Selectivity for the methoxycarbonylation of ethylene versus CO-ethylene copolymerization with catalysts based on C4-bridged bidentate phosphines and phospholes

Article abstract of DOI:10.1016/S0022-328X(01)01180-9

The C₄-bridged phospholes 11,12-bis(2,3,4,5-tetramethylphospholylmethyl)-9,10-dihydro-9,10-ethano- anthracene (cis, 4a; trans, 4b) and diphosphines 11,12-bis(diphenylphosphinomethyl)-9,10-dihydro-9,10-ethano-anthracene (cis, 5a; trans, 5b) and their corresponding palladium complexes [(P-P)PdCl₂] (6a-d) have been prepared and characterized. Single-crystal X-ray analyses of 6a-d have been undertaken and they reveal that 4a and b and 5a and b coordinate in a bidentate manner forming seven-membered chelate rings with natural bite angles between 98.62 and 100.30°. The palladium-catalyzed carbonylation of ethylene has been studied using 4a and b and 5a and b. Catalyst mixtures generated from 4a and b, palladium acetate and methanesulfonic acid are selective for the copolymerization of ethylene with carbon monoxide, generating low molecular weight polymers. Surprisingly, the activity of catalyst systems based on cis-(4a) is markedly higher than that based on its trans-isomer, 4b. The marked influence of the nature of the four-carbon tether is highlighted by comparative catalyst testing with [{1,4-bis(2,3,4,5-tetramethylphospholyl)butane}Pd(OAc)₂], which rapidly decomposes under the conditions used for copolymerization. In contrast, under identical conditions catalyst mixtures formed from 5a and b show a marked dependence of the selectivity on the stereochemistry of the ethano-anthracene tether, the former generating a low molecular weight copolymer while the latter generate mainly methyl propanoate. Interestingly, polyketone generated from catalysts based on bisphosphole 4a has a markedly higher average molecular weight than that formed using its diphenylphosphino counterpart, 5a.

Full text of DOI:10.1016/S0022-328X(01)01180-9

Journal of Organometallic Chemistry 640 (2001) 182–196
www.elsevier.com/locate/jorganchem
Selectivity for the methoxycarbonylation of ethylene versus
COꢀethylene copolymerization with catalysts based on C4-bridged
bidentate phosphines and phospholes
Simon Doherty a,*, Edward G. Robins a, Julian G. Knight b,*, Colin R. Newman b,
Barrie Rhodes b, Paul A. Champkin b, William Clegg b
a School of Chemistry, Da6id Keir Building, The Queen’s Uni6ersity of Belfast, Stranmillis Road, Belfast BT4 5AG, UK
b Department of Chemistry, Bedson Building, The Uni6ersity of Newcastle upon Tyne, Newcastle upon Tyne NE1 7RU, UK
Received 4 July 2001; accepted 2 August 2001
Abstract
The C4-bridged phospholes 11,12-bis(2,3,4,5-tetramethylphospholylmethyl)-9,10-dihydro-9,10-ethano-anthracene (cis, 4a; trans,
4b) and diphosphines 11,12-bis(diphenylphosphinomethyl)-9,10-dihydro-9,10-ethano-anthracene (cis, 5a; trans, 5b) and their
corresponding palladium complexes [(PꢀP)PdCl2] (6a–d) have been prepared and characterized. Single-crystal X-ray analyses of
6a–d have been undertaken and they reveal that 4a and b and 5a and b coordinate in a bidentate manner forming
seven-membered chelate rings with natural bite angles between 98.62 and 100.30°. The palladium-catalyzed carbonylation of
ethylene has been studied using 4a and b and 5a and b. Catalyst mixtures generated from 4a and b, palladium acetate and
methanesulfonic acid are selective for the copolymerization of ethylene with carbon monoxide, generating low molecular weight
polymers. Surprisingly, the activity of catalyst systems based on cis-(4a) is markedly higher than that based on its trans-isomer,
4b. The marked influence of the nature of the four-carbon tether is highlighted by comparative catalyst testing with [{1,4-
bis(2,3,4,5-tetramethylphospholyl)butane}Pd(OAc)2], which rapidly decomposes under the conditions used for copolymerization.
In contrast, under identical conditions catalyst mixtures formed from 5a and b show a marked dependence of the selectivity on
the stereochemistry of the ethano-anthracene tether, the former generating a low molecular weight copolymer while the latter
generate mainly methyl propanoate. Interestingly, polyketone generated from catalysts based on bisphosphole 4a has a markedly
higher average molecular weight than that formed using its diphenylphosphino counterpart, 5a. © 2001 Elsevier Science B.V. All
rights reserved.
Keywords: Diphosphines; Copolymerization; Ethylene; Carbon monoxide; Methoxycarbonylation; Selectivity
a-aryl propionic acid derivatives [3], precursors to non-
steroidal anti-inflammatory drugs, and alkoxycarbony-
lation has been used to synthesize oxygenated terpene
derivatives, which find commercial applications in
flavoring compositions and perfumes [4]. When the
substrate is ethylene and the reaction medium
methanol, palladium-catalyzed carbonylation results ei-
ther in the alternating copolymerization of carbon
monoxide and ethylene, to generate polyketone, or
methoxycarbonylation to give methyl propanoate (Eq.
(1)). The alternating copolymerization of olefins with
carbon monoxide has attracted interest because of the
useful properties of polyketone, which has been her-
alded as a new engineering thermoplastic, a tough,
1. Introduction
The palladium-catalyzed carbonylation of unsatu-
rated substrates such as alkenes and alkynes is a power-
ful tool in organic synthesis and has been used to
prepare a range of important compounds [1]. For in-
stance, intramolecular versions of this reaction have
been used to synthesize five- and six-membered g-lac-
tones from the corresponding yne–ol [2]; enantioselec-
tive hydroxycarbonylation has been used to prepare
* Corresponding authors. Fax: +44-28-90665297 (S.D.); +44-
191-2226929 (J.G.K.).
E-mail addresses: s.doherty@qub.ac.uk (S. Doherty), j.g.knight@
newcastle.ac.uk (J.G. Knight).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)01180-9

S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
183
flexible, wear-resistant, high impact strength polymer
which has been developed into commercial products [5],
while methyl propanoate is a potential intermediate in
the manufacture of acrylic esters, large volume chemical
intermediates which are produced annually on a multi-
million tonne scale [6].
2,3-bis(diphenylphosphinomethyl)norbornane deriva-
tives are selective for the production of copolymer while
that based on endo, exo-2,3-bis(diphenylphosphi-
nomethyl)norbornane is selective for methyl propan-
oate [13]. Herein, we report that the bidentate phospho-
les cis- and trans-11,12-bis(2,3,4,5-tetra m e t h y l p h o s-
pholylmethyl)-9,10-dihydro-9,10-ethano-anthracene
form catalysts that are selective for copolymer whereas
the selectivity of their diphenylphosphino counterparts
depends on the stereochemistry of the anthracene
tether, catalysts based on trans-11,12-bis( d i p h e n y l-
phosphinomethyl)-9,10-dihydro-9,10-ethano-anthracene
giving mainly methyl propanoate while cis-11,12-bis-
(diphenylphosphinomethyl)-9,10-dihydro-9,10-ethano-
anthracene is selective for polyketone.
(1)
Since the initial discovery that catalysts based on
palladium acetate and a bidentate phosphine were
highly active and selective for copolymerization, selec-
tivity has commonly been believed to be determined by
ligand denticity, catalysts based on bidentate phosphi-
nes giving high molecular weight copolymer and those
based on monodentate phosphines giving methyl
propanoate. Indeed, a range of systems based on a
palladium or nickel precursor and a bidentate phospho-
rus or nitrogen donor have been shown to catalyze the
copolymerization of a-olefins with carbon monoxide
[7]. However, in a recent study, workers at INEOS
Acrylics discovered that the bidentate phosphine 1,2-
bis(di-tert-butylphosphinomethyl)benzene in combina-
tion with Pd2(dba)3 forms a catalyst mixture that is
2. Experimental
2.1. General comments
All manipulations involving air-sensitive materials
were carried out in an inert atmosphere glove box or
using standard Schlenk line techniques under an atmo-
sphere of nitrogen or argon. Diethyl ether and hexane
were distilled from potassium–sodium alloy, THF from
potassium, and CH2Cl2 and pyridine from CaH2. Deu-
teriochloroform was pre-dried with CaH2, and then
exceptionally active (440 000
g
product (mol
cat)−1 h−1) and highly selective (99.98%) for methyl
propanoate production [8]. Heaton and co-workers
have used a combination of multinuclear NMR spec-
troscopy and 13C-labeling to show that methoxycar-
bonylation of ethene most likely occurs via a hydride
rather than the alternative methoxycarbonyl cycle [9].
Further evidence in support of the hydride mechanism
has been provided by the isolation and characterization
of the 3-oxopentyltriphenylphosphine cation, a by-
product formed during the catalytic methoxycarbonyla-
tion of ethene [10], and the isolation of
palladium-hydride and -acyl complexes, key intermedi-
ates in the catalytic hydroesterification of styrene [11].
In contrast, we have recently found that catalyst mix-
tures based on the o-xylyl-tethered bidentate phosp-
,
vacuum transferred and stored over 4 A molecular
sieves. 1,1,1,3,3,3-Hexafluoropropan-2-ol was pur-
chased from Aldrich and used as received. 1H-, 31P- and
13C-NMR spectra were recorded in JEOL LAMBDA
500 or Bruker AC 200, AMX 300 and DRX 500
machines. Gas chromatographic analysis was carried
out using a Pye Unicam Series 104 instrument using the
following conditions: initial temperature 50 °C for 2
min, final temperature 150 °C, ramp rate 12 °C min−1
,
injection temperature 200 °C, detector temperature
200 °C, carrier gas N2 55 ml min−1, column: 2 m×4
mm id, 20% Carbowax 20M on Gas-Chem Q 85-100
mesh. The diphosphines cis- and trans-{11,12-
bis(2,3,4,5-tetramethylphospholylmethyl)-9,10-dihydro-
9,10-ethano-anthracene} (4a and b) and 11,12-bis-
(diphenylphosphinomethyl)-9,10-dihydro-9,10-ethano-
anthracene (5a and b) were prepared from the corre-
sponding dimesylates [14] as described below. 1-Phenyl-
2,3,4,5-tetramethyl phosphole was prepared according
to a previously published method [15].
hole,
1,2-bis(2,3,4,5-tetramethylphospholylmethyl)-
benzene, are highly active and selective for the produc-
tion of polyketone, demonstrating that replacement of
di-tert-butyl groups for the 2,3,4,5-tetramethylphospho-
lyl fragment affects a dramatic change in selectivity
[12]. The disparate selectivity of these two catalyst
systems, both containing an o-xylyl tether, suggests that
the origin of selectivity in the palladium-catalyzed car-
boxylation of ethylene may not depend solely on
monodentate versus bidentate coordination.
In an extension of our preliminary studies, we have
investigated the potential of related C4-bridged diphos-
phines to catalyze the carboxylation of ethylene and
found that those based on cis- and trans-1,2-bis-
(diphenylphosphinomethyl)cyclohexane and endo, endo-
2.1.1. Synthesis of 9,10-dihydro-9,10-ethano-
anthracene-11,12-dicarboxylic acid-anhydride (1a)
Anhydrous aluminum chloride (41.14 g, 309 mmol)
was added portion-wise to a suspension of anthracene
(50.0 g, 281 mmol) in CH2Cl2 (500 ml) and cooled to
0 °C. A CH2Cl2 solution of maleic anhydride (27.51 g,
281 mmol) was added in portions to maintain the

