Ligand and solvent effects in the alternating copolymerization of carbon monoxide and olefins by palladium-diphosphine catalysis

16

Organometallics 2002, 21, 16-33

Articles

Liga n d a n d Solven t Effects in th e Alter n a tin g

Cop olym er iza tion of Ca r bon Mon oxid e a n d Olefin s by

P a lla d iu m -Dip h osp h in e Ca ta lysis

Claudio Bianchini,* Hon Man Lee, Andrea Meli, Werner Oberhauser,

Maurizio Peruzzini, and Francesco Vizza

Istituto per lo Studio della Stereochimica ed Energetica dei Composti di Coordinazione

(ISSECC)-CNR, Via J acopo Nardi 39, 50132 Firenze, Italy

Received August 9, 2001

The substitution of two hydrogen atoms by methyl groups in the 1,2 positions of 1,2-bis-

(diphenylphosphino)ethane (dppe) gives meso- and rac-2,3-bis(diphenylphosphino)butane

(meso-2,3-dppb and rac-2,3-dppb). The corresponding Pd(II) complexes Pd(OTs)₂(meso-2,3-

dppb) and Pd(OTs)₂(rac-2,3-dppb) are effective catalyst precursors for the alternating

copolymerization and terpolymerization of carbon monoxide with ethene and ethene/propene

in MeOH with productivities that are higher than those of the unsubstituted dppe catalyst

Pd(OTs)₂(dppe) even by a factor of 10 (OTs ) p-toluenesulfonate). It has been found that

the low productivity of the dppe-based catalyst in MeOH is due to the autoionization of the

precursor Pd(OAc)₂(dppe) in MeOH to give the catalytically inactive bis-chelate species [Pd-

(dppe)₂](OAc)₂and palladium acetate. In an attempt to evaluate and rationalize the effective

ligand control on the intrinsic catalytic activity, the methyl complexes [Pd(Me)(MeCN)(P-

P)]PF₆have been synthesized and employed in CH₂Cl₂to catalyze the alternating carbon

monoxide/ethene copolymerization. The intrinsic activity of the three precursors decreases

in the order [Pd(Me)(MeCN)(meso-2,3-dppb)]⁺> [Pd(Me)(MeCN)(rac-2,3-dppb)]⁺> [Pd(Me)-

(MeCN)(dppe)]⁺. High-pressure NMR experiments and the determination of activation

barriers of migratory insertions agree to indicate the relative stability of the â-chelate ring

in [Pd(CH₂CH₂C(O)Me)(P-P)]⁺as the factor that controls the copolymerization rate in aprotic

solvents. The impact of the different diphosphines on both productivity and intrinsic catalytic

activity has been attributed to the different stereochemical rigidity of the Pd(P-P) five-

membered metallarings. The â-chelate complexes [Pd(CH₂CH₂C(O)Me)(P-P)]PF₆with

diphosphine ligands containing two carbon atoms between the phosphorus donors have been

isolated for the first time and employed to study the chain-transfer by protonolysis, which

proceeds via the enolate mechanism. It has been shown that the chain-transfer products

2+

[Pd(OH)(P-P)]₂do not represent a dead end for the alternating CO/ethene copolymeri-

zation.

In tr od u ction

and widespread commercialization of polyketones (e.g.,

Carilon from Shell^1aand Ketonex from BP^1b) is ham-

pered, however, by their still imperfect stability, in

particular when exposed to high temperature for ex-

tended periods or subjected to repeated cycles of melting

and solidification. Moreover, the exclusive use of pal-

ladium catalysts to achieve high productivity provides

inherent drawbacks due to the high cost of this noble

metal as well as its tendency to plate out during both

catalyst preparation and copolymer workup. Both cata-

lyst composition and material stability can therefore be

improved, which is demonstrated also by the continuous

discovery of more and more efficient catalysts in differ-

ent phase variation systems as well as the increasing

availability of detailed mechanistic information on the

copolymerization process.²

Copolymers and terpolymers of carbon monoxide and

ethene or ethene/propene are not only low-cost thermo-

plastics but also polymeric materials characterized by

unique chemical and physical properties as well as

environmentally friendly nature.¹Perfectly alternating

polyketones are actually photo- and biodegradable,

completely impermeable to hydrocarbons, and easily

modifiable by various chemical procedures; they can be

processed by means of conventional techniques into

films, sheets, plates, fibers, and shaped articles for

domestic use and for parts in car industry. The rapid

(1) (a) Shell, Carilon Thermoplastic Polymers, Information Sheet,

1994. (b) Eur. Plast. News 1995, October, 57. (c) Ash, C. E. Int. J .

Polym. Mater. 1995, 30, 1. (d) Ash, C. E.; Flood, J . E. ACS Proc. 1997,

76, 110.

Publication on Web 12/11/2001

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 17

Sch em e 1

Sch em e 2

Sch em e 3

and Pd(TFA)₂(rac-2,4-bdpp) (3) have been found to

catalyze the CO/ethene copolymerization in MeOH,

yielding 16, 24, and 17 kg of polyketone (g of Pd)^-1in 3

h, respectively (TFA ) trifluoroacetate).⁴

Since the discovery of the effectiveness of palladium-

1,3-bis(diphenylphosphino)propane (dppp) catalysts in

1984 at Shell Research,³many different types of diphos-

phine ligands have been designed and successfully

employed to catalyze the alternating CO/ethene copo-

lymerization in conjunction with palladium(II) salts

through the general mechanism shown in Scheme 1.^2b,c

Common ligand variations have generally involved

substitution(s) at either the phenyl rings or the satu-

rated carbon backbone.⁴With respect to the latter issue,

it has been recently found that the introduction of alkyl

substituents in the 2-position of the carbon backbone

of dppp does not significantly improve the catalytic

performance of the corresponding Pd(II) precursors.⁴In

contrast, a considerable increase in productivity (kg of

polyketone (g of Pd)^-1) has been obtained by introducing

methyl groups in both 1-positions of the dppp backbone,

particularly with R,S (S,R) stereochemistry as in meso-

It was suggested that both electronic and steric factors

contribute to increase the productivity of the meso-2,4-

bdpp 2,^4abut a clear-cut explanation of the activity trend

exhibited by the palladium(II) precursors 1-3 was not

given. Accordingly, it was not possible to rule out that

the higher activity of 2 might simply be due to a

fortuitous combination of electronic and steric effects.

Recently, however, other reports have appeared in

which palladium(II) catalysts with meso-diphosphines

have been shown to give higher copolymerization activ-

ity as compared to catalysts bearing similar diphosphine

ligands but devoid of meso structure. Consiglio and co-

workers have communicated that cationic [Pd(solvent)₂-

(P-P)]²⁺complexes are much more active in both CO/

propene and CO/ethene copolymerization when the P-P

ligand contains methyl groups with either (S,R) or (R,S)

stereochemistry as in meso-2,4-bdpp or meso-(C₆H₄)(CH-

(Me)PPh₂)₂(Scheme 3).⁵It has also been reported that

the ligand cis,trans,cis-1,2,3,4-tetrakis(diphenylphos-

phino)cyclobutane (cyclo-tetraphos) with pseudo-meso-

structure (Scheme 3) forms palladium(II) catalysts for

CO/ethene copolymerization which are more efficient

than those with 1,2-bis(diphenylphosphino)ethane (dppe)

by 1 order of magnitude.⁶

2,4-bis(diphenylphosphino)pentane

(meso-2,4-bdpp)

(Scheme 2).⁴Under comparable experimental condi-

tions, Pd(TFA)₂(dppp) (1), Pd(TFA)₂(meso-2,4-bdpp) (2),

(2) (a) Sen, A. Acc. Chem. Res. 1993, 26, 303. (b) Drent, E.;

Budzelaar, P. H. M. Chem. Rev. 1996, 96, 663. (c) Drent, E.; van

Broekhoven, J . A. M.; Budzelaar, P. H. M. In Applied Homogeneous

Catalysis with Organometallic Compounds; Cornils, B., Herrmann, W.

A., Eds.; VCH: Weinheim, 1996; Vol. 1, p 333. (d) Sommazzi, A.;

Garbassi, G. Prog. Polym. Sci. 1997, 22, 1547. (e) Nozaki, K.; Hijama,

T. J . Organomet. Chem. 1999, 576, 248. (f) Bianchini, C.; Meli, A.

Coord. Chem. Rev. 2001, 225, 35.

(3) (a) Drent, E. Eur. Pat. Appl. 121965A2, 1984; Chem. Abstr. 1985,

102, 46423. (b) Drent, E.; van Broekhoven, J . A. M.; Doyle, M. J .

Organomet. Chem. 1991, 417, 235.

(4) (a) Bianchini, C.; Lee, H. M.; Meli, A.; Moneti, S.; Vizza, F.;

Fontani, M.; Zanello, P. Macromolecules 1999, 32, 4183. (b) Bianchini,

C.; Lee, H. M.; Barbaro, P.; Meli, A.; Moneti, S.; Vizza, F. New J . Chem.

1999, 23, 929.

Intrigued by the increasing evidence of a meso effect

enhancing the productivity of palladium-diphosphine

(5) (a) Sesto, B.; Consiglio, G. ISHC12, Stockholm, 2000, Abstracts,

p 97. (b) Sesto, B.; Consiglio, G. J . Am. Chem. Soc. 2001, 123, 4097.

(6) Bianchini, C.; Lee, H. M.; Meli, A.; Oberhauser, W.; Vizza, F.;

Bru¨geller, P.; Haid, R.; Langes, C. J . Chem. Soc., Chem. Commun.

2000, 777.

18 Organometallics, Vol. 21, No. 1, 2002

Bianchini et al.

Sch em e 4

catalysts, we decided to verify the existence/generality

of such a phenomenon and, if confirmed, to try to

understand its chemical underpinnings. Our attention

was immediately drawn by the dppe ligand, which

differs from dppp by the presence of two CH₂spacers

separating the diphenylphosphino groups and also

forms palladium(II) catalysts for the alternating CO/

ethene copolymerization that are remarkably less active

than those with dppp.²Then, we synthesized the dia-

stereomeric ligands meso-2,3-bis(diphenylphosphino)-

butane (meso-2,3-dppb) and rac-2,3-bis(diphenylphos-

phino)butane (rac-2,3-dppb), which differ from dppe by

the presence of two methyl groups in the 1,2 positions

of the ligand backbone (Scheme 4).

F igu r e 1. ORTEP drawing of 4‚CH₂Cl₂. Phenyl rings of

dppe are omitted for clarity.

Various palladium(II) precursors have been prepared

with dppe, rac-2,3-dppb, and meso-2,3-dppb and have

been employed to catalyze the copolymerization of CO

and ethene in either MeOH or CH₂Cl₂. The existence

of a ligand effect, particularly of a favorable meso effect,

controlling the intrinsic catalytic activity was actually

observed also for this series of ligands, and therefore a

detailed in situ and model study of the copolymerization

process was carried out.

Resu lts

F igu r e 2. ORTEP drawing of 5. Phenyl rings of meso-2,3-

dppb are omitted for clarity.

Syn th esis a n d Ch a r a cter iza tion of P d (OAc)₂(P -

P ) (OAc ) CH₃CO₂; P -P ) d p p e, 4; m eso-2,3-d p p b,

5; r a c-2,3-d p p b, 6). The bis-acetate complex Pd(OAc)₂-

(dppe) (4) has been synthesized by the reaction of Pd(II)

acetate in CH₂Cl₂with a stoichiometric amount of dppe,

while the preparation of the meso-2,3-dppb and rac-2,3-

dppb complexes Pd(OAc)₂(meso-2,3-dppb) (5) and Pd-

(OAc)₂(rac-2,3-dppb) (6) was best accomplished by scav-

enging the chloride ligands from the bis-chloride

derivatives with silver acetate (Scheme 5). Pure samples

of meso-2,3-dppb and rac-2,3-dppb were obtained by

standard organic reactions; rac-(CH₃CH(OMs))₂(OMs

) mesyl) was isolated by fractional recrystallization of

an isomeric mixture of (CH₃CH(OMs))₂and then reacted

with KPPh₂to give rac-2,3-dppb.

The solid-state structures of 4, 5, and 6 have been

determined by single-crystal X-ray analyses (Figures

1-3, Tables 1 and 2). In all compounds the palladium

center is coordinated by a chelating diphosphine and

by two cis acetate ions in an almost regular square-

planar geometry. No major metrical and angular dif-

ferences are apparent in the three structures, except for

the dihedral angle defined by the atoms C1,1, P1, P2,

and C1,3, which is significantly smaller (10.42(15)°) for

the meso complex 5 than for the other two (4, 20.71(67)°;

6, 19.15(40)°). No particular contact involving the

hydrogen atoms of the phenyl rings and the palladium

centers has been observed for all compounds. The

chelate metallarings are puckered with one axial and

one equatorial CMe group for the meso diasteromer and

two equatorial for the rac.

F igu r e 3. ORTEP drawing of 6‚H₂O. Phenyl rings of rac-

2,3-dppb are omitted for clarity.

observed (see below).⁷In this solvent, the ³¹P{¹H} NMR

spectra show singlets at δ 58.9 (4), 61.8 (5), and 60.5

(6) at room temperature. Chelate ring flipping in the

meso complex is therefore quite rapid to interchange

axial and equatorial CMe groups, so that a pseudo-

symmetry plane makes the phosphorus and methyl

groups magnetically equivalent down to -60 °C. Below

this temperature, the dynamic process is frozen out and

the molecule, like in the solid state, loses any symmetry

(AM pattern with broad singlets at δ 65.6 and 60.4 for

(7) (a) J arrett, P. S.; Sadler, P. J . Inorg. Chem. 1991, 30, 2098. (b)

Angulo, I. M.; Bouwman, E.; Lutz, M.; Mul, W. P.; Spek, A. L. Inorg.

Chem. 2001, 40, 2073. (c) Angulo, I. M.; Bouwman, E.; Lok, S. M.; Lutz,

M.; Mul, W. P.; Spek, A. L. Eur. J . Inorg. Chem. 2001, 1465.

