc. for C27H58O4P2 (508.27): C, 63.75; H,
100 ml of a saturated solution of NH4Cl in water. The
aqueous layer was extracted with diethyl ether (3×30
ml). Subsequently the organic phase was dried with
anhydrous Na2SO4 and the solvent was removed in
vacuo. Finally the crude product was distilled under
reduced pressure. Analytical data of a selected diphos-
phine ligand are presented in Section 4.7. Full data sets
for all ligands are summarized in Ref. [20].
11.49. Found: C, 63.67; H, 11.39%. IR (film): w(OH)
3314 cm−1 31P{1H}-NMR (CDCl3): l −30.6 (s).
.
13C{1H}-NMR (CDCl3): l 61.6 (s, CH2OH), 32.0 (s,
CH2CH2OH), 30.7 (d, 1J(PC)=10.8 Hz, C3), 28.4
2
(N=22.2 Hz1 [34], C2), 26.4 (d, J(PC)=10.8 Hz, C4),
3
25.3 (d, J(PC)=12.1 Hz, C5), 25.0 (s, CH2(CH2)2OH),
2
21.8 (t, J(PC)=13.0 Hz, C1) (for atom labeling see
Scheme 1).
4.7. 1,3-Bis(dimethylphosphino)propane (1l) [35]
4.4. 1,3-Bis[di(n-heptyl)phosphino]propane (1r)
1,3-Bis(dichlorophosphino)propane (7.52 g, 30.6
mmol) and CH3MgCl (50 ml of a 3 M solution in THF,
150 mmol) were reacted to give 1l. Yield: 4.57 g (91%),
b.p. 79–81°C (13 mbar). MS (EI, 70 eV): m/z 164.0
[M+]. 31P{1H}-NMR (CDCl3): l −51.3 (s). 13C{1H}-
NMR (CDCl3): l 33.7 (N=20.6 Hz1 [34], C2), 22.0 (t,
1,3-Diphosphinopropane (907 mg, 8.40 mmol) and
1-heptene (3.62 g, 36.9 mmol) were reacted to give 1r.
Yield: 4.08 g (97%). MS (EI, 70 eV): m/z 500.3 [M+].
Anal. Calc. for C31H66P2 (500.31): C, 74.35; H, 13.28.
Found: C, 73.99; H, 13.00%. 31P{1H}-NMR (CDCl3): l
−30.4 (s). 13C{1H}-NMR (CDCl3): l 31.8 (s, CH2),
31.5 (d, 1J(PC)=10.8 Hz, C3), 29.2–28.7 (m, C2+
1
2J(PC)=13.8 Hz, C1), 13.9 (d, J(PC)=12.8 Hz, CH3)
(for atom labeling see Scheme 1).
2
3
CH2), 27.2 (d, J(PC)=11.5 Hz, C4), 25.9 (d, J(PC)=
12.8 Hz, C5), 22.7–22.3 (m, C1+CH2), 14.1 (s, CH3)
(for atom labeling see Scheme 1).
4.8. General procedure for the preparation of the
diacetatodiphosphinepalladium(II) complexes 2a–k
4.5. 1,3-Bis(dichlorophosphino)propane [18]
To a stirred solution of palladium(II) acetate in 150
ml of acetonitrile a solution of 1a–k in 150 ml of
ethanol was added at 25°C. Subsequently the solvents
were removed under reduced pressure. Finally the
diphospalladium(II) complexes 2a–k were precipitated
from ethanol with n-hexane and dried in vacuo. Ana-
lytical data of a selected complex are presented in
Section 4.9. Full data sets for all complexes are summa-
rized in Ref. [33].
To a cooled solution (−30°C) of H2P(CH2)3PH2
(10.56 g, 98 mmol) in 150 ml of dichloromethane
trichloromethyl chloroformate (52.53 g, 266 mmol) was
added dropwise. The mixture was warmed to 25°C and
stirred for 6 h. The solvent and phosgene residues were
distilled into a cooled flask (−196°C). The crude,
brown product was distilled under reduced pressure
(86–88°C,
3
mbar) to give 20.5
g
(85%) of
Cl2P(CH2)3PCl2. MS (EI, 70 eV): m/z 245.8 [M+].