184
S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
reaction mixture at 0 °C. After warming to room tem-
perature (r.t.) and stirring overnight, the suspension
was carefully poured onto water (750 ml), the organic
layer separated, and the aqueous layer extracted with
CH2Cl2 (3×75 ml). The organic extracts were com-
bined, washed with water (200 ml), dried over MgSO4
and the solvent removed to afford the desired anhy-
2H, CH2OH), 1.66 (m, 2H, ethano bridge CꢀH).
13C{1H}-NMR (125.65 MHz, CDCl3, l): 143.0–123.4
(C6H4), 66.3 (s), 46.4(s), 46.1 (s).
2.1.5. Synthesis of cis-11,12-bis(methanesulfonyl-
methyl)-9,10-dihydro-9,10-ethano-anthracene (3a)
Methanesulfonyl chloride (13.71 g, 119.7 mmol) was
added to a stirred solution of 2a (13.06 g, 49.0 mmol) in
dry pyridine (130 ml) at such a rate to maintain the
temperature below 5 °C. After stirring for further 1 h,
the dark suspension was poured onto 1000 ml of 2 N
HCl and stirred for 10 min. The solid was collected by
filtration, washed with water (3×50 ml) and recrystal-
lized from MeOH to give mesylate 3a in 67% yield (13.8
g). 1H-NMR (500.0 MHz, CDCl3, l): 7.27 (dd, 2H,
1
dride, 1a, as a tan solid in 87% yield (67.67 g). H-
NMR (500.0 MHz, CDCl3, l): 7.32 (dd, 2H, JHH=3.4,
6.8 Hz, C6H4), 7.26 (dd, 2H, JHH=3.6, 6.7 Hz, C6H4),
7.12 (m, 4H, C6H4), 4.76 (s, 2H, bridgehead CꢀH), 3.45
(s, 2H, ethano bridge C-H). 13C{1H}-NMR (125.65
MHz, CDCl3, l): 170.4 (s, CO), 140.6 (s, C6H4), 138.0
(s, C6H4), 127.7 (s, C6H4), 127.1 (s, C6H4), 125.2 (s,
C6H4), 124.3 (s, C6H4).
J
HH=3.3, 6.1 Hz, C6H4), 7.24 (dd, 2H, JHH=3.3, 6.1
2.1.2. Synthesis of trans-9,10-dihydro-9,10-ethano-
anthracene-dicarboxylic acid-(11,12)-diethyl ester (1b)
trans-9,10-Dihydro-9,10-ethano-anthracene-dicar-
boxylic acid-(11,12)-diethyl ester (1b) was prepared
from diethyl fumarate and isolated in 82% yield accord-
Hz, C6H4), 7.1 (dd, 2H, JHH=3.0, 5.1 Hz, C6H4), 7.07
(dd, 2H, JHH=3.3, 4.7 Hz, C6H4), 4.33 (s, 2H, bridge-
head CꢀH), 3.92 (dd, 2H, JHH=5.5, 10.0 Hz, CH2),
3.65 (dd, 2H, JHH=10.0 Hz, CH2), 2.87 (s, 6H,
CH3SO3), 2.44 (br s, 2H, ethano bridge CꢀH).
1
ing to the procedure described above for 1a. H-NMR
(500.0 MHz, CDCl3, l): 7.00–7.28 (m, 8H, C6H4), 4.65
(s, 2H, bridgehead CꢀH), 3.99 (m, 4H, CH2), 3.34 (s,
2H, ethano bridge CꢀH), 1.14 (t, 6H, JHH=7.2 Hz,
CH3).
2.1.6. Synthesis of trans-11,12-bis(methanesulfonyl-
methyl)-9,10-dihydro-9,10-ethano-anthracene (3b)
trans - 11,12 - Bis(methanesulfonylmethyl) - 9,10 - dihy-
dro-9,10-ethano-anthracene was prepared in 67% yield
according to the procedure described above for 3a.
1H-NMR (500.0 MHz, CDCl3, l): 7.25 (m, 4H, C6H4),
7.10 (m, 4H, C6H4), 4.28 (s, 2H, bridgehead CꢀH), 3.83
(dd, 2H, J=5.8, 10.1 Hz, CH2), 3.67 (t, 2H, JHH=10.1
Hz, CH2), 2.93 (s, 6H, CH3SO3), 1.79 (m, 2H, ethano
bridge CꢀH). 13C{1H}-NMR (125.65 MHz, CDCl3, l):
142.12 (s), 141.3 (s), 126.8 (s), 126.4 (s), 125.6 (s), 123.9
(s), 67.9 (s), 37.6 (s), 37.4 (s) 30.9 (s), 25.6 (s).
2.1.3. Synthesis of cis-11,12-bis(hydroxymethyl)-9,10-
dihydro-9,10-ethano-anthracene (2a)
A solution of 1a (36.47 g, 132 mmol) in THF (100
ml) was added dropwise via cannula to a stirred suspen-
sion of lithium aluminum hydride (20.0 g, 527 mmol) in
THF (300 ml) at 0 °C. After the addition was com-
plete, the mixture was heated at reflux overnight. After
cooling to r.t., the suspension was diluted with THF
(600 ml) and quenched by addition of water (20 ml),
followed by NaOH (8.0 g in 20 ml of water) and finally
water (20 ml). After hydrolysis was complete, the re-
sulting mixture was filtered, the solids washed with
THF (3×100 ml) and the solvent removed in vacuo.
Crystallization of the crude product from EtOH at r.t.
2.1.7. Synthesis of cis-11,12-bis(2,3,4,5-tetramethyl-
phospholylmethyl)-9,10-dihydro-9,10-ethano-anthracene
(4a)
A solution of 1-phenyl-2,3,4,5-tetramethylphosphole
(4.35 g, 20.1 mmol) in THF (40 ml) was stirred with
lithium powder (0.55 g, 79.2 mmol) under an argon
atmosphere for 2 h, after which the solution turned
intense red. After filtering, the phenyl lithium
by-product was quenched by addition of anhydrous
aluminum chloride (0.89 g, 6.70 mmol) to an ice-cold
solution of the reaction mixture. After stirring at 0 °C
for 30 min, a solution of 3a (3.80 g, 9.0 mmol) in THF
was added dropwise. The reaction mixture was allowed
to warm to r.t. and then heated at reflux overnight.
After cooling to r.t., the solvent was removed under
reduced pressure and the resulting gummy residue
extracted into CH2Cl2 (100 ml), filtered, washed with
degassed water (3×50 ml) and the organic phase
separated and dried over MgSO4. After filtration, the
solvent was removed to give a white solid which was
1
yielded diol 2a in 61% yield (21.5 g). H-NMR (200.1
MHz, CDCl3, l): 6.97–7.24 (m, 8H, C6H4), 4.13 (s, 2H,
bridgehead CꢀH), 3.44 (dd, 2H, CH2OH), 3.22 (dd, 2H,
CH2OH), 2.74 (br s, 2H, CH2OH), 2.26 (m, 2H, ethano
bridge CꢀH).
2.1.4. Synthesis of trans-11,12-bis(hydroxymethyl)-
9,10-dihydro-9,10-ethano-anthracene (2b)
trans-11,12-Bis-(hydroxymethyl)-9,10-dihydro-9,10-
ethano-anthracene was isolated in 78% yield according
1
to the procedure described above for 2a. H-NMR (500
MHz, CDCl3, l): 7.90 (dd, 2H, JHH=3.3, 6.4 Hz,
C6H4), 7.40 (dd, 2H, JHH=3.0, 6.4 Hz, C6H4), 4.16 (s,
2H, bridgehead CꢀH), 3.41 (dd, 2H, JHH=6.4, 3.6 Hz,
CH2), 3.02 (dd, 2H, JHH=2.8, 7.9 Hz, CH2), 2.0 (br s,