All the bis-acetate complexes are quite stable in CH₂-

Cl₂, where no autoionization to bis-chelate species was

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 19

Sch em e 5

Ta ble 1. Cr ysta l Da ta a n d Str u ctu r e Refin em en ts for 4‚CH₂Cl₂, 5, a n d 6‚H₂O

4‚CH₂Cl₂

₃₁H₃₂Cl₂0₄P₂Pd

5

6‚H₂O

C₃₂H₃₆0₅P₂Pd

empirical formula

fw, g

cryst color

cryst size, mm

cryst syst

space group

unit cell dimens

a, Å

C

C₃₂H₃₄0₄P₂Pd

650.93

707.85

white

668.99

white

pale yellow

0.55 × 0.25 × 0.20

monoclinic

P2₁/c

0.50 × 0.10 × 0.45

0.45 × 0.20 × 0.05

monoclinic

P2₁/n

triclinic

P1h

11.711(2)

19.818(3)

14.380(4)

8.191(5)

8.798(1)

15.665(7)

22.102(5)

b, Å

c, Å

13.173(2)

14.404(4)

104.45(2)

90.80(3)

R, deg

â, deg

112.80(1)

100.32(1)

γ, deg

94.48(3)

volume, Å³

Z

3076.7(19)

4

1499.5(12)

2

2996.8(15)

4

density(calc), g/cm³

abs coeff, mm^-1

F(000)

1.528

0.906

1440

1.442

0.760

668

1.482

0.753

1336

radiation/wavelength, Å

θ range, deg

index ranges

Mo KR/0.71069

1.85-22.97

-12 < h < 11

0 < k < 21

0 < l < 15

Mo KR/0.71069

2.85-24.97

-9 < h < 9

-15 < k < 15

0 < l < 17

5473

5245 [R(int) ) 0.0361]

full-matrix least-squares on F²

5245/0/310

1.049

Mo KR/0.71069

2.28-22.97

-9 < h < 9

0 < k < 17

0 < l < 24

no. of reflns collected

no. of unique reflns

refinement method

no. of data/restraints/params

goodness-of-fit on F²

final R indices [I > 2σ(Ι)]

R₁

4457

4275

4261 [R(int) ) 0.0355]

full-matrix least-squares on F²

4261/0/302

0.890

4155 [R(int) ) 0.0550]

full-matrix least-squares on F²

4155/0/314

1.022

0.0556

0.1717

0.0462

0.1087

0.0490

0.0931

wR₂

R indices (all data)

R₁

0.0845

0.0860

0.1317

wR₂

0.1985

0.1237

0.1126

largest diff peak and

0.936 and -1.331

0.888 and -1.313

0.574 and -0.444

hole, e/Å³

Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles

In conclusion, NMR spectroscopy shows the meso com-

plex to adopt in ambient temperature solution a time-

averaged preferred conformation containing a pseudo-

symmetry plane with two axial and two equatorial

phenyl groups located as shown in Scheme 6, while the

Pd(rac-2,3-dppb) ring would be fixed into a twisted

conformation with two pseudoaxial and two pseudo-

equatorial phenyl substituents diagonally positioned

with respect to the Pd(P-P) plane.^8,9

To estimate the basicity of the palladium center in 4,

5, and 6, the homologous series of bis-cyanide com-

pounds Pd(CN)₂(P-P) (P-P ) dppe, meso-2,3-dppb, rac-

2,3-dppb) were prepared by reaction of Pd(CN)₂with the

corresponding P-P ligand.^10,11IR spectroscopy showed

(d eg) for 4‚CH₂Cl₂, 5, a n d 6‚H₂O

4‚CH₂Cl₂

5

6‚H₂O

Pd-P(1)

Pd-P(2)

Pd-O(1)

Pd-O(3)

2.218(2)

2.225(2)

2.078(7)

2.108(6)

2.221(2)

2.072(3)

2.075(4)

2.212(2)

2.232(2)

2.087(5)

P(1)-Pd-P(2)

O(3)-Pd-P(2)

O(1)-Pd-O(3)

O(3)-Pd-P(1)

O(1)-Pd-P(1)

O(1)-Pd-P(2)

85.24(9)

172.67(19)

89.7(3)

87.46(19)

167.8(2)

97.5(2)

85.44(6)

84.81(7)

173.66(10)

92.95(14)

89.86(11)

177.82(11)

92.42(11)

94.19(14)

91.68(19)

178.95(15)

89.34(15)

173.53(15)

P_Aand P_M, respectively). The presence of a C₂axis in

the rac isomer 6 does not allow one to estimate its

molecular flexibility by variable-temperature NMR

spectroscopy. It is well known, however, that the pres-

ence of the methyl substituents makes the metallaring

in 6 more rigid than in the dppe derivative 4, in which

the Pd(dppe) metallaring rapidly interconverts in solu-

tion from one chiral twisted conformation to the other.⁸

(8) Fryzuk, M. D.; Bosnich B. J . Am. Chem. Soc. 1977, 99, 6262.

(9) (a) MacNeil, P. A.; Roberts, N. K.; Bosnich B. J . Am. Chem. Soc.

1981, 103, 2273. (b) Bakos, J .; To´th, I.; Heil, B.; Szalontai, G.; Pa´rka´nyi,

L.; Fu¨lo¨p, V. J . Organomet. Chem. 1989, 370, 263.

(10) Bemi, L.; Clark, H. C.; Davies, J . A.; Fyfe, C. A.; Wasylishen,

R. E. J . Am. Chem. Soc. 1982, 104, 438.

(11) Tolman, C. A. J . Am. Chem. Soc. 1970, 92, 2953.

20 Organometallics, Vol. 21, No. 1, 2002

Bianchini et al.

Sch em e 6

lymerization reactions catalyzed by 4, 5, and 6 were

studied in situ by means of high-pressure NMR spec-

troscopy (HPNMR) in 10 mm OD sapphire tubes. Mass

transfer of gases from the headspace of the 10 mm NMR

tubes is generally efficient enough to replenish the

solution which is being depleted of the reagents by the

catalyst.^4,12In the NMR experiments was employed a

palladium concentration higher than that in the batch

reactions (ca. 10^-2vs 10^-4M) in order to acquire well-

resolved NMR spectra in a short residence time.¹²The

amount of the gaseous reagents contained in the head-

space of the tubes was large enough to maintain a

relatively high concentration of gases in solution during

and after each experiment (¹H NMR evidence for ethene,

singlet at δ 5.3).^4,12

Ta ble 3. ³¹P {¹H} NMR a n d Selected IR Da ta for

P d (CN)₂(P -P ) Com p lexes

A sequence of ³¹P{¹H} NMR spectra relative to the

reaction catalyzed by the dppe-based catalyst in MeOH-

d₄is shown in Figure 4.

a

P-P

³¹P{¹H} NMR (δ)

ν (CN) (cm^-1

)

dppe

53.8^b

59.4^c

57.6^c

2139, 2134

2142, 2134

2138

meso-2,3-dppb

rac-2,3-dppb

The first spectrum, recorded after dissolving 4 in

MeOH-d₄at room temperature (trace a), showed the

partial conversion of the precursor (δ 63.9) to the bis-

chelate complex cation [Pd(dppe)₂]²⁺(7) (δ 58.8), which

was identified by its independent synthesis.¹³The

addition of the co-reagents TsOH (5 equiv) and BQ (10

equiv) resulted in the complete transformation of re-

sidual 4 into the bis-tosylate derivative Pd(OTs)₂(dppe)^4a

(8) (δ 75.4), while the concentration of 7 did not change

(trace b). After 1 h at room temperature, no variation

of the NMR picture was observed (trace c). The NMR

tube was pressurized with 40 bar of carbon monoxide/

ethene (1:1) at room temperature (trace d) and then

heated to 85 °C for 1 h. As shown by the spectrum

acquired at room temperature (trace e), the concentra-

tion of 8 decreased remarkably, while that of 7 in-

creased. Black palladium metal and polyketone were

found in the HPNMR tube.

Unlike the dppe precursor 4, complexes 5 and 6 did

not form any bis-chelate complex upon dissolution in

MeOH-d₄under either nitrogen or CO/ethene copolym-

erization conditions. The NMR behavior of 5 is il-

lustrated in Figure 5. The only diphosphine-Pd complex

seen by NMR spectroscopy under copolymerization

conditions (i.e., in the presence of both comonomers)

gave a singlet at δ 78.1 (traces b-d) that was observed

also when isolated Pd(OTs)₂(meso-2,3-dppb)^4a(9) was

dissolved in MeOH. In line with previously reported

studies with palladium precursors stabilized by dppp-

like diphosphines,^4athe resonance at δ 78.1 was at-

tributed to a catalyst resting state containing “Pd(P-

In KBr. In CDCl₃. ^cIn DMF-d₇.

a

b

the stretching frequency ν(CN) to be invariant within

the bis-cyanide complexes (Table 3), which suggests that

the electron density at palladium does not appreciably

change by substituting either rac-2,3-dppb or meso-2,3-

dppb for dppe.^10,11This is also consistent with the close

³¹P{¹H} NMR chemical shifts of 4, 5, and 6.

Alter n a tin g Cop olym er iza tion (Ter p olym er iza -

tion ) of Eth en e (Eth en e/P r op en e) a n d Ca r bon

Mon oxid e Ca ta lyzed by P d (OTs)₂(P -P ) in MeOH.

Au tocla ve Exp er im en ts. Compounds 4, 5, and 6 have

been employed as catalyst precursors to copolymerize

CO and ethene or ethene/propene in MeOH applying

standard experimental conditions for this kind of reac-

tion.²p-Toluenesulfonic acid (TsOH) and 1,4-benzo-

quinone (BQ) were employed as protic acid and oxidant,

respectively.^2,3

As shown in Table 4, both 5 and 6 give rise to efficient

catalysts for the formation of copolymers and terpoly-

mers with productivities (entries 3, 4 and 7, 8) that are

slightly lower than those of the standard catalyst 1^3,4

but higher than those obtained with the unsubstituted

dppe derivative 4 up to 1 order of magnitude. Notably,

the catalyst derived from meso 5 is more efficient than

that derived from the rac diasteromer 6, in particular

for the terpolymerization reaction (6.8 vs 4.5 kg of

terpolymer (g of Pd)^-1).

The overall productivity trend exhibited by 4, 5, and

6 reflects a ligand-induced activity decreasing in the

order meso-2,3-dppb > rac-2,3-dppb . dppe, which is

analogous to that observed for the dppp-like precursors

1, 2, and 3 under comparable conditions.⁴

(12) (a) Bianchini, C.; Herrera, V.; J ime´nez, M. V.; Meli, A.; Sa´nchez-

Delgado, R. A.; Vizza, F. J . Am. Chem. Soc. 1995, 117, 8567. (b)

Bianchini, C.; Fabbri, D.; Gladiali, S.; Meli, A.; Pohl, W.; Vizza, F.

Organometallics 1996, 15, 4604. (c) Bianchini, C.; J ime´nez, M. V.; Meli,

A.; Moneti, S.; Patinec, V.; Vizza, F. Organometallics 1997, 16, 5696.

(d) Bianchini, C.; Meli, A.; Patinec, V.; Sernau, V.; Vizza, F. J . Am.

Chem. Soc. 1997, 119, 4945. (e) Bianchini, C.; Casares, J . A.; Meli, A.;

Sernau, V.; Vizza, F.; Sa´nchez-Delgado, R. A. Polyhedron 1997, 16,

3099. (f) Bianchini, C.; Meli, A.; Moneti, S.; Oberhauser, W.; Vizza, F.

Organometallics 1998, 17, 2636. (g) Bianchini, C.; Masi, D.; Meli, A.;

Peruzzini, M.; Vizza, F.; Zanobini, F. Organometallics 1998, 17, 2495.

(h) Bianchini, C.; Meli, A.; Moneti, S.; Oberhauser, W.; Vizza, F.;

Herrera, V.; Fuentes, A.; Sa´nchez-Delgado, R. A. J . Am. Chem. Soc.

1999, 121, 7071. (i) Bianchini, C.; Lee, H. M.; Meli, A.; Vizza, F.

Organometallics 2000, 19, 849.

The polyketones produced with 5 and 6 appeared as

fluffy off-white solids, while a gray powder was obtained

using 4. Irrespective of the catalyst precursor, the H

1

and ¹³C{¹H} NMR spectra of the copolymers showed the

presence of both ketone and ester end groups, limiting

viscosity numbers (LVN) in the range 0.25-0.35 dL g^-1

and average molecular weights (M_n) in the range 6-8

kg mol^-1(Table 4).²

(13) (a) Lutz, M. E.; Patrick, J . M.; Raston, C. L.; Twiss, P.; White,

A. H. Aust. J . Chem. 1984, 37, 2193. (b) Luo, H.-K.; Kou, Y.; Wang,

X.-W.; Li, D.-G. J . Mol. Catal. 2000, 151, 91. (c) Bianchini, C.;

Mantovani, G.; Meli, A.; Oberhauser, W.; Bru¨ggeler, P.; Stampfl, T. J .

Chem. Soc., Dalton Trans. 2001, 690.

Alter n a tin g Cop olym er iza tion of Eth en e a n d

Ca r bon Mon oxid e Ca ta lyzed by P d (OTs)₂(P -P ) in

MeOH. In Situ HP NMR Stu d y. The CO/ethene copo-

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 21

Ta ble 4. Alter n a tin g Co- a n d Ter p olym er iza tion of Ca r bon Mon oxid e w ith Eth en e a n d Eth en e/P r op en e

Ca ta lyzed by P d (OTs)₂(P -P ) in MeOH^a

copol.

prod.^b

LVN

M_n

mp

terpol.

prod.^b

LVN

M_n

mp

(°C)

propene

(%)^c

complex

P-P ligand

dppp

dppe

meso-2,3-dppb

rac-2,3-dppb

entry

(dLg^-1

)

(kgmol^-1

)

(°C) entry

(dL g^-1

)

(kg mol^-1

)

1

4

5

6

1

2

3

4

16.2

1.1

11.1

8.8

0.61

0.25

0.29

0.28

14.2

6.7

7.7

256

251

254

252

5

6

7

8

16.0

1.0

6.8

0.60

0.27

0.26

0.31

14.1

7.2

6.9

224

227

225

227

6

5

6

5

7.7

4.5

8.0

a

Conditions: catalyst (0.01 mmol), MeOH (100 mL), BQ (0.8 mmol), TsOH (0.2 mmol), initial p(C₂H₄) (20 bar), initial p(CO) (20 bar),

temperature (85 °C), time (3 h), stirring rate (1400 rpm). In the terpolymerization reactions 20 g of propene was introduced in the autoclave

b

just before pressurizing with the C₂H₄/CO mixture. Productivity expressed as kg of polymer (g of Pd)^-1

.

^cAmount of propene incorporated

in the polymeric chains as calculated by ¹H NMR integration. The units CH₂C(O)CH(Me)CH₂and CH₂C(O)CH₂CH(Me) are in 1:1 ratio

in all polymeric samples (¹³C{¹H} NMR integration).

Sch em e 7

may occur through several pathways including sponta-

neous degradation of palladium hydrides²and, espe-

cially in the presence of oxygenated coligands, an inner-

sphere metal reduction.¹⁴

The deactivation of the dppe precursor proceeds via

the autoionization mechanism illustrated in Scheme 7,

which is well known for Ni(OAc)₂(dppe).⁷In a polar

solvent such as MeOH (ꢀ ) 32.6), ³¹P{¹H} NMR has

shown that the autoionization equilibrium is largely

shifted to the right, namely, toward the formation of

the catalytically inactive bis-chelate complex 7 and of

palladium acetate (trace a in Figure 4). Consistent with

F igu r e 4. ³¹P{¹H} NMR study (sapphire tube, MeOH-d₄,

20 °C, 81.01 MHz) of CO/C₂H₄copolymerization assisted

by 4: (a) dissolving 4 in MeOH-d₄under nitrogen; (b)

a reversible autoionization reaction, when all MeOH-

d₄was removed in vacuo from a solution such as that

giving trace a in Figure 4 and the solid residue was

dissolved in CD₂Cl₂(ꢀ ) 9.1),⁷the precursor 4 was re-

formed quantitatively.

immediately after the addition of both TsOH and BQ; (c)

after 1 h; (d) after the tube was pressurized with 40 bar of

CO/C₂H₄(1:1); (e) spectrum acquired at room temperature

after 1 h heating to 85 °C.