1H-NMR (CDCl3): l 2.51–2.37 (4H, m, Cl2PCH2),
2.24–2.05 (2H, m, Cl2PCH2CH2). 31P{1H}-NMR
(CDCl3): l 191.2 (s). 13C{1H}-NMR (CDCl3): l 43.3–
4.9. Diacetato{1,3-bis[di(7-hydroxyheptyl)phosphino]-
propane}palladium(II) (2f)
1f (400 mg, 0.709 mmol) was reacted with palla-
dium(II) acetate (159 mg, 0.709 mmol) to give 2f. Yield:
475 mg (85%). MS (FAB, 50°C): m/z 729.6 [M+−
OAc]. Anal. Calc. for C35H72O8P2Pd (788.83): C, 53.26;
H, 9.19. Found: C, 53.50; H, 9.30%. 31P{1H}-NMR
(CD3OD): l 19.6 (s). 13C{1H}-NMR (CD3OD): l 178.8
(s, CH3CO2), 62.9 (s, CH2OH), 33.7 (s, CH2CH2OH),
32.4 (N=15.5 Hz1 [34], C3), 30.2 (s, CH2), 27.0–26.8
(m, C2+2 CH2), 26.0 (s, CH2), 23.7 (s, CH3CO2) (for
atom labeling see Scheme 1). IR (KBr): w(OH) 3446,
was(CH3COO−) 1586, ws(CH3COO−) 1406, w(CꢀOH)
42.4 (N=53.9 Hz1 [34], Cl2PꢀCH2), 17.2 (t, J(PC)=
2
14.1 Hz, Cl2PCH2CH2) (for atom labeling see Scheme
1).
Caution: The evolving gases CO and HCl should be
bubbled through a trap containing sodium hydroxide
solution and then released into the hood by means of a
waste-gas line. Trichloromethyl chloroformate is ex-
tremely toxic and corrosive, a gas mask should be used
for safety. All contaminated glassware has to be washed
with an aqueous NaOH solution.
1069 cm−1
.
4.6. General procedure for the preparation of the
phosphine ligands 1l,m,7
4.10. General procedure for the preparation of the
dichlorodiphosphinepalladium(II) complexes 3l–x
Stirred solutions of Cl2P(CH2)3PCl2 and the corre-
sponding Grignard reagent in each 50 ml of THF were
reacted for 3 h at r.t. The mixture was hydrolyzed with
Stirred solutions of the corresponding phosphine 1l–
x and (PhCN)2PdCl2 in each 20 ml of dichloromethane
were reacted for 2 h at r.t. Subsequently the solvent was
removed in vacuo. Finally the complexes 2l–x were
1
1 AXX% pattern, N = ꢀ J(PC) + 3J(PC) ꢀ

E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
185
washed with cold n-hexane (−30°C, 3×20 ml) and
recrystallized from DMSO/acetonitrile (3l), acetonitrile/
toluene (3m), toluene (3n,x), toluene/n-hexane (3m–
t,v,w) or n-hexane (3u) to give colorless, crystalline
complexes. Analytical data of a selected complex are
presented in Section 4.11. Full data sets for all com-
plexes are summarized in Ref. [20].
4.14. Copolymerization of ethene with carbon monoxide
A solution of the corresponding diphospalladium(II)
complex 2a–k (0.01 mmol, 0.1 mmol of HBF4, 30 ml of
water or methanol), was placed in a magnetically
stirred steel autoclave, which was charged with 30 bar
of ethene and carbon monoxide each. The complex
solutions of 4l–x (30 ml of CH2Cl2, 2 ml of methanol)
were treated in the same way. The reaction temperature
was adjusted by a thermostat. Finally the gases were
released automatically after 2 h. The puffy, colorless
copolymer was collected by filtration and dried in
vacuo. A reproducibility of about 10% was found.
4.11. Dichloro{1,3-bis[di(n-heptyl)phosphino]propane}-
palladium(II) (3r)
1r (560 mg, 1.120 mmol) was reacted with
(PhCN)2PdCl2 (391 mg, 0.993 mmol) to give 3r. Yield:
578 mg (86%), m.p. 158°C. MS (FD, 30°C): m/z 642.6
[M+−Cl]. Anal. Calc. for C31H66Cl2P2Pd (677.69): C,
54.91; H, 9.81; Cl, 10.46; Pd, 15.71. Found: C, 54.62;
H, 9.74; Cl, 10.32; Pd, 16.03%. 31P{1H}-NMR (CDCl3):
l 19.2 (s). 13C{1H}-NMR (CDCl3): l 31.6 (s, CH2),
31.1 (N=14.9 Hz1 [34], C3), 28.9 (s, CH2), 28.2 (N=
34.9 Hz1 [34], C2), 25.4 (s, C5), 22.5 (s, CH2), 21.3–20.7
(m, C1+C4), 14.0 (s, CH3) (for atom labeling see
Scheme 1).