S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
185
washed with 40:60 petroleum–ether (3×40 ml) and
crystallized from CH2Cl2 at −78 °C to afford 4a as a
white solid in 68% yield (3.13 g). 1H-NMR (500.0 MHz,
CDCl3, l): 7.34–7.02 (m, 8H, C6H4), 4.52 (s, 2H,
bridgehead CH), 2.12 (br m, 2H, ethano bridge CꢀH),
2.01 (d, 6H, J=10.4 Hz, phosphole-CH3), 1.84 (s, 6H,
phosphole-CH3), 1.82 (d, 6H, J=10.4 Hz, phosphole-
CH3), 1.81 (s, 6H, phosphole-CH3), 1.41 (m, 2H,
ml, 22 mmol) with rapid stirring. After warming to r.t.
and stirring for an additional 30 min, the solution was
cooled in an ice bath and a THF solution of cis-11,12-
bis(methanesulfonylmethyl)-9,10-dihydro-9,10-ethano-
anthracene (4.1 g, 10 mmol) added dropwise. The reac-
tion mixture was left to warm to r.t. and stirred
overnight during which time there was a gradual color
change from deep orange–red to pale yellow. The
solvent was removed under reduced pressure and the
resulting oily residue was dissolved in Et2O (100 ml)
and washed with degassed water (30–40 ml). The or-
ganic phase was separated, dried over MgSO4, filtered
and the solvent removed to afford cis-11,12-bis-
(diphenylphosphinomethyl)-9,10-dihydro-9,10-ethano-
CHaHb), 1.08 (m, 2H, CH Hb). 31P{1H}-NMR (202.0
a
MHz, CDCl3, l): 6.67 (s, phosphole-P). 13C{1H)-NMR
(125.65 MHz, CDCl3, l): 144.0 (s, C6H4), 142.1 (d,
JPC=76.0 Hz, phosphole ring), 140.9 (s, C6H4), 135.1
(d, PC=108.1 Hz, phosphole ring), 126.1–123.2
J
(C6H4), 48.0 (d, JPC=10.3 Hz, bridgehead CꢀH), 39.8
(s, ethano bridge CꢀH), 25.7 (d, JPC=17.6 Hz, CH2),
14.3 (d, JPC=21.7 Hz, CH3), 13.2 (d, JPC=21.6 Hz,
CH3).
1
anthracene (5a) in 54% yield (3.2 g). H-NMR (500.0
MHz, CDCl3, l): 6.95–7.80 (m, 28H, C6H5 and C6H4),
4.54 (s, 2H, bridgehead CꢀH), 1.95 (br s, 2H, ethano
bridge CꢀH), 1.40 (m, 4H, CH2). 31P{1H}-NMR (202.0
MHz, CDCl3, l): −17.9. 13C{1H}-NMR (125.65 MHz,
CDCl3, l): 143.6–123.6 (C6H4 and C6H5), 49.9 (d,
2.1.8. Synthesis of trans-11,12-bis((2,3,4,5-tetramethyl-
phospholylmethyl)-9,10-dihydro-9,10-ethano-anthracene
(4b)
JPC=9.4 Hz), 38.4 (d, JPC=14.6 Hz), 30.1 (d, JPC
=
Compound 4b was prepared from trans-11,12-bis-
(methanesulfonylmethyl) - 9,10 - dihydro - 9,10 - ethano-
anthracene in 79% yield according to the procedure
described above. 1H-NMR (500.0 MHz, CDCl3, l):
7.24–7.01 (m, 8H, C6H4), 4.16 (s, 2H, bridgehead
CꢀH), 2.00 (d, 6H, J=10.6 Hz, phosphole-CH3), 1.84
(m, 6H, phosphole-CH3), 1.82 (d, 6H, J=10.6 Hz,
phosphole-CH3), 1.80 (s, 6H, phosphole-CH3), 1.45 (dd,
2H, J=14.4, 2.1 Hz, CHaHb), 1.35 (m, 2H, ethano
bridge CꢀH), 1.42 (m, 2H, CHaHb), 1.28 (m, 4H, CH2).
31P{1H}-NMR (202.0 MHz, CDCl3, l): 6.57 (s, phosp-
hole-P). 13C{1H)-NMR (125.65 MHz, CDCl3, l): 144.5
(s, C6H4), 142.6 (d, JPC=33.6 Hz, phosphole ring),
140.5 (s, C6H4), 134.6 (d, JPC=101.8 Hz, phosphole
ring), 126.2 (s, C6H4), 126.1 (s, C6H4), 125.7 (s, C6H4),
123.2 (s, C6H4), 48.8 (d, JPC=7.3 Hz, bridgehead
CꢀH), 47.4 (d, JPC=10.3 Hz, ethano bridge CꢀH), 29.8
(d, JPC=19.5 Hz, CH2), 14.3 (d, JPC=22.6 Hz, phosp-
hole-CH3), 13.4 (d, JPC=21.7 Hz, phosphole-CH3).
12.3 Hz).
2.1.11. Synthesis of trans-11,12-bis(diphenylphosphino-
methyl)-9,10-dihydro-9,10-ethano-anthracene (5b)
Compound 5b was isolated as a white solid from the
reaction
between
trans-11,12-bis(methanesulfonyl-
methyl)-9,10-dihydro-9,10-ethano-anthracene
lithium diphenylphosphide according to the procedure
described above for 5a. H NMR (500.0 MHz, CDCl3,
l): 6.9–7.60 (m, 28H, C6H4 and C6H5), 4.26 (s, 2H,
bridgehead CꢀH), 1.80 (m, 2H, ethano bridge CꢀH),
1.52 (m, 4H, CH2). 13C{1H}-NMR (125.65 MHz,
CDCl3, l): 143.8–123.3 (C6H5 and C6H4), 48.9 (d,
and
1
J
J
PC=13.4 Hz), 45.6 (dd, JPC=19.7 Hz), 34.6 (dd,
PC=10.3 Hz).
2.1.12. Synthesis of [{cis-11,12-bis-(2,3,4,5-
tetramethylphospholylmethyl)-9,10-dihydro-9,10-ethano-
anthracene}PdCl2] (6a)
A solution of [(cycloocta-1,5-diene)PdCl2] (0.27 g,
0.95 mmol) in CH2Cl2 was treated with a CH2Cl2
solution of 11,12-bis-(2,3,4,5-tetramethylphospholyl-
methyl)-9,10-dihydro-9,10-ethano-anthracene (0.48 g,
0.95 mmol) and stirred vigorously for ca. 30 min. The
reaction mixture was filtered, the solvent removed un-
der vacuum and the resulting yellow solid washed with
n-hexane (ꢀ20 ml) and dried under vacuum to give 6a
as a pale yellow solid in 73% yield (0.48 g). Crystalliza-
tion from a concentrated CH2Cl2 solution layered with
2.1.9. Synthesis of 1,4-bis(2,3,4,5-tetramethyl-
phospholyl)butane (4c)
Compound 4c was prepared from 1,4-dibromobutane
in 45% yield according to the procedure described
1
above. H-NMR (500.0 MHz, CDCl3, l): 1.90 (d, 12H,
J=10.3 Hz, phosphole-CH3), 1.81 (s, 12H, phosphole-
CH3), 1.58 (br m, 4H, CH2), 1.08 (br m, 4H, CH2).
31P{1H}-NMR (202.0 MHz, CDCl3, l): 12.28 (s,
phosphole-P).
1
n-hexane gave X-ray quality crystals of 6a. H-NMR
2.1.10. Synthesis of cis-11,12-bis(diphenylphosphino-
methyl)-9,10-dihydro-9,10-ethano-anthracene (5a)
A solution of diphenylphosphine (3.86 ml, 22 mmol)
in THF (15 ml) was cooled to −78 °C and was treated
with a 2.5 M solution of butyl lithium in hexanes (8.93
(500.0 MHz, CDCl3, l): 7.2 (dd, 2H, J=5.5, 3.0 Hz,
C6H4), 7.15 (dd, 2H, J=5.5, 3.7 Hz, C6H4), 7.10 (dd,
2H, J=5.2, 3.1 Hz, C6H4), 7.07 (dd, 2H, J=5.2, 3.1
Hz, C6H4), 3.93 (s, 2H, bridgehead CꢀH), 2.49 (m, 2H,
ethano bridge CꢀH), 2.08 (d, 6H, J=11.9 Hz, phosp-

186
S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
hole-CH3), 1.89 (s, 6H, phosphole-CH3), 1.87 (d, 6H,
J=11.9 Hz, phosphole-CH3), 1.84 (s, 6H, phosphole-
CH3), 1.38 (m, 4H, CH2). 31P{1H}-NMR (202.0 MHz,
CDCl3, l): 42.3 (s, phosphole-P). Anal. Calc. for
C34H40Cl2P2Pd: C, 59.38; H, 5.86. Found: C, 59.63; H,
5.79%.
C6H5), 7.30 (m, 8H, C6H4/C6H5), 7.25 (dt, 4H, JHH=
2.7, 7.9 Hz, C6H4/C6H5), 7.05 (m, 8H, C6H4/C6H5), 6.90
(d, 2H, JHH=6.7 Hz, C6H4/C6H5), 3.36 (s, 2H, bridge-
head CꢀH), 2.35 (dt, 2H, JHH=4.3, 5.6 Hz, ethano
bridge CꢀH). 31P{1H}-NMR (202.0 MHz, CDCl3, l):
29.7. Anal. Calc. for C43H38Cl4P2Pd: C, 59.72; H, 4.42.
Found: C, 59.80; H, 4.77%.
Compounds 6b–d were prepared according to the
procedure described above.
2.1.16. Synthesis of [{1,4-bis(2,3,4,5-tetramethylphos-
pholyl)butane}PdMeCl] (6e)
2.1.13. Synthesis of [{trans-11,12-bis-(2,3,4,5-
tetramethylphospholylmethyl)-9,10-dihydro-9,10-ethano-
anthracene}PdCl2] (6b)
A solution of [(cycloocta-1,5-diene)PdMeCl] (0.250 g,
1.0 mmol) in CH2Cl2 (10 ml) was treated with a CH2Cl2
solution (7–10 ml) of 1,4-bis(2,3,4,5-tetramethylphos-
pholyl)butane (0.334 g, 1.0 mmol) and stirred vigor-
ously for ca. 3–4 h. The reaction mixture was filtered,
the residue washed with hexane (2×10 ml) to give 6e
as a pale yellow powder in 68% yield (0.39 g). 1H-NMR
(200.1 MHz, CDCl3, l): 1.91 (br s, 4H, CH2), 1.84 (br
s, 12H, phosphole-CH3), 1.80 (br, 16H, phosphole-
CH3+CH2), 1.08 (br, 3H, PdꢀCH3). 31P{1H}-NMR
(202.0 MHz, CDCl3, l): 41.7 (s).
Compound 6b was isolated as pale yellow X-ray
quality crystals in 68% yield by slow diffusion of n-hex-
ane into a concentrated CHCl3 solution at r.t. 1H-NMR
(500.0 MHz, CDCl3, l): 7.13 (dd, 2H, J=7.7, 1.6 Hz,
C6H4), 7.10 (dd, 2H, J=7.0, 1.2 Hz, C6H4), 7.07 (dd,
2H, J=7.7, 1.6 Hz, C6H4), 7.13 (dd, 2H, J=7.0, 1.2
Hz, C6H4), 3.86 (s, 2H, bridgehead CꢀH), 2.07 (d, 6H,
J=11.9 Hz, phosphole-CH3), 1.87 (s, 6H, phosphole-
CH3), 1.86 (d, 6H, J=11.9 Hz, phosphole-CH3), 1.80
(s, 6H, phosphole-CH3), 1.74 (m, overlapping 2H,
CHaHb, 2H, ethano bridge C-H), 1.54 (m, 2H, CHaHb).
31P{1H}-NMR (202.0 MHz, CDCl3, l): 41.8 (s, phosp-
hole-P). 13C{1H)-NMR (125.65 MHz, CDCl3, l): 143.4
(s, C6H4), 138.1 (s, C6H4), 132.1 (d, JPC=67.0 Hz,
phosphole ring), 130.2 (d, JPC=51.2 Hz, phosphole
ring), 126.9 (s, C6H4), 126.4 (s, C6H4), 125.6 (s, C6H4),
123.1 (s, C6H4), 51.8 (s, bridgehead CꢀH), 45.7 (s,
ethano bridge CꢀH), 38.5 (d, JPC=23.6 Hz, CH2), 14.2
(d, JPC=19.5 Hz, phosphole-CH3), 13.8 (d, JPC=26.2
Hz, phosphole-CH3). Anal. Calc. for C34H40Cl2P2Pd·
3CHCl3: C, 42.47; H, 4.14. Found: C, 42.83; H, 3.94%.
2.1.17. Synthesis of [{cis-11,12-bis(2,3,4,5-tetramethyl-
phospholylmethyl)-9,10-dihydro-9,10-ethano-
anthracene}Pd(OAc)2] (7a)
A suspension of palladium acetate (0.23 g, 1.06
mmol) in toluene (ꢀ1–2 ml) was treated with a tol-
uene solution (ꢀ5–7 ml) of cis-11,12-bis-(2,3,4,5-tetra-
methylphospholylmethyl)-9,10-dihydro-9,10-ethano-
anthracene (0.54 g, 1.06 mmol) and stirred vigorously
for ca. 30 min during which time a pale solid
precipitated from solution. The reaction mixture was
filtered and the precipitate was washed with toluene
(2×3 ml) and hexane (3×10 ml), and dried under
vacuum to give 7a as a yellow powder in 48% yield
2.1.14. Synthesis of [{cis-11,12-bis(diphenylphos-
phinomethyl)-9,10-dihydro-9,10-ethano-anthracene}-
PdCl2] (6c)
Compound 6c was isolated as pale yellow crystals in
74% yield by slow diffusion of Et2O into a concentrated
CH2Cl2 solution. X-ray quality crystals can be grown
by slow diffusion of n-hexane into a concentrated
1
(0.36 g). H-NMR (500.0 MHz, CDCl3, l): 7.07–7.21
(m, 8H, C6H4), 3.93 (s, 2H, bridgehead CꢀH), 2.57 (m,
2H, ethano bridge CꢀH), 2.15 (d, 6H, J=11.3 Hz,
phosphole-CH3), 1.84 (m, 18H, phosphole-CH3), 1.80
(br s, 6H, CH3CO2), 1.43 (m, 4H, CH2). 31P{1H}-NMR
(202.0 MHz, CDCl3, l): 44.5 (s, phosphole-P).
Catalyst precursors 7c and d were prepared using a
procedure similar to that described for 7a.
1
CHCl3 solution at r.t. H-NMR (500.0 MHz, CDCl3,
l): 7.18–7.50 (m, 28H, C6H5 and C6H4), 3.60 (s, 2H,
bridgehead CꢀH), 1.88 (m, 2H, ethano bridge CꢀH),
1.18 (br s, 4H, CH2). 31P{1H}-NMR (202.0 MHz,
CDCl3, l): 32.3. Anal. Calc. for C42H36Cl2P2Pd: C,
64.70; H, 4.65. Found: C, 64.29; H, 4.88%.
2.1.18. Synthesis of [{cis-11,12-bis(diphenylphosphino-
methyl)-9,10-dihydro-9,10-ethano-anthracene}-
Pd(OAc)2] (7c)
Compound 7c was isolated as a yellow powder in
69% yield. 1H-NMR (500.0 MHz, CDCl3, l): 7.20–7.58
(m, 28H, C6H5 and C6H4), 3.86 (s, 2H, bridgehead
CꢀH), 2.28 (m, 2H, ethano bridge CꢀH), 1.42–1.70 (m,
6H, CH3), 0.70 (m, 4H, CH2). 31P{1H}-NMR (202.0
MHz, CDCl3, l): 24.1.
2.1.15. Synthesis of [{trans-11,12-bis(diphenylphos-
phinomethyl)-9,10-dihydro-9,10-ethano-anthracene}-
PdCl2] (6d)
Compound 6d was isolated as pale yellow X-ray
quality crystals in 68% yield by slow diffusion of n-hex-
ane into a concentrated CH2Cl2 solution at r.t. 1H-
NMR (500.0 MHz, CDCl3, l): 7.70 (dd, 4H, JHH=8.0,
11.6 Hz, C6H4/C6H5), 7.50 (t, 2H, JHH=7.6 Hz, C6H4/