The autoionization reaction of 4 in MeOH is very

rapid and is not inhibited by the presence of BQ, TsOH,

CO, C₂H₄, or CO/C₂H₄, as shown by batch reactions

performed applying a different order of addition of the

reagents. However, palladium acetate is not stable in

methanol under copolymerization conditions in the

absence of ligands and degrades to palladium metal,²

which accounts for the fast and complete deactivation

of the dppe-based catalyst.

Syn th esis of th e Meth yl Aceton itr ile Com p lexes

[P d (Me)(MeCN)(P -P )]P F ₆a n d th eir Use a s Ca ta -

lyst P r ecu r sor s in th e Alter n a tin g Cop olym er iza -

tion of Eth en e a n d Ca r bon Mon oxid e in CH₂Cl₂.

The new methyl acetonitrile palladium complexes [Pd-

(Me)(MeCN)(meso-2,3-dppb)]PF₆(11) and [Pd(Me)(Me-

CN)(rac-2,3-dppb)]PF₆(12) were synthesized following

a procedure previously reported for the dppe derivative

[Pd(Me)(MeCN)(dppe)]PF₆(10).¹⁵This involves the treat-

ment of the neutral methyl chloride complexes PdCl-

(Me)(P-P) dissolved in a CH₂Cl₂/MeCN solution (10:1,

v:v) with AgPF₆as chloride scavenger. Concentration

under reduced pressure and addition of diethyl ether/

n-hexane gave analytically pure samples of [Pd(Me)-

(MeCN)(P-P)]PF₆, whose unambiguous identification

was achieved by both ³¹P{¹H} (AM patterns) and ¹H

F igu r e 5. ³¹P{¹H} NMR study (sapphire tube, MeOH-d₄,

20 °C, 81.01 MHz) of CO/C₂H₄copolymerization assisted

by 5: (a) dissolving 5 in MeOH-d₄under nitrogen; (b)

immediately after the addition of both TsOH and BQ; (c)

after the tube was pressurized with 40 bar of CO/C₂H₄(1:

1); (d) spectrum acquired at room temperature after 1 h

heating to 85 °C.

P)²⁺” moieties stabilized by OTs^-ligands, eventually in

rapid exchange with solvent molecules.

On standing in solution at 85 °C, also the catalysts

generated by 5 and 6 underwent degradation, but this

was much slower than that of the dppe system and also

different in nature. Indeed, after 3 h in copolymerization

conditions, no bis-chelate complex analogous to 7 was

seen, while traces of palladium metal were actually

observed. Catalyst deactivation using Pd(II) precursors

(14) (a) Amatore, C.; Carre´, E.; J utand, A.; M’Barki, M. A. Orga-

nometalllics 1995, 14, 1818. (b) Amatore, C.; J utand, A.; Thuilliez, A..

Organometalllics 2001, 20, 3241.

(15) Dekker, G. P. C. M.; Elsevier, C. J .; Vrieze, K.; van Leeuwen,

P. W. N. M. Organometallics 1992, 11, 1598.

22 Organometallics, Vol. 21, No. 1, 2002

Bianchini et al.

Ta ble 5. Alter n a tin g Cop olym er iza tion of Ca r bon Mon oxid e w ith Eth en e Ca ta lyzed by

a

[P d (Me)(MeCN)(P -P )]P F ₆a n d [P d (OH)(P -P )]₂(P F ₆)₂in CH₂Cl₂

copolymer productivity^c

e

initial rate^b

M_n

mp^e

(°C)

complex

10

11

12

P-P ligand

dppe

meso-2,3-dppb

rac-2,3-dppb

dppe

(g h^-1

)

0.5 h

3 h

5 h

3 h^d

(kg mol^-1

)

2.2

3.3

2.9

0.6

1.0

0.8

0.3

0.6

1.9

3.0

2.3

2.5

3.9

3.2

2.3

3.6

2.9

56

53

58

255

253

254

16

cis/trans-17

meso-2,3-dppb

-

a

Conditions: catalyst (0.01 mmol), CH₂Cl₂(100 mL), initial p(C₂H₄) (20 bar), initial p(CO) (20 bar), temperature (70 °C), stirring rate

(1400 rpm). Expressed as g of comonomers consumed per hour. ^cExpressed as kg of polymer (g of Pd)^-1

.

In the presence of BQ (0.8

b

d

mmol). ^eData for the copolymers obtained in the reactions lasting 3 h.

F igu r e 6. Gas uptake against time during the copolym-

erization reactions catalyzed by the methyl acetonitrile

complexes 10-12 in CH₂Cl₂.

F igu r e 7. ³¹P{¹H} NMR study (sapphire tube, CD₂Cl₂,

81.01 MHz) of CO/C₂H₄copolymerization assisted by 10:

(a) dissolving 10 in CD₂Cl₂under nitrogen at room tem-

perature; (b) after the tube was pressurized with 40 bar of

CO/C₂H₄(1:1) at room temperature; (c) at 50 °C (first

spectrum); (d) after 15 min at 70 °C.

NMR spectroscopy (singlets at ca. δ 2.2 for the coordi-

nated MeCN ligand).

Complexes 10-12 were employed as catalyst precur-

sors for the copolymerization of CO and ethene in CH₂-

Cl₂(distilled over CaH₂) at 70 °C in the absence of both

organic oxidant and protic acid, which are typical

conditions for reactions catalyzed by initiators contain-

ing a Pd-alkyl moiety in aprotic solvents.¹⁶The results

of reactions lasting 0.5, 3, and 5 h are reported in Table

5, which also contains data in the presence of organic

oxidant as well as initial copolymerization rates. These

have been estimated measuring the gas consumption

within the first 10 min (Figure 6).

A perusal of Table 5 reveals that the intrinsic activity,

indicated by the initial rate values, decreased in the

ligand order meso-2,3-dppb > rac-2,3-dppb > dppe. All

the catalyst systems underwent slow deactivation with

time as a result of palladium metal formation. Consis-

tently, the addition of BQ increased the productivity as

in the reactions in MeOH, which means that, also in

CH₂Cl₂, catalyst deactivation occurs by palladium(II)

reduction.

In Situ HP NMR Stu d y of th e Alter n a tin g Cop o-

lym er iza tion of Eth en e a n d Ca r bon Mon oxid e

Ca ta lyzed by [P d (Me)(MeCN)(P -P )]P F ₆in Eith er

Dr y or Wet CD₂Cl₂. The CO/C₂H₄copolymerization

reactions catalyzed by 10, 11, and 12 were studied in

situ by means of HPNMR spectroscopy in 10 mm OD

sapphire tubes applying comparable experimental con-

ditions to those of the batch reactions. Selected ³¹P{¹H}

NMR spectra relative to the reactions catalyzed by the

dppe and meso-2,3-dppb precursors 10 and 11 in CD₂-

Cl₂distilled over CaH₂are reported in Figures 7 and 8,

respectively.

The ³¹P{¹H} NMR spectrum of the precursor 10 is

shown in Figure 7 (trace a). Immediately after pressur-

izing the tube to 40 bar with a 1:1 CO/C₂H₄mixture at

room temperature, polyketone formation occurred while

10 converted to the â-chelates [Pd(CH₂CH₂C(O)-chain)-

(dppe)]⁺(13) (AM pattern at δ 57.7 and 37.0, J (PP) )

29.2 Hz) where “chain” is the propagating copolymer

(trace b) (Scheme 1). By ¹H NMR integration, the

average repeating units of the chain have been esti-

As is generally observed in copolymerization reactions

in CH₂Cl₂,^2,16the productivities were lower than those

obtained in MeOH solution with comparable palladium

precursors, while the molecular weights were much

larger (50-60 vs 6-8 kg mol^-1) and almost independent

of the diphosphine ligand (Tables 4 and 5). The large

M_n’s did not allow us to determine the nature of the end

1

groups by H NMR spectroscopy in 1,1,1,3,3,3-hexafluo-

ropropan-2-ol-d₂(HFIP-d₂). On the basis of previous

studies on co-oligomers, it is likely that the polyketone

material contains vinyl and ketone end groups, formed

by â-H chain transfer and Pd-H initiation, respec-

tively.¹⁶

(16) (a) Koide, Y.; Bott, S. G.; Barron, A. R. Organometalllics 1996,

15, 2213. (b) Reddy, K. R.; Chen, C.-L.; Hung Liu, Y.-H.; Peng, S.-M.;

Chen, J .-T.; Liu, S.-T. Organometalllics 1999, 18, 2574. (c) Schwarz,

J .; Herdtweck, E.; Herrmann, W. A. Organometalllics 2000, 19, 3154.

(d) Barlow, G. K.; Boyle, J . D.; Cooley, N. A.; Ghaffar, T.; Wass, D. F.

Organometalllics 2000, 19, 1470. (e) Braunstein, P., Fryzuk, M. D.;

Le Dall, M.; Naud, F.; Rettig, S.; Speiser, F. J . Chem. Soc., Dalton

Trans. 2000, 1067.

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 23

F igu r e 9. ³¹P{¹H} NMR study (sapphire tube, CD₂Cl₂,

81.01 MHz) of CO/C₂H₄copolymerization assisted by 10

in the presence of H₂O: (a) after heating a solution of 10

under 40 bar of CO/C₂H₄(1:1) to 50 °C; (b) at 70 °C (first

spectrum); (c) after 1 h at 70 °C.

F igu r e 8. ³¹P{¹H} NMR study (sapphire tube, CD₂Cl₂,

81.01 MHz) of CO/C₂H₄copolymerization assisted by 11:

(a) dissolving 11 in CD₂Cl₂under nitrogen at room tem-

perature; (b) after the tube was pressurized with 40 bar of

CO/C₂H₄(1:1) at room temperature; (c) at 70 °C (first

spectrum); (d) after 45 min at 70 °C.

fied by comparison with either authentic specimens or

structurally similar compounds (vide infra) prepared by

independent methods.

The copolymerization reactions in CD₂Cl₂catalyzed

by 10, 11, and 12 were similarly followed by HPNMR

spectroscopy in the presence of added water (10 µL) to

study the chain-transfer mechanism by protonolysis and

ultimately gain insight into the possible paths to

catalyst deactivation in solution.

Figure 9 shows a sequence of spectra relative to the

copolymerization reaction catalyzed by 10 in wet CD₂-

Cl₂. The first spectrum at 50 °C (trace a) was rather

similar to that in dry solvent (Figure 7, trace c) except

for the presence of some broad resonances in the 60-

70 ppm region. The presence of water accelerated the

degradation of the â-chelates, as no trace of the latter

compounds was visible already in the first spectrum at

70 °C (trace b). Formed in their place were the bis-

chelate 7‚PF₆and other products, among which only the

µ-OH binuclear complex [Pd(OH)(dppe)]₂²⁺(16) (singlet

at δ 64.2) could be unambiguously identified. With time,

16 disappeared and the bis-chelate 7‚PF₆remained the

only well-defined palladium complex (trace c after 1 h

at 70 °C).

mated to be 8.^2,17The â-chelates were the only phos-

phorus-containing compounds visible on the NMR time

scale at 50 °C (trace c). With time, however, they slowly

decomposed to give palladium metal and the catalyti-

cally inactive bis-chelate complex 7‚PF₆(singlet at δ 57.2

overlapping the low-field resonance of the â-chelates).

Increasing the temperature to 70 °C accelerated the

decomposition of the â-chelates that disappeared com-

pletely after 15 min at 70 °C, leaving 7‚PF₆as the

predominant palladium complex (trace d). The large

noise of the spectrum did not rule out the formation of

other unidentified species, in fact.

The copolymerization reactions catalyzed by 11 (Fig-

ure 8) or 12 gave a similar NMR picture. The ³¹P{¹H}

NMR spectrum at room temperature under nitrogen of

precursor 11 (AM pattern with δ 64.8 and 42.7) is shown

in trace a. Pressurizing the tube to 40 bar with a 1:1

mixture of carbon monoxide/ethene at room temperature

transformed 11 into the â-chelate complexes of the

formula [Pd(CH₂CH₂C(O)-chain)(meso-2,3-dppb)]⁺(14)

(AM pattern with signals at δ 61.0 and 41.3 and J (PP)

) 35.1 Hz, trace b; number of average repeating units

) 9).^2,17The appearance of the â-chelates at room

temperature coincided with the formation of the copoly-

mer product. When the temperature was increased to

70 °C, the intensity of the â-chelate signal decreased

(trace c), which was essentially attributed to catalyst

degradation.²The formation of the known bis-chelate

complexes^13ccis-[Pd(meso-2,3-dppb)₂]²⁺(cis-15) and trans-

[Pd(meso-2,3-dppb)₂]²⁺(trans-15) (singlets at δ 59.7 and

58.2) became appreciable after ca. 10 min at 70 °C and

coincided with the formation of palladium metal. The

bis-chelate complexes were the predominant phosphorus-

containing species after 45 min at 70 °C (trace d).

The â-chelates 13 and 14^2,17-19and the binuclear

complexes 7‚PF₆,^13acis-15, and trans-15^13cwere identi-

The overall ³¹P{¹H} HPNMR pictures obtained for the

copolymerization reactions catalyzed by 11 and 12 in

wet CD₂Cl₂were similar to each other. The variable-

temperature spectra for the reaction of 11 are reported

in Figure 10. The most striking difference between the

reaction catalyzed by 10 and that catalyzed by 11 is that

the µ-OH binuclear complexes, cis- and trans-[Pd(OH)-

2+

(meso-2,3-dppb)]₂(cis-17, trans-17) (broad resonance

centered at δ 67.8 comprising two singlets, vide infra),

were formed already at 50 °C before any trace of the

bis-chelates cis- and trans-15 appeared in the spectrum

(trace a). With time, the â-chelates slowly converted into

the µ-OH complexes (in a separate experiment, the

HPNMR tube was removed from the probe-head at this

stage: the formation of the unsoluble copolymer was

confirmed, while no trace of palladium metal was seen).

Increasing the temperature to 70 °C accelerated the

conversion process. Indeed, in the first spectrum at 70

°C, the µ-OH complexes were the largely predominant

palladium compounds in solution (trace b). At 70 °C,

the µ-OH complexes started decomposing to give pal-

(17) Reddy, K. R.; Surekha, K.; Lee, G.-H.; Peng, S.-M.; Chen, J .-

T.; Liu, S.-T. Organometalllics 2001, 20, 1292.

(18) (a) Ozawa, F.; Hayashi, T.; Koide, H.; Yamamoto, A. J . Chem.

Soc., Chem. Commun. 1991, 1469. (b) Dekker, G. P. C. M.; Elsevier,

C. J .; Vrieze, K.; van Leeuwen, P. W. N. M.; Roobeek, C. F. J .

Organomet. Chem. 1992, 430, 357. (c) Zuideveld, M. A.; Kamer, P. C.