4.15. Copolymerization of h-olefins with carbon
monoxide
The copolymerization of propene, 1-hexene and 1-
tetradecene with carbon monoxide was carried out in a
250 ml mechanically stirred steel autoclave with electri-
cal heating, and air cooling. The evacuated autoclave
was charged with a solution of the corresponding dica-
tionic palladium(II) complex 4m–x (0.01 mmol) in
CH2Cl2 (70 ml), 1-hexene or 1-tetradecen (30 ml) and
CH3OH (2 ml) as activator and 60 bar of carbon
monoxide. In the case of propene the catalyst solution
(100 ml of CH2Cl2) was saturated with the olefin (20°C,
20 min) and pressurized with 51 bar of carbon monox-
ide. At the end of the reaction period the gases were
released automatically. The 1-alkene/carbon monoxide
copolymers were obtained by evaporation of the
solvent.
The biphasic experiments were carried out in the
same reactor set-up as the ethene/CO copolymeriza-
tions. The aqueous catalyst solution of 2a–k (0.01
mmol of 2a–k, 0.1 mmol of HBF4, 30 ml of water, 0.3
g of the corresponding emulsifier) was reacted with 20
ml of the 1-alkene (1-hexene, 1-decene, 6-hexenol, 10-
undecenol) and 60 bar of CO at 60°C. In the case of
propene 20 ml of toluene were saturated at 20°C and
applied as organic phase. The polyketones were col-
lected by filtration and washed with methanol.
4.12. General procedure for the preparation of the
dicationic bis(acetonitrile)-diphospalladium(II)
complexes 4l–x
To a solution of the corresponding dichloropalla-
dium(II) complex 3l–x in 10 ml of CH2Cl2 a solution of
AgBF4 in 10 ml of CH3CN was added and the mixture
was stirred for 1 h at r.t. Subsequently the suspension
was centrifuged, decanted and the solvent was removed
in vacuo. The crude product was dissolved in CH2Cl2
and centrifuged again to remove traces of AgCl. Finally
the solvent was removed under reduced pressure to give
the dicationic complexes 4l–x. Analytical data of a
selected complex are presented in Section 4.13. Full
data sets for all complexes are summarized in Ref. [20].
4.13. Bis(acetonitrile){1,3-bis[di(n-heptyl)phosphino]-
propane}palladium(II) bis(tetrafluoroborate) (4r)
4.16. CO/ethene/1-alkene terpolymerization reactions
3r (370 mg, 0.546 mmol) was reacted with AgBF4
(211 mg, 1.09 mmol) to give 4r. Yield: 457 mg (97%),
m.p. 124°C (dec.). MS (FAB, 50°C): m/z 776.5 [M+
−BF4]. Anal. Calc. for C35H72B2F8N2P2Pd (862.43): C,
48.72; H, 8.41; N, 3.25; Pd, 12.35. Found: C, 48.40; H,
8.63; N, 2.96; Pd, 12.88%. IR (KBr): w(CN) 2323, 2294,
Terpolymerization reactions were performed in a 300
ml autoclave, equipped with a mechanical stirrer, by the
following procedure: 100 ml of degassed water, 20 g
(0.24 mol) of 1-hexene or 35 g (0.49 mol) of N-vinyl
formamide and the emulsifier (see Table 6) were stirred
under N2 to form a stable emulsion. The emulsion was
then transferred to the autoclave, under N2. A solution
of the catalyst complex (0.02 mmol) and cocatalyst
para-toluene sulfonic acid (0.2 mmol) in 3 ml of
methanol was added, the autoclave sealed and purged
with N2 and CO/ethene gas (1:1 mixture) twice. Then
w(BF4) 1063 cm−1 31P{1H}-NMR (CD3CN): l 24.9 (s).
.
13C{1H}-NMR (CD3CN): l 32.4 (s, CH2), 31.7 (N=
29.2 Hz1 [34], C3), 29.6 (s, CH2), 27.3 (N=40.6 Hz1
[34], C2), 25.6 (s, C4), 23.4 (s, CH2), 19.5–18.7 (m,
C1+C5), 14.5 (s, CH3) (for atom labeling see Scheme
1).