S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
187
2.1.19. Synthesis of [{trans-11,12-bis(diphenyl-
phosphinomethyl)-9,10-dihydro-9,10-ethano-
anthracene}Pd(OAc)2] (7d)
mmol) for ca. 15 min before removing the solvent
under reduced pressure and redissolving the residue in
MeOH before adding CH3SO3H. The reaction mixture
was pressurized to 20 bar with an equimolar mixture of
CO and ethylene and heated at 90 °C for 2 h. The
reaction was quenched by release of CO–ethylene pres-
sure and the polymer was isolated by filtration, dried
and weighed and the filtered solution analyzed by gas
chromatography. Provided the catalyst mixtures are
generated with the exclusion of oxygen using Schlenk
techniques, the same results could be obtained with
catalyst prepared by both methods, i.e. using complexes
either prepared in advance or generated in CH2Cl2,
prior to polymerization.
Compound 7d was isolated as a yellow powder in
62% yield. 1H-NMR (500.0 MHz, CDCl3, l): 6.50–7.95
(m, 28H, C6H4 and C6H5), 3.73 (br s, 2H, bridgehead
CꢀH), 2.28 (s, 2H, ethano bridge CꢀH), 1.85 (m, 6H,
CH3), 1.48 (s, 4H, CH2). 31P{1H}-NMR (202.0 MHz,
CDCl3, l): 23.1. 13C{1H}-NMR (125.65 MHz, CDCl3,
l): 144–123 (C6H4 and C6H5), 50.9 (bridgehead CꢀH),
44.1 (ethano bridge CꢀH), 36.6 (d, JPC=14.1 Hz,
CH2).
2.2. Polymerization procedure
Polymerizations were conducted in MeOH in a 300
ml Parr bench-mounted reactor. The catalyst precursors
[(PꢀP)Pd(OAc)2] were prepared according to the proce-
dure described above. In a typical procedure, 0.085
mmol of [(PꢀP)Pd(OAc)2] was dissolved in 60 ml of
anhydrous MeOH, 0.85 mmol of CH3SO3H added and
the solution transferred to a 300 ml autoclave under N2.
Alternatively, the catalyst can be generated prior to
charging the reactor by stirring a CH2Cl2 solution of
diphosphine (0.085 mmol) and palladium acetate (0.085
2.3. Crystal structure determination of 6a–d
All measurements were made on a Bruker AXS
SMART 1K CCD area-detector diffractometer using
graphite-monochromated Mo–Ka radiation (u=
,
0.71073 A) and narrow frame exposures (0.3° in ꢀ).
Cell parameters were refined from all strong reflections
in each data set. Intensities were corrected semi-empiri-
cally for absorption, based on symmetry-equivalent and
repeated reflections. No significant intensity decay was
Table 1
Crystal data and structure refinement parameters for compounds 6a–d
Compound
6a·2CH2Cl2
6b·3CHCl3
6c·2CHCl3
6d·CH2Cl2
C43H38Cl4P2Pd
864.87
160
Empirical formula
Formula weight
Temperature (K)
Crystal system
Space group
Unit cell dimensions
C36H44Cl6P2Pd
857.75
160
C37H43Cl11P2Pd
1046.00
160
Monoclinic
P21/n
C44H38Cl8P2Pd
1018.68
160
Monoclinic
P21/c
Triclinic
Monoclinic
P21/c
(
P1
,
a (A)
11.1072(6)
12.1233(7)
16.9939(10)
69.628(2)
87.862(2)
67.256(2)
1965.83(19)
2
11.2206(4)
17.9819(6)
22.2659(7)
12.6042(6)
19.7955(9)
17.6945(8)
14.4531(10)
16.2930(12)
16.8625(12)
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
V (A )
93.742(1)
105.300(2)
104.921(1)
3
,
4483.0(3)
4
1.168
0.60×0.50×0.50
28.4
4258.4(3)
4
1.046
0.40×0.30×0.20
28.6
3837.0(5)
4
0.877
0.75×0.65×0.55
28.5
Z
Absorption coefficient (mm−1
Crystal size (mm)
qmax (°)
)
0.986
0.50×0.40×0.40
28.5
Reflections measured
Unique reflections
Reflections with F2\2|(F2)
Rint (on F2)
16194
8560
7752
0.0271
414
0.0667
0.1452
1.304
37645
10547
9397
0.0231
468
0.0674
0.1655
1.061
36140
10081
6856
0.0615
496
0.0854
0.2504
1.108
26221
8943
7589
0.0332
463
0.0536
0.1313
1.059
No. of parameters
a
R
[F2\2|(F2)]
b
Rw (all data)
Goodness-of-fit c (S)
−3
,
Largest difference peak and hole (e A
)
1.99 and −1.61
2.18 and −2.18
2.06 and −1.82
2.86 and −3.34
a Conventional R=ꢀꢁꢁFoꢁ−ꢁFcꢁꢁ/ꢀꢁFoꢁ for ‘observed’ reflections having Fo2\2|(F2o).
b [ꢀ(wF2o−Fc2)2/ꢀw(Fo2)2]1/2 for all data.
c Goodness-of-fit=[ꢀw(Fo2−F2c)2/(no. of unique reflections−no. of parameters)]1/2
.

188
S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
Scheme 2. (i) One equivalent AlCl3, diethylfumarate; (ii) LiAlH4,
ether; (iii) MsCl, pyridine; (iv) LiPPh2 or (2,3,4,5-tetramethyl phos-
pholyl)lithium in THF.
Scheme 1. (i) One equivalent AlCl3, maleic anhydride; (ii) LiAlH4,
ether; (iii) MsCl, pyridine; (iv) LiPPh2 or (2,3,4,5-tetramethyl phos-
pholyl)lithium in THF.
late was chosen since all attempts to convert the diol
into the corresponding bromomethyl derivative resulted
in the formation of intractable solids. Dropwise addi-
tion of a THF solution of lithium 2,3,4,5-tetramethyl
phospholyl or diphenylphosphide into a THF solution
of the mesylate gave 4a and b and 5a and b, respec-
tively, as evidenced by the disappearance of the intense
color associated with the lithium salt.
observed. The structures were solved by direct methods
(Patterson synthesis for 6d) and refined on F2 values for
all unique data by full-matrix least-squares. Table 1
gives further details. All non-hydrogen atoms were
refined anisotropically. Hydrogen atoms located in dif-
ference maps were constrained with a riding model;
U(H) was set at 1.2 (1.5 for methyl groups) times Ueq
for the parent atom. The solvent molecule in 6d is
disordered. The largest residual peaks are close to Pd
and Cl atoms; these and relatively high displacement
parameters for solvent molecules may indicate further
unresolved disorder. Programs used were SHELXTL [16]
for structure solution, refinement and molecular graph-
ics, Bruker AXS SMART (control) and SAINT (integra-
tion) and local programs [17].
The 31P{1H}-NMR spectra of 4a and b and 5a and b
each contain a single resonance in the region expected
1
for a uncoordinated phosphine or phosphole. The H-
NMR spectrum of 4a contains four signals associated
with the methyl groups attached to the phosphole rings,
two doublets at l 2.01 (JPH=10.4 Hz) and 1.82 (JPH
=
10.4 Hz) and two singlets at l 1.84 and 1.81, as well as
a singlet at l 4.53 belonging to the protons attached to
the central six-membered ring, a complex multiplet at l
1.41 for the diastereotopic methylene protons, and a
multiplet at l 1.08 corresponding to the protons at-
3. Results and discussion
1
tached to the ethane bridge. The H-NMR spectrum of
3.1. Synthesis and coordination chemistry of
anthracene-bridged diphosphines and phospholes 4a and
b and 5a and b
4b shows a similar set of resonances and is consistent
with the expected formulation. A full spectroscopic
assignment of 4a and b has been undertaken using
1
1
1H–1H COSY, H{31P} and H–13C heteronuclear sin-
Ligands 4a and b and 5a and b were prepared
according to the procedure outlined in Schemes 1 and
2, by reacting the corresponding dimesylate with
2,3,4,5-tetramethylphospholyl lithium and lithium
diphenyl phosphide, respectively. The anhydride 9,10-
dihydrido-9,10-ethano-anthracene-11,12-dicarboxylic
acid-anhydride (1a) and diester trans-9,10-dihydrido-
9,10-ethano anthracene-dicarboxylic acid-(11,12)-di-
ethyl ester (1b) were isolated in near quantitative yields
from the Diels–Alder cycloaddition of anthracene with
maleic anhydride and diethyl fumarate, respectively, in
the presence of a stoichiometric amount of aluminum
trichloride. Reduction of the anhydride and diester with
lithium aluminum hydride gave the corresponding diol
2, which was converted into its mesylate 3 by reaction
with methane sulfonyl choride in pyridine. The mesy-
gle- and multiple-bond correlation studies.
The bis(phosphole), 1,4-bis(2,3,4,5-tetramethylphos-
pholyl)butane (4c) [18] has also been prepared in order
to undertake comparative catalyst testing with systems
based on the more common C4-bridged diphosphine
1,4-bis(diphenylphosphino)butane (dppb), to defini-
tively assess the influence of the phospholyl ring on
catalyst activity and selectivity. Phosphole 4c was pre-
pared by addition of a tetrahydrofuran solution of
lithium 2,3,4,5-tetramethylphospholyl into a solution of
1,4-dibromobutane according to previously published
1
procedures (Eq. (2)). The H- and 13C{1H}-NMR spec-
tra are consistent with the expected formulation, the
former containing a doublet at l 1.90 (JPH=10.3 Hz)
and a singlet at l 1.81, belonging to the 2,5- and
3,4-methyl groups attached to the phosphole ring, re-

