The synthesis and late transition metal chemistry of 7-aza-N-indolyl phosphines and the activity of their palladium complexes in CO-ethene co-polymerisation

Article abstract of DOI:10.1039/b309882f

The 7-aza-N-indolyl phosphine ligands PR₂(N₂C₇H₅) (L¹, R = Ph; L², R = NC₄H₄) were prepared in a two-step process involving treatment of 7-azaindole with BuLi, followed by reaction of the lithium salt with PClR₂. L¹ and L² react with [MCl₂(cod)] (M = Pd, Pt) to give [PdCl₂(L-κ²P,N)] (1a, L = L¹; 1b, L = L²) or [PtCl₂(L-κ²P,N)] (2a, L = L¹; 2b, L = L²) and with [Rh(μ-Cl)(cod)]₂ in the presence of CO to give [RhCl(CO)(L-κ²P,N)] (3a, L = L¹; 3b, L = L²). Crystal structures for 1a,b and 3a,b are reported, and structural and spectroscopic evidence confirm that L² is a poorer σ-donor/better π-acceptor than L¹. The complexes [PdClMe(L-κ²P,N)] (4a, L = L¹; 4b, L = L²), prepared from [PdClMe(cod)], react with AgOTf to yield [PdMe(OTf)(L-κ²P,N)] (5a, L = L¹; 5b, L = L²). Complexes 5a,b are active catalysts for the co-polymerisation of CO and ethene, with activities similar to previously reported catalysts containing P,N-donor ligands. From the stepwise insertion reactions of CO and ethene with 5a,b, the acyl complexes [Pd{C(O)Me}(OTf)(L-κ²P,N)] (7a, L = L¹; 7b, L = L²) and alkyl complexes [Pd{CH₂CH₂C(O)-Me-κ²C,O} (L-κ²P,N)]OTf (8a, L = L¹; 8b, L = L²) have been isolated and crystallographically characterised, and the acyl complexes [Pd{C(O)CH₂CH₂C(O)Me-κ²C,O} (L-κ²P,N)]OTf (9a, L = L¹; 9b, L = L²) have been spectroscopically characterised. Reactions of 7a and 9a with MeOH gave methyl acetate and methyl 4-ketopentanoate respectively, with formation of palladium metal and conversion of the remaining palladium to [Pd(L¹-κ²P,N)₂](OTf)₂ 10 which has been crystallographically characterised.

Full text of DOI:10.1039/b309882f

View Article Online / Journal Homepage / Table of Contents for this issue
The synthesis and late transition metal chemistry of
7-aza-N-indolyl phosphines and the activity of their palladium
complexes in CO–ethene co-polymerisation
Andrew D. Burrows,* Mary F. Mahon and Maurizio Varrone
Department of Chemistry, University of Bath, Claverton Down, Bath, UK BA2 7AY.
E-mail: a.d.burrows@bath.ac.uk
Received 15th August 2003, Accepted 8th October 2003
First published as an Advance Article on the web 23rd October 2003
The 7-aza-N-indolyl phosphine ligands PR2(N2C7H5) (L1, R = Ph; L2, R = NC4H4) were prepared in a two-step
process involving treatment of 7-azaindole with BuLi, followed by reaction of the lithium salt with PClR2. L1 and L2
react with [MCl2(cod)] (M = Pd, Pt) to give [PdCl2(L-κ2P, N )] (1a, L = L1; 1b, L = L2) or [PtCl2(L-κ2P, N )] (2a, L = L1;
2b, L = L2) and with [Rh(µ-Cl)(cod)]2 in the presence of CO to give [RhCl(CO)(L-κ2P, N )] (3a, L = L1; 3b, L = L2).
Crystal structures for 1a,b and 3a,b are reported, and structural and spectroscopic evidence confirm that L2 is a
poorer σ-donor/better π-acceptor than L1. The complexes [PdClMe(L-κ2P, N )] (4a, L = L1; 4b, L = L2), prepared
from [PdClMe(cod)], react with AgOTf to yield [PdMe(OTf )(L-κ2P, N )] (5a, L = L1; 5b, L = L2). Complexes 5a,b
are active catalysts for the co-polymerisation of CO and ethene, with activities similar to previously reported
catalysts containing P, N -donor ligands. From the stepwise insertion reactions of CO and ethene with 5a,b, the
acyl complexes [Pd{C(O)Me}(OTf )(L-κ2P, N )] (7a, L = L1; 7b, L = L2) and alkyl complexes [Pd{CH2CH2C(O)-
Me-κ2C,O}(L-κ2P, N )]OTf (8a, L = L1; 8b, L = L2) have been isolated and crystallographically characterised, and the
acyl complexes [Pd{C(O)CH2CH2C(O)Me-κ2C,O}(L-κ2P, N )]OTf (9a, L = L1; 9b, L = L2) have been spectroscopically
characterised. Reactions of 7a and 9a with MeOH gave methyl acetate and methyl 4-ketopentanoate respectively,
with formation of palladium metal and conversion of the remaining palladium to [Pd(L1-κ2P, N )2](OTf )2 10 which
has been crystallographically characterised.
salt on treatment with BuLi, followed by reaction of this
Introduction
intermediate with either PClPh2 or PCl(NC4H4)2. These reac-
Mixed-donor ligands, containing a combination of hard and
tions are shown in Scheme 1, together with the numbering
soft donor sites, have a number of important applications in
scheme used for the 7-azaindolyl group. It is also possible to
both coordination chemistry and catalysis including use in
synthesise L1 and L2 using the reaction between PClR2 and
cross-couplings,1 ethene oligomerisations2 and Heck-type
7-azaindole in the presence of a base such as NEt3 or DBU
(1,8-diazabicyclo[5.4.0]undec-7-ene). However, this proved to
be a poor preparative route due to difficulties in separating
the desired ligand from by-products and unreacted starting
reactions.3 Recently, we have become interested in N-pyrrolyl
phosphine ligands in which the presence of one or more P–N
bonds to pyrrole rings serves to make the phosphine a strong
π-acceptor.4 Addition of a hard donor atom to such a ligand
materials.
provides a means of exaggerating the electronic difference
between the donor atoms, and we have previously reported the
synthesis and chemistry of the keto-functionalised N-pyrrolyl
phosphines PPh2{NC4H3C(O)Me-2}5 and P(NC4H4)2{NC4-
H3C(O)Me-2}.6 As anticipated, the ligands were observed to
become more electron-withdrawing with the increasing number
of pyrrolyl substituents.
N-Pyrrolyl phosphine ligands containing nitrogen function-
alities have received only limited attention. Ligands containing
amine and imine groups have been described and used in
Scheme 1
rhodium-catalysed hydrosilylation reactions7 and nickel-
catalysed cross-coupling reactions,8 and recently oxazoline-9
The 31P{1H} NMR spectra of L1 and L2 consisted of singlets,
and cyano-functionalised10 N-pyrrolyl phosphines have been
at δ 33.4 and δ 72.2 respectively. The observed downfield shift of
reported. In this paper the syntheses of the P, N -donor ligands
L2 with respect to L1 is consistent with the greater electron
7-aza-N-indolyldiphenylphosphine L1 and 7-aza-N-indolyldi-
withdrawing character of N-pyrrolyl substituents compared to
N-pyrrolylphosphine L2 are presented, along with the reactions
phenyl substituents which leads to increased deshielding of the
of these ligands with palladium(), platinum() and rhodium()
centres. In order to evaluate the effects of the anticipated
1
phosphorus nucleus. The H NMR spectra were assigned on
the basis of H–1H and H–31P correlation experiments. In the
spectrum of L1 the signal for H4 was observed as a doublet of
unresolved multiplets due to the presence of a small 5JHP coup-
1
1
electronic difference between these ligands, preliminary studies
on CO–ethene co-polymerisation catalysed by the methyl
palladium complexes of L1 and L2 are also reported. Since
phenyl and N-pyrrolyl groups are isosteric,11 any difference in
the structure or reactivity of transition metal complexes con-
taining L1 and L2 can be attributed to the difference in the
electronic character of the phosphorus donors.
3
4
5
ling in addition to JHH and JHH coupling. This JHP coupling
was present in all the 7-aza-N-indolyl phosphorus derivatives
studied (vide infra) and small JHP couplings were also observed
in the signals of H2 and H3.
The 1H NMR spectrum of L2 showed distinct resonances for
the pyrrolyl ring protons, with the signals for the α and β
protons appearing as a pseudo quintet and a pseudo triplet
respectively. The signal for H4 appeared as a doublet of doublet
of doublets due to the coupling with H5 and H6 and the
Synthesis of the 7-aza-N-indolyl phosphines L1 and L2
The ligands L1 and L2 were synthesised in a two-step process
involving conversion of 7-azaindole to the lithium 7-azaindolyl
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
4718

View Article Online
5
Table 1 Selected bond lengths (Å) and angles (Њ) for complexes 1a,b
phosphorus atom. The presence of JHP was confirmed by the
1H{31P} NMR spectrum in which the signal for H4 collapsed to
a doublet of doublets. Signals in the 13C{1H} NMR spectra
were assigned through 1H–13C correlation experiments, with
quaternary carbons assigned on the basis of their chemical
shifts.