J .; vanLeeuwen, P. W. N. M.; Klusener, P. A. A.; Stil, H. A.; Roobeek,

C. F. J . Am. Chem. Soc. 1998, 120, 7977.

(19) Shultz, C. S.; Ledford, J .; DeSimone, J . M.; Brookhart, M. J .

Am. Chem. Soc. 2000, 122, 6351.

24 Organometallics, Vol. 21, No. 1, 2002

Bianchini et al.

F igu r e 10. ³¹P{¹H} NMR study (sapphire tube, CD₂Cl₂,

81.01 MHz) of CO/C₂H₄copolymerization assisted by 11

in the presence of H₂O: (a) after heating a solution of 11

under 40 bar of CO/C₂H₄(1:1) to 50 °C; (b) at 70 °C (first

spectrum); (c) after 1 h at 70 °C.

F igu r e 11. ³¹P{¹H} NMR study (sapphire tube, CD₂Cl₂,

20 °C, 81.01 MHz) on the stability of the â-chelate 18 in

CD₂Cl₂under nitrogen: (a) after dissolving 18 in CD₂Cl₂;

(b) after 1.5 h; (c) after 3.5 h; (d) after 15 h.

Sch em e 8

Sch em e 9

F igu r e 12. ³¹P{¹H} NMR study (sapphire tube, CD₂Cl₂,

20 °C, 81.01 MHz) on the stability of the â-chelate 18 in

wet CD₂Cl₂under nitrogen: (a) after dissolving 18 in wet

CD₂Cl₂; (b) after 1.5 h; (c) after 3.5 h; (d) after 15 h.

ladium metal and the bis-chelates cis-15 and trans-15,

which were the only well-defined palladium products

after 1 h (trace c).

CH₂C(O)Me)(dppp)]SbF₆.¹⁹The chloride acyl complexes

PdCl(COMe)(P-P) (P-P ) dppe, 27; meso-2,3-dppb, 28;

rac-2,3-dppb, 29) were quantitatively generated in CD₂-

Cl₂at room temperature by carbonylation (20 bar CO,

40 °C) of the chloride methyl precursors PdCl(Me)(P-

P) in HPNMR tubes and then transformed into the

desired â-chelates by treatment with AgSbF₆, as chlo-

ride scavenger, under a stream of C₂H₄(Scheme 9).

All the new compounds (acyl acetonitrile, acyl carbo-

nyl, chloride acyl, â-chelates) were straightforwardly

Mod el Ch a in -Tr a n sfer Rea ction s Usin g th e P a l-

la d iu m (II) â-Ketoa lk yl Ch ela tes. The â-chelate com-

plexes [Pd(CH₂CH₂C(O)Me)(P-P)]PF₆(P-P ) dppe, 18;

meso-2,3-dppb, 19; rac-2,3-dppb, 20) were isolated in the

solid state through the procedure illustrated in Scheme

8, and their thermal behavior in both dry and wet CH₂-

Cl₂was studied. The synthesis involves the conversion

of the methyl acetonitrile complexes 10-12 to the

corresponding acyl carbonyl derivatives [Pd(COMe)-

(CO)(P-P)]PF₆(P-P ) dppe, 21; meso-2,3-dppb, 22; rac-

2,3-dppb, 23) by reaction with 15 bar CO in CH₂Cl₂at

room temperature. To the solution of each acyl carbonyl

complex in a 10 mm OD sapphire NMR tube was added

a small amount of MeCN (20 µL) at -40 °C. After NMR

spectroscopy showed the complete transformation of the

acyl carbonyl complexes to the acyl acetonitrile deriva-

tives [Pd(COMe)(MeCN)(P-P)]PF₆(P-P ) dppe, 24;

meso-2,3-dppb, 25; rac-2,3-dppb, 26), the NMR tube was

purged with C₂H₄at -40 °C until the â-chelate com-

plexes were quantitatively obtained. At this point, all

the solvent was removed under reduced pressure at -20

°C to give a solid product that was washed with

n-pentane and dried.

1

identified by ³¹P{¹H} and H NMR spectroscopy (Tables

6 and 7). Comparisons with the spectra of similar or

related complexes described in the literature have

substantially contributed to validate our structural

assignments.^2,15-20

Irrespective of the synthetic procedure, [Pd(CH₂-

CH₂C(O)Me)(dppe)]⁺was found to be rather stable at

room temperature in dry CD₂Cl₂, where it disappeared

completely in ca. 15 h. Formed in its place were free

vinyl methyl ketone and methyl ethyl ketone in a ca.

5:1 ratio (¹H NMR, GC/MS), traces of Pd metal, and

several unknown palladium complexes (Figure 11).

When 18 was dissolved in wet CD₂Cl₂(10 µL water),

its decomposition gave exclusively the µ-OH binuclear

complex 16 and methyl ethyl ketone (Figure 12); traces

of methyl vinyl ketone were detected by GC/MS.

Alternatively, the â-chelate complexes were prepared

-

in situ as SbF₆salts following a procedure recently

described by Brookhart and co-workers for [Pd(CH₂-

(20) Toth, I.; Elsevier, C. J . J . Am. Chem. Soc. 1993, 115, 10388.

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 25

Ta ble 6. ³¹P {¹H} NMR Ch em ica l Sh ifts (p p m ), (Mu ltip licity), a n d [P P Cou p lin g Con sta n ts] (Hz) for th e

P a lla d iu m Com p lexes [P d (X)(Y)(P -P )]^0/+in CD₂Cl₂Solu tion s a t {T} (°C)

X, Y )

CH₂CH₂-

C(O)Me

X ) CO(CH₂)₂-

COMe

X ) Me

X )Me

X ) COMe

Y ) CO

X ) COMe

Y ) NCMe

X ) Me

Y ) Cl

X ) COMe

Y ) Cl

P-P

dppe

Y ) NCMe

Y ) CO

61.2 (d) [27.0] 61.5 (d) [27.1]

39.5 (d) {rt}

64.8 (d) [34.0] 63.0 (d) [32.9]

40.0 (d) [50.2] 38.4 (d) [44.2] 59.9 (d) [27.2] 40.0 (d) [45.1] 57.4 (d) [29.6]

46.5 (d) {-80} 36.2 (d) {rt} 23.7 (d) {rt} 31.9 (d) {rt} 22.2 (d) {rt} 36.8 (d) {rt}

44.0 (d) [57.4] 45.1 (d) [53.6] 62.8 (d) [32.0] 42.8 (d) [48.7] 61.2 (d) [35.7]

40.6 (d) [50.8]

35.9 (d) {rt}

44.1 (d) [60.7]

meso-

2,3-dppb

42.7 (d) {rt}

63.5 (d) [40.7] 60.0 (d) [44.1]

49.2 (d) {-80} 40.3 (d) {rt}

36.1 (d) {rt}

36.8 (d) {rt}

58.3 (d) {rt}

41.0 (d) {rt}

41.5 (d) {-40}

47.4 (d) [64.8]

rac-

2,3-dppb

47.2 (d) [66.8] 49.0 (d) [64.7] 63.8 (d) [38.0] 49.4 (d) [54.8] 61.7 (d) [36.8]

42.8 (d) {rt}

44.5 (d) {-80} 38.3 (d) {rt}

36.8 (d) {rt}

38.8 (d) {rt}

32.4 (d) {rt}

40.7 (d) {rt}

38.8 (d) {-40}

Ta ble 7. ¹H NMR Ch em ica l Sh ifts (p p m ), (Mu ltip licity), a n d [Cou p lin g Con sta n ts] (Hz) for Selected

Reson a n ces of th e P a lla d iu m Com p lexes [P d (X)(Y)(P -P )]^0/+in CD₂Cl₂Solu tion s a t {T} (˚C)

X )Me

X ) COMe X ) COMe

X ) Me

Y ) Cl

X ) COMe

Y ) Cl

X, Y )

X ) (CO)(CH₂)₂COMe

Y ) CO

P-P

dppe

Y ) NCMe

Y ) CO

Y ) NCMe

CH₂CH₂COMe

0.62 (dd) Me 0.78 (brs) 2.05 (s)

2.08 (s)

0.78 (dd) Me 2.05 (s)

1.74 (ddd) PdCH₂

not detected

[J (HP) 2.0,

7.3] {rt}

Me {-80} COMe {rt} COMe {rt} [J (HP) 3.0,

COMe {rt} [J (HP) 8.2, 2.1;

J (HH) 6.5]

8.0] {rt}

2.53 (s) COMe 3.31

(dd) CH₂CO [J (HP)

9.0; J (HH) 7.0] {rt}

meso-

2,3-dppb [J (HP) 2.1,

7.5] {rt}

0.39 (dd) Me 055 (brs) 2.10 (s)

2.10 (s)

Me {-80} COMe {rt} COMe {rt} [J (HP) 3.0,

8.1] {rt}

0.55 (dd) Me 1.90 (s)

1.85 (ddd) PdCH₂

COMe {rt} [J (HP) 8.0, 2.0;

J (HH) 6.8] 2.53 (s)

1.80 (s) COMe 2.40

(m) CH₂CO

2.75 (m)

COMe 3.30 (dd)

CH₂CO [J (HP) 8.5;

Pd(CO)CH₂{-40}

J (HH) 6.8] {rt}

rac-

0.57 (dd) Me 0.75 (brs) 2.04 (s)

2.08 (s)

Me {-80} COMe {rt} COMe {rt} [J (HP) 3.1,

8.2] {rt}

0.55 (dd) Me 2.03 (s)

1.60 (m) PdCH₂

1.50 (s) COMe 2.10

(m) CH₂CO 2.35

2,3-dppb [J (HP) 2.0,

7.3] {rt}

COMe {rt} 2.50 (s) COMe

3.20 (m) CH₂CO {rt} (m) Pd(CO)CH₂{-40}

Sch em e 10

In comparison to 18, the â-chelates 19 and 20 showed

a slightly decreased stability in CD₂Cl₂, where they

decomposed within 10 h, yielding more methyl vinyl

ketone than methyl ethyl ketone (ca. 4:1). In the

presence of added water, only traces of methyl vinyl

ketone were detected by GC/MS, while methyl ethyl

ketone was produced in almost quantitative yield to-

gether with the µ-OH complexes cis-17 and trans-17 in

a ca. 1:5 ratio.

The thermal behavior of the â-chelates in CD₂Cl₂was

also studied at 50 and 70 °C; identical results were

obtained except for faster decomposition rates. The

isolated â-chelates were also employed to catalyze the

CO/ethene copolymerization in either dry or wet CD₂-

Cl₂in HPNMR tubes. The NMR picture was identical

to those reported in Figures 7, 8, and 9. In particular,

the ³¹P{¹H} NMR resonances of the â-chelates became

broader as the NMR tube was pressurized to 40 bar with

CO/C₂H₄. Such a broadening was attributed to the

formation of different â-chelates of the general formula

[Pd(dppe)(CH₂CH₂C(O)-chain)]⁺, where chain is a propa-

gating co-oligomeric chain.^2,17

Th e Role of th e µ-OH Com p lexes in Ca ta lysis.

Interestingly, the µ-OH complexes do not represent a

dead end for the alternating copolymerization in CH₂-

Cl₂. Under the experimental conditions reported in

Table 5, isolated 16 and cis/trans-17 were effective

catalysts for the CO/ethene copolymerization. The µ-OH

complexes were not as efficient as their mononuclear

precursors 10 and 11 however (Table 5), which we

ascribe to the occurrence of an induction period. Inde-

pendent reactions of 16 and 17 in CH₂Cl₂with either

C₂H₄or CO showed that only the latter substrate was

able to degrade the µ-OH complexes at 70 °C, yielding

the bis-chelate complexes 7‚PF₆and cis/trans-15, re-

spectively.

By analogy with CO/ethene copolymerization reac-

tions in water, the binuclear structure of the µ-OH

complexes is apparently destroyed by CO to give mono-

nuclear Pd-OH initiators that may start the propaga-

tion step with either Pd-COOH or Pd-H species

(Scheme 10).^21-23

Gen er a tion a n d Migr a tor y In ser tion Ba r r ier s of

[P d (Me)(CO)(P -P )]P F ₆. In an attempt to rationalize

the different catalytic performance of the dppe, meso-

2,3-dppb, and rac-2,3-dppb precursors in CH₂Cl₂dis-

tilled over CaH₂, the migratory insertion barriers

relative to reactions occurring during the propagation

of the alternating CO/ethene copolymerization have

been determined by NMR spectroscopy for various

model intermediates.

The methyl carbonyl precursors [Pd(Me)(CO)(P-P)]-

PF₆(P-P ) dppe, 30; meso-2,3-dppb, 31; rac-2,3-dppb,

32) were generated in HPNMR tubes by pressurizing

CD₂Cl₂solutions of the corresponding methyl acetoni-

trile compounds with CO (15 bar) at -117 °C. ³¹P{¹H}

and H NMR spectra showed the only presence of the

methyl carbonyl complexes 30-32 also at -80 °C. Upon

1

(21) J iang, Z.; Sen, A. Macromolecules 1994, 27, 7215.

(22) (a) Verspui, G.; Papadogianakis, G., Sheldon, R. A. J . Chem.

Soc., Chem. Commun. 1998, 401. (b) Verspui, G.; Schanssema, F.;

Sheldon, R. A. Appl. Catal. 2000, 198, 5.

(23) Bianchini, C.; Lee, H. M.; Meli, A.; Moneti, S.; Patinec, V.;

Petrucci, G.; Vizza, F. Macromolecules 1999, 32, 3859.

26 Organometallics, Vol. 21, No. 1, 2002

Bianchini et al.

Ta ble 8. Exp er im en ta l Activa tion Ba r r ier s a n d Tem p er a tu r es for Migr a tor y In ser tion s

[Pd(Me)(CO)(P-P)]⁺[Pd(CH₂CH₂C(O)Me)(P-P)]⁺

P-P

∆G^q(kcal/mol)

t_1/2(min)

T (°C)

t_1/2(min)

p(CO) (bar)

dppe

meso-2,3-dppb

rac-2,3-dppb

16.9(1)

16.1(1)

17.5(1)

12

3

62

-40

20

-40

15

97

480

20

Sch em e 11

Sch em e 13

Sch em e 12

at 20 °C was the â-chelate ring opened by CO. At this

temperature, the reaction went further on and the

formation of the carbonyl acyl was accompanied by that

of several products, including Pd metal.

slow heating of the NMR probe-head, the latter com-

plexes transformed selectively into the corresponding

carbonyl acyl derivatives [Pd(COMe)(CO)(P-P)]PF₆(21-

23) (Scheme 11).

The meso-2,3-dppb and rac-2,3-dppb acyl carbonyl

complexes started to form at -50 °C. The highest

temperature required for CO insertion was observed for

the dppe derivative (-40 °C).

Kinetics of transformation of the methyl carbonyl

complexes with dppe, meso-2,3-dppb, and rac-2,3-dppb

ligands into the corresponding acyl carbonyl products

were studied by ³¹P{¹H} NMR spectroscopy at -40 °C,

where the reactions were sufficiently fast to allow for a

reliable determination of the half-life times (t_1/2).