S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
189
spectively, and two broad high-field multiplets associ-
ated with the methylene groups of the butane tether.
Unexpectedly, addition of a dichloromethane solu-
tion of 1,4-bis(2,3,4,5-tetramethylphospholyl)butane to
a solution of [{cycloocta-1,5-diene}PdCl2] resulted in
the immediate formation of a red precipitate which was
too insoluble to characterize using conventional spec-
troscopic techniques. Reaction of 1,4-bis(2,3,4,5-te-
(2)
tramethylphospholyl)butane
with
[{cycloocta-1,5-
diene}PdMeCl] in dichloromethane resulted in the for-
mation of a colorless solution from which an off-white
solid was isolated by precipitation. Unexpectedly, the
room temperature 31P{1H}-NMR spectrum of this
product contains a single sharp resonance at l 41.7 and
not the two signals that would be predicted for inequiv-
alent phosphole groups trans to methyl and chloride.
The presence of a single 31P resonance is consistent with
a trans arrangement of methyl and chloride, indicating
Addition of a dichloromethane solution of 4a and b
and 5a and b into a dichloromethane solution of
[{cycloocta-1,5-diene}PdCl2] resulted in a gradual color
change from yellow to orange with the formation of
[(PꢀP)PdCl2] (PꢀP=4a, 6a; PꢀP=4b, 6b; PꢀP=5a, 6c;
PꢀP=5b, 6d), in yields of up to 70% (Eqs. (3) and (4)).
The 31P{1H}-NMR spectra of 6a–d each contain sharp
singlets characteristic of a coordinated diphosphine. In
1
the H-NMR spectrum of 6a, four signals are found,
that
[{1,4-bis(2,3,4,5-tetramethylphospholyl)butane}-
two doublets at l 2.08 and 1.87 (JPH=11.9 Hz) and
two singlets at l 1.89 and 1.84, which belong to the 2,5-
and 3,4-methyl groups of the phosphole ring, respec-
tively. The magnitude of JPH is similar to that in
1-phenyl 2,3,4,5-tetramethylphosphole and its palla-
dium complex [19]. Complex multiplets centered at l
1.38 and 2.49 are consistent with diastereotopic methyl-
PdMeCl] most likely exists as a polymer/oligomer (Eq.
(5)). The alternative explanation that 4c coordinates in
a bidentate manner and that the two phosphole groups
are rendered equivalent by a rapid fluxional process has
been eliminated by undertaking a variable temperature
31P{1H}-NMR study which showed no evidence of line
broadening.
(5)
ene protons and the protons attached to the ethano
bridge, respectively, and a singlet at l 3.93 belongs to
the bridgehead protons, i.e. those attached to the cen-
tral six-membered ring. In addition, the aromatic pro-
The catalyst precursors [(PꢀP)Pd(OAc)2] (7a, 7c and
d) were prepared by the addition of a concentrated
solution of the corresponding diphosphine to a rapidly
stirred suspension of [Pd3(OAc)6] [20] to give a light
colored precipitate which was filtered, washed with
toluene and hexane and dried to afford the desired
compounds as spectroscopically pure solids. Attempts
to prepare 7b resulted in the formation of a mixture of
compounds; therefore, the catalyst precursor was gener-
ated in situ prior to catalyst testing. The 31P{1H}- and
1H-NMR spectroscopic characteristics of these com-
pounds bear a close similarity to those of their dichlo-
ride counterparts.
tons appear as
a
characteristic set of four
well-separated doublets of doublets at l 7.20, 7.15, 7.10
and 7.07.
(3)
(4)
3.2. X-ray crystal structures of anthracene-bridged
diphosphine and phosphole complexes 6a–d
The lack of structural data for palladium(II) com-
plexes of bidentate phospholes prompted us to under-
take single-crystal X-ray analyses of 6a and b. Single
crystals of 6a·2CH2Cl2 and 6b·3CHCl3 suitable for X-
ray analysis were grown from a dichloromethane solu-

190
S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
Table 2
tion layered with hexane and a chloroform solution
layered with hexane, respectively. Perspective views of
the molecular structures of 6a and b together with the
atomic numbering scheme are illustrated in Figs. 1 and
2, respectively, and a selection of relevant bond lengths
and angles is listed in Table 2 while crystal data are
provided in Table 1. Since the structures of both com-
pounds are based on bidentate phospholes bridged by a
C4-ethano anthracene tether and are clearly related,
they will be described together. In both 6a and b, the
coordination sphere around palladium is close to
square planar, as indicated by the dihedral angle of 2.9°
(6a) and 8.5° (6b) between the planes containing
P(1)Pd(1)P(2) and Cl(1)Pd(1)Cl(2), and 4a and b both
form seven-membered chelate rings. The PdꢀP bond
,
Selected bond distances (A) and bond angles (°) for 6a and b
6a
6b
Bond lengths
Pd(1)ꢀCl(1)
Pd(1)ꢀCl(2)
Pd(1)ꢀP(1)
Pd(1)ꢀP(2)
P(1)ꢀC(19)
P(1)ꢀC(22)
C(19)ꢀC(20)
C(20)ꢀC(21)
C(21)ꢀC(22)
P(2)ꢀC(27)
P(2)ꢀC(30)
C(27)ꢀC(28)
C(28)ꢀC(29)
C(29)ꢀC(30)
2.3553(13)
2.3662(14)
2.2392(14)
2.2455(14)
1.809(6)
1.803(5)
1.348(8)
1.490(8)
1.341(8)
1.810(6)
1.799(6)
1.345(8)
1.485(10)
1.352(9)
Pd(1)ꢀCl(1)
Pd(1)ꢀCl(2)
Pd(1)ꢀP(1)
Pd(1)ꢀP(2)
P(1)ꢀC(19)
P(1)ꢀC(22)
C(19)ꢀC(20)
C(20)ꢀC(21)
C(21)ꢀC(22)
P(2)ꢀC(27)
P(2)ꢀC(30)
C(27)ꢀC(28)
C(28)ꢀC(29)
C(29)ꢀC(30)
2.3500(13)
2.3400(12)
2.2498(11)
2.2506(11)
1.799(5)
1.797(5)
1.334(7)
1.476(7)
1.350(7)
1.776(5)
1.800(5)
1.363(7)
1.479(8)
1.329(7)
,
lengths in 6a [Pd(1)ꢀP(1)=2.2392(14) A, Pd(1)ꢀP(2)=
,
2.2455(14) A] are similar to those in 6b [Pd(1)ꢀP(1)=
Bond angles
,
,
2.2498(11) A, Pd(1)ꢀP(2)=2.2506(11) A], slightly
longer than those in the phosphole complex [{1,2-
bis(2,3,4,5-tetramethylphospholylmethyl)benzene}-
P(1)ꢀPd(1)ꢀP(2)
Cl(1)ꢀPd(1)ꢀCl(2)
P(1)ꢀPd(1)ꢀCl(1)
P(2)ꢀPd(1)ꢀCl(2)
Pd(1)ꢀP(2)ꢀC(30)
Pd(1)ꢀP(2)ꢀC(1)
Pd(1)ꢀP(2)ꢀC(27)
C(30)ꢀP(2)ꢀC(1)
C(1)ꢀP(2)ꢀC(27)
Pd(1)ꢀP(1)ꢀC(18)
Pd(1)ꢀP(1)ꢀC(19)
Pd(1)ꢀP(1)ꢀC(22)
C(18)ꢀP(1)ꢀC(22)
C(18)ꢀP(1)ꢀC(19)
C(19)ꢀP(1)ꢀC(22)
98.62(5)
91.63(5)
84.66(5)
80.05(5)
114.9(2)
124.34(18)
114.2(2)
99.9(3)
104.4(3)
122.36(18)
116.55(19)
115.00(18)
100.4(2)
104.0(3)
93.8(3)
P(1)ꢀPd(1)ꢀP(2)
Cl(1)ꢀPd(1)ꢀCl(2)
P(1)ꢀPd(1)ꢀCl(1)
P(2)ꢀPd(1)ꢀCl(2)
Pd(1)ꢀP(2)ꢀC(30)
Pd(1)ꢀP(2)ꢀC(1)
Pd(1)ꢀP(2)ꢀC(27)
C(30)ꢀP(2)ꢀC(1)
C(1)ꢀP(2)ꢀC(27)
Pd(1)ꢀP(1)ꢀC(18)
Pd(1)ꢀP(1)ꢀC(19)
Pd(1)ꢀP(1)ꢀC(22)
100.30(4)
89.63(5)
83.28(4)
87.11(4)
114.94(17)
121.84(16)
115.68(18)
99.7(2)
106.3(2)
124.78(15)
112.92(16)
117.14(16)
C(18)ꢀP(1)ꢀC(22) 104.3(2)
C(18)ꢀP(1)ꢀC(19)
C(19)ꢀP(1)ꢀC(22)
98.4(2)
93.5(2)
PdCl2] [12] but within the range expected for palladium
complexes of diphosphines, including [Pd(BIPHOS)Cl2]
[21] and [Pd(DMPP)2Cl2] [22]. The distances and angles
in the phosphole rings are similar to those previously
reported for related compounds, including [Pd-
(bidmpp)2][BF4]2 [23], [Pd(BIPHOS)Cl2] [21], trans-
[Pd(DMPP)2X2] [24], [Pd(TMBA)(DMPP)Cl] [25] and
[Pd(TMBA)(DMPP)(NO3)] [26] and are consistent with
localized double and single bonds. The two phosphole
rings in 6a are essentially planar and approximately
parallel forming a dihedral angle of 20.2° and oriented
perpendicular to the PdP2Cl2 plane. In contrast, the
dihedral angle between the phosphole rings in 6b is
considerably greater (33.6°) presumably reflecting the
torsional twist enforced by the trans geometry at the
ethane carbon–carbon bond bridging the anthracene
fragment. The natural bite angles of 100.30 (1) and
98.62(5)°, for 6a and 6b, respectively, are greater than
that in [PdCl2(dppb)] [27] and much closer to the value
reported in the palladium dichloride complex of the
C4-tethered diphosphine 2,2-dimethyl-4,5-bis(diphenyl-
Fig. 1. Molecular structure of [{cis-11,12-bis(2,3,4,5-tetramethylphos-
pholylmethyl)-9,10-dihydro-9,10-ethano-anthracene}PdCl2] (6a). Hy-
drogen atoms and the CH2Cl2 molecules of crystallization have been
omitted.
Fig. 2. Molecular structure of [{trans-11,12-bis(2,3,4,5-tetramethyl-
phospholylmethyl)-9,10-dihydro-9,10-ethano-anthracene}PdCl2] (6b).
Hydrogen atoms and the CHCl3 molecules of crystallization have
been omitted.
phosphinomethyl)-1,3-dioxolane
(ÚPPdP=96.78°)
[28]. In both 6a and b, the Cl(1)Pd(1)Cl(2) angles are
close to 90° [Cl(1)ꢀPd(1)ꢀCl(2)=91.63(5)°, 6a;