186
E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
4.17. X-ray structural determination of 4n
the pressure and temperature were set to 80 bar and
90°C using CO/ethene gas (1:1 mixture). After 3 h at
constant temperature and pressure, the autoclave was
vented and the reaction mixture cooled to r.t. The
polymer suspension was filtered, washed with
methanol and acetone and dried overnight at 80°C in
vacuum.
Single crystals of 4n were obtained by slow diffusion
of toluene into a concentrated solution of 4n in acetoni-
trile. The crystal was mounted on a glass fiber and
transferred to a P4 Siemens diffractometer, using
graphite-monochromated Mo–Ka radiation. The lattice
constants were determined by 18 precisely centered
high-angle reflections and refined by least-squares
methods. The structure was solved by direct methods
[36] and refined by full-matrix least-squares on F2 using
SHELXTL-97 [37]. All non-hydrogen atoms were refined
anisotropically (based on F2). Hydrogen atoms were
placed in calculated positions (d(CH3)=0.98;
In case of CO/ethene/propene terpolymerization re-
actions a 3.5 l autoclave equipped with a mechanical
stirrer and a catalyst injection port, was evacuated and
filled with N2 in two cycles. Then 1000 ml of degassed
water (containing 15 g of K30®, sodium alkylsul-
fonates, 40 weight percentage in water) and 150 g
(3.57 mol) of propene were introduced to the auto-
clave. The mixture was purged with CO/ethene gas
(1:1 mixture). Then the pressure and temperature were
set to 80 bar and 90°C using CO/ethene gas (1:1
mixture). Under these conditions, the reaction was
started by the injection of the catalyst solution (0.02
mmol 2e and 0.175 mmol para-toluene sulfonic acid in
a mixture of 10 ml of water and 3 ml of methanol).
After 5 h at constant temperature and pressure, the
autoclave was vented and the reaction mixture was
cooled to r.t. The polymer suspension was filtered,
washed with methanol and acetone and dried overnight
at 80°C in vacuum.
,
d(CH2)=0.99 A). The final cell parameters and specific
data collection parameters for 4n are summarized in
Table 7.
5. Supplementary material
Crystallographic data for the structure reported in
this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publi-
cation no. CCDC-137749. Copies of the data can be
obtained free of charge on application to CCDC, 12
Union Road, Cambridge, CB2 1EZ, UK (fax: +44-
1223-336033; e-mail: deposit@ccdc.cam.ac.uk).
Table 7
Crystal data and refinement details for complex 4n
Acknowledgements
4n
M.S. acknowledges the Land Baden-Wu¨rttemberg
for the award of
a Ph.D. Fellowship (Landes-
Formula
C19H40B2F8N2P2Pd
graduiertenfo¨rderungsgesetz). Generous support of this
research by BASF Aktiengesellschaft, the Fonds der
Chemischen Industrie, Frankfurt/Main, Germany, and
by Degussa AG is gratefully acknowledged. We thank
Professor Dr P.C. Schmitt, Universita¨t Tu¨bingen and
his co-worker Heike Bauer for making available a DSC
instrument. The authors acknowledge J. Adrian, I.
Hennig, W. Heckmann and S. Lehmann of BASF
Aktiengesellschaft for analyzing the polymers by GPC,
DSC, scanning electron microscopy, and NMR
experiments.
Formula weight
Color
Crystal dimensions (mm)
Crystal system
Space group
638.49
Colorless
0.3×0.2×0.1
Monoclinic
P21/n (no. 14)
11.318(2)
12.402(7)
21.066(3)
103.710(13)
2872.6(17)
4
,
a (A)
,
b (A)
,
c (A)
i (°)
3
,
V (A )
Z
Dcalc (g cm−3
T (°C)
)
1.476
−100
F(000) (e)
1304
0.820
4.6–54.92
22 833
6550
v(Mo–Ka) (mm−1
2q Limits (°)
)
References
No. of reflections measured
No. of unique data with I]2|(I)
No. of variables
[1] (a) E. Drent, P.H.M. Budzelaar, Chem. Rev. 96 (1996) 663. (b)
A. Sen, Acc. Chem. Res. 16 (1993) 303. (c) F.C. Rix, M.
Brookhart, P.S. White, J. Chem. Am. Soc. 118 (1996) 4746. (d)
P. Margel, T. Ziegler, J. Am. Chem. Soc. 118 (1996) 4746.
[2] (a) A.S. Abu-Surrah, G. Eckert, W. Pechhold, W. Wilke, B.
Rieger, Macromol. Rapid Commun. 17 (1996) 559. (b) E.A.