As previously observed by Brookhart and co-workers

for [Pd(Me)(CO)(dppp)]BAr′₄,¹⁹the rate of conversion of

the methyl carbonyl complexes was independent of the

CO pressure (5-25 bar) and followed first-order kinet-

ics. The observed rate constants are therefore the true

rate constants, and the ∆G^qvalues associated with the

migratory insertion of the methyl carbonyl complexes

could be straightforwardly calculated from the t_1/2

values. The data obtained show the energy barrier to

increase in the ligand order meso-2,3-dppb < dppe <

rac-2,3-dppb (Table 8). The activation barriers for

insertion reported for the dppe derivative nicely fit

previous data for [Pd(Me)(CO)(dppe)]SO₃CF₃(16.6 kcal

mol^-1at -38 °C).¹⁵

Gen er a tion a n d Ca r bon yla tion of â-Ch ela tes

[P d (CH₂CH₂C(O)Me)(P -P )]SbF ₆. The â-chelate com-

plexes [Pd(CH₂CH₂C(O)Me)(P-P)]SbF₆(P-P ) dppe,

18SbF₆; meso-2,3-dppb, 19SbF₆; rac-2,3-dppb, 20SbF₆)

were prepared as outlined in Scheme 9.¹⁹

After each â-chelate was identified by NMR spectros-

copy, the HPNMR tube was removed from the spec-

trometer and was purged with Ar to remove free ethene.

The tube was cooled to ca. -117 °C, pressurized with

20 bar CO, and then inserted into the probe-head

precooled at -90 °C. The conversion of the â-chelates

to the carbonyl acyl complexes [Pd(CO)(COCH₂CH₂C-

(O)Me)(P-P)]SbF₆(33-35) was followed by variable-

temperature ³¹P{¹H} NMR spectroscopy (Scheme 12).

The formation of the meso-2,3-dppb and rac-2,3-dppb

carbonyl acyl complexes started to occur at -40 °C,

while the dppe derivative was much more stable, as only

The rates of conversion of the â-chelates to the

carbonyl acyl complexes 33-35 were evaluated as half-

life times obtained from the decay (and increase) of the

phosphorus resonances at appropriate temperatures:

-40 °C for the meso-2,3-dppb and rac-2,3-dppb com-

plexes 34 and 35; 20 °C for the dppe complex 33. The

activation barriers for the process could not be calcu-

lated straightforwardly from the t_1/2values as the

reaction rates proved to be dependent on the CO

pressure in the range investigated (15-25 bar). The

observed CO dependence suggests that the rate-limiting

step is related to the opening of the metallacycle by CO

(steps a, b in Scheme 13) rather than to the following

migratory insertion of the alkyl carbonyl complex that

should be independent of the CO pressure (step c).

Remarkably, the highest energy barrier required to

disrupt the metallacycle structure was shown by com-

plex 33, which contains dppe, namely, the ligand that

originates the least efficient catalysts in both MeOH and

CH₂Cl₂.

Attem p ted Deter m in a tion of th e Migr a tor y In -

ser t ion Ba r r ier s of [P d (COMe)(C₂H ₄)(P -P )]P F ₆.

The acyl acetonitrile derivatives 24-26 were prepared

in situ following the procedure illustrated in Scheme 8.

To CD

₂Cl₂solutions of the acyl carbonyl complexes 21-

23 was added 30 µL of CD₃CN at -30 °C. The reaction

mixtures were purged with Ar at -30 °C to remove

excess and coordinated CO. As a result, the acyl aceto-

nitrile complexes were formed quantitatively.

In an attempt to generate acyl ethene complexes of

the general formula [Pd(COMe)(C₂H₄)(P-P)]⁺, the acyl

acetonitrile complexes were reacted with C₂H₄(15 bar)

at -110 °C. In no case, however, was an acyl ethene

intermediate intercepted, the â-chelate complexes being

rapidly and quantitatively formed (Scheme 14), which

is consistent with a slower replacement of MeCN by

C₂H₄as compared to the migratory insertion of the acyl

ethene intermediate. Alternatively, one may suggest

that this ethylene intermediate is in rapid equilibrium

with the acetonitrile complex and that the equilibrium

lies to the left. A quite similar reactivity pattern has

been reported by Brookhart and co-workers for the

reaction of isolated [Pd(Me)(CO)(dppp)]⁺with C₂H₄.¹⁹

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 27

Sch em e 14

compared to meso-2,3-dppb and rac-2,3-dppb, which may

ultimately facilitate Pd-P unfastening.^{2f,3,16a,24,25}In-

deed, in the series of the palladium(II) complexes 4, 5,

and 6, which, by the way, exhibit comparable electron

density at palladium (Table 3) as well as Pd-P dis-

tances (Table 2), the P-Pd-P bite angles cannot play

any role, as they are practically identical within the

three complexes (Table 2).

Indirect support for an inverse correlation between

ligand rigidity and autoionization in diphosphine pal-

ladium(II) complexes is provided by the previous obser-

vation that Pd(OAc)₂(o-dppbe) and Pd(OAc)₂(cis-dppen)

containing the rigid o-bis(diphenylphosphino)benzene

and cis-bis(diphenylphosphino)ethylene ligands, respec-

tively, do not undergo autoionization in MeOH, where,

unlike dppe, they give rise to quite efficient CO/ethene

copolymerization catalysts.⁶Moreover, it is worth men-

tioning that Vrieze and co-workers have demonstrated

that backbone rigidity of dinitrogen ligands disables

dissociation of one ligating nitrogen.²⁴

Sch em e 15

Following a procedure reported by these authors, we

have also tried to determine the activation barriers to

the migratory insertion of [Pd(COMe)(C₂H₄)(P-P)]⁺by

scavenging the chloride ligand from the acyl chloride

derivatives 27, 28, and 29 with AgSbF₆in the presence

of ethene (15 bar) at -120 °C (Scheme 15). All our

attempts were unsuccessful, as only the â-chelates were

seen by NMR spectroscopy at -120 °C, which is con-

sistent with an activation barrier lower than 12 kcal

mol^-1. A ∆G^qvalue of 12.3(1) kcal mol^-1has been

reported by Brookhart for [Pd(COMe)(C₂H₄)(dppp)]-

SbF₆.¹⁹

It is noteworthy that the trend of the activation

barriers of migratory insertion reactions at the metal

center in the Pd^II(dppp) system is in nice agreement

with ours, in particular as regards the barriers for alkyl

carbonyl complexes that, irrespective of the diphosphine,

are higher than those for acyl ethene compounds.¹⁹

Alter n a tin g Cop olym er iza tion in CH₂Cl₂by [P d -

(Me)(MeCN)(P -P ]⁺P r ecu r sor s. The copolymeriza-

tion reactions catalyzed by the Pd(OAc)₂(P-P) precur-

sors in MeOH are clearly biased by the solvent that

selectively damages the dppe catalyst, thus preventing

one from making reliable comparisons of the intrinsic

activities. In MeOH, therefore, there is no way to either

confirm or discard the existence of a ligand effect, in

particular of a meso effect, in the alternating copolym-

erization of CO/olefin by palladium catalysts stabilized

by diphosphine ligands containing two carbon spacers.

Unlike MeOH, dry CH₂Cl₂behaves as an innocent

solvent in the CO/ethene copolymerization. Accordingly,

the results obtained in CH₂Cl₂(batch reactions, in situ

HPNMR experiments, and kinetic studies) can be used

to rationalize both the intrinsic activity and the pro-

ductivity that decrease in the order [Pd(Me)(MeCN)-

Discu ssion

(meso-2,3-dppb)]⁺> [Pd(Me)(MeCN)(rac-2,3-dppb)]⁺

>

Alter n a tin g Cop olym er iza tion in MeOH by P d -

(OAc)₂(P -P ) P r ecu r sor s. The HPNMR experiments

(Figures 4 and 5) show that the much lower productivity

in MeOH of the dppe precursor 4 as compared to both

5 and 6 (Table 4) is substantially determined by the

solvent. Already at room temperature, MeOH rapidly

transforms 4 into the catalytically inactive bis-chelate

7 and palladium acetate, which gives palladium metal

on long standing in solution under copolymerization

reaction conditions. The fact that no apparent degrada-

tion of the meso-2,3-dppb and rac-2,3-dppb catalysts

occurs within 1 h of copolymerization indicates that the

methyl substituents on the ligand backbone inhibit

somehow the formation of bis-chelates species and

unstable palladium acetate.

In the absence of added diphosphine, the formation

of bis-chelates from Pd(OAc)₂(P-P) should proceed via

acetate decoordination (assisted by solvation), followed

by a complex rearrangement involving monodentate

diphosphine ligands. Consistent with this mechanism,

the autoionization of M(OAc)₂(dppe) complexes (M )

Ni,⁷Pd) has been found to be favored by a high dielectric

constant of the solvent as well as a small P-Pd-P bite

angle; that is, the M(OAc)₂(dppp) complexes do not

undergo ionization even in MeOH.⁷Within this context,

we suggest that the propensity to autoionization of 4 is

related to the higher skeletal flexibility of dppe as

[Pd(Me)(MeCN)(dppe)]⁺(Table 5). Since the barriers to

migratory insertion of the methyl carbonyl complexes

follow a different order ([Pd(Me)(CO)(rac-2,3-dppb)]⁺>

[Pd(Me)(CO)(dppe)]⁺> [Pd(Me)(CO)(meso-2,3-dppb)]⁺),

they should not control the propagation rate, which is

consistent with previous experimental studies.^2,15In

contrast, it is very likely that the opening/CO migratory

insertion of the â-chelates constitutes the rate-limiting

step or a step close to it in the catalytic cycle. Indeed,

the HPNMR experiments in catalytic conditions show

the â-chelates to be resting states of the catalysis

(Figures 7 and 8), which is again in agreement with

previous studies.^2,13b,26Although the activation barriers

relative to the opening/carbonylation of the â-chelates

could not be calculated due to the complex dependence

on CO pressure (Scheme 13), there is little doubt that

dppe forms the most stable â-chelate (catalyst resting

state), while meso-2,3-dppb forms the least stable â-che-

late, which accounts for the observed intrinsic activity

trend. Within this context, it is worth mentioning that

an inverse relationship between catalytic activity and

(24) Van Asselt, R.; Gielens, E. G. C.; Rulke, E. R.; Vrieze, K.;

Elsevier, C. J . J . Am. Chem. Soc. 1994, 116, 977.

(25) Xu, F. Y.; Zhao, A. X.; Chien, J . C. W. Macromol. Chem. 1993,

194, 2579.

(26) Mul, W. P.; Oosterbeek, H.; Beitel, G. A.; Kramer, G.-J .; Drent,

E. Angew. Chem., Int. Ed. 2000, 39, 1848.

28 Organometallics, Vol. 21, No. 1, 2002

Bianchini et al.

Sch em e 16

Sch em e 17

â-chelate stability has been previously proposed by

Barron to account for the higher catalytic activity of

[PdCl(COBu^t)(dppp)] vs [PdCl(COBu^t)(dppe)] in the

alternating copolymerization of CO/ethene in CH₂Cl₂.^16a

Why the rigid diphosphines meso- and rac-2,3-dppb

form â-chelate complexes that are less kinetically stable

than that with dppe is hard to state in the absence of a

detailed kinetic study as well as a theoretical analysis

of the energy profile associated with the rupture of the

â-ketoalkyl ring by CO in the dppe, meso-2,3-dppb, and

rac-2,3-dppb complexes. We are inclined to ascribe the

easier opening of the â-chelates with meso- and rac-2,3-

dppb to their greater backbone rigidity that would

destabilize the five-coordinate transition state for the

turnover-limiting step.²⁷

ketone increases with the water concentration in CH₂-

Cl₂, compounds A and B are presumably in equilibrium

with each other; in contrast, we can neither verify nor

exclude that the â-chelate is in equilibrium with A.

Indeed, neither A nor B was seen by NMR spectroscopy

in experiments performed in dry and wet CD₂Cl₂in the

temperature range from -60 to 30 °C.

Con clu sion

The productivity of a catalyst and its intrinsic activity

are terms frequently misused or confused for each other,

although they signify different concepts. The catalytic

productivity is related to the amount of product ob-

tained, and as such, it is determined by a complex web

of factors, the most important of which is the catalyst

durability in the chosen experimental conditions. In

contrast, the intrinsic activity is related to the rate of

catalyst formation (initiation step) and to the rate of a

single catalytic cycle. The intrinsic activity is therefore

an indirect measure of the activation barrier involved

in the rate-limiting step.

We have shown here that the productivity of Pd-

(OAc)₂(P-P) precursors (P-P ) dppe, meso-2,3-dppb,

rac-2,3-dppb) in the alternating CO/ethene copolymer-

ization in MeOH is dramatically affected by the solvent

which selectively damages the dppe-based catalyst. This

does not happen in CH₂Cl₂, which behaves as an

innocent solvent and therefore allows for the evaluation

of the intrinsic activity of the palladium catalysts

investigated. Batch catalytic reactions, high-pressure

NMR experiments in catalytic conditions, and the

determination of activation barriers of migratory inser-

tions agree to indicate the relative kinetic stability of

the â-chelate ring in [Pd(CH₂CH₂C(O)Me)(P-P)]⁺as the

factor that controls the copolymerization rate in CH₂-

Cl₂by alkyl palladium(II) precursors. A study in progress

in this laboratory confirms the existence of a similar

intrinsic activity/â-chelate stability relationship also for

palladium catalysts containing the diphosphine ligands

dppp, rac-2,4-bdpp, and meso-2,4-bdpp.³⁰

Besides Consiglio⁵and ourselves,^4,6a relation between

productivity in CO/ethene polyketone and backbone

rigidity of the chelating diphosphine has also been

observed by Doherty for reactions catalyzed in MeOH

by palladium(II) complexes containing bis(phospholyl)

ligands.²⁸

A directly proportional correlation between chelate

ligand rigidity and polymer productivity has been also

reported by Keim and Schultz for the homogeneous

polymerization of ethene by nickel(II)-(P-O) cata-

lysts.²⁹

Ch a in -Tr a n sfer by P r oton olysis. Termination by

protonolysis in Pd-diphosphine-catalyzed CO/ethene

copolymerizations has been demonstrated by Van Leeu-

wen to occur via a mechanism involving the conversion

of â-chelates [Pd(CH₂CH₂C(O)Me)(P-P)]⁺to palladium

enolates B via PdH(η²-vinyl ketone) complexes (A)

(Scheme 16).^18c

For P-P ) dppp or 1,1′-diphenylphosphino ferrocene,

the enolate B is a long-living species that undergoes

chain transfer by protonolysis with MeOH by regiose-

lective proton delivery at the C₂carbon atom to give Pd-

OMe and methyl ethyl ketone.

In light of the HPNMR experiments in dry and wet

CD₂Cl₂, it is apparent that also the â-chelates with dppe,

meso-2,3-dppb, and rac-2,3-dppb undergo chain transfer

by the van Leeuwen mechanism (Scheme 17).