1a
1b
Pd(1)–N(2)
Pd(1)–P(1)
Pd(1)–Cl(1)
Pd(1)–Cl(2)
P(1)–N(1)
2.044(2)
2.2144(6)
2.3444(6)
2.2875(7)
1.717(2)
Pd–N(1)
Pd–P
Pd–Cl(1)
Pd–Cl(2)
P(1)–N(2)
P(1)–N(3)
P(1)–N(4)
2.0541(13)
2.1867(4)
2.3446(4)
2.2899(4)
1.6871(14)
1.6681(14)
1.6737(14)
Both L1 and L2 are air and moisture sensitive and also sus-
ceptible to alcoholysis, reacting in methanol to give 7-azaindole
and either PPh2OMe or P(NC4H4)2OMe. The observed selective
methanolysis of the functionalised pyrrolyl substituent in L2
is similar to behaviour observed for P(NC4H4)2{NC4H3C(O)-
Me-2},12 and may simply be a consequence of the functional-
ised pyrrolyl group being a better leaving group. However, it
is also possible that hydrogen bonding between the pyridyl
nitrogen atom and the incoming protic nucleophile acts as a
directing force, as has been proposed to rationalise the mois-
ture-sensitivity of N-pyrazolyl phosphines.13 Since N-indolyl14
and N-carbazolyl15,16 phosphines have greater tolerance to
hydrolysis, the sensitivity of L1 or L2 towards protic reagents is
unlikely to be related to steric effects.
N(2)–Pd(1)–P(1)
N(2)–Pd(1)–Cl(1)
P(1)–Pd(1)–Cl(2)
Cl(2)–Pd(1)–Cl(1)
N(2)–Pd(1)–Cl(2)
P(1)–Pd(1)–Cl(1)
85.47(6)
92.22(6)
88.82(2)
93.48(2)
174.29(6)
177.64(2)
N(1)–Pd–P
N(1)–Pd–Cl(1)
P–Pd–Cl(2)
Cl(2)–Pd–Cl(1)
N(1)–Pd–Cl(2)
P–Pd–Cl(1)
85.44(4)
93.33(4)
87.341(15)
93.878(14)
172.78(4)
175.551(14)
The molecular structures of complexes 1a and 1b are shown
in Figs. 1 and 2 respectively, and selected bond distances and
angles are presented in Table 1. The crystal structures con-
firmed the chelating coordination mode of the ligands, with
both complexes showing distorted square planar coordination
geometries. The azaindolyl rings are flat, and the ligands
Palladium(II) and platinum(II) complexes of L1 and L2
The reaction of one equivalent of L1 or L2 with [PdCl2(cod)]
gave the complexes [PdCl2(L1-κ2P, N )] 1a and [PdCl2(L2-κ2P, N )]
1b as yellow crystalline materials in good yields. The analogous
reactions with [PtCl2(cod)] led to the complexes [PtCl2-
(L1-κ2P, N )] 2a and [PtCl2(L2-κ2P, N )] 2b, as colourless crystalline
materials, also in high yield. All four compounds are soluble in
chlorinated solvents, and stable in both the solid state and solu-
tion. This indicates that coordination of L1 or L2 in a chelate
manner reduces the instability of the P–N bond, as previously
observed with the ketophosphines PPh2{NC4H3C(O)Me-2}5
and P(NC4H4)2{NC4H3C(O)Me-2}.6 This is supported by the
observation that compounds 1a,b and 2a,b are not susceptible
to hydrolysis.
The 31P{1H} NMR spectra of 1a,b both consisted of single
phosphorus resonances, at δ 80.9 and δ 70.5 respectively,
whereas the 31P{1H} NMR spectra of 2a,b both comprised
singlets, at δ 55.4 and δ 47.3, with 195Pt satellites. The difference
1
in the value of JPPt observed for 2b (1JPPt 5409 Hz) and 2a
(1JPPt 4125 Hz) follows the general trend for which an increase
in the electronegativity of the substituents on the phosphorus
atom causes an increase in the metal–phosphorus coupling con-
stant.17 In these and other complexes (vide infra) the difference
in δ between analogous complexes of L1 and L2 is much less
than in the free ligands, consistent with greater π-backbonding
in complexes of L2 which serves to reduce the deshielding of the
phosphorus nucleus.
Fig.
1
Molecular structure of [PdCl2(L1)] 1a with the thermal
ellipsoids shown at the 30% probability level.
1
As for the free ligands, the H NMR spectra of complexes
1
1a,b and 2a,b were assigned on the basis of H–1H correlation
experiments. Coordination of the pyridyl nitrogen atom to the
metal centre causes the signal for H6 of the 7-azaindolyl ring to
be shifted downfield with respect to the free ligands L1 or L2
(∆δ 0.70–0.98). A similar downfield shift for the α protons of
coordinated pyridine moieties has been reported and used as
1
evidence for coordination of the nitrogen atom.18 In the H
NMR spectra of the platinum complexes 2a,b the signal for H6
appeared as an unresolved doublet of multiplets with broad
195Pt satellites.
In the 1H NMR spectrum of 1b, the signal for H6 was
4
observed as a doublet of pseudo triplets at δ 9.06 with JHP
through the metal of 1.2 Hz, and the signal for H4 observed as a
5
doublet of doublet of doublets with JHP 1.2 Hz. The presence
1
of these long-range H–31P couplings was confirmed by per-
1
forming selectively decoupled H NMR experiments. Thus the
1H{31P} NMR spectrum of 1b showed the resonance for H4 as a
1
doublet of doublets, whilst the H NMR spectrum recorded
with the signal for H6 selectively decoupled also showed this
resonance as a doublet of doublets.
Fig.
2
Molecular structure of [PdCl2(L2)] 1b with the thermal
ellipsoids shown at the 30% probability level.
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4719

View Article Online
possess crystallographic bite angles of 85.47(6)Њ for L1 in 1a and
85.44(4)Њ for L2 in 1b. These values are larger than those gener-
ally observed in phosphine–imine and –amine19 complexes con-
taining five-membered chelate rings, reflecting the inflexibility
of the five-membered chelate rings in complexes of L1 and L2.
The Pd–P bond distance in 1b [2.1867(4) Å] is shorter than
that observed in 1a [2.2144(6) Å], in agreement with the
increased double bond character which arises from the greater
degree of π-backbonding to L2. As expected there are no sig-
nificant differences in the steric requirements of L1 and L2 as
evidenced by the half crystallographic cone angles of the
pyrrolyl (80Њ and 60Њ) and phenyl (79Њ and 61Њ) groups. The
mutually cis chlorides have different Pd–Cl bond lengths reflect-
ing the larger trans influence of the phosphorus donor over the
nitrogen donor. The sums of the angles, Σθ, around the nitrogen
atoms in the 7-azaindolyl fragments are similar in 1a,b and
within the range 358Њ–360Њ, with the exception of N(4) in 1b for
which Σθ is 352.8Њ.
Rhodium(I) complexes of L1 and L2
Fig. 3 Molecular structure of [RhCl(CO)(L1)] 3a with the thermal
ellipsoids shown at the 30% probability level.
The reaction of [Rh(µ-Cl)(cod)]2 with two equivalents of either
L1 or L2 in the presence of CO gave the rhodium() carbonyl
complexes[RhCl(CO)(L1-κ2P, N )]3aand[RhCl(CO)(L2-κ2P, N )]
3b in good yields. Both complexes were isolated as yellow crys-
tals and fully characterised on the basis of multinuclear NMR
and IR spectroscopy, microanalysis and X-ray crystallography.
The 31P{1H} NMR spectra of 3a,b consisted of doublets with
that for 3a at δ 104.1 (1JPRh 180 Hz) and that for 3b at δ 101.9
(1JPRh 242 Hz). In both cases the resonances are downfield
relative to the free ligand, and the greater value of 1JPRh for 3b is
consistent with the anticipated π-acceptor abilities of the phos-
1
phorus ligands. These relatively large values of JPRh suggest
that the complexes both assume coordination arrangements in
which the phosphorus atom lies trans to the chloride rather
than the carbonyl.
The 1H NMR spectra were as expected, and as with 1a,b and
2a,b the chelating coordination mode of the ligands causes the
signals for H6 of the 7-azaindolyl rings to be shifted downfield
with respect to those of the free ligands. The 13C{1H} NMR
spectra confirmed the geometry suggested by the 31P{1H} NMR
spectra, with the small values of 2JCP reflecting the mutually cis
coordination of the carbonyl and phosphorus, as anticipated
for the thermodynamically favoured products.20 The carbonyl
stretching frequencies in the IR spectra were observed at 1992
cmϪ1 for 3a and 2017 cmϪ1 for 3b. As with the coupling constant
information, this provides clear evidence that L2 has a lower
σ-donor/higher π-acceptor ability than L1.
X-ray structural analyses for complexes 3a and 3b confirmed
the stereochemistry at the rhodium centres proposed on the
basis of the NMR data. The molecular structures are illustrated
in Figs. 3 and 4 and selected bond lengths and angles are
reported in Table 2. For 3a, two crystallographically independ-
ent molecules were present in the unit cell, though due to their
similarity only one molecule is depicted.
The metal centres in 3a,b display distorted square planar
geometry with the carbonyl ligand occupying a trans position
relative to the nitrogen. The crystallographic bite angles for
L1 and L2, 84.41(8)Њ and 84.57(7)Њ for 3a and 84.27(4)Њ for 3b,
are very similar to those in the palladium complexes 1a,b. The
Rh–P distance is shorter in 3b [2.1646(4) Å] than in 3a
[2.1977(9) Å and 2.2108(8) Å], and the observed difference
in the Rh–P bond distances between 3b and 3a is similar to
that reported between the complexes trans-[RhCl(CO){P(NC4-
H4)3}2] [Rh–P 2.282(4) Å] and trans-[RhCl(CO){PPh3}2]
[av. Rh–P 2.325 Å].4
Fig. 4 Molecular structure of [RhCl(CO)(L2)] 3b with the thermal
ellipsoids shown at the 30% probability level.