S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
191
Cl(1)ꢀPd(1)ꢀCl(2)=89.63(5)°, 6b] and the large natural
Table 3
Selected bond distances (A) and bond angles (°) for 6c and d
,
bite angle manifests itself in compression of the two
P(1)ꢀPd(1)ꢀCl(1) and P(2)ꢀPd(1)ꢀCl(2) angles which
are smaller than 90° [P(1)ꢀPd(1)ꢀCl(1)=84.66(5)°,
P(2)ꢀPd(1)ꢀCl(2)=85.05(5)°, 6a; P(2)ꢀPd(1)ꢀCl(2)=
87.11(4)°, P(1)ꢀPd(1)ꢀCl(1) 83.28(4)°, 6b]. Structural
data suggest that bidentate phospholes behave in much
the same manner as a conventional diphosphine, as
s-donor p-acceptors.
6c
6d
Bond lengths
Pd(1)ꢀCl(1)
Pd(1)ꢀCl(2)
Pd(1)ꢀP(1)
Pd(1)ꢀP(2)
P(1)ꢀC(1)
P(2)ꢀC(18)
C(1)ꢀC(2)
C(2)ꢀC(17)
C(17)ꢀC(18)
2.3472(19)
2.3618(18)
2.2527(18)
2.2543(19)
1.835(7)
1.835(7)
1.538(9)
1.569(10)
1.533(10)
Pd(1)ꢀCl(1)
Pd(1)ꢀCl(2)
Pd(1)ꢀP(1)
Pd(1)ꢀP(2)
P(1)ꢀC(30)
P(2)ꢀC(17)
C(17)ꢀC(15)
C(15)ꢀC(16)
C(16)ꢀC(30)
2.3420(10)
2.3533(9)
2.2621(10)
2.2911(10)
1.828(4)
1.848(4)
1.518(5)
1.551(5)
1.540(5)
Single-crystal X-ray analyses of 6c and d have also
been undertaken in order to compare with related
complexes of C4-bridged diphosphines including endo,
endo
and
exo,
endo-[{2,3-bis(diphenylphosphi-
Bond angles
nomethyl)norbornane}PdCl2]. The molecular structures
of 6c and 6d are shown in Figs. 3 and 4, respectively,
and a selection of relevant bond lengths and angles is
listed in Table 3 while crystal data are provided in
Table 1. Both complexes have the expected square
planar coordination around palladium, the dihedral
P(1)ꢀPd(1)ꢀP(2)
Cl(1)ꢀPd(1)ꢀCl(2)
P(1)ꢀPd(1)ꢀCl(1)
P(2)ꢀPd(1)ꢀCl(2)
P(2)ꢀPd(1)ꢀCl(1)
P(1)ꢀPd(1)ꢀCl(2)
Pd(1)ꢀP(1)ꢀC(1)
Pd(1)ꢀP(1)ꢀC(37)
Pd(1)ꢀP(1)ꢀC(31)
C(1)ꢀP(1)ꢀC(37)
C(1)ꢀP(1)ꢀC(31)
99.78(7)
90.04(7)
84.79(7)
85.82(7)
170.79(8)
173.82(7)
120.2(2)
112.0(2)
109.6(3)
101.5(3)
104.0(3)
P(1)ꢀPd(1)ꢀP(2)
Cl(1)ꢀPd(1)ꢀCl(2)
P(1)ꢀPd(1)ꢀCl(1)
P(2)ꢀPd(1)ꢀCl(2)
P(2)ꢀPd(1)ꢀCl(1)
P(1)ꢀPd(1)ꢀCl(2)
Pd(1)ꢀP(1)ꢀC(30)
Pd(1)ꢀP(1)ꢀC(37)
Pd(1)ꢀP(1)ꢀC(31)
C(30)ꢀP(1)ꢀC(37)
C(30)ꢀP(1)ꢀC(31)
C(31)ꢀP(1)ꢀC(37)
Pd(1)ꢀP(2)ꢀC(18)
Pd(1)ꢀP(2)ꢀC(24)
Pd(1)ꢀP(2)ꢀC(17)
C(18)ꢀP(2)ꢀC(17)
C(18)ꢀP(2)ꢀC(24)
C(17)ꢀP(2)ꢀC(24)
100.25(3)
89.41(4)
81.23(4)
89.04(3)
170.59(4)
170.63(4)
120.88(13)
112.15(13)
110.99(13)
100.39(18)
101.55(18)
109.86(18)
113.87(13)
103.76(13)
127.33(13)
97.76(18)
107.74(19)
106.18(18)
C(31)ꢀP(1)ꢀC(37) 108.6(3)
Pd(1)ꢀP(2)ꢀC(18)
Pd(1)ꢀP(2)ꢀC(25)
Pd(1)ꢀP(2)ꢀC(19)
C(18)ꢀP(2)ꢀC(19) 102.6(4)
C(18)ꢀP(2)ꢀC(25) 100.5(3)
C(19)ꢀP(2)ꢀC(25) 106.9(4)
123.1(2)
114.8(3)
107.4(2)
angle between PdP2 and PdCl2 planes being 8.5 and
9.4°, for 6c and 6d, respectively. The structures clearly
show that diphosphines 5a and b bind in a chelating
manner to form seven-membered rings. While the PdꢀP
bond lengths in 6c are essentially equivalent
Fig. 3. Molecular structure of [{cis-11,12-bis(diphenylphosphi-
nomethyl)-9,10-dihydro-9,10-ethano-anthracene}PdCl2] (6c). Hydro-
gen atoms and the CHCl3 molecules of crystallization have been
omitted.
,
,
[Pd(1)ꢀP(1)=2.2527(18) A; Pd(1)ꢀP(2)=2.2543(19) A],
and in the range expected for complexes of diphosphi-
nes, there is a marked difference between the PdꢀP
,
bond lengths in 6d [Pd(1)ꢀP(1)=2.2621(10) A,
,
,
Pd(1)ꢀP(2)=2.2911(10) A); D(PdꢀP)=0.0290 A], indi-
cating that the interaction of one of the diphenylphos-
phino groups to the palladium center is significantly
weaker than the other. This difference is substantially
greater than the corresponding difference in 6b, for
,
which D(PdꢀP)=0.0008 A. While the ClꢀPdꢀP angles
in 6c show a significant compression from 90° and are
similar [Cl(1)ꢀPd(1)ꢀP(1)=84.79(7)°, Cl(2)ꢀPd(1)ꢀ
P(2)=85.82(7)°], there is a significant difference be-
tween the corresponding angles in 6d [P(1)ꢀ
Pd(1)ꢀCl(1)=81.23(4)°,
P(2)ꢀPd(1)ꢀCl(2)=89.04°,
DÚ(PPdCl)=7.81°]. A similar distortion is clearly evi-
dent in the structure of 6b, although the difference
DÚ(PPdCl)=3.83° is not as pronounced as that in 6d.
The natural bite angles P(1)PdP(2) of 99.78(7) and
100.25(3)° for 6c and 6d, respectively, are in the range
Fig. 4. Molecular structure of [{trans-11,12-bis(diphenylphosphi-
nomethyl)-9,10-dihydro-9,10-ethano-anthracene}PdCl2] (6d). Hydro-
gen atoms and the CH2Cl2 molecule of crystallization have been
omitted.

192
S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
expected for palladium complexes of four-carbon teth-
ered diphosphines and are comparable to that reported
for [Pd(dppf)Cl2] [29].
of diphenylphosphino-substituted phosphines form cat-
alysts that are selective for copolymerization [7,30].
Notably, in his early studies Drent showed a marked
dependence of polymer productivity and molecular
weight on the length of the tether in the bidentate
phosphines Ph2P(CH2)nPPh2 (n=1–6). In particular,
he showed that catalysts formed from palladium acetate
and 1,3-bis(diphenylphosphino)propane (dppp) gave
maximum rates (6000 g polymer (g of Pd)−1 h−1) and
polymer molecular weights (n=180), whereas those
3.3. Catalytic studies
Catalysts formed from
a combination of cis-
11,12-bis(diphenylphosphinomethyl)-9,10-dihydro-9,10-
ethano-anthracene (5a) and palladium acetate are active
for the copolymerization of ethene and carbon monox-
ide, generating low molecular weight polymers with an
activity of 13 300 g polymer (g of Pd)−1 h−1. The
results of the polymerization experiments and polymer
properties are summarized in Table 4. Copolymeriza-
tion reactions were carried out in methanol under 10
bar each of CO and ethene and after 2 h the reactor
was cooled, the polymer isolated by filtration and the
reaction mixture analyzed by gas chromatography. The
IR spectra of the copolymers (recorded as a KBr disc)
contain a CO stretching band at 1692 cm−1. As an
example, the 13C{1H}-NMR spectrum of copolymer
based on
a
four-carbon tether (n=4, 1,4-bis(-
diphenylphosphino)butane) were less active (2300 g
polymer (g of Pd)−1 h−1) and gave polymer with a
significantly lower molecular weight (n=45) [7b]. In
order to relate the results obtained using catalyst mix-
tures formed from 4a–b and 5a–b with those previ-
ously reported for dppb, we performed copoly-
merization reactions using [Pd(dppb)(OAc)2] under our
conditions (Table 4, entry 7) and found that the activity
and average degree of polymerization compares favor-
ably with catalyst mixtures formed with 5a. We have
not made any attempt to optimize catalyst conditions,
either via the use of oxidizing additives such as 1,4-ben-
zoquinone or by varying the Brønsted acid. Rather, we
have relied on comparative results obtained using dppb
as a standard under our conditions in order to relate
the results of our studies to those reported in the
literature.
generated
using
cis-11,12-bis-(diphenylphosphino-
methyl) - 9,10 - dihydro - 9,10 - ethano - anthracene–
Pd(OAc)2, recorded in 1,1,1,3,3,3-hexafluoroisopropyl
alcohol, contains signals associated with the methylene
and carbonyl carbons at l 35.0 and 212.1 ppm, respec-
tively, and no signals were observed in the region
expected for double insertion of C2H4 into the polymer
[5]. In addition, resonances at l 217.0 (CH3CH2CO),
176.4 (MeOCO), 6.5 (CH3CH2), 52.0 (CH2OCH3) and
27.5 (CH2ꢀC(O)OMe) are clearly visible and corre-
spond to ketone and ester end groups on the polymer.
End group analysis of the 13C-NMR spectrum has been
used to determine the average degree of polymerization
(n) and it has revealed that the polyketone is a short
chain polymer with n=40 (Table 4, entry 5).
Under similar conditions, catalyst mixtures generated
from
palladium
acetate
and
trans-11,12-bis-
(diphenylphosphinomethyl)-9,10-dihydro-9,10-ethano-
anthracene (5b) produce methyl propanoate (Table 4,
entry 6), with an activity of 7300 g methyl propanoate
(mol cat)−1 hr−1 together with polyketone (2,800 g
polymer (mol Pd)−1 h−1, n=20), giving a selectivity
greater than 70%. The selectivity of this catalyst is
perhaps not surprising since we have previously shown
that the selectivity of catalysts based on endo, exo- and
The formation of polyketone using a catalyst mixture
based on 5a is perhaps not surprising since the majority
Table 4
Summary of ethylene carbon monoxide copolymerization results a
c
Entry
Ligand
Mass of polymer
(g) b
Productivity/g polyketone
n
Mass of
propanoate
(g) b,d
Productivity d/g propanoate
(mol cat)−1 h−1
(mol cat)−1 h−1
1
2
3
4
5
6
7
4a
4b
4c
3.5
1.6
0
20 600
9500
0
80
50
–
–
–
–
0.20
–
-
-
–
phosphole e
0
0
–
1200
–
7300
–
5a
5b
dppb
2.3
0.44
4.5
13 300
2800
26 700
40
20
45
1.25
–
a All reactions were performed in methanol pressurized to 20 bar with an equimolar mixture of CO–C2H4 at 90 °C for 2 h.
b Average mass of product over three runs.
c The average degree of polymerization (n) estimated by integration of the end group signals in the 13C{1H}-NMR spectra.
d Determined from GC analysis of the reaction mixture using n-decane as an internal standard.
e 1-Phenyl-2,3,4,5-tetramethylphosphole.