The production of both methyl ethyl ketone and µ-OH

complexes in wet CH₂Cl₂does imply the formation of

enolates of type B. In turn, the production of methyl

vinyl ketone in dry CH₂Cl₂implies the intermediacy of

hydride η²-olefin complexes of type A. Since we have

observed that the rate of production of methyl ethyl

The â-chelate complexes [Pd(CH₂CH₂C(O)Me)(P-P)]-

PF₆with diphosphine ligands containing two carbon

atoms between the phosphorus donors have been iso-

lated for the first time^18aand employed to study the

chain transfer by protonolysis, which proceeds via the

enolate mechanism reported by van Leeuwen for the

dppp-based catalysts.^18cIt has been shown that µ-OH

(27) Shultz, C. S.; DeSimone, J . M.; Brookhart, M. Organometallics

2001, 20, 16.

(28) Doherty, S.; Eastman, G. R.; Tooze, R. P.; Scanlan, T. H.;

Williams, D.; Elsegood, M. R. J .; Clegg, W. Organometallics 1999, 18,

3558.

2+

complexes of the formula [Pd(OH)(P-P)]₂are active

(30) Bianchini, C.; Lee, H. M.; Meli, A.; Oberhauser, W.; Vizza, F.

Manuscript in preparation.

(29) Keim, W.; Schultz, R. P. J . Mol. Catal. 1994, 92, 21.

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 29

m eso-(CH₃CHP P h ₂)₂(m eso-2,3-d p p b). A 60 mL sample

of a THF solution of 5.7 g of meso-(CH₃CH(OMs))₂(23.23 mmol)

was slowly added to 100 mL of a THF solution of KPPh₂

dioxane adduct (22 g, 55.00 mmol) at 0 °C. The mixture was

allowed to stir overnight at room temperature. The solvent

was removed completely under vacuum. Then 100 mL of water

was added to give a white solid. The solid was filtered, washed

with petroleum ether, and dried under vacuum overnight.

Yield: 85%. Anal. Calcd for C₂₈H₂₈P₂: C, 78.86; H, 6.62.

Found: C, 78.15; H, 6.51. ³¹P{¹H} NMR (CD₂Cl₂): δ -7.2 (s).¹H

NMR (CD₂Cl₂): δ 1.1, (pseudo quartet, J (HP) ) J (HH) ) 7.0

Hz, 6H, CHCH₃), 2.6 (m, 2H, CHCH₃), 7.2-7.5 (m, 20H, Ph).

r a c-(CH₃CHP P h ₂)₂(r a c-2,3-d p p b). The compound was

prepared in a similar manner as meso-2,3-dppb. Anal. Calcd

for C₂₈H₂₈P₂: C, 78.86; H, 6.62. Found: C, 78.42; H, 6.58. ³¹P-

catalyst precursors for the alternating CO/ethene copo-

lymerization.

Exp er im en ta l Section

All reactions and manipulations were carried out under an

atmosphere of nitrogen by using Schlenk-type techniques.

Diethyl ether and tetrahydrofuran (THF) were distilled from

LiAlH₄. Reagent grade methanol or CH₂Cl₂, freshly distilled

from CaH₂, was used in the copolymerization reactions. All

the other reagents and solvents were used as purchased from

Aldrich, Fluka, or Strem. The palladium complexes PdCl(Me)-

(COD) (COD ) cycloocta-1,5-diene),³¹PdCl₂(dppe),¹⁵[PdCl-

(Me)(dppe)],¹⁵and [Pd(Me)(MeCN)(dppe)]PF₆¹⁵were prepared

following literature methods. All of the isolated solid samples

were collected on sintered-glass frits and washed with ap-

propriate solvents before being dried under a stream of

nitrogen. Copolymerization reactions were performed with a

250 mL stainless steel autoclave, constructed at the ISSECC-

CNR (Firenze, Italy), equipped with a magnetic drive stirrer

and a Parr 4842 temperature and pressure controller. The

autoclave was connected to a gas reservoir to maintain a

constant pressure over all the catalytic reactions. Deuterated

solvents for NMR measurements, except as stated otherwise,

were dried over molecular sieves. ¹H, ¹³C{¹H}, and ³¹P{¹H}

NMR spectra were obtained on either a Bruker ACP 200

(200.13, 50.32, and 81.01 MHz, respectively) or a Bruker DPX

300 spectrometer (300.13, 75.49, and 121.51 MHz, respec-

tively). All chemical shifts are reported in ppm (δ) relative to

tetramethylsilane, referenced to the chemical shifts of residual

solvent resonances (¹H, ¹³C) or 85% H₃PO₄(³¹P). The 10 mm

sapphire NMR tube was purchased from Saphikon, Milford,

NH, while the titanium high-pressure charging head was

constructed at the ISSECC-CNR (Firenze, Italy).³²Caution:

Since high gas pressures are involved, safety precautions must

be taken at all stages of studies involving high-pressure NMR

tubes. Elemental analyses were performed using a Carlo Erba

Model 1106 elemental analyzer. Infrared spectra were recorded

on a Perkin-Elmer 1600 Series FT-IR spectrophotometer.

Molecular weights of selected polyketone materials were

measured on a Waters gel permeation chromatograph equipped

with a differential refractometer. Either pure HFIP or a

mixture of trichlorobenzene/phenol was used as the solvent,

and polystyrene standards were used to calibrate the instru-

ment. Viscosity measurements (LVN) were carried out in

m-cresol at 60 °C in a standard capillary viscosity-measuring

device. Melting points were determined in glass capillaries

under air.

m eso-(CH₃CH(OMs))₂. Methanesulfonyl chloride (18 mL,

0.23 mol) was added dropwise to a CH₂Cl₂solution (50 mL) of

meso-2,3-butanediol (10.0 g, 0.11 mol) and NEt₃(33.5 mL, 0.24

mol). The solution was allowed to stir overnight. The organic

layer was washed with water (2 × 100 mL), separated, and

concentrated to dryness under reduced pressure. The crude

solid residue was recrystallized with CH₂Cl₂and petroleum

ether to give a white solid. Yield: 71%. ¹H NMR (CDCl₃): δ

1.4 (d, J (HH) ) 6.5 Hz, 6 H, CHCH₃), 3.1 (s, 6 H, SO₂CH₃),

4.9 (m, 2 H, CH).

r a c-(CH₃CH(OMs))₂. A pure sample of rac-(CH₃CH(OMs))₂

was obtained by fractional recrystallization of an isomeric

mixture of (CH₃CH(OMs))₂from CH₂Cl₂and ethanol. Large

crystals of rac-(CH₃CH(OMs))₂separated first. Yield: 10%. ¹H

NMR (CDCl₃): δ 1.5 (d, J (HH) ) 5.5 Hz, 6 H, CHCH₃), 3.1 (s,

6 H, SO₂CH₃), 4.8 (m, 2 H, CH). The preparation of the

isomeric mixture of (CH₃CH(OMs))₂was similar to that of

meso-(CH₃CH(OMs))₂and started with 40.0 g of a isomeric

mixture of 2,3-butanediol (0.444 mol). Yield: 83%.

1

{¹H} NMR (CD₂Cl₂): δ -9.4 (s). H NMR (CD₂Cl₂): δ 1.2 (dd,

J (HP) ) 14.0 Hz, J (HH) ) 7.0 Hz, 6 H, CHCH₃), 2.4 (m, 2 H,

CHCH₃), 7.2-7.4 (m, 20H, Ph).

P d Cl₂(m eso-2,3-d p p b). A mixture of 400 mg of meso-2,3-

dppb (0.939 mmol) and 167 mg of PdCl₂(0.939 mmol) in 20

mL of N,N-dimethylformamide (DMF) solution was stirred for

12 h. During this time, a yellow solid precipitated. The solution

was first concentrated to ca. 4 mL under reduced pressure,

and then 30 mL of petroleum ether was slowly added. The

formed solid was filtered off, washed with petroleum ether,

and dried in a stream of nitrogen. Yield: 85%. Anal. Calcd for

C

₂₈H₂₈Cl₂P₂Pd: C, 55.70; H, 4.67. Found: C, 55.23; H, 4.57.

1

³¹P{¹H} NMR (CD₂Cl₂): δ 69.5 (s). H NMR (CD₂Cl₂): δ 0.98

(dd, J (HP) ) 15.2 Hz, J (HH) ) 7.1 Hz, 6H, CHCH₃), 3.00 (m,

2H, CHCH₃), 7.45-7.63 and 7.78-8.02 (m, 20H, Ph).

P d Cl₂(r a c-2,3-d p p b). A mixture of 300 mg of rac-2,3-dppb

(0.704 mmol) and 125 mg of Pd(II) chloride (0.704 mmol) in

20 mL of DMF was stirred for 12 h. During the reaction, a

pink solid precipitated. The solution was concentrated to ca.

4 mL under reduced pressure, and then 30 mL of petroleum

ether was slowly added. The precipitate was filtered off.

Recrystallization from CH₂Cl₂gave a white solid. Yield: 63%.

Anal. Calcd for C₂₈H₂₈Cl₂P₂Pd: C, 55.70; H, 4.67. Found: C,

55.52; H, 4.71. ³¹P{¹H} NMR (CD₂Cl₂): δ 66.2 (s). ¹H NMR

(CD₂Cl₂): δ 1.05 (dd, J (HP) ) 13.1 Hz, J (HH) ) 6.3 Hz, 6H,

CHCH₃), 2.43 (m, 2H, CHCH₃), 7.52-7.83 and 7.90-8.01 (m,

20H, Ph).

P d (OAc)₂(d p p e) (4). To 10 mL of a methanol solution of

169 mg of Pd(II) acetate (0.753 mmol) was added a CH₂Cl₂

solution of 300 mg of dppe (0.753 mmol). The orange-red

mixture was allowed to stir for 20 min. Then the volume was

reduced to ca. 1 mL under reduced pressure. Petroleum ether

(30 mL) was slowly added to give an orange-red solid of 4,

which was filtered, washed with petroleum ether, and dried

in a stream of nitrogen. Yield: 35%. Anal. Calcd for C₃₀H₃₀O₄P₂-

Pd: C, 57.84; H, 4.85. Found: C, 57.62; H, 4.78. ³¹P{¹H} NMR

(CD₂Cl₂): δ 58.9 (s). ¹H NMR (CD₂Cl₂): δ 1.66 (s, 6H, CO₂-

CH₃), 2.23 (m, 4H, CH₂), 7.52-7.83 (m, 20H, Ph). ¹³C{¹H} NMR

(CD₂Cl₂): δ 23.5 (s, CO₂CH₃), 27.5 (AXX′ system, m, CH₂),

aromatic carbons [129.6 (s), 131.4 (s), 132.8 (s), 134.1 (s)], 176.6

(s, CO₂CH₃). Crystals of 4^.CH₂Cl₂suitable for an X-ray

structure analysis were obtained by slow crystallization of 4

from CH₂Cl₂/petroleum ether under nitrogen at room temper-

ature.

P d (OAc)₂(m eso-2,3-d p p b) (5). To a 20 mL CH₂Cl₂solution

of 280 mg of PdCl₂(meso-2,3-dppb) (0.464 mmol) was added

194 mg of silver acetate (1.16 mmol). After 5 h stirring, the

mixture was passed through a column of Celite to remove AgCl

formed. The solution was concentrated to ca. 2 mL under

vacuum. Then 30 mL of diethyl ether was added at 0 °C to

give 5 as a pale yellow solid. Yield: 70%. Anal. Calcd for

C

₃₂H₃₄O₄P₂Pd: C, 59.04; H, 5.26. Found: C, 58.95; H, 5.31.

³¹P{¹H} NMR (CD₂Cl₂): 20 °C, δ 61.8 (s); -80 °C, δ 65.6 (br

s), 60.4 (br s); coalescence temperature at -70 °C. ¹H NMR

(31) Ruelke, R. E.; Han, I. M.; Elsevier: C. J .; Vrieze, K.; van

Leeuwen, P. W. N. M.; Roobeek, C. F.; Zoutberg, M. C.; Wang, Y. F.;

Stam, C. H. Inorg. Chim. Acta 1990, 169, 5.

3

(CD₂Cl₂): δ 1.55 (s, 6H, CO₂CH₃), 1.05 (dd, J (HH) ) 7.0 Hz,

³J (HP) ) 15.0 Hz, 6H, CHCH₃), 2.83 (m, 2H, CHCH₃), 7.4-

30 Organometallics, Vol. 21, No. 1, 2002

Bianchini et al.

1

8.1 (m, 20H, aromatic protons). ¹³C{¹H} NMR (CD₂Cl₂): δ 24.0

(s, CO₂CH₃), 12.0 (s, CHCH₃), 39.5 (AXX′ system, m, CHCH₃),

aromatic carbons [129.5 (s), 132.6 (s), 133.9 (s), 135.8 (s)], 177.0

(s, CO₂CH₃). Crystals suitable for an X-ray structure analysis

were obtained by slow crystallization of 5 from CH₂Cl₂/diethyl

ether solutions under nitrogen at room temperature.

(d, J (PP) ) 38.0 Hz), 38.8 (d). H NMR (CD₂Cl₂): δ 0.55 (dd,

³J (HP) ) 3.0 Hz, ³J (HP) ) 8.1 Hz, 3H, PdCH₃), 1.05 (dd,

³J (HH) ) 7.3 Hz, ³J (HP) ) 12.4 Hz, 3H, CHCH₃), 1.10 (dd,

³J (HH) ) 7.0 Hz, ³J (HP) ) 13.2 Hz, 3H, CHCH₃), 2.05 (m,

1H, CHCH₃), 2.51 (m, 1H, CHCH₃), 7.3-8.1 (m, 20H, aromatic

protons).

P d (OAc)₂(r a c-2,3-d p p b) (6). This complex was obtained

as a white solid in 70% yield following a procedure analogous

to that used for 5 starting from PdCl₂(rac-2,3-dppb). Anal.

Calcd for C₃₂H₃₄O₄P₂Pd: C, 59.04; H, 5.26. Found: C, 58.79;

H, 5.21. ³¹P{¹H} NMR (CD₂Cl₂, 20/-80 °C): δ 60.5 (s). ¹H NMR

(CD₂Cl₂): δ 1.48 (s, 6H, CO₂CH₃), 1.04 (m, 6H, CHCH₃), 2.31

(br s, 2H, CHCH₃), 7.57-8.15 (m, 20H, aromatic protons). ¹³C-

{¹H} NMR (CD₂Cl₂): δ 24.0 (s, CO₂CH₃), 14.0 (AXX′ system,

[P d (Me)(MeCN)(P -P )]P F ₆(P -P ) m eso-2,3-d p p b, 11;

r a c-2,3-d p p b, 12). A solid sample of AgPF₆(0.13 g, 0.50 mmol)

was added to a magnetically stirred solution of the appropriate

PdCl(Me)(P-P) complex (0.45 mmol) in a 10:1, v/v mixture of

CH₂Cl₂/MeCN. After 30 min stirring at room temperature,

AgCl was removed by filtration on Celite. The resultant

colorless solution was concentrated to ca. 2 mL under reduced

pressure. Addition of a 1:10 v/v mixture of diethyl ether/n-

hexane (20 mL) led to the precipitation of a brownish solid,

which was filtered off, washed with petroleum ether, and dried

in a nitrogen stream. Yield: 50-55%. 11. Anal. Calcd for

1

m, ²J (CP) + J (CP) ) 14.7 Hz, CHCH₃), 36.4 (AXX′ system,

m, CHCH₃), aromatic carbons [129.6 (s), 132.0 (s), 133.5 (s),

137.4 (s)], 176.0 (s, CO₂CH₃). Crystals of 6‚H₂O suitable for

an X-ray structure analysis were obtained by slow crystalliza-

tion of 6 from CH₂Cl₂/diethyl ether under nitrogen at room

temperature.