Klop, B.J. Lommerts, J. Veurink, J. Aerts, R.R. van Puijen-
broek, J. Polym. Sci. Part B: Polym. Phys. 33 (1995) 315.
310
S
R1
wR2
1.958
0.0379
0.0939
a
b
a R1=SꢀꢀFoꢀ−ꢀFcꢀꢀ/SꢀFoꢀ.
b
W
R2 =ꢀS[w(F2o−F2c)2]/S[w(F2o)2]]0.5
.

E. Lindner et al. / Journal of Organometallic Chemistry 602 (2000) 173–187
187
[3] (a) A. Sen, Z. Jiang, J.T. Chen, Macromolecules 22 (1989) 2012.
(b) S. Siddhart, Z. Jiang, A. Sen, Polym. Prep. 34 (1993) 378.
[4] (a) F.Y. Xu, C.W. Jame, Macromolecules 26 (1993) 3485. (b)
M.D.E. Forbes, Macromolecules 27 (1994) 1020.
[5] (a) E. Drent (Shell) Eur. Patent Appl. 121 956 (1984) [Chem.
Abstr. 102 (1985) 46 432]. (b) E. Drent, J.A.M. van Broekhoven,
M.J. Doyle, J. Organomet. Chem. 417 (1991) 235. (c) E. Drent
(Shell) Eur. Patent Appl. 229 408 (1986) [Chem. Abstr. 108
(1988) 6617].
[6] (a) A. Sen, T.-W. Lai, J. Am. Chem. Soc. 104 (1982) 3520. (b) A.
Sen, T.-W. Lai, Organometallics 3 (1984) 866.
[7] F.Y. Xu, A.X. Zhao, J.C.W. Chien, Makromol. Chem. 194
(1993) 2579.
[8] (a) J.C.W. Chien, A.X. Zhao, F.Y. Xu, Polym. Bull. 28 (1992)
315. (b) A.S. Abu-Surrah, R. Wursche, B. Rieger, Macromol.
Chem. Phys. 198 (1997) 1197. (c) H.A. Klok, P. Eibeck, M.
Schmid, A.S. Abu-Surrah, M. Mo¨ller, B. Rieger, Macromol.
Chem. Phys. 198 (1997) 2768.
[9] S. Kacker, J.S. Kim, A. Sen, Angew. Chem. Int. Ed. Engl. 37
(1998) 1251.
[10] L.M. Wojcinski II, M.T. Boyer, A. Sen, Inorg. Chim. Acta 270
(1998) 8.
[11] (a) S. Mecking, L.K. Johnson, L. Wang, M. Brookhart, J. Am.
Chem. Soc. 120 (1998) 888. (b) B.L. Small, M. Brookhart,
A.M.A. Bennett, J. Am. Chem. Soc. 120 (1998) 4049.
[12] G.J.P. Britovsek, V.C. Gibson, B.S. Kimberley, P.J. Maddox,
S.J. McTavish, G.A. Solan, A.J.P. White, D.J. Williams, Chem.
Commun. (1998) 849.
[13] (a) A. Batistini, G. Consiglio, U. Suter, Angew. Chem. Int. Ed.
Engl. 31 (1992) 303. (b) M. Barsacchi, A. Batistini, G. Consiglio,
U.W. Suter, Macromolecules 25 (1992) 3604.
[14] (a) Z. Jiang, A. Sen, J. Am. Chem. Soc. 117 (1995) 4455. (b) M.
Sperrle, G. Consiglio, J. Am. Chem. Soc. 117 (1995) 12 130. (c)
A. Aeby, G. Consiglio, Helv. Chim. Acta 81 (1998) 35. (c) K.
Nozaki, N. Sato, Y. Tonomura, M. Yasutomi, H. Takaya, T.
Hiyama, T. Matsubara, N. Koga, J. Am. Chem. Soc. 119 (1997)
12 779.
[15] (a) B. Cornils, W. Herrmann, Aqueous-Phase Organometallic
Catalysis, Wiley-VCH, Weinheim, 1998. (b) Z. Jiang, A. Sen,
Macromolecules 27 (1994) 7215. (c) G. Verspui, G. Papado-
gianakis, R.A. Sheldon, Chem. Commun. (1998) 401. (d) C.
Bianchini, H.M. Lee, A. Meli, S. Moneti, V. Patinec, G.
Petrucci, F. Vizza, Macromolecules 32 (1999) 3859.
[16] L. Maier, Helv. Chim. Acta 49 (1966) 842.