Synthesis of methylpalladium complexes of L1 and L2 and inser-
tion of CO and ethene
The complexes [PdClMe(L1-κ2P, N )] 4a and [PdClMe(L2-
κ2P, N )] 4b were prepared by the addition of one equivalent of
[PdClMe(cod)] to dichloromethane solutions of either L1 or L2,
and [PdMe(OTf )(L4-κ2P, N )] 5a and [PdMe(OTf )(L2-κ2P, N )] 5b
were prepared by chloride abstraction with AgOTf from 4a,b
respectively. Compounds 4a,b and 5a,b were isolated in almost
quantitative yield as air and moisture sensitive crystalline
materials and fully characterised. Each of the 31P{1H} NMR
spectra contained a singlet, shifted significantly downfield
relative to the corresponding free ligand. Coordination of
the nitrogen atom can be inferred by the chemical shift for the
7-azaindolyl H6 protons, which are downfield from those of the
free ligands, albeit with a smaller displacement than observed
for 1a,b and 2a,b (∆δ 0.12–0.37).
The appearance of the resonances for the methyl protons are
characteristic of the coordination geometries in these com-
plexes. Those in the 1H NMR spectra of 4a,b and 5a appeared
as doublets with 3JHP between 1.6 and 4.4 Hz whereas that of 5b
was observed as a singlet. The low magnitude of 3JHP indicates
a cis arrangement of the phosphorus atom and methyl group
and therefore formation of the isomer in which the ligands with
The crystallographic half cone angles for the phenyl (av. 70.4Њ)
and pyrrolyl (av. 68.2Њ) substituents indicate that these are iso-
steric, and Σθ around the nitrogen atoms are very close to 360Њ
with the exception of Σθ around N(3) in 3b which is 351.9Њ.
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4720

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) for complexes 3a,b
3a (molecule 1)
3a (molecule 2)
3b
Rh(1)–C(1)
Rh(1)–P(1)
Rh(1)–N(1)
Rh(1)–Cl(1)
P(1)–N(2)
1.821(4)
2.1977(9)
2.127(3)
2.3752(8)
1.712(3)
Rh(2)–C(21)
Rh(2)–P(2)
Rh(2)–N(3)
Rh(2)–Cl(2)
P(2)–N(4)
1.821(3)
2.2108(8)
2.109(3)
2.3620(8)
1.722(3)
Rh(1)–C(1)
Rh(1)–P(1)
Rh(1)–N(1)
Rh(1)–Cl(2)
P(1)–N(2)
P(1)–N(3)
P(1)–N(4)
1.8305(19)
2.1646(4)
2.1184(15)
2.3620(4)
1.7006(15)
1.6915(15)
1.6837(16)
N(1)–Rh(1)–P(1)
P(1)–Rh(1)–C(1)
C(1)–Rh(1)–Cl(1)
Cl(1)–Rh(1)–N(1)
N(1)–Rh(1)–C(1)
P(1)–Rh(1)–Cl(1)
84.41(8)
90.30(11)
93.18(11)
92.04(8)
173.92(13)
176.18(3)
N(3)–Rh(2)–P(2)
P(2)–Rh(2)–C(21)
C(21)–Rh(2)–Cl(2)
Cl(2)–Rh(2)–N(3)
N(3)–Rh(2)–C(21)
P(2)–Rh(2)–Cl(2)
84.57(7)
93.30(10)
90.93(10)
91.28(7)
177.67(12)
173.06(3)
N(1)–Rh(1)–P(1)
P(1)–Rh(1)–C(1)
C(1)–Rh(1)–Cl(2)
Cl(2)–Rh(1)–N(1)
N(1)–Rh(1)–C(1)
P(1)–Rh(1)–Cl(2)
84.27(4)
92.05(6)
91.54(6)
92.12(4)
176.30(7)
175.852(16)
the largest trans influence are cis to each other. The lower value
3
for JHP for complexes of L2 is in agreement with this being a
poorer σ-donor than L1. The signals for the methyl group in the
13C{1H} NMR spectra appeared as singlets or as doublets with
small 2JCP confirming this geometry.
The reactions of complexes 4a,b and 5a,b with CO in CD2Cl2
at ambient temperature were followed by 31P{1H} and 1H NMR
spectroscopy. The insertion of CO into the Pd methyl bond of
4a,b and 5a,b produced the palladium acyl complexes [PdCl-
{C(O)Me}(L1-κ2P, N )] 6a, [PdCl{C(O)Me}(L2-κ2P, N )] 6b,
[Pd{C(O)Me}(OTf )(L1-κ2P, N )] 7a and [Pd{C(O)Me}(OTf )-
(L2-κ2P, N )] 7b respectively. These complexes were isolated as
crystalline materials and fully characterised by NMR and IR
spectroscopy. In addition the structures of 7a and 7b were
confirmed crystallographically.
Complexes 6a,b and 7a,b are air and moisture sensitive,
decomposing when exposed to moisture to yield palladium
metal. This feature prevented satisfactory elemental analyses
for 6b and 7a,b from being obtained. The CO insertion reac-
tions are irreversible, as evidenced by the fact that both the
31P{1H} and 1H NMR spectra of 6a,b and 7a,b did not undergo
any significant changes for more than 24 h on removal of un-
reacted CO.
The reactions of the chloro methyl complexes 4a,b to pro-
duce 6a,b were slow, requiring more than 24 h to reach comple-
tion. In contrast 5a,b exhibited reaction times of a few minutes,
which parallels previous results showing that the use of more
weakly-coordinating anions causes an increase in the rate of
CO insertion.21
Fig. 5 Molecular structure of [Pd{C(O)Me}(OTf )(L1)] 7a with the
thermal ellipsoids shown at the 30% probability level.
and 31P{1H}, 13C{1H} and 1H NMR spectra were recorded
both at ambient temperature and Ϫ70 ЊC. The 31P{1H} NMR
spectrum showed the phosphorus resonance for 7a as a
2
doublet centred at δ 65.5 with JCP 6 Hz, whereas the 13C{1H}
NMR spectrum showed an intense doublet at δ 220.5 with
1
2JCP 6 Hz, assigned to the acyl group of 7a. The H NMR
spectrum showed the protons of the methyl group as a
4
The complexes 6a,b and 7a,b afforded very similar spectro-
scopic data, suggesting structural similarities. The IR spectra
showed strong absorptions in the range 1697–1729 cmϪ1 which
doublet of doublets centred at δ 2.28 with JHP 1.2 Hz and
2JHC 6.0 Hz. These data are consistent with 13C-labelled 7a
and indicate the absence, under these reaction conditions,
of significant amounts of any palladium acyl carbonyl
complexes.
are typical for Pd-acyl C᎐O stretches. In the 31P{1H} NMR
᎐
spectra significant upfield shifts characteristic of the insertion
of CO in the Pd–Me bond were observed. The 1H NMR spectra
showed significant downfield shifts for the resonances of the
methyl group and little change in the aromatic region. In the
13C{1H} NMR spectra the signals for the methyl groups were
significantly shifted compared with those of the parent pal-
ladium methyl complexes and appeared as doublets with 3JCP in
the range 29–46 Hz. The resonances for the carbonyl group
were observed in the expected range as doublets with small 2JCP
coupling constants. The magnitude of 3JCP and 2JCP suggest that
the acyl group is cis to the phosphorus.
The structures of complexes 7a and 7b were confirmed by
X-ray crystallography. Crystals suitable for X-ray diffraction
studies were obtained by slow diffusion of hexane into CD2Cl2
solutions under an atmosphere of CO. The crystal structures of
7a,b are shown in Figs. 5 and 6 with selected bond distances and
angles given in Table 3. Complex 7a crystallised with three
independent molecules in the asymmetric unit but for clarity
reasons only one of these is depicted in Fig. 5.
The solid state structures are consistent with the solution
structures, showing the palladium centres in distorted square
planar coordination geometries with the acyl group and the
phosphorus atom occupying relative cis positions. Deviations
from idealised square planar coordination geometries are
small as evidenced by the values of the cis and trans angles at
the palladium centres. The Pd–N bond distances are longer in
the structures of 7a,b than in those of 1a,b, which reflects the
increased trans effect of the acyl group compared with chloride.
The Pd–C distances in 7a,b are both normal for palladium acyl
bonds trans to nitrogen atom, and the acyl groups adopt
1
In monitoring the CO insertion reaction by H and 31P{1H}
NMR spectroscopy no signals apart from those of the starting
materials and the products were observed, indicating that the
necessary pre- or post- insertion isomerisation reactions, one of
which must have occurred to give the observed product, are too
fast to be detected at ambient temperature.22
In an effort to detect the presence of palladium acyl carbonyl
complexes such as [Pd{C(O)Me}(CO)(L1-κ2P, N )]OTf,
a
CD2Cl2 solution of 5a was exposed to an atmosphere of 13CO,
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4721

View Article Online
Table 3 Selected bond lengths (Å) and angles (Њ) for complexes 7a,b
appears as a triplet due to the coupling with the other methyl-
ene group (3JHH 6.2 Hz), whereas the latter was assigned to the
CH2 group bonded to the palladium and appears as a triplet of
doublets due to an additional coupling with the phosphorus
atom (3JHP 1.2 Hz). The methyl protons were observed as a
singlet at δ 2.49. The 1H NMR spectrum of 8b presented similar
features to that of 8a though the signals for the protons of
the methylene groups appeared as distinct but unresolved
multiplets.