C

₃₁H₃₄F₆NP₃Pd: C, 50.73; H, 4.67; N, 1.91. Found: C, 50.54;

H, 4.59; N, 1.82. ³¹P{¹H} NMR (CD₂Cl₂): δ 64.8 (d, J (PP) )

34.0 Hz), 42.7 (d). ¹H NMR (CD₂Cl₂): δ 0.57 (dd, ³J (HP) ) 1.9

[P d (d p p e)₂](OAc)₂(7). To a stirred CH₂Cl₂solution (5 mL)

of dppe (0.60 g, 1.5 mmol) was added a solution of Pd(II)

acetate (0.17 g, 0.75 mmol) in MeOH (20 mL) at room

temperature. After 20 min, the volume was reduced to ca. 5

mL under reduced pressure. Petroleum ether (10 mL) was

slowly added to give 7 as a red solid that was filtered off,

washed with petroleum ether, and dried in a stream of

nitrogen. Yield: 75%. Anal. Calcd for C₅₆H₅₄O₄P₄Pd: C, 65.86;

H, 5.33. Found: C, 66.14; H, 5.25.

3

Hz, J (HP) ) 7.3 Hz, 3H, PdCH₃), 0.85 (dd, J (HH) ) 7.3 Hz,

³J (HP) ) 14.6 Hz, 3H, CHCH₃), 1.14 (dd, ³J (HH) ) 7.1 Hz,

³J (HP) ) 14.3 Hz, 3H, CHCH₃), 2.21 (s, 3H, MeCN), 2.80 (m,

1H, CHCH₃), 2.90 (m, 1H, CHCH₃), 7.3-7.9 (m, 20H, aromatic

protons). IR (Nujol mull, KBr plates): (MeCN) 2319, 2287

cm^-1. 12. Anal. Calcd for C₃₁H₃₄F₆NP₃Pd: C, 50.73; H, 4.67;

N, 1.91. Found: C, 50.81; H, 4.71; N, 1.83. ³¹P{¹H} NMR (CD₂-

1

Cl₂): δ 63.5 (d, J (PP) ) 40.7 Hz), 42.8 (d). H NMR (CD₂Cl₂):

δ 0.39 (dd, ³J (HP) ) 2.1 Hz, ³J (HP) ) 7.5 Hz, 3H, PdCH₃),

1.09 (dd, ³J (HH) ) 6.3 Hz, ³J (HP) ) 13.2 Hz, 6H, CHCH₃),

2.15 (s, 3H, MeCN), 2.50 (m, 2H, CHCH₃), 7.4-78.0 (m, 20H,

aromatic protons). IR (Nujol mull, KBr plates): (MeCN) 2319,

[P d (d p p e)₂](P F ₆)₂(7‚P F ₆). To a stirred solution of PdCl₂-

(dppe) (0.29 g, 0.5 mmol) and dppe (0.20 g, 0.5 mmol) in 20

mL of CH₂Cl₂was added 2 equiv of AgPF₆(1 mmol). After 30

min, the mixture was passed through a column of Celite to

remove the formed AgCl. The solution was concentrated to ca.

5 mL under vacuum. Addition of diethyl ether led to the

precipitation of 7^.PF₆in 85% yield. Anal. Calcd for C₅₂H₄₈F₁₂P₆-

Pd: C, 52.34; H, 4.05. Found: C, 52.13; H, 4.02.

2287 cm^-1

.

[P d (OH)(P -P )]₂(P F ₆)₂(P -P ) d p p e, 16; m eso-2,3-d p p b,

cis-/tr a n s-17). Solid AgPF₆(0.18 g, 0.72 mmol) was added to

a CH₂Cl₂(30 mL) solution of the appropriate PdCl₂(P-P)

complex (0.33 mmol) at room temperature. After AgCl was

removed by filtration, KOBu^t(41 mg, 0.33 mmol) dissolved in

1 mL of water was added under vigorous stirring. After 30

min, the solvent was removed under reduced pressure and the

solid residue was treated with 20 mL of CH₂Cl₂. The CH₂Cl₂

phase was passed through Celite, and the resulting clear

yellow solution was concentrated to 5 mL under reduced

pressure. Addition of a 1:1 mixture of diethyl ether and

n-hexane led to the precipitation of a yellow microcrystalline

solid, which was filtered off, washed with diethyl ether, and

dried in a stream of nitrogen. Yield: 40-60%. 16. Anal. Calcd

for C₅₂H₅₀F₁₂O₂P₆Pd₂: C, 46.83; H, 3.78. Found: C, 46.95; H,

Au t oion iza t ion E q u ilib r iu m b et w een 4 a n d 7. NMR

Exp er im en t. A 5 mm NMR tube was charged with a solution

of 4 (12 mg, 1.8 × 10^-2mmol) in MeOH-d₄(1 mL) at room

temperature under nitrogen. ³¹P{¹H} and ¹H NMR spectra

showed the large (80%) conversion of 4 into 7. Subsequently,

the solvent was evaporated under vacuum, and the remaining

1

residue was dissolved in 1 mL of CD₂Cl₂. ³¹P{¹H} and H NMR

spectra showed that 4 had re-formed quantitatively.

P d (CN)₂(P -P ) (P -P ) d p p e, m eso-2,3-d p p b , r a c-2,3-

d p p b). To 0.126 mmol of diphosphine in 5 mL of DMF was

added 0.126 mmol of Pd(CN)₂. The resulting clear yellow

solution was allowed to stir overnight. Slow addition of 30 mL

of diethyl ether gave a colorless solid that was filtered off,

washed with diethyl ether, and dried under vacuum overnight.

³¹P{¹H} NMR and selected IR data for the cyanide complexes

are reported in Table 3.

3.68. IR (Nujol mull, KBr plates): (ν_OH) 3591 cm^{-1 31}P-

.

{¹H}NMR (CD₂Cl₂): δ 64.2 (s). ¹H NMR (CD₂Cl₂): δ -1.69 (s,

OH). cis-/tr a n s-17. Anal. Calcd for C₅₆H₅₈F₁₂O₂P₆Pd₂: C,

48.40; H, 4.21. Found: C, 48.00; H, 4.11. IR (Nujol mull, KBr

plates): (ν_OH) 3595 cm^-1

δ₂68.2 (s) in a ratio of 85:15. H NMR (CD₂Cl₂): δ₁+δ₂-1.71

.

³¹P{¹H}NMR (CD₂Cl₂): δ₁67.7 (s),

P d Cl(Me)(P -P ) (P -P ) m eso-2,3-d p p b, r a c-2,3-d p p b).

To a stirred solution of PdCl(Me)(COD) (0.13 g, 0.50 mmol) in

THF (20 mL) was added 1 equiv of diphosphine (0.50 mmol)

at room temperature. Almost immediately an off-white product

began to precipitate. After 20 min, diethyl ether was added to

the reaction mixture to complete the precipitation of the

product, which was filtered off, washed with petroleum ether,

and dried in a nitrogen stream. Yield: 85-95%. P d Cl(Me)-

(m eso-2,3-d p p b). Anal. Calcd for C₂₉H₃₁ClP₂Pd: C, 59.71; H,

5.36. Found: C, 59.43; H, 5.31. ³¹P{¹H} NMR (CD₂Cl₂): δ 62.8

1

(s, OH).

[P d (CH₂CH₂C(O)Me)(P -P )]P F ₆(P -P ) d p p e, 18; m eso-

2,3-d p p b, 19; r a c-2,3-d p p b, 20). In a typical experiment, a

solution of 0.035 mmol of the methyl acetonitrile complex 10

(11 or 12) in 2 mL of CD₂Cl₂was transferred into a 10 mm

sapphire HPNMR tube and pressurized to 15 bar with CO at

room temperature. ³¹P{¹H} and ¹H NMR spectra recorded at

room temperature showed the quantitative conversion of 10

into the corresponding acyl carbonyl derivative 21 (22 or 23)

(selected ³¹P{¹H} and ¹H NMR data are reported in Tables 6

and 7, respectively). After the tube was immersed into a cooling

bath at -40 °C and depressurized to ambient pressure 20 µL

of MeCN was added by a syringe under nitrogen. ³¹P{¹H} and

¹H NMR spectra recorded at -40 °C showed the quantitative

conversion of the acyl carbonyl derivatives into the corre-

1

(d, J (PP) ) 32.0 Hz), 36.8 (d). H NMR (CD₂Cl₂): δ 0.55 (dd,

³J (HP) ) 3.1 Hz, ³J (HP) ) 8.2 Hz, 3H, PdCH₃), 0.70 (dd,

³J (HH) ) 7.3 Hz, ³J (HP) ) 14.4 Hz, 3H, CHCH₃), 1.21 (dd,

³J (HH) ) 6.9 Hz, ³J (HP) ) 13.0 Hz, 3H, CHCH₃), 2.71 (m,

2H, CHCH₃), 7.3-8.1 (m, 20H, aromatic protons). P d Cl(Me)-

(r a c-2,3-d p p b). Anal. Calcd for C₂₉H₃₁ClP₂Pd: C, 59.71; H,

5.36. Found: C, 59.54; H, 5.30. ³¹P{¹H} NMR (CD₂Cl₂): δ 63.8

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 31

sponding acyl acetonitrile derivatives [Pd(COMe)(MeCN)(P-

employed in the above-reported CO/C₂H₄copolymerization.

Propene (20 g) was introduced into the autoclave after the

catalyst solution with the autoclave cooled at 4 °C. The

autoclave was charged with 20 bar of CO and 20 bar of C₂H₄.

After the contents of the autoclave had been brought to 85 °C,

the pressure was then maintained at ca. 55 bar by introducing

under pressure a 1:1 CO/C₂H₄mixture. Anal. Calcd for

(COCH₂CH₂)_n: C, 64.3; H, 7.2. Found: C, 64.1; H, 7.1. IR

(powder sample in KBr pellet): 3391 (w), 2912 (m), 1694 (vs),

1

P)]PF₆(24-26) (selected ³¹P{¹H} and H NMR data for these

complexes are reported in Tables 6 and 7, respectively).

Bubbling ethene for 2 min into the solutions of the latter

complexes maintained at -40 °C led to the quantitative

formation of the â-chelate complexes. The final solution was

transferred from the HPNMR tube into a flask immersed in a

bath at -20 °C. Evaporation of the solvent under reduced

pressure gave a solid product, which was washed with n-

1408 (s), 1333 (s), 1259 (m), 1056 (s), 811 (m), 592 (m) cm^-1

.

1

pentane and dried (selected ³¹P{¹H} and H NMR data for 18-

¹H NMR (HFIP-d₂): δ 3.73 (s, CH₂CO₂CH₃), 3.05 (m, CH₂-

CHCH₃), 2.81 (s, CH₂COCH₂), 2.55 (q, J (HH) ) 7.4 Hz, COCH₂-

CH₃), 1.15 (d, J (HH) ) 6.5 Hz, CH₂CHCH₃), 1.09 (t, J (HH) )

7.4 Hz, COCH₂CH₃). ¹³C{¹H} NMR (HFIP/C₆H₆, 9:1, v/v): δ

217.1 (COCH₂CH₃), 214.1 (CH₂COCHCH₃), 212.8 (CH₂COCH₂),

176.6 (CH₂CO₂CH₃), 52.6 (CH₂CO₂CH₃), 45.0 (COCH₂CHCH₃),

41.3 (COCH₂CHCH₃), 35.7 (CH₂COCH₂), 34.5 (CH₂COCHCH₃),

27.9 (CH₂CO₂CH₃), 15.7 (CH₂COCHCH₃), 6.5 (COCH₂CH₃).

20 are reported in Tables 6 and 7, respectively). 18. Anal. Calcd

for C₃₀H₃₁F₆OP₃Pd: C, 49.98; H, 4.33. Found: C, 49.55; H,

4.25. IR (Nujol mull, KBr plates): (νCdO) 1624 cm^-1. 19. Anal.

Calcd for C₃₂H₃₅F₆OP₃Pd: C, 51.32; H, 4.71. Found: C, 50.99;

H, 4.65. IR (Nujol mull, KBr plates): (νCdO) 1621 cm^-1. 20.

Anal. Calcd for C₃₂H₃₅F₆OP₃Pd: C, 51.32; H, 4.71. Found: C,

51.11; H, 4.61. IR (Nujol mull, KBr plates): (νCdO) 1622 cm^-1

.

Cop olym er iza t ion of E t h en e a n d Ca r b on Mon oxid e

Ca ta lyzed by P d (OTs)₂(P -P ) in MeOH. A. Au tocla ve

Exp er im en ts. MeOH (100 mL) was introduced by suction into

a 250 mL autoclave containing 86.5 mg of BQ (0.8 mmol), 38.0

mg of TsOH (0.2 mmol), and 0.01 mmol of catalyst precursor

previously evacuated by a vacuum pump. The autoclave was

charged with a 1:1 CO/C₂H₄mixture to 40 bar at room

temperature. After the contents of the autoclave had been

brought to 85 °C, the pressure was then maintained at ca. 45

bar by continuous feeding of an equimolar mixture of CO and

C₂H₄from a gas reservoir. The reaction mixture was then

stirred (1400 rpm) for the required reaction time. The reaction

was stopped by cooling the autoclave to room temperature by

means of an ice-water bath. After the unreacted gases were

released, the formed insoluble ethene-carbon monoxide co-

polymer was filtered off, washed with methanol, and dried in

a vacuum oven at 70 °C overnight. Anal. Calcd for (COCH₂-

CH₂)_n: C, 64.3; H, 7.2. Found: C, 64.1; H, 7.1. IR (powder

sample in KBr pellet): 3391 (w), 2912 (m), 1694 (vs), 1408 (s),

Cop olym er iza t ion of E t h en e a n d Ca r bon Mon oxid e

Ca ta lyzed by th e Meth yl Aceton itr ile Com p lexes 10-12

in CH₂Cl₂. A. Au tocla ve Exp er im en ts. Under a nitrogen

atmosphere, a 100 mL solution of CH₂Cl₂(freshly distilled

under nitrogen from CaH₂) containing the catalyst precursor

(0.01 mmol) was introduced into a 250 mL autoclave equipped

with a magnetic drive stirrer and a temperature and pressure

controller. The autoclave was charged with 20 bar of CO and

20 bar of C₂H₄. After the contents of the autoclave had been

brought to 70 °C, the pressure was maintained at ca. 40 bar

by continuous feeding of an equimolecular mixture of C₂H₄and

CO from a high-pressure gas reservoir connected to the

autoclave. A pressure gauge connected to the gas reservoir was

employed to determine the gas consumption during the reac-

tions. The reaction mixture was then stirred (1400 rpm) for

the desired time. The reaction was stopped by cooling the

autoclave to room temperature using an ice-water bath. After

the unreacted gases were released, the insoluble copolymer

was filtered off, washed with CH₂Cl₂, and dried in a vacuum

oven at 70 °C overnight. Anal. Calcd for (COCH₂CH₂)_n: C, 64.3;

H, 7.2. Found: C, 64.1; H, 7.1. IR (powder sample in KBr

pellet): 3391 (w), 2912 (m), 1694 (vs), 1408 (s), 1333 (s), 1259

1

1333 (s), 1259 (m), 1056 (s), 811 (m), 592 (m) cm^-1. H NMR

(HFIP-d₂): δ 3.73 (s, CH₂CO₂CH₃), 2.83 (s, CH₂COCH₂), 2.55

(q, J (HH) ) 7.4 Hz, COCH₂CH₃), 1.09 (t, J (HH) ) 7.4 Hz,

COCH₂CH₃). ¹³C{¹H} NMR (HFIP/CDCl₃, 9:1, v/v): δ 217.2

(COCH₂CH₃), 213.0 (CH₂COCH₂), 176.6 (CH₂CO₂CH₃), 52.6

(CH₂CO₂CH₃), 36.1 (CH₂COCH₂), 27.9 (CH₂CO₂CH₃), 7.1

(COCH₂CH₃).