[20] M. Schmid, Ph.D. Thesis, University of Tu¨bingen, Germany,
1999. Complexes 2l and 4m crystallize in the monoclinic space
group P21/c and orthorhombic space group Pnma, respectively.
[21] C. Nachtigal, B. Steuer, W. Preetz, Chem. Ber. 129 (1996) 1531.
Further details are available on request from Fachinformation-
szentrum Karsruhe, Gesellschaft fu¨r wissenschaftlich-technische
Information mbH, D-76344 Eggenstein-Leopoldshafen, on quot-
ing the depository number CSD-405693.
[22] A.S. Abu-Surrah, R. Wursche, B. Rieger, G. Eckert, W. Pech-
hold, Macromolecules 29 (1996) 4806.
[23] E. Renaud, R.B. Russell, S. Fortier, S.C. Brown, C.B. Baird, J.
Organomet. Chem. 419 (1991) 403.
[24] (a) P.J. Stang, D.H. Cao, G.T. Poulter, A.M. Arif, Organometal-
lics 14 (1995) 1110. (b) M. Sperrle, V. Gramlich, G. Consiglio,
Organometallics 15 (1996) 5196. (c) E. Lindner, M. Schmid, P.
Wegner, C. Nachtigal, M. Steimann, R. Fawzi, Inorg. Chim.
Acta 296 (1999) 103.
[25] Y.M. Chung, K.K. Kang, W.S. Ahn, P.K. Lim, J. Mol. Catal.
Part A: 137 (1999) 23.
[26] (a) A. Wakker, H. Krikor, Kunststoffe 86 (1996) 1318. (b) H.
Seifert, Kunststoffe 88 (1996) 1154. (c) J.G. Bonner, A.K. Pow-
ell, Polym. Mater. Sci. Eng. 74 (1997) 108. (d) A. Wakker,
Carilon™ Aliphatic Polyketones: Progress in Application Devel-
opment of this New General Purpose, Engineering Thermoplas-
tic ETP 98 World Congress, June 22–24, Zu¨rich, 1998.
[27] (a) O. Chiantore, M. Lazzari, F. Ciardelli, S. De Vito, Macro-
molecules 30 (1997) 2589. (b) G.A. Holt, J.E. Spruiell, Antec 96
(1996) 1780. (c) S. De Vito, F. Ciardelli, E. Benedetti, E.
Bramanti, Polym. Adv. Technol. 8 (1997) 53.
[28] (a) Z. Jiang, G.M. Dahlen, A. Sen, in: T.C. Chung (Ed.), New
Advances in Polyolefins, Plenum, New York, 1993, p. 47. (b) F.
Garbassi, A. Sommazzi, Polym. News 20 (1994) 201. (c) A.J.
Waddon, N.R. Karttunen, A.J. Lesser, Macromolecules 32
(1999) 423.
[29] H. Hanecka, in: J. Houben, Th. Weyl, E. Mu¨ller (Eds.), Metho-
den der Organischen Chemie, vol. 8, Thieme, Stuttgart, 1952.
[30] T.-S. Chou, J.-J. Yuang, C.H. Tsao, J. Chem. Res. (1985) 18.
[31] A. Dirschel, F. Erne, Microchim. Acta (1961) 866.
[32] (a) W. Scho¨niger, Microchim. Acta (1955) 123. (b) W. Scho¨niger,
Microchim. Acta (1956) 869.
[33] J. Wald, Ph.D. Thesis, University of Tu¨bingen, Germany, 1999.
[34] R.K. Harris, Can. J. Chem. 42 (1964) 2275.
[35] H.H. Karsch, A. Appelt, Z. Naturforsch. Teil B 38 (1983) 1399.
[36] G.M. Sheldrick, SHELXS 5.03, Program for Crystal Structure
Solution, University of Go¨ttingen, Germany, 1995; Acta Crystal-
logr. Sect. A 51 (1995) 33.
[17] E.R. Kolodny, M. Grayson, R. Myers, (American Cyanamid
Co.) FR 1487270 (1967) [Chem. Abstr. 68 (1968) 70 455].
[18] K. Sommer, Z. Anorg. Allg. Chem. 376 (1970) 37.
[19] Z. Csa´kai, R. Skoda-Fo¨ldes, L. Kolla´r, Inorg. Chim. Acta 286
(1999) 93.
[37] G.M. Sheldrick, SHELXTL V5.03, Program for Crystal Structure
Refinement, University of Go¨ttingen, Germany, 1995.
.