7b
7a
Molecule 1
Molecule 2
Molecule 3
Pd–P
Pd–N
Pd–O
Pd–C
P–Na
2.2106(6)
2.193(2)
2.1345(16)
1.994(3)
1.718(2)
2.1796(12)
2.190(4)
2.151(4)
1.989(5)
1.694(4)
2.1851(11)
2.196(4)
2.119(4)
1.993(5)
1.684(4)
2.1813(11)
2.205(4)
2.143(3)
1.990(5)
1.697(4)
The 13C{1H} NMR spectra of 8a,b were assigned on the basis
of 1H–13C NMR correlation experiments. The only unexpected
feature is the presence in both spectra of a 4JCP coupling of 3 Hz
between the phosphorus atom and the carbon of the methyl
group. The IR spectra showed strong ν(CO) absorptions at 1631
cmϪ1 for 8a and 1628 cmϪ1 for 8b, which are shifted to lower
frequency relative to those for the parent acyl complexes (7a,
1711 cmϪ1; 7b 1729 cmϪ1) indicating coordination of the
ketonic oxygen.
P–Pd–N
N–Pd–O
O–Pd–C
C–Pd–P
P–Pd–O
N–Pd–C
85.89(5)
88.53(7)
95.59(8)
89.83(7)
173.67(5)
174.85(9)
85.39(10)
92.02(14)
90.60(18)
91.96(15)
169.50(10)
177.35(18)
84.59(10)
93.17(14)
90.95(17)
91.76(13)
171.44(11)
174.90(17)
84.17(10)
97.39(12)
86.60(15)
91.71(13)
175.93(8)
175.49(15)
a Azaindolyl nitrogen atoms.
Crystals of 8a and 8b suitable for X-ray crystallographic
studies were obtained by the slow diffusion of hexane into
CD2Cl2 solutions saturated with ethene. The structures of 8a
and 8b are shown in Figs. 7 and 8 while selected bond distances
and angles are given in Table 4. The metal centre in both com-
plexes exhibits a distorted square planar coordination geometry
with the coordinated ketonic oxygen trans to the phosphorus.
The most significant difference between the structures is the
Pd–P bond length, which is longer in 8a than in 8b by 0.03 Å,
in a similar manner to the other structures in this paper
(vide supra).
Fig. 6 Molecular structure of [Pd{C(O)Me}(OTf )(L2)] 7b with the
thermal ellipsoids shown at the 30% probability level.
orientations approximately perpendicular to the metal
coordination planes.
As observed in the X-ray crystal structures of 1a,b and 3a,b,
substitution of the two phenyl rings with two N-pyrrolyl rings
leads to a decrease in the M–P bond distance by approximately
0.03 Å. Σθ around the nitrogen atoms in both structures are
close to 360Њ in agreement with their formal sp2 hybridisations.
However, as observed in the structures of 1b and 3b, one of
the nitrogen atoms of the pyrrolyl substituents in 7b has a
significant pyramidal distortion (mean 353Њ).
Fig. 7 Molecular structure of [Pd{CH2CH2C(O)Me-κ2C,O}(L1)]OTf
8a with the thermal ellipsoids shown at the 30% probability level and
the anion omitted for clarity.
The insertion of ethene in the palladium acyl bonds of
complexes 7a,b produced the alkyl complexes [Pd{CH2CH2C-
(O)Me-κ2C,O}(L1-κ2P, N )]OTf 8a and [Pd{CH2CH2C(O)-
Me-κ2C,O}(L2-κ2P, N )]OTf 8b. These reactions were slower than
the CO insertions, requiring several hours to reach completion.
The conversions of 7a into 8a and 7b into 8b were monitored
by NMR spectroscopy, which indicated that, as for the CO
insertion reactions, the pre- or post- isomerisations necessary to
generate the observed isomers are too fast to be detected.
The configurations of 8a,b were determined by multinuclear
NMR and IR spectroscopy and confirmed by X-ray single
crystal diffraction studies. The 31P{1H} NMR spectra of 8a,b
showed singlets, at δ 84.3 and δ 91.1 respectively, which are
8a and 8b were further reacted with CO to yield [Pd{C(O)-
CH2CH2C(O)Me-κ2C,O}(L1-κ2P, N )]OTf 9a and [Pd{C(O)-
CH2CH2C(O)Me-κ2C,O}(L2-κ2P, N )]OTf 9b. Both reactions
were monitored by NMR spectroscopy and found to be con-
siderably slower than the CO insertion reactions into 5a,b,
requiring several hours to reach completion. This decrease in
reactivity has been observed before for similar palladium
complexes and rationalised on the basis of the 5-membered
metallocyclic ring needing to open for CO to coordinate.23
However, it is notable that analogous complexes to 8a,b
containing phosphine–imine ligands do not react with CO in
dichloromethane, requiring the presence of acetonitrile to
facilitate reaction.24 In contrast to the reactions of 5a,b with
CO, the reactions of 8a,b with CO are reversible. Consequently
9a,b were only observed in solutions saturated with CO.
1
shifted to low field with respect to 7a,b. In the H NMR spec-
trum of 8a, the chemically inequivalent methylene groups gave
rise to two distinctive signals at δ 3.16 and δ 1.99. The former
was assigned to the CH2 group bonded to the acyl group and
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4722

View Article Online
Table 4 Selected bond lengths (Å) and angles (Њ) for complexes 8a,b
both starting materials and products, the applied pressures of
both monomers were kept low. Before exposure to ethene,
excess CO was removed by freezing the sample with liquid
nitrogen and applying a reduced pressure. Conversions were
8a
8b
Pd(1)–C(20)
Pd(1)–O(1)
Pd(1)–N(2)
Pd(1)–P(1)
C(22)–O(1)
P(1)–N(1)
2.029(4)
2.102(2)
2.137(3)
2.1808(10)
1.238(4)
1.716(3)
Pd(1)–C(1)
Pd(1)–O(1)
Pd(1)–N(1)
Pd(1)–P(1)
C(3)–O(1)
P(1)–N(2)
P(1)–N(3)
P(1)–N(4)
2.027(5)
2.102(3)
2.148(4)
2.1532(15)
1.250(6)
1.692(4)
1.694(4)
1.666(4)
1
determined by comparing the integrals in the 31P{1H} and H
NMR spectra for the insertion products with those of the par-
ent compounds. The results indicated that the insertion of CO
into the palladium methyl bond of 5b is approximately 1.3
times faster than in that of 5a, whereas ethene insertion into the
palladium acyl bond in 7a is approximately 3 times faster than
in 7b. Although the former finding is not likely to be significant,
the latter result is consistent with the greater stability of the Pd–
O bond trans to the more electron withdrawing L2 thus reducing
the rate of anion loss in 7b over 7a.
N(2)–Pd(1)–P(1)
C(20)–Pd(1)–O(1)
O(1)–Pd(1)–N(2)
C(20)–Pd(1)–P(1)
O(1)–Pd(1)–P(1)
C(20)–Pd(1)–N(2)
85.69(9)
82.88(13)
97.98(11)
93.46(11)
175.92(7)
179.11(13)
N(1)–Pd(1)–P(1)
C(1)–Pd(1)–O(1)
O(1)–Pd(1)–N(1)
C(1)–Pd(1)–P(1)
O(1)–Pd(1)–P(1)
C(1)–Pd(1)–N(1)
85.31(12)
83.57(19)
97.80(16)
93.33(16)
174.43(10)
178.6(2)
Catalytic CO ethene co-polymerisation experiments
Following on from the insertion studies, a series of catalytic test
reactions for the CO–ethene co-polymerisation23,25 catalysed by
5a,b was undertaken. While this reaction is normally carried
out with P, P-26 or N,N-27 donor ligands, there have been a
number of reports on the activity of complexes containing P, N -
donor ligands in this reaction.19,24,28 Although less active than
catalysts with homotopic ligands, use of these heterotopic
ligands has allowed a number of key intermediates to be iso-
lated and crystallographically characterised. The activities
(Table 5) are expressed in grams of polyketone (PK) per gram
of palladium based on the mass of the insoluble material iso-
lated at the end of each run. Pure white precipitates were
obtained only in the experiments performed in pure dichloro-
methane (runs 1 and 4). All the other experiments produced off-
white materials owing to the presence of some palladium metal.
Analysis of the filtrate from runs 2, 3 and 5 by 1H NMR
spectroscopy did not reveal detectable quantities of methyl-
propanoate—the product of the hydromethoxycarbonylation
of ethene. Catalytic reactions performed in pure methanol did
not produce any PK or methylpropanoate and led only to the
decomposition of the catalysts.
Characterisation of the obtained polymeric materials was
performed by 13C{1H} NMR and IR spectroscopy and by
measurement of the melting points. The IR spectra of the
polymeric materials isolated from all the runs were very similar
and contained an intense carbonyl absorption band at 1695
cmϪ1 in agreement with previous results.23 The melting points
were observed at temperatures around 250 ЊC and are in the
range reported in the literature (240–260 ЊC) for polymers with
a CO content of 50%, alternating structures and high molecular
weights.29 The 13C{1H} NMR spectra contained resonances at
δ 213.1 and δ 36.1 with intensities in the ratio 1:2, which can be
assigned to the methylene and carbonyl carbons of PK
respectively.
Low intensity signals were observed in the range δ 0–50,
assignable to the terminal groups of the polymeric chain. The
intensity ratios indicated a molecular mass of approximately
23000.30 13C{1H} NMR spectra for PK isolated from the reac-
tions in the presence of methanol showed an increase in inten-
sity of the signals from the end groups, implying a decrease in
the molecular weights of the obtained polymers.
These reactions demonstrate that 5a,b are active catalysts for
the CO–ethene co-polymerisation, with 5b affording a slightly
higher activity than 5a. The polymeric materials isolated from
the reaction catalysed by both complexes have similar molecu-
lar weights. The obtained activities for 5a,b are comparable
with those reported for palladium catalysts modified with other
heteroditopic bidentate ligands, though low if compared with
the activities obtained using palladium complexes modified
with diphosphine ligands.19,24,28 An increase in the pressure of
the monomers as well as the presence of an acid co-catalyst
(run 6) resulted in an increase in the quantity of the isolated
polymers.
