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

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

^aInstitut fu¨r Anorganische Chemie der Uni6ersita¨t Tu¨bingen, Auf der Morgenstelle 18, D-72076 Tu¨bingen, Germany

^bBASF 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=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)₃PH₂or 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)₂with the ligands 1a–k in ethanol–acetonitrile.

Treatment of PdCl₂(NCC₆H₅)₂with 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-SO₃catalyst the water-soluble complexes

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)

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)₂PdCl₂is subjected to a CH₂Cl₂solution 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 AgBF₄in 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)₃PH₂and

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 HBF₄was 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. ^ORTEPplot 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)₂or P(CH₂OH)₂groups 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)₂groups 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 (4l4t). Ultra high

molecular weights of about M_w=420 000 g mol⁻¹are

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 (M_w=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

T_P(°C)

t_P(h)

Activity ^d

Solution viscosity p_red(ml g⁻¹

)

M_w(kg mol⁻¹

)

D

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

MeOH

60

80

2

2904

1945

2601

2581

2379

2577

2481

1899

913

1479

1770

1608

1585

80

n.d.

86.64

82.34

113.96

n.d.

119.01

n.d.

316.92

314.41

554.73

n.d.

342.24

n.d.

78.79

n.d.

261.1

81.9

n.d.

119.6

92.8

1013.4

251.5

n.d.

2.5

4.5

n.d.

2.5

3.6

2.5

1.6

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.

241.2

275.2

171.5

263.7

246.5

316.0

10.1

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

108.8

82.9

128.5

159.5

106.2

H₂O ^h

2j

H₂O ^h

545

n.d.

^aThe copolymerizations were carried out with 10⁻⁵mol 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.

^bPolymerization temperature.

^cPolymerization time.

^dActivity in g polymer×(g Pd×h)⁻¹

.

^eMolecular weight determined by GPC.

^fD=M_w/M_n.

^g30 ml of CH₂Cl₂and 2 ml of MeOH.

^hWith 10⁻⁴mol 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

M_w(kg mol⁻¹

)

D

1

2

3

4

5

6

30

0

2

0

2

0

300

0

300

0

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

^aCopolymerization with 2e and ten equivalents of HBF₄at 80°C, 30 bar of ethene and CO each, for 2 h.

^bPotassium dodecylsulfate.

^cActivity in g polymer×(g Pd×h)⁻¹

.

^dMolecular weight determined by GPC.

^eD=M_w/M_n.

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 M_w/M_nwere 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 \5×10⁵g 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

T_P(°C)

t_P(h)

Activity ^d

TON ^e(cycles)

M_w(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

1-Hexene

1-Tetradecene

25

60

25

24

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

n.d. ⁱ

n.d.

ꢀ100

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

n.d.

ꢀ100

n.d.

75

4

5.4

38.0

42.5

33.6

4

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

n.d.

ꢀ100

95

72

^aPropene: 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.

^bPolymerization temperature.

^cPolymerization time.

^dActivity in g polymer×(g Pd×h)⁻¹

.

^eTurnover number.

^fMolecular weight determined by GPC.

^gD=M_w/M_n.

^hH–T: content of head-to-tail connected units.

ⁱn.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

T_P(°C)

t_P(h)

Activity ^e

TON ^f(cycles)

M_w(kg mol⁻¹

)

D

H–T ⁱ(%)

1

2

3

4

5

6

7

8

9

1-Hexene

Propene ^j

1-Decene

5-Hexenol

10-Undecenol

–

60

25

60

4

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

N

C

A

ꢀ100

2.5

3.3

3.1

5.5

3.6

4.9

4.4

48

980

794

2340

1186

ꢀ100

^aThe copolymerizations were carried out in 30 ml of an aqueous solution of 2f (10⁻⁵mol) with ten equivalents of HBF₄, 2 ml of methanol,

20 ml of the corresponding oleﬁn and 60 bar of CO.

^bA: anionic surfactant potassium dodecylsulfate; N: neutral surfactant polyethylene glycol-600; C: cationic surfactant benzalkonium te-

traﬂuoroborate.

^cPolymerization temperature.

^dPolymerization time.

^eActivity in g polymer×(g Pd×h)⁻¹

.

^fTurnover number.

^gMolecular weight determined by GPC.

^hD=M_w/M_n.

ⁱH–T: content of head-to-tail connected units.

^jThe 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 T_mwere

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)

M_w

D

Solution viscosity p_red

(kg mol⁻¹

)

(ml g⁻¹

)

/

1

2

3

4

5

6

7

8

2e

2c

2f

–

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

(D), 0.75

–

127

112

124

120

65

161

174

60

n.d.

10.5 (bimodal)

10.0 (bimodal)

2.8

n.d.

9

2h

10

11

12

(dppp-SO₃)Pd

(dppp-SO₃)PdTos₂

600

980

59

56

(D), 0.75

n.d.

^aAll 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

(

^bThe emulsiﬁer (A) was sodium dodecylsulfate, emulsiﬁer (D) was K30^®(mixture of sodium alkylsulfonates, 40 wt.% in water, product of Bayer AG).

^cActivity in g polymer×(g Pd×h)⁻¹

.

^dDetermined by NMR spectroscopy.

173

^eDetermined by gel-permeation chromatography (GPC) against PMMA standard samples.

^fn.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 p_redwere 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. PdCl₂was 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 NaBF₄in 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/C₆D₆as 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 ZrO₂rotors)

and 3–3.5 kHz (7 mm ZrO₂rotors). 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 Na₂SO₄and 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 31}P{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,

w_as(CH₃COO⁻) 1586, w_s(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)₃PCl₂and 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)₂PdCl₂in 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

¹AXX% 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 CH₂Cl₂a solution of

AgBF₄in 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 N₂to 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 N₂and CO/ethene gas (1:1 mixture) twice. Then

w(BF₄) 1063 cm^{−1 31}P{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–K_aradiation. 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 F²using

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 N₂in 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

D_calc(g cm⁻³

T (°C)

)

1.476

−100

F(000) (e)

1304

0.820

4.6–54.92

22 833

6550

v(Mo–K_a) (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

^aR₁=SꢀꢀF_oꢀ−ꢀF_cꢀꢀ/SꢀF_oꢀ.

b

W

R₂=ꢀS[w(F²_o−F²_c)²]/S[w(F²_o)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, ^SHELXS5.03, 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, ^SHELXTLV5.03, Program for Crystal Structure

Reﬁnement, University of Go¨ttingen, Germany, 1995.

.

Article Doi

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

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

Source and publish data:

Authors:

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

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

Products guided by the article

R&D Labs maybe for 28240-70-2

Relevant to this article

Hot Product