S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
193
Fig. 5. Perspective views showing the torsional angles in the four-carbon tether of (a) [{endo, exo-bis(diphenylphosphinomethyl)norbornane}-
PdCl2]; (b) [{trans-11,12-bis(diphenylphosphinomethyl)-9,10-dihydro-9,10-ethano-anthracene}PdCl2]. Phenyl rings on P(1) and P(2) have been
omitted for clarity.
endo, endo-bis(diphenylphosphinomethyl)norbornane is
strongly influenced by the ligand stereochemistry [13].
Catalysts based on endo, exo-bis(diphenylphosphi-
nomethyl)norbornane are highly selective for the pro-
duction of methyl propanoate (\90%) while those
of dimer and higher oligomers. Clearly, the production
of polyketone (n=14) and methyl propanoate with
\90% selectivity is not consistent with the presence of
a single active catalyst and is most likely due to the
presence of two distinct catalytic species operating in
parallel, one in which the diphosphine acts in a biden-
tate manner, to generate polyketone, and the other in
which one arm of the diphosphine dissociates and
effectively behaves as a monodentate ligand to generate
methyl propanoate. Although it is well known that the
tendency of bidentate phosphines to open and act as
bridging ligands increases with increasing natural bite
angle [31], the bite angle of 100.30(1)° associated
with [trans-11,12-bis(diphenylphosphinomethyl)-9,10-
dihydro-9,10-ethano-anthracene}PdCl2] (6b) is similar
to that formed by endo, endo-bis(diphenylphosphi-
nomethyl)norbornane, (ÚPPdP=100.53(6)°), which is
exclusive for the production of polyketone, during
which it presumably acts in a bidentate manner, i.e. if
selectivity is determined by monodentate versus biden-
tate coordination, it is unlikely to depend solely on the
natural bite angle but on a subtle combination of
factors. Comparison of the structures of 6a–d reveals
that the difference in the PdꢀP bond lengths in 6d
formed
from
endo,
endo-bis(diphenylphosphi-
nomethyl)norbornane generate polyketone. By analogy,
catalysts based on trans-11,12-bis(diphenylphosphi-
nomethyl)-9,10-dihydro-9,10-ethano-anthracene, which
bears a close stereochemical similarity to endo, exo-bis-
(diphenylphosphinomethyl)norbornane in that both
contain a conformationally rigid four-carbon tether
with a torsional twist about C2ꢀC3, would be predicted
to be selective for methyl propanoate. The torsional
twist in the four-carbon tethers of [{trans-11,12-bis-
(diphenylphosphinomethyl)-9,10-dihydro-9,10-ethano-
anthracene}PdCl2]
C(30)C(16)C(15)=76.5°)
(torsional
and
ÚC(17)C(15)C(16)ꢀ
[{endo, exo-bis-
(diphenylphosphinomethyl)norbornane}PdCl2]
(ÚC(8)C(7)C(1)ꢀC(21)C(1)C(7)=76.4°) are shown in
Fig. 5, which clearly illustrates the stereochemical simi-
larity between these diphosphines. In contrast, cis-
11,12-bis(diphenylphosphinomethyl)-9,10-dihydro-9,10-
ethano-anthracene more closely resembles endo, endo-
bis(diphenylphosphinomethyl)norbornane in the stereo-
chemical arrangement of the four-carbon tether
(torsional angles of 6.4 and 0°, respectively) and, not
surprisingly, both form catalysts that are selective for
polyketone. In addition to producing methyl
propanoate, we noted that catalysts based on endo,
,
(DPdꢀP=0.029 A) is significantly greater than those in
,
,
,
6a–c (6a, 0.0063 A; 6b, 0.0008 A; 6c, 0.0016 A) which
supports the proposal that trans-11,12-bis(diphenyl-
phosphinomethyl)-9,10-dihydro-9,10-ethano-anthracene
forms a second distinct catalytic species via dissociation
of the less strongly bound diphenylphosphino group.
The generation of polyketone and methyl propanoate
using trans-11,12-bis(diphenylphosphinomethyl)-9,10-
dihydro-9,10-ethano-anthracene clearly supports our
preliminary observations and the suggestion that the
stereochemistry of the four-carbon tether could play a
role in influencing catalyst selectivity. Although the
average degree of polymerization of 20 for catalysts
based on trans-11,12-bis(diphenylphosphinomethyl)-
exo-bis(diphenylphosphinomethyl)norbornane
also
produce polyketone with an average degree of polymer-
ization of 14. If catalyst selectivity results from a large
difference in the rate of chain termination versus propa-
gation, the average degree of polymerization would be
expected to be significantly less than 14, since methyl
propanoate is generated after only a single turnover
and thus there would be rapidly diminishing amounts

194
S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
9,10-dihydro-9,10-ethano-anthracene is notably lower
than that of its cis counterpart (n=45), it is still fully
consistent with the presence of two distinct catalytic
species.
cene does not behave in a similar manner and generate
methyl propanoate. Further studies are clearly required
to establish the origin of this selectivity.
To evaluate the effect of the phosphole substitutent,
in combination with a four-carbon tether, [{1,4-
bis(2,3,4,5-tetramethylphospholyl)butane}Pd(OAc)2]
was prepared and used to catalyze the copolymerization
of C2H4 and CO. In a comparative experiment, under
the same conditions as those used with 5a and b,
catalysts generated from 1,4-bis(2,3,4,5-tetramethyl-
phospholyl)butane are inactive for either copolymeriza-
tion or methoxycarbonylation (Table 4, entry 3), which
appears to be due to catalyst instability, as evidenced
by the rapid formation of a dark colored precipitate
characteristic of decomposition. The poor performance
of 1,4-bis(2,3,4,5-tetramethylphospholyl)butane con-
trasts sharply with the high activity of catalysts based
on dppb and further highlights the marked influence of
the nature of the substituents attached to phosphorus in
determining catalyst activity/stability. However, spec-
troscopic studies have shown that 1,4-bis(2,3,4,5-te-
tramethylphospholyl)butane does not act in a bidentate
manner but most likely bridges two metal centers to
form polymers/oligomers (vide infra), a factor which
will also influence catalyst performance. In addition,
comparison of the performance of catalysts based
The dramatic difference in selectivity between cata-
lysts based on 1,2-bis(2,3,4,5-tetramethylphospholyl-
methyl)benzene and 1,2-bis(di-tert-butylphosphino-
methyl)benzene, which gave copolymer and methyl
propanoate, respectively, prompted us to examine the
activity and selectivity of catalysts formed from cis- and
trans - 11,12 - bis(2,3,4,5 - tetramethylphospholylmethyl)-
9,10-dihydro-9,10-ethano-anthracene (4a–b). Under
our conditions, catalyst mixtures based on cis-11,12-
bis(2,3,4,5-tetramethylphospholylmethyl)-9,10-dihydro-
9,10-ethano-anthracene and trans-11,12-bis(2,3,4,5-te-
tramethylphospholylmethyl)-9,10-dihydro-9,10-ethano-
anthracene gave polyketone with a productivity of
20 000 and 9500 g polymer (mol cat)−1 h−1, respec-
tively (Table 4, entries 1 and 2). The copolymer gener-
ated using catalysts formed from phospholes 4a and b
separated during the reaction as gray, high melting
powders (temperature range 250–260 °C). The average
degree of polymerization has been estimated by integra-
tion of the end group resonances in the solution
13C{1H}-NMR spectrum. Interestingly, polyketone gen-
erated using a catalyst mixture based on 4a has an
unexpectedly high molecular weight (n=80), clearly
evident from the exceptionally low intensity of the end
group resonances, while polymer generated using 4b is
significantly shorter (n=50) and much closer to the
molecular weight of polyketone produced with catalyst
based on 1,4-bis(diphenylphosphino)butane [7b].
The most surprising aspect of these studies is the
selectivity of trans-11,12-bis(2,3,4,5-tetramethylphos-
pholylmethyl)-9,10-dihydro-9,10-ethano-anthracene for
copolymerization whereas its diphenylphosphino coun-
terpart gives mainly methyl propanoate. This is consis-
tent with the previous observation in which
1,2 - bis(2,3,4,5 - tetramethylphospholylmethyl)benzene
and 1,2-bis(di-tert-butylphosphinomethyl)benzene are
selective for copolymer and methyl propanoate, respec-
tively. Such a dramatic change in selectivity for biden-
tate ligands may be due to steric protection of the axial
sites offered by the phospholyl groups, which prevents
chain termination via an associative process. In addi-
tion, the phospholyl fragment resembles 1,10-phenan-
throline and 2,2-bipyridine, which form catalysts that
are selective for copolymerization, in that they provide
limited effective shielding of the active sites trans to
the bidentate ligand. However, if, as suggested
earlier, trans-11,12-bis(diphenylphosphinomethyl)-9,10-
dihydro-9,10-ethano-anthracene forms two active spe-
cies, one in which the phosphine acts in a monodentate
manner the other as a bidentate ligand, it is difficult to
understand why trans-11,12-bis(2,3,4,5-tetramethyl-
phospholylmethyl)-9,10-dihydro-9,10-ethano-anthra-
on
trans-11,12-bis(diphenylphosphinomethyl)-9,10-
dihydro-9,10-ethano-anthracene and trans-11,12-bis-
(2,3,4,5-tetramethylphospholyl)-9,10-dihydro-9,10-
ethano-anthracene further reinforces the idea that, for
diphosphines that act as bidentate ligands, selectivity is
influenced by the substituents attached to phosphorus.
The key discovery in polyketone chemistry involving
replacement of the monodentate phosphine PPh3 for
the chelating bidentate dppp which resulted in a change
in selectivity from methyl propanoate to polyketone
prompted us to investigate the dependence of catalyst
selectivity on ligand denticity, by conducting catalyst
testing using a combination of palladium acetate and
the monodentate phosphole 1-phenyl-2,3,4,5-tetra-
methylphosphole. Under the same conditions as those
used for 4a and b and 5a and b, catalysts based on
monodentate phospholes are inactive for copolymeriza-
tion and show only low activity for alkoxycarbonyla-
tion (Table 4, entry 4), confirming the observation that
monodentate ligands do not form catalysts that are
selective for polyketone.
In conclusion, following our preliminary studies in
which we demonstrated that catalyst systems based on
the C4-bridged diphosphine endo, endo-2,3-bis-
(diphenylphosphinomethyl)norbornane are selective for
copolymerization of ethylene with carbon monoxide
while those formed from its exo, endo-2,3-bis-
(diphenylphosphinomethyl)norbornane counterpart are
highly selective for the production of methyl