Fig. 8 Molecular structure of [Pd{CH2CH2C(O)Me-κ2C,O}(L2)]OTf
8b with the thermal ellipsoids shown at the 30% probability level and
the anion omitted for clarity.
The 31P{1H} NMR spectra of 9a,b both consisted of singlets,
shifted to high field with respect to 8a,b. The 1H NMR spectra
showed the signal for the 7-azaindolyl H6 proton at δ 8.42 as a
broad doublet for 9a and at δ 8.45 as a doublet of doublets for
9b. In both cases the methylene protons were observed as a
multiplet between δ 2.7 and δ 2.9. The only signals observed
in the NMR spectra were those assigned to the products and
starting materials, so as for the insertions of CO into 5a,b and
ethene into 7a,b, the pre or post isomerisation processes are too
fast to be detected.22 The IR spectra showed two distinctive
carbonyl stretching frequencies (9a, 1711, 1622 cmϪ1; 9b 1716,
1660 cmϪ1) consistent with the presence of coordinated and
uncoordinated carbonyl groups.
Following the studies on the stepwise migratory insertion
reactions, CO and ethene were bubbled simultaneously into a
CD2Cl2 solution of 5a and the reaction monitored by NMR
spectroscopy. The spectra showed that CO and ethene insertion
reactions progress consecutively. Within a few minutes 5a is
converted into 7a, which over the course of 2 h reacts with
ethene to produce 8a. Continued passage of CO and ethene
through the solution did not result in the formation of 9a which
evidently requires higher CO pressure and/or longer reaction
times to be formed.
In order to determine the relative rates of insertion for the
complexes of L1 and L2, equimolar amounts of 5a,b were dis-
solved in CD2Cl2, exposed to an atmosphere of CO and ethene
and the reactions monitored by NMR spectroscopy. Since it
was of interest to record spectra that contain the signals for
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4723

View Article Online
Table 5 Catalytic results for complexes 5a,b in the co-polymerisation of CO and ethenea
Run
Complex
n/mmol
p(CO)/psi
p(C2H4)/psi
T/ЊC
t/h
MeOH
PK/mg
g(PK)/g(Pd)
1
2
3
5a
5a
5a
0.017
0.061
0.075
20
30
40
20
30
40
50
25
25
24
24
48
0
70
150
330
38
22
82
300 µl
10 ml
4
5
5b
5b
0.047
0.042
20
20
20
20
50
25
20
24
0
320
240
64
55
500 µl
6
5bb
0.045
20
20
50
20
0
700
146
a Solvent CH2Cl2 (50 ml). b 5 µl of HSO3CF3 added.
Table 6 Selected bond lengths (Å) and angles (Њ) for complex 10
Methanol-induced termination reactions
The catalytic experiments demonstrated that methanol has
a negative effect in the co-polymerisation of CO and ethene
catalysed by 5a,b. However as the molecular weight of the
polymers obtained from the experiments performed with small
amounts of methanol remained quite high, the decrease in
activity could be due to the lack of an efficient chain transfer
process with the termination reactions not producing species
able to start a new catalytic cycle. This may also explain why
methylpropanoate is not produced in the catalytic experiments
performed in the presence of methanol and why the catalytic
reactions in pure methanol ended with complete catalyst
decomposition.
Pd(1)–P(1)
Pd(1)–P(2)
Pd(1)–N(2)
Pd(1)–N(4)
P(1)–N(1)
P(2)–N(3)
2.2602(11)
2.2481(12)
2.127(4)
2.115(3)
1.705(4)
1.705(4)
N(4)–Pd(1)–N(2)
N(2)–Pd(1)–P(1)
N(4)–Pd(1)–P(2)
P(2)–Pd(1)–P(1)
N(2)–Pd(1)–P(2)
N(4)–Pd(1)–P(1)
97.73(14)
83.85(10)
84.56(11)
94.50(4)
172.32(11)
174.87(10)
According to the proposed mechanism for the CO–ethene
co-polymerisation in methanol,23 two termination reactions are
possible: protonolysis of the palladium alkyl bond of com-
plexes similar to [Pd{CH2CH2C(O)Me-κ2C,O}(L1-κ2P, N )]OTf
8a or methanolysis of a palladium acyl bond of complexes simi-
lar to [Pd{C(O)Me}(OTf )(L1-κ2P, N )] 7a and [Pd{C(O)CH2-
CH2C(O)Me-κ2C,O}(L1-κ2P, N )]OTf 9a. In order to establish
which of the two termination processes is active in the CO–
ethene co-polymerisation experiments catalysed by complexes
5a,b, the reactions of 7a, 8a and 9a with approximately equiv-
alent amounts of methanol in CD2Cl2 at ambient temperature
were followed by multinuclear NMR spectroscopy. The 1H
NMR spectra for the reactions of 7a and 9a showed that
methanol is quantitatively converted to methyl acetate and
methyl 4-keto-pentanoate respectively. Both reactions were
slow, taking more than 24 h to reach completion, and pro-
ceeded with the precipitation of significant amounts of
1
palladium black. On the contrary, the H and 31P{1H} NMR
spectra for 8a showed no significant variation after 2 days. In
the 31P{1H} NMR spectra for the reactions of 7a and 9a with
MeOH, the disappearance of the signals for the starting
material is accompanied by the appearance of a new singlet
resonance at δ 90.2 thus indicating that both complexes are
converted quantitatively into a new palladium complex 10. The
1H NMR spectra for 10 showed complex signals in the aromatic
region, which were difficult to interpret, but no signal in the
alkyl region.
Filtration of the solution to remove palladium, followed by
layering with hexane gave crystalline material suitable for X-ray
crystallography, which subsequently revealed the identity of 10
to be the dicationic palladium complex cis-[Pd(L1-κ2P, N )2]-
(OTf )2. The molecular structure of 10ؒ1.5H2O is shown in
Fig. 9 and selected bond lengths and angles are reported in
Table 6. The palladium centre has a distorted square planar
coordination geometry with the L1 ligands coordinated in a cis
orientatation. The Pd–N bond distances of 2.115(3) Å and
2.127(4) Å are appreciably longer than the Pd–N bond distance
in 1a, indicating the higher trans influence of the diphenyl-
phosphino group compared to chloride in addition to steric
effects. The two planes of the 7-aza-indolyl rings are not
coplanar but twisted by approximately 22Њ, which is probably a
consequence of the steric repulsion between the H6 hydrogen
atoms on the azaindolyl rings. A separation between the two
Fig. 9 Molecular structure of [Pd(L1)2](OTf )2 10 with the thermal
ellipsoids shown at the 30% probability level and the anions omitted for
clarity.
hydrogen atoms of 2.15 Å is observed, whereas in a hypo-
thetical planar structure this distance would have been 1.5 Å.
In order to confirm that cis-[Pd(L1)2](OTf )2 is the compound
observed in solution, this and the analogous platinum complex
cis-[Pt(L1-κ2P, N )2](OTf )2 11 were synthesised in a more logical
manner. Thus the reaction of [PdCl2(cod)] with two equivalents
of L1 in the presence of AgOTf gave 10 whereas the reaction of
2a with one equivalent of L1 in the presence of AgOTf gave 11.
The product from the [PdCl2(cod)] reaction exhibited identical
spectra to 10, whereas the 31P{1H} NMR spectrum of 11 con-
sisted of a singlet at δ 62.8 with 195Pt satellites (1JPPt 3601 Hz).
The high value of 1JPPt is indicative of a cis arrangement of the
two phosphorus atoms in a similar manner to that observed in
10.
The methanolysis of 7b proceeded in an analogous manner
to that of complex 7a, converting methanol into methyl acetate
on the basis of 1H NMR spectroscopy. The 31P{1H} NMR
spectra showed the progressive disappearance of the signal
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4724

View Article Online
Scheme 2
assigned to 7b and appearance of a broad singlet at δ 129.6
decomposition of a palladium hydride intermediate produces a
which was tentatively assigned to [Pd(L2)2](OTf )2, though
attempts to isolate this complex were unsuccessful.
bis(chelate) complex 10 that is responsible for the obtained low
activities. A mechanism for the decomposition pathway has
been proposed, and 10 has been isolated and crystallographic-
ally characterised.
Since the reaction of 7a with methanol proceeded with the
conversion of methanol into methyl acetate, this suggests the
initial step in the conversion of 7a to 10 involves formation of a
palladium hydride complex. 1H NMR spectra of the crude
reaction mixture contained signals in the hydride region though
unfortunately this intermediate could not be fully characterised
or isolated because of its ready decomposition into 10 and
palladium black. This finding begged the question of whether
or not 10 itself was active in the CO–ethene co-polymerisation;
however the failure of 10 to react with CO or ethene either in
the presence or absence of methanol indicated that this is not
so.
Although the electronic difference between L1 and L2 is mani-
fested in structural parameters such as M–P and P–N bond
lengths, it does not lead to significant differences in the catalytic
activity of their palladium methyl complexes in CO–ethene
co-polymerisation. However, differences in reactivity from
phosphine–imine compounds have been observed.