(m), 1056 (s), 811 (m), 592 (m) cm^-^{1 1}H NMR (HFIP-d₂): δ

.

2.85 (s, CH₂C(O)CH₂). ¹³C{¹H} NMR (HFIP-d₂): δ 213.6

(CH₂C(O)CH₂), 36.0 (CH₂C(O)CH₂).

A series of reactions were carried in the presence of BQ (0.87

g, 0.8 mmol). The composition and molecular weight of the

polyketone products were analogous to those obtained in the

absence of oxidant.

B. HP NMR Exp er im en ts. The reactions were followed by

variable-temperature ³¹P{¹H} and ¹H NMR spectroscopy.

Details of the procedure are exemplified for 5. A sequence of

selected ³¹P{¹H} NMR spectra is reported in Figure 5. A 10

mm sapphire HPNMR tube was charged with a solution of 5

(12 mg, 1.8 × 10^-2mmol) in MeOH-d₄(2 mL) under nitrogen

and then placed into the NMR probe at room temperature (³¹P-

{¹H} NMR singlet at δ 66.9, trace a). Addition of TsOH (0.09

mmol) and BQ (0.18 mmol) led to the conversion of 5 into Pd-

(OTs)₂(meso-2,3-dppb) (9) (downfield shift of the ³¹P{¹H} NMR

resonance to δ 78.1, trace b). Pressurizing with a 1:1 mixture

of carbon monoxide/ethene to 40 bar at room temperature

caused no change in both the ¹H and ³¹P{¹H} NMR spectra

(trace c). Complex 9 was the only phosphorus-containing

species visible by NMR also during the copolymerization

reaction (1 h at 85 °C). The tube was then allowed to cool to

room temperature. A comparison between the ³¹P{¹H} NMR

spectrum of this sample (trace d) and the spectrum acquired

at room temperature before the catalytic reaction (trace c)

showed no decrease in the intensity of the signal of 9. Once

the tube was removed from the probe-head, the copolymer

appeared as an off-white solid. Appreciable degradation of the

meso-2,3-dppb-derived catalyst to palladium metal was ob-

served only in parallel experiments lasting 3 h at 85 °C.

B. HP NMR Exp er im en ts. Details of the procedure are

exemplified for 11. A 10 mm sapphire HPNMR tube was

charged with a solution of the methyl acetonitrile complex (12

mg, 1.8 × 10^-2mmol) in 2 mL of CD₂Cl₂(freshly distilled from

CaH₂) under nitrogen and then placed into the NMR probe at

room temperature. After ³¹P{¹H} and ¹H NMR spectra were

acquired, the tube was pressurized with a 1:1 mixture of CO/

C₂H₄to 40 bar at room temperature. The reactions were

followed by variable-temperature ³¹P{¹H} and ¹H NMR spec-

troscopy for 30 min at 50 °C and 1 h at 70 °C. Afterward, the

tube was cooled to room temperature, and the internal

pressure was released. Palladium metal was found in the NMR

tube together with the copolymer.

A series of analogous experiments were carried out in the

presence of 10 µL of water.

Cop olym er iza t ion of E t h en e a n d Ca r bon Mon oxid e

Ca ta lyzed by th e µ-OH Com p lexes 16 a n d cis-/tr a n s-17

in CH₂Cl₂. Au tocla ve Exp er im en ts. The catalytic activity

of the µ-OH binuclear complexes was tested in 30 min

experiments following a procedure analogous to that employed

for the methyl acetonitrile complexes.

Ter p olym er iza tion of Eth en e/P r op en e a n d Ca r bon

Mon oxid e Ca ta lyzed by P d (OTs)₂(P -P ) in MeOH. Au to-

cla ve E xp er im en t s. The procedure was similar to that

Th er m a l Sta bility of th e â-Ch ela tes 18 a n d 19 in Dr y

a n d Wet CD₂Cl₂. Details of the procedure are exemplified for

32 Organometallics, Vol. 21, No. 1, 2002

Bianchini et al.

18. A 5 mm NMR tube was charged with a solution of 18 (0.01

mmol) in 1 mL of either dry or wet CD₂Cl₂(10 µL of H₂O) under

nitrogen at room temperature. The reactions were followed by

nitrogen thermostat bath (-117 °C) and charged with 20 bar

of CO. After this procedure, the tube was placed into the probe-

head precooled to -90 °C. ³¹P{¹H} and ¹H NMR spectra showed

that the â-chelate complexes were still present. The probe

temperature was gradually increased, and ³¹P{¹H} and ¹H

NMR spectra were recorded at each intermediate temperature.

Increasing the temperature converted the â-chelate complex

into the carbonyl acyl derivative [Pd(COCH₂CH₂C(O)Me)(CO)-

1

³¹P{¹H} and H NMR spectroscopy by acquiring NMR spectra

every hour. Sequences of selected ³¹P{¹H} NMR spectra are

reported in Figures 11 (dry CD₂Cl₂) and 12 (wet CD₂Cl₂). The

organic products formed during the reactions were detected

by GC and GC/MS analysis of the final reaction solutions.

Experiments carried out at 50 and 70 °C gave identical

results except for faster chemical transformations.

1

(P-P)]SbF₆(selected ³¹P{¹H} and H NMR data of the carbonyl

acyl complexes 33-35 are reported in Tables 6 and 7,

respectively). At the conversion temperature, the decrease in

concentration of the methyl carbonyl complex was followed by

³¹P{¹H} NMR spectroscopy. Spectra were taken at intervals

of 5-10 min, depending on conditions. The reaction was

followed for 2-3 half-lives.

Analogous experiments in the pressure range from 15 to 25

bar showed that the t_1/2values decrease with increasing

pressure.

In Situ Ca r bon yla tion of th e Meth yl Aceton itr ile

Com p lexes 10-12. HP NMR Exp er im en ts. In a typical

experiment, the methyl acetonitrile complex (0.035 mmol) was

dissolved in 2 mL of CD₂Cl₂(distilled over CaH₂under

nitrogen), and the resulting solution was then transferred into

1

a 10 mm sapphire tube. After ³¹P{¹H} and H NMR spectra of

this sample were acquired at room temperature, the NMR

1

probe-head was cooled to -80 °C. ³¹P{¹H} and H NMR spectra

[P d (CH₂CH₂C(O)Me)(P -P )]SbF ₆/[P d (COCH₂CH₂C(O)-

Me)(CO)(P -P )]SbF ₆Equ ilibr ia . Solutions of the carbonyl

acyl complexes 33-35, prepared as reported above in HPNMR

tubes, were immersed in a thermostat bath at -20 °C. CO was

released, and then couples of vacuum-nitrogen cycles were

applied. ³¹P{¹H} and ¹H NMR spectra of these samples

acquired at -20 °C let us follow the transformation of the

carbonyl acyl complexes into the corresponding â-chelates

18SbF₆-20SbF₆.

were recorded at this temperature. The sapphire tube was

removed from the spectrometer, immersed into an ethanol/

liquid nitrogen thermostat bath (-117 °C), and charged with

15 bar of CO. The tube was then placed into the probe-head

precooled to -80 °C. Irrespective of the diphosphine ligand,

the ³¹P{¹H} and ¹H NMR spectra showed the quantitative

conversion of the starting methyl acetonitrile complex into the

corresponding methyl carbonyl derivative [Pd(Me)(CO)(P-P)]-

1

PF₆(selected ³¹P{¹H} and H NMR data of the methyl carbonyl

Attem p ted Deter m in a tion of th e Migr a tor y In ser tion

Ba r r ier s of [P d (COMe)(C₂H₄)(P -P )]BF ₄. CD₂Cl₂solutions

of the carbonyl acyl complexes 21-23, prepared as reported

above in HPNMR tubes, were immersed in a thermostat bath

at -30 °C. CO was released, and then 30 µL of CD₃CN was

added. The reaction mixtures were purged with Ar to remove

excess and coordinated CO. ³¹P{¹H} and ¹H NMR spectra of

these samples showed the quantitative formation of the acyl

acetonitrile derivatives [Pd(COMe)(MeCN)(P-P)]PF₆(24-26)

(selected ³¹P{¹H} and ¹H NMR data are reported in Tables 6

and 7, respectively). The tubes were pressurized with C₂H₄

(15 bar) at -110 °C. In no case were acyl ethene intermediates

of the general formula [Pd(COMe)(C₂H₄)(P-P)]⁺intercepted,

the â-chelate complexes 18-20 being rapidly and quantita-

tively formed.

X-r a y Da ta Collection a n d Str u ctu r e Deter m in a tion

of 4‚CH₂Cl₂, 5, a n d 6‚H₂O. Suitable crystals were sealed in

a glass capillary and transferred to a Enraf-Nonius CAD4

diffractometer. Data were collected at room temperature. A

set of 25 carefully centered reflections having 6.5° e θ e 10.0°

were used to determine the lattice constants. The intensities

of three standard reflections were measured every 2 h for

orientation and intensity control. This procedure revealed no

decay of intensity. The data were corrected for Lorentz and

polarization effects. Atomic scattering factors with anomalous

dispersion correction were taken from X-ray crystallography

tables.³³Absorption correction was applied via ψ scan. The

computational work was performed with a DIGITAL DEC

5000/200 workstation using the program SHELX-93.³⁴The

structures were solved by direct methods using the SIR92

program,³⁵and all of the non-hydrogen atoms were found

through a series of F_oFourier maps. Refinement was done by

full-matrix least-squares calculations, initially with isotropic

thermal parameters and finally with anisotropic thermal

parameters for all the atoms but the hydrogens. The phenyl

complexes 30-32 are reported in Tables 6 and 7, respectively).

The probe temperature was gradually increased, and ³¹P{¹H}

and ¹H NMR spectra were recorded at each intermediate

temperature. Increasing the temperature converted each

methyl carbonyl complex into the corresponding carbonyl acyl

derivative [Pd(COMe)(CO)(P-P)]PF₆(21-23). At the conver-

sion temperature, the decrease in concentration of the methyl

carbonyl complex was followed by ³¹P{¹H} NMR spectroscopy.

Spectra were taken at intervals of 5-10 min, depending on

conditions. Each reaction was followed for 2-3 half-lives.

Experiments were repeated with varying pressures of CO (5-

25) to determine the dependence of the reaction rate on the

CO concentration. The results indicated that the rate of the

reaction is independent of CO. According to first-order kinetics,

the free energy of activation was calculated as follows: ∆G^q)

RT(ln k_r-ln kT/h) with k_r) ln 2/t_1/2

.

Gen er a tion a n d Ca r bon yla tion of th e â-Ch ela te Com -

p lexes 18SbF ₆-20SbF ₆. HP NMR Exp er im en ts. In a typical

experiment, the chloride methyl complex PdCl(Me)(P-P)

(0.035 mmol) was dissolved in 2 mL of deoxygenated CD₂Cl₂,

and the resulting solution was transferred into a 10 mm

sapphire tube and then pressurized to 20 bar of CO at room

temperature. The formation of the chloride acyl complexes

containing dppe, meso-2,3-dppb, or rac-2,3-dppb required

1

heating to 40 °C for 2 h (selected ³¹P{¹H} and H NMR data of

the chloride acyl complexes 27-29 are reported in Tables 6

and 7, respectively). Once the chloride acyl complex was

obtained, the tube was removed from the spectrometer and

the excess CO was released. The tube was then immersed into

a thermostat bath at -20 °C, and nitrogen was bubbled

throughout the solution for 15 min to eliminate any trace of

CO. Ethene was bubbled into the solution maintained at -20

°C for 5 min, and then a solid sample of AgSbF₆(0.08 mmol)

was added to the solution. The immediately formed AgCl

settled on the bottom of the tube. Irrespective of the diphos-

1

phine ligand, the ³¹P{¹H} and H NMR spectra of this sample

(32) Bianchini, C.; Meli, A.; Traversi, A. Ital. Pat. FI A000025, 1997.

(33) (a) Wilson, A. J . C. International Tables for X-Ray Crystal-

lography; Kluwer: Dordrecht, The Netherlands, 1992; p 219. (b)

Wilson, A. J . C. International Tables for X-Ray Crystallography;

Kluwer: Dordrecht, The Netherlands, 1992; p 500.

(34) Sheldrick, G. M. SHELX-93 Program for Crystal Structure

Refinement; University of Go¨ttingen: Go¨ttingen, Germany, 1993.

(35) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;

Burla, M. C.; Polidori, G.; Camalli, M. J . Appl. Crystallogr. 1994, 27,

435.

showed the quantitative conversion of the chloride acyl

complex into the corresponding â-chelate complex [Pd(CH₂-

CH₂C(O)Me)(P-P)]SbF₆. Afterward, the NMR probe-head was

cooled to -90 °C, and ³¹P{¹H} and ¹H NMR spectra were

recorded at this temperature. The sapphire tube was removed

from the probe, and nitrogen was bubbled throughout the

solution at room temperature to eliminate any trace of ethene.

The sapphire tube was immersed into an ethanol/liquid

Palladium-Diphosphine Catalysis

Organometallics, Vol. 21, No. 1, 2002 33

rings were treated as rigid bodies with D_6hsymmetry, and

hydrogen atoms were introduced at calculated positions.

Su p p or tin g In for m a tion Ava ila ble: Tables of bond

lengths and angles and positional and thermal param-

eters for compounds 4‚CH₂Cl₂, 5, and 6‚H₂O. This material

is available free of charge via the Internet at

http://pubs.acs.org.

Ack n ow led gm en t. Thanks are due to the European

Commission for contract no. HPRN CT 2000-00010 and

to COST Action D17 for sponsoring the Working Group

0007/2000.

OM010727B

Article Doi

Ligand and solvent effects in the alternating copolymerization of carbon monoxide and olefins by palladium-diphosphine catalysis

DOI: 10.1021/om010727b

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om010727b

Full text of DOI:10.1021/om010727b

Products guided by the article

R&D Labs maybe for 399039-62-4

Relevant to this article

Hot Product