S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
195
[2] (a) C.G. Chardarin, S.L. Wov, R.C. Clark, C.H. Hathcock,
Tetrahedron Lett. (1976) 1769;
propanoate, we have now shown that catalysts based
on cis- and trans-11,12-bis(diphenylphosphinomethyl)-
9,10-dihydro-9,10-ethano-anthracene show similar se-
lectivity patterns. Single-crystal X-ray structure
determinations have revealed only minor differences in
the bite angles of these phosphines and clearly selectiv-
ity must be determined by a combination of factors,
one of which appears to be the stereochemistry of the
C4-tether. Substitution of the diphenylphosphino
groups in trans-11,12-bis(diphenylphosphinomethyl)-
9,10-dihydro-9,10-ethano-anthracene for the 2,3,4,5-te-
tramethylphospholyl fragment results in a dramatic
reversal of selectivity and catalysts based on trans-
11,12-bis(2,3,4,5-tetramethylphospholyl)-9,10-dihydro-
9,10-ethano-anthracene are selective for the production
of polyketone. In the case of phosphole-based catalysts,
the nature of the four-carbon tether appears to play an
important role in determining stability since catalysts
formed from 1,4-bis(2,3,4,5-tetramethylphospholyl)-
butane decompose rapidly under conditions of copoly-
merization. Clearly, evaluation of the factors that influ-
ence catalyst selectivity, activity and polymer molecular
weight is not straightforward and further studies are
required to fully investigate the origin of these effects.
In particular, investigations into counterion effects, the
preparation of cationic methyl and methoxycarbonyl
precatalysts and variations in the electronic and steric
properties of the phospholyl fragment are currently
under way.
(b) T.F. Murray, E.G. Samsel, V. Varma, J.R. Norton, J. Am.
Chem. Soc. 193 (1981) 7520;
(c) T.F. Murray, V. Varma, J.R. Norton, J. Org. Chem. 43
(1978) 353;
(d) B. El Ali, H. Alper, J. Org. Chem. 56 (1991) 5357.
[3] H. Alper, N. Hamel, J. Am. Chem. Soc. 112 (1990) 2803.
[4] (a) S.E. Houssame, L.E. Firdoussi, S. Allaoud, A. Karim, Y.
Castanet, A. Mortreux, J. Mol. Catal. A. 168 (2001) 15;
(b) L.E. Firdoussi, A. Bagga, S. Allaoud, B. Ait Allal, A. Karim,
Y. Castanet, A. Mortreux, J. Mol. Catal. A. 135 (1998) 11.
[5] (a) E. Drent, P.H.M. Budzelaar, Chem. Rev. 96 (1996) 663;
(b) A. Sen, Acc. Chem. Res. 26 (1993) 303;
(c) K. Nozaki, T. Hiyama, J. Organomet. Chem. 576 (1999) 248;
(d) B. Milani, G. Mestroni, Comments Inorg. Chem. 20 (1999)
301;
(e) M. Sperrle, G. Consiglio, Chem. Ber. 130 (1997) 1557;
(f) A. Vavasori, L. Toniolo, J. Mol. Catal. A 110 (1996) 13;
(g) E. Drent, J.A.M. van Broekhoven, P.H.M. Budzelaar, in: B.
Cornils, W.A. Hermann (Eds.), Applied Homogeneneous Catal-
ysis with Organometallic Compounds, vol. 1, VCH, Weinheim,
1996 (333 pp.).
[6] (a) E. Drent, Eur. Patent Appl., EP 0 495 548 (1992);;
(b) E. Drent, Eur. Patent Appl., EP 0 495 547 (1992);;
(c) E. Drent, Eur. Patent Appl., WO 97/03943 (1997);;
(d) G.R. Eastham, R.P. Tooze, X.L. Wang, K. Whiston, (ICI)
WO 96/19434 (1996);;
(e) G.W. Parshall, S.D. Ittel, Homogeneous Catalysis: The Ap-
plications and Catalysis by Soluble Transition Metal Complexes,
Wiley, New York, 1992.
[7] For examples of diphosphines see: (a) E. Drent, Eur. Pat. Appl.
121 965 A2, 1984;
(b) E. Drent, J.A.M. van Broekhoven, M.J. Doyle, J.
Organomet. Chem. 417 (1991) 235;
(c) M. Barsacchi, A. Batistini, G. Consiglio, U.W. Suter, Macro-
molecules 25 (1992) 3604;
(d) M. Sperrles, G. Consiglio, J. Am. Chem. Soc. 117 (1995)
12130;
4. Supplementary material
(e) Z. Jiang, S.E. Adams, A. Sen, Macromolecules 27 (1994)
2694;
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC nos. 116121, 116122, 116123, and
116124 for compounds 6d, 6c, 6b and 6a, respectively.
Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (Fax: +44-1223-336033;
e-mail: deposit@ccdc.cam.ac.uk or www: http://
www.ccdc.cam.ac.uk).
(f) Z. Jiang, A. Sen, J. Am. Chem. Soc. 117 (1995) 4455;
(g) C. Bianchini, H.M. Lee, A. Meli, S. Moneti, F. Vizza, M.
Fontani, P. Zanello, Macromolecules 32 (1999) 4183;
(h) C. Bianchini, H.M. Lee, A. Meli, W. Oberhauser, F. Vizza,
P. Bruggeller, R. Haid, C. Langes, Chem. Commun. (2000) 777;
(i) G. Verspini, F. Schanserra, R.A. Sheldon, Angew. Chem. Int.
Ed. Engl. 39 (2000) 803;
For examples of dintrogen donors see: (j) F.C. Rix, M.
Brookhart, P.S. White, J. Am. Chem. Soc. 118 (1996) 4746;
(k) B. Milani, A. Anzilutti, L. Vicentini, A. Sessanta Santi, E.
Zangrando, S. Geremia, G. Mestroni, Organometallics 16 (1997)
5064;
(l) B.A. Markies, D. Kruis, M.H.P. Rietveld, K.A.N. Verberk, J.
Boersma, H. Koojima, M.T. Lakin, A.L. Speck, G. van Koten,
J. Am. Chem. Soc. 117 (1995) 117, 5263;
(m) M. Brookhart, F.C. Rix, J.M. DeSimone, J.C. Barborak, J.
Am. Chem. Soc. 114 (1992) 5894;
(n) M. Brookhart, M.I. Wagner, G.G.A. Balavoine, H.A. Had-
dou, J. Am. Chem. Soc. 1126 (1994) 3641;
(o) B. Milani, L. Vicentini, A. Sonmazzi, F. Garbassi, E.
Chiarparin, E. Zangrando, G. Mestroni, J. Chem. Soc. Dalton
Trans. (1996) 3139;
Acknowledgements
We gratefully acknowledge funding from the Univer-
sity of Newcastle upon Tyne, EPSRC (E.G.R., P.A.C.,
equipment to W.C.) and INEOS Acrylics (P.A.C.), and
thank Johnson Matthey for loans of palladium salts.
References
(p) B. Milani, E. Alessio, G. Mestroni, E. Zangrando, L. Ran-
daccio, G. Consiglio, J. Chem. Soc. Dalton Trans. (1996) 1021;
(q) M. Brookhart, M.I. Wagner, J. Am. Chem. Soc. 118 (1996)
7219;
[1] J. Tsuji, Palladium Reagents and Catalysts. Innovations in Or-
ganic Synthesis, Wiley, New York, 1995.

196
S. Doherty et al. / Journal of Organometallic Chemistry 640 (2001) 182–196
[21] M. Gougygou, O. Tissot, J.-C. Daran, G.G.A. Balavoine,
(r) L.K. Johnson, S. Mecking, M. Brookhart, J. Am. Chem. Soc.
118 (1996) 267.
Organometallics 16 (1997) 1008.
[22] J.J. MacDougall, J.H. Nelson, F. Mathey, J.J. Mayerle, Inorg.
Chem. 19 (1980) 709.
[23] T. Kojima, K. Saeki, K. Ono, M. Ohba, Y. Matsuda, Chem.
Commun. (1997) 1697.
[8] W. Clegg, G.R. Eastham, M.R.J. Elsegood, R.P. Tooze, X.L.
Wang, K. Whiston, Chem. Commun. (1999) 1877.
[9] G.R. Eastham, B.T. Heaton, J.A. Iggo, R.P. Tooze, R. Why-
man, S. Zucchini, Chem. Commun. (2000) 609.
[24] W.L. Wilson, J. Fischer, R.E. Wasylishen, K. Eichele, V.J.
Catalano, J.H. Frederick, J.H. Nelson, Inorg. Chem. 35 (1996)
1486.
[25] N. Gul, J.H. Nelson, Tetrahedron 26 (2000) 71.
[26] N. Gul, J.H. Nelson, Organometallics 19 (2000) 91.
[27] V.D. Makhaev, Z.M. Dzhabieva, S.V. Konovalikhin, O.A. D’Y-
achenko, G.P. Belov, Koord. Khim. 22 (1996) 598.
[28] V. Gramlich, G. Consiglio, Helv. Chim. Acta 62 (1979) 1016.
[29] T. Hayashi, M. Konishi, Y. Kabori, M. Kumada, T. Higuchi, K.
Mirotsu, J. Am. Chem. Soc. 106 (1984) 158.
[30] (a) S.J. Dossett, A. Gillon, A.G. Orpen, J.S. Fleming, P.G.
Pringle, D.F. Wass, M.D. Jones, Chem. Commun. (2001) 699;
(b) G.K. Barlow, J.D. Boyle, N.A. Cooley, T. Ghaffar, D.F.
Wass, Organometallics 19 (2000) 1470.
[31] (a) P. Dierkes, P.W.M.N. van Leeuwen, J. Chem. Soc. Dalton
Trans. (1999) 1519;
[10] R.P. Tooze, K. Whiston, A.P. Malyan, M.J. Taylor, N.W.
Wilson, J. Chem. Soc. Dalton Trans (2000) 3441.
[11] A. Seayad, S. Jayasree, K. Damodarao, L. Toniolo, R.V.
Chaudhari, J. Organomet. Chem. 601 (2001) 100.
[12] S. Doherty, G.R. Eastham, R.P. Tooze, T.S. Scanlan, D.
Williams, M.R.J. Elsegood, W. Clegg, Organometallics 18 (1999)
3558.
[13] J.G. Knight, S. Doherty, A. Harriman, E.G. Robins, M.
Betham, G.R. Eastham, R.P. Tooze, M.R.J. Elsegood, P.
Champkin, W. Clegg, Organometallics 19 (2000) 4957.
[14] K. Weinges, W. Sipos, J. Klein, J. Deuter, H. Irngartinger,
Chem. Ber. 120 (1987) 5.
[15] P.J. Fagan, W.A. Nugent, J.C. Calabrese, J. Am. Chem. Soc.
116 (1994) 1880.
[16] G.M. Sheldrick, SHELXTL User Manual, Version 5, Bruker AXS
Inc, Madison, WI, 1994.
[17] SMART and SAINT software for CCD diffractometers, Bruker
AXS Inc., Madison, WI, 1994.
(b) A.R. Sanger, J. Chem. Soc. Dalton Trans. (1977) 1971;
(c) G. Consiglio, C.A.S. Nefkens, Tetrahedron: Asymmetry 1
(1990) 417;
[18] P. Gradoz, D. Baudry, M. Ephritikhine, F. Nief, F. Mathey, J.
Chem. Soc. Dalton Trans. (1992) 3047.
(d) M. Portnoy, Y. Ben-David, D. Milstein, J. Organomet.
Chem. 503 (1995) 149;
[19] J.J. MacDougall, J.H. Nelson, F. Mathey, J. Mayerle, J. Inorg.
Chem. 19 (1980) 709.
(e) G.P.C.M. Dekker, C.J. Elsevier, K. Vrieze, P.W.M.N. van
Leeuwen, Organometallics 11 (1992) 1598;
[20] T.A. Stephenson, S.M. Morehouse, A.R. Powell, J.P. Heffer, G.
Wilkinson, J. Chem. Soc. (1965) 3632.
(f) N.A. Al-Salem, H.D. Empsall, R. Markham, B.L. Shaw, B.
Weeks, J. Chem. Soc. Dalton Trans. (1979) 1972.