Experimental
General experimental
The formation of palladium hydride complexes through
methanolysis of palladium acyl complexes has been reported
previously.31 In particular the reaction of methanol with
[PdCl{C(O)Me}(bdpp)] [bdpp = 2,4-bis(diphenylphosphino)-
pentane] in the presence of CO resulted in the formation of
methyl acetate and hydrogen as well as a 1:1 mixture of
[PdCl2(bdpp)] and [Pd2(µ-H)(µ-CO)(bdpp)2]Cl.32 A similar
palladium hydride dimer has been isolated and crystallo-
graphically characterised by using the ligand 1,3-bis(di-iso-
propylphosphino)propane.33
The formation of 10 from the reaction of 7a with methanol
can be tentatively rationalised as shown in Scheme 2. The
methanolysis of the acyl palladium bond in 7a generates an
unstable monomeric or dimeric palladium hydride complex that
rapidly decomposes with the formation of 10, palladium metal
and hydrogen. The observed unreactivity of 10 with CO or
ethene together with the plausible assumption that reaction of
ethene with the palladium hydride complex generated by the
methanolysis of complexes similar to 7a or 9a is slower than its
decomposition makes an efficient chain transfer process for the
CO–ethene co-polymerisation catalysed by complex 5a unavail-
able. This helps to rationalise the results obtained in the
catalytic test reactions.
Reactions were routinely carried out using Schlenk-line tech-
niques under pure dry dinitrogen or argon, using dry dioxygen-
free solvents unless noted otherwise. Microanalyses (C, H and
N) were carried out by Mr Alan Carver (University of Bath
Microanalytical Service). Infrared spectra were recorded on a
Nicolet 510P spectrometer as KBr pellets, Nujol mulls on KBr
discs or in solutions using KBr cells. NMR spectra were
recorded on JEOL EX-270, Varian Mercury 400 and Bruker
Avance 300 spectrometers referenced to TMS or 85% H3PO4.
36
The complexes [PdCl2(cod)],34 [PtCl2(cod)],35 [Rh(µ-Cl)(cod)]2
and [PdClMe(cod)]37 were prepared by standard literature
methods. Pyrrole was dried over molecular sieves and distilled
before use, and triethylamine was distilled over potassium
before use. PCl(NC4H4)2 was prepared as described
previously.15,38
Synthesis of 7-aza-N-indolyldiphenylphosphine L1
A solution of BuLi in hexane (5.0 cm3, 2.5 M) was added drop-
wise to a solution of 7-azaindole (1.50 g, 12.7 mmol) in THF–
hexane at Ϫ78 ЊC. The reaction mixture was allowed to warm to
ambient temperature and stirred for further 30 min. The yellow
solution was added dropwise to a cold hexane solution of
PClPh2 (2.75 g, 12.5 mmol). After 4 h stirring, the white precipi-
tate was removed by filtration to give a colourless solution.
After evaporation of the solvent under reduced pressure the
colourless residue was triturated in hexane to give a white pow-
der. The product was recrystallised from dichloromethane–
hexane to give L1. Yield: 2.70g (72%). Calc. for C19H15N2P: C,
75.5; H, 5.00; N, 9.27%. Found: C, 75.1; H, 4.99; N, 9.02%.
31P{1H} NMR (161.8 MHz, CDCl3): δ 33.4 (s). 1H NMR (399.8
Conclusions
P, N -Donor ligands containing π-accepting N-pyrrolyl groups
can be prepared from 7-azaindole in a two-step process involv-
ing lithiation followed by reaction with a chlorophosphine.
These ligands readily coordinate to palladium(), platinum()
and rhodium(), and both spectroscopic and structural evidence
confirm that the di(N-pyrrolyl)-substituted ligand L2 is a
stronger π-acceptor/weaker σ-donor than the diphenyl-substi-
tuted ligand L1.
The controlled stepwise insertion reactions of CO and ethene
in the palladium methyl complexes 5a,b provided the acyl
palladium complexes 7a,b and alkyl complexes 8a,b which were
fully characterised. Preliminary catalytic experiments for the
CO–ethene co-polymerisation showed that 5a,b are both active
catalysts with 5b affording slightly higher activities. The pres-
ence of methanol causes a decrease in activity or catalyst
decomposition, and studies on the termination reactions caused
by methanol suggested that under the catalytic conditions the
3
4
MHz, CDCl3): δ 8.40 (dd, 1H, JHH 4.8 Hz, JHH 1.6 Hz, H6),
7.91 (dm, 1H, 3JHH 8.0 Hz, H4), 7.40–7.33 (m, 10H, Ho, Hm, Hp),
7.12 (dd, 1H, 3JHH 4.8, 8.0 Hz, H5), 6.96 (dd, 1H, 3JHH 3.6 Hz,
3
4
3JHP 1.8 Hz, H2), 6.57 (dd, 1H, JHH 3.6 Hz, JHP 1.2 Hz, H3).
13C{1H} NMR (75.5 MHz, CDCl3): δ 152.9 (d, 2JCP 16 Hz, C8),
143.9 (s, C6), 136.6 (d, 1JCP 14 Hz, Ci), 132.6 (d, 2JCP 20 Hz, Co),
130.3 (d, 2JCP 7 Hz, C2), 130.0 (s, Cp), 129.2 (s, C4), 129.0 (d, 3JCP
7 Hz, Cm), 122.4 (s, C9), 117.1 (s, C5), 104.9 (s, C3).
Synthesis of 7-aza-N-indolyl-di-N-pyrrolyl phosphine L2
A solution of BuLi in hexane (2.0 cm3, 2.5 M) was added drop-
wise to a solution of 7-azaindole (0.61 g, 5.2 mmol) in THF–
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4725

View Article Online
1
hexane at Ϫ78 ЊC. The reaction mixture was allowed to warm to
ambient temperature and stirred for further 30 min. The yellow
solution was added dropwise to a cold solution of PCl(NC4H4)2
(1.03 g, 5.2 mmol) in THF–hexane. After 2 h stirring, the white
precipitate which had formed was removed by filtration to give
a colourless solution. After evaporation of the solvent under
reduced pressure, the colourless residue was washed with cold
hexane to obtain L2 as a colourless oil. Yield: 1.05 g (72%).
31P{1H} NMR (161.8 MHz, CDCl3): δ 72.2 (s). 1H NMR (399.8
MHz, CDCl3): δ 8.36 (dd, 1H, JHH 4.8 Hz, JHH 1.6 Hz, H6),
7.91 (ddd, 1H, 3JHH 8.0 Hz, 4JHH 1.6 Hz, 5JHP 1.0 Hz, H4), 7.16
(dd, 1H, 3JHH 4.8, 8.0 Hz, H5), 6.89 (ps. quin, 4H, J 2.0 Hz, Hα),
6.82 (dd, 1H, 3JHH 3.7 Hz, 3JHP 2.4 Hz, H2), 6.61 (dd, 1H, 3JHH
3.7 Hz, 4JHP 0.6 Hz, H3), 6.40 (ps. t, 4H, J 2.0 Hz, Hβ). 13C{1H}
NMR (100.5 MHz, CDCl3): δ 151.5 (d, 2JCP 15 Hz, C8), 143.9 (s,
C6), 129.1 (s, C4), 126.8 (d, 2JCP 4.6 Hz, C2), 122.7 (d, 2JCP 15 Hz,
Cα), 122.4 (s, C9), 117.7 (s, C5), 113.0 (d, 3JCP 4 Hz, Cβ), 106.0 (s,
C3).
31P{1H} NMR (121.5 MHz, CD2Cl2): δ 47.3 (s, JPPt 5409 Hz).
1H NMR (399.8 MHz, CD2Cl2): δ 9.34 (dm, 1H, JHH 6.0 Hz,
3
3JHPt 41 Hz, H6), 8.26 (ddd, 1H, 3JHH 8.0 Hz, 4JHH 1.2 Hz, 5JHP
2.4 Hz, H4), 7.42 (dd, 1H, 3JHH 6.0, 8.0 Hz, H5), 7.37 (dd, 3JHH
3.6 Hz, 3JHP 2.8 Hz, 1H, H2), 7.28 (dd, 1H, 3JHH 3.2, 4JHP 0.4 Hz,
H3), 7.15 (ps. quin, 4H, Hα), 6.53 (ps. t, 4H, Hβ).
Synthesis of [RhCl(CO)(L1-ꢀ2P,N)] 3a
[Rh(µ-Cl)(cod)]2 (0.295 g, 0.60 mmol) was added to a dichloro-
methane solution of L1 (0.362 g, 1.20 mmol). The solution was
stirred for 2 h with CO gas bubbling through it. Upon addition
of hexane a yellow precipitate was formed, which was separated
by filtration, dried under reduced pressure and recrystallised
from dichloromethane–hexane. Yield: 0.461 g (82%). Calc. for
C20H15ClN2OPRh: C, 51.3; H, 3.23; N, 5.98%. Found: C, 50.9;
H, 3.22; N, 5.99%). 31P{1H} NMR (161.8 MHz, CD2Cl2):
δ 104.1 (d, 1JPRh 180 Hz). 1H NMR (399.8 MHz, CD2Cl2): δ 9.03
(dm, 1H, 3JHH 5.2 Hz, H6), 8.06 (dd, 1H, 3JHH 8.0 Hz, 4JHH 1.2
Hz, H4), 7.76–7.70 (m, 4H, Hm), 7.58–7.55 (m, 2H, Hp), 7.52–
7.47 (m, 4H, Ho), 7.25–7.22 (m, 2H, H2, H5), 6.92 (d, 1H, 3JHH
3.6 Hz, H3). 13C{1H} NMR (100.5 MHz, CD2Cl2): δ 187.1 (dd,
1JCRh 70 Hz, 2JCP 23 Hz, CO), 156.4 (d, 2JCP 17 Hz, C8), 142.9 (s,
C6), 132.3 (s, Cp), 132.2 (s, C4), 132.1 (d, 2JCP 15 Hz, Co), 131.4
(d, 1JCP 3 Hz, Ci), 129.1 (d, 3JCP 11 Hz, Cm), 125.3 (d, 2JCP 7 Hz,
3
4
Synthesis of [PdCl2(L1-ꢀ2P,N)] 1a
[PdCl2(cod)] (0.230 g, 0.81 mmol) was added to a dichloro-
methane solution of L1 (0.244 g, 0.81 mmol). The solution was
stirred for 2 h, then half of the solvent was evaporated under
reduced pressure. Upon addition of hexane a yellow precipitate
was formed, which was filtered, washed with hexane and dried
under reduced pressure. The product was recrystallised from di-
chloromethane–hexane. Yield 0.379 g (98%). Calc. for C19H15-
Cl2N2PPdؒCH2Cl2: C, 42.6; H, 3.04; N, 4.96%. Found: C, 42.5;
H, 3.04; N, 4.96%. 31P{1H} NMR (161.8 MHz, CDCl3): δ 80.9
(s). 1H NMR (399.8 MHz, CDCl3): δ 9.15 (dm, 1H, 3JHH 4.8 Hz,
3
C2), 120.0 (d, JCP 6 Hz, C9), 117.5 (s, C5), 110.7 (s, C3). IR
(CH2Cl2, cmϪ1): 2005 [s, ν(CO)]. IR (KBr, cmϪ1): 1992 [s,
ν(CO)].
Synthesis of [RhCl(CO)(L2-ꢀ2P,N)] 3b
As for 3a using [Rh(µ-Cl)(cod)]2 (0.352 g, 0.71 mmol) and L2
(0.400 g 1.43 mmol). Yield: 0.533 g (84%). Calc. for C16H13-
Cl2N4OPRh: C, 43.0; H, 2.93; N, 12.5%. Found: C, 43.1; H,
2.91; N, 12.6%. 31P{1H} NMR (161.8 MHz, CD2Cl2): δ 101.9
(d, 1JPRh 242 Hz). 1H NMR (399.8 MHz, CD2Cl2): δ 8.94 (dm,
1H, 3JHH 5.6 Hz, H6), 8.13 (ddd, 1H, 3JHH 8.0 Hz, 4JHH 1.2 Hz,
5JHP 2.0 Hz, H4), 7.39 (t, 1H, 3JHH 3.2 Hz, 3JHP 5.6 Hz, H2), 7.32
(dd, 1H, 3JHH 5.6, 8.0 Hz, H5), 7.09 (ps. quin, 4H, Hα), 6.97 (d,
3
H6), 8.15 (dm, 1H, JHH 8.1 Hz, H4), 7.90–7.84 (m, 4H, Hm),
7.70–7.64 (m, 2H, Hp), 7.58–7.52 (m, 4H, Ho), 7.32 (dd, 1H,
3
3
3JHH 4.8, 8.1 Hz, H5), 7.17 (dd, 1H, JHH 3.7 Hz, JHP 2.0 Hz,
H2), 7.01 (d, 1H, 3JHH 3.7 Hz, H3).
Synthesis of [PdCl2(L2-ꢀ2P,N)] 1b
As for 1a using [PdCl2(cod)] (0.335 g, 1.2 mmol) and L2 (0.336 g
1.2 mmol). Yield 0.526 g (98%). Calc. for C15H13Cl2N4PPd: C,
39.4; H, 2.86; N, 12.2%. Found: C, 39.0; H, 2.74; N, 11.9%.
31P{1H} NMR (161.8 MHz, CD2Cl2): δ 70.5 (s). 1H NMR
(399.8 MHz, CD2Cl2): δ 9.06 (dt, 1H, 3JHH 5.6 Hz, 4JHH 1.2 Hz,
4JHP 1.2 Hz, H6), 8.21 (ddd, 1H, 3JHH 8.0 Hz, 4JHH 2.8 Hz, 5JHP
1.2 Hz, H4), 7.43 (dd, 1H, 3JHH 5.6, 8.0 Hz, H5), 7.34 (ps. t, 1H,
1H, JHH 3.2 Hz, H3), 6.49 (m, 4H, Hβ). 13C{1H} NMR (100.5
3
MHz, CD2Cl2): δ 186.1 (dd, 1JCRh 70 Hz, 2JCP 23 Hz, CO), 155.8
2
2
(d, JCP 25 Hz, C8), 143.2 (s, C6), 133.4 (s, C4), 124.7 (d, JCP
2
3
8 Hz, C2), 124.0 (d, JCP 10 Hz, Cα), 120.4 (d, JCP 10 Hz, C9),
118.8 (s, C5), 115.4 (d, 3JCP 10 Hz, Cβ), 112.1 (d, 3JCP 2 Hz, C3).
IR (CH2Cl2, cmϪ1): 2027 [s, ν(CO)]. IR (KBr, cmϪ1): 2017 [s,
ν(CO)].
3
3JHH 3.2 Hz, JHP 3.2 Hz, H2), 7.17 (ps. quin, Hα), 7.09 (d, 1H,
3JHH 3.2 Hz, H3), 6.55 (ps. t, 4H, Hβ).
Synthesis of [PtCl2(L1-ꢀ2P,N)] 2a
Synthesis of [PdClMe(L1-ꢀ2P,N)] 4a
[PdClMe(cod)] (0.406 g, 1.53 mmol) was added to a dichloro-
methane solution of L1 (0.463 g, 1.53 mmol). The pale yellow
solution was stirred for 2 h and half of the solvent was elimin-
ated under reduced pressure. Hexane was added to precipitate a
white powder which was isolated by filtration, washed with
hexane then diethyl ether and dried under reduced pressure.
Yield: 0.680 g (97%). Calc. for C20H18ClN2PPdؒ1/3CH2Cl2: C,
50.1; H, 3.86; N, 5.75%. Found: C, 49.8; H, 3.94; N, 5.75%.
31P{1H} NMR (161.8 MHz, CD2Cl2): δ 83.0 (s). 1H NMR
(399.8 MHz, CD2Cl2): δ 8.75 (dd, 1H, 3JHH 5.2 Hz, 4JHH 1.6 Hz,
[PtCl2(cod)] (0.418 g, 1.1 mmol) was added to a dichloro-
methane solution of L1 (0.338 g, 1.1 mmol). The solution was
stirred for 2 h and half of the solvent was evaporated under
reduced pressure. Upon addition of hexane a colourless precipi-
tate was formed, which was filtered, washed with hexane and
dried under reduced pressure. The product was recrystallised
from dichloromethane–hexane. Yield: 0.620 g (98%). Calc. for
C19H15Cl2N2PPtؒ1/2CH2Cl2: C, 38.4; H, 2.64; N, 4.59%. Found:
C, 38.7; H, 2.69; N, 4.67%. 31P{1H} NMR (161.8 MHz,
3
4
5
H6), 8.05 (dt, 1H, JHH 8.0 Hz, JHH 1.6 Hz, JHP 1.6 Hz, H4),
7.68–7.63 (m, 4H, Hm), 7.61–7.56 (m, 2H, Hp), 7.52–7.46 (m,
4H, Ho), 7.26 (dd, 1H, 3JHH 5.2, 8.0 Hz, H5), 7.18 (dd, 1H, 3JHH
4.0 Hz, 3JHP 2.0 Hz, H2) 6.84 (d, 1H, 3JHH 4.0 Hz, H3), 0.75 (d,
1
CD2Cl2): δ 55.4 (s, JPPt 4125 Hz). 1H NMR (399.8 MHz,
CD2Cl2): δ 9.33 (dd, 1H, 3JHH 5.7 Hz, 4JHH 1.2 Hz, 3JHPt 39 Hz,
3
4
5
H6), 8.21 (dt, 1H, JHH 7.8 Hz, JHH 1.2 Hz, JHP 1.2 Hz, H4),
7.83–7.78 (m, 4H, Hm), 7.67–7.60 (m, 2H, Hp), 7.58–7.50 (m,
4H, Ho), 7.31 (dd, 1H, 3JHH 7.8, 5.7 Hz, H5), 7.25 (dd, 1H, 3JHH
3.6 Hz, 3JHP 2.0 Hz, H2), 7.21 (dd, 1H, 3JHH 3.6 Hz, 4JHP 0.8 Hz,
H3).
3H, JHP 4.4 Hz, Me). 13C{1H} NMR (75.5 MHz, CD2Cl2)
3
2
2
δ 155.2 (d, JCP 15 Hz, C8), 142.8 (s, C6), 133.2 (d, JCP 14 Hz,
Co), 133.0 (d, 1JCP 2 Hz, Ci), 131.7 (s, Cp), 130.0 (s, C4), 129.6 (d,
3JCP 11 Hz, Cm), 126.2 (d, 2JCP 7 Hz, C2), 120.8 (d, 3JCP 6 Hz, C9),
118.6 (s, C5), 110.8 (s, C3), Ϫ5.8 (d, 2JCP 5 Hz, Me).
Synthesis of [PtCl2(L2-ꢀ2P,N)] 2b
As for 2a using [PtCl2(cod)] (0.374 g, 1.0 mmol) and L2 (0.280 g,
1.0 mmol). Yield: 0.535 g (98%). Calc. for C15H13Cl2N4PPt : C,
33.0; H, 2.40; N, 10.3%. Found: C, 32.8; H, 2.36; N, 10.2%.
Synthesis of [PdClMe(L2-ꢀ2P,N)] 4b
As for 4a using [PdClMe(cod)] (1.40 g, 5.3 mmol) and L2
(1.47 g, 5.3 mmol). Yield 1.93 g (84%). Calc. for C16H16ClN4-
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4726

View Article Online
PPd: C, 44.0; H, 3.69; N, 12.8%. Found: C, 43.4; H, 3.55; N,
Synthesis of [PdCl{C(O)Me}(L2-ꢀ2P,N)] 6b
12.7%. 31P{1H} NMR (161.8 MHz, CD2Cl2): δ 89.3 (s). 1H
Compound 4b (0.050 g, 0.11 mmol) was transferred in a NMR
tube fitted with a Young valve and dissolved in a minimum
amount of CD2Cl2. A CO pressure of approximately 20 psi was
applied and the NMR tube was shaken vigorously. After 2 h
31P{1H} NMR spectroscopy indicated a conversion of approx-
imately 40%. 31P{1H} NMR (161.8 MHz, CD2Cl2): δ 78.9 (s).
3
NMR (399.8 MHz, CD2Cl2): δ 8.73 (dd, 1H, JHH 5.2 Hz,
4JHH 0.8 Hz, H6), 8.11 (ddd, 1H, 3JHH 7.2 Hz, 4JHH 0.8 Hz, 5JHP
3
1.2 Hz, H4), 7.36 (dd, 1H, JHH 5.6, 7.2 Hz, H5), 7.35 (d, 1H,
3JHH 4.4 Hz, H2), 7.03 (ps. quin, 4H, Hα), 6.92 (d, 1H, 3JHH 4.4
3
Hz, H3), 6.49 (m, 4H, Hβ), 0.95 (d, 3H, JHP 3.6 Hz, Me).
13C{1H} NMR (100.5 MHz, CD2Cl2) δ 154.2 (d, 2JCP 20 Hz, C8),
143.0 (s, C6), 132.3 (s, C4), 124.8 (d, 2JCP 7 Hz, C2), 124.0 (d, 2JCP
1H NMR (399.8 MHz, CD2Cl2): δ 8.53 (bd, 1H, JHH 4.8 Hz,
3
H6), 8.12 (m, 1H, H4), 7.40 (m, 1H, H5), 7.37 (m, 1H, H3), 7.05
8 Hz, Cα), 120.7 (d, 2JCP 7 Hz, C9), 119.6 (s, C5), 115.6 (d, 3JCP
Hz, Cβ), 111.9 (d, 2JCP 3 Hz, C3), Ϫ4.6 (d, 2JCP 5 Hz, Me).
8
3
(ps. quin, 4H, Hα), 6.90 (d, 1H, JHH 3.6 Hz, H2), 6.51 (m, 4H,
Hβ), 2.09 (s, 3H, Me). 13C{1H} NMR (100.5 MHz, CD2Cl2):
2
δ 219.3 (s, CO), 153.8 (d, JCP 18 Hz, C8), 142.9 (s, C6), 132.5
Synthesis of [PdMe(OTf)(L1-ꢀ2P,N)] 5a
(s, C4), 125.1 (d, 2JCP 8 Hz, C2), 124.0 (d, 2JCP 11 Hz, Cα), 121.0
3
3
(d, JCP 4 Hz, C9), 119.7 (s, C5), 115.8 (d, JCP 9 Hz, Cβ), 111.9
AgOTf (0.095 g, 0.37 mmol) was added to a dichloromethane
solution of 4a (0.170 g, 0.37 mmol). After 4 h stirring the sus-
pension was filtered to remove AgCl and the filtrate evaporated
to dryness. The resulting pale yellow powder was washed with
hexane and dried under reduced pressure. Yield: 0.192 g (91%).
Calc. for C21H18F3N2O3PPdS: C, 44.0; H, 3.17; N, 4.89%.
Found: C, 43.8; H, 3.16; N, 4.76%. 31P{1H} NMR (161.8 MHz,
(s, C3), 39.6 (d, JCP 32 Hz, Me). IR (CH2Cl2, cmϪ1): 1715
3
[s, ν(CO)].
Synthesis of [Pd{C(O)Me}(OTf)(L1-ꢀ2P,N)] 7a
Compound 5a (0.050 g, 0.087 mmol) was transferred into
a NMR tube fitted with a Young valve and dissolved in a
minimum amount of CD2Cl2. A CO pressure of approximately
20 psi was applied and the NMR tube was shaken vigorously.
After ∼5 min 31P{1H} NMR spectroscopy showed complete
conversion to 7a. The contents of the NMR tube were trans-
ferred into a Schlenk and a sample was taken to record the IR
spectrum. The solution was diluted with dichloromethane and
layered with hexane. Slow diffusion of hexane into the di-
chloromethane solution gave 7a as colourless crystals suitable
for X-ray diffraction studies. 31P{1H} NMR (161.8 MHz,
1
CD2Cl2): δ 84.6 (s). H NMR (399.8 MHz, CD2Cl2): δ 8.53
3
3
(d, 1H, JHH 5.6 Hz, H6), 8.07 (d, 1H, JHH 7.9 Hz, H4), 7.82–
7.46 (m, 6H, Hm, Hp), 7.54 (m, 4H, Ho), 7.32 (dd, 1H,
3
3
3JHH 5.6, 7.9 Hz, H5), 7.16 (dd, 1H, JHH 2.4 Hz, JHP 3.6 Hz,
3
3
H2), 6.86 (d, 1H, JHH 3.6 Hz, H3), 0.85 (d, 3H, JHP 1.6 Hz,
Me). 13C{1H} NMR (67.9 MHz, CD2Cl2): δ 154.8 (d, JCP
2
15 Hz, C8), 142.9 (s, C6), 133.7 (s, Cp), 133.3 (d, 2JCP 15 Hz, Co),
3
132.5 (s, Ci), 130.0 (d, JCP 12 Hz, Cm), 129.9 (s, C4), 126.4 (d,
2JCP 6 Hz, C2), 121.1 (d, JCP 6 Hz, C9), 119.1 (s, C5), 111.4 (s,
3
C3), Ϫ1.2 (d, JCP 4 Hz, Me). 19F{1H} NMR (376.2 MHz,
1
2
CD2Cl2): δ 65.5 (s). H NMR (399.8 MHz, CD2Cl2): δ 8.46 (d,
CD2Cl2): δ –78.9 (s).
3
3
1H, JHH 5.2 Hz, H6), 8.11 (d, 1H, JHH 8.0 Hz, H4), 7.71–7.64
(m, 6H, Ho, Hp), 7.57–7.52 (m, 4H, Hm), 7.30 (dd, 1H, 3JHH 5.2,
8.0 Hz, H5), 7.20 (dd, 1H, 3JHH 3.6 Hz, 3JHP 2.4 Hz, H2), 6.86 (d,
Synthesis of [PdMe(OTf)(L2-ꢀ2P,N)] 5b
3
4
1H, JHH 3.6 Hz, H3), 2.27 (d, 3H, JHP 1.2 Hz, Me). 13C{1H}
NMR (100.5 MHz, CD2Cl2): δ 220.1 (d, 2JCP 6 Hz, CO), 152.5
(d, 2JCP 14 Hz, C8), 142.8 (s, C6), 133.5 (d, 1JCP 2 Hz, Ci), 133.1
(d, 2JCP 15 Hz, Co), 132.6 (s, Cp), 129.9 (d, 3JCP 12 Hz, Cm), 127.6
(s, C4), 126.6 (d, 2JCP 6 Hz, C2), 121.6 (d, 3JCP 5 Hz, C9), 119.0 (s,
As for 5a using AgOTf (0.095 g, 0.37 mmol) and 4b (0.163 g,
0.37 mmol). Yield: 0.202 g, (98%). Calc. for C17H16F3N4O3-
PPdS: C, 37.1; H, 2.93; N, 10.2%. Found: C, 37.1; H, 2.98; N,
10.1%. 31P{1H} NMR (161.8 MHz, CD2Cl2): δ 90.1 (s). 1H
3
NMR (399.8 MHz, CD2Cl2): δ 8.48 (bd, 1H, JHH 5.6 Hz,
3
3
3
C5), 110.8 (s, C3), 36.5 (d, JCP 29 Hz, Me). 19F{1H} NMR
H6), 8.16 (dm, 1H, JHH 8.0 Hz, H4), 7.40 (dd, 1H, JHH 5.6,
3
3
(376.2 MHz, CD2Cl2): δ Ϫ78.7 (s). IR (CD2Cl2, cmϪ1): 1711 [s,
ν(CO)].
8.0 Hz, H5), 7.36 (ps. t, 1H, JHH 3.6 Hz, JHP 3.6 Hz, H2),
7.04 (ps. quin, 4H, Hα), 6.95 (d, 1H, 3JHH 3.6 Hz, H3), 6.54 (m,
4H, Hβ), 1.04 (s, 3H, Me). 13C{1H} NMR (100.5 MHz,
CD2Cl2): δ 153.8 (d, 2JCP 19 Hz, C8), 143.4 (s, C6), 133.1 (s, C4),
Synthesis of [Pd{C(O)Me}(OTf)(L2-ꢀ2P,N)] 7b
125.0 (d, 2JCP 6 Hz, C2), 124.2 (d, 2JCP 9 Hz, Cα), 121.2 (d, 3JCP
8
4
As for 7a starting from 5b (0.050 g, 0.091 mmol). 31P{1H} NMR
3
3
(161.8 MHz, CD2Cl2): δ 79.1 (s). 1H NMR (399.8 MHz,
Hz, C9), 120.2 (s, C5), 116.4 (d, JCP 9 Hz, Cβ), 112.5 (d, JCP
Hz, C3), 0.6 (s, Me). 19F{1H} NMR (376.2 MHz, CD2Cl2):
3
3
CD2Cl2): δ 8.37 (bd, 1H, JHH 5.2 Hz, H6), 8.11 (dm, 1H, JHH
8.0 Hz, H4), 7.42 (ps. t, 1H, 3JHH 3.5 Hz, 3JHP 3.5 Hz, H2), 7.39
δ Ϫ78.5 (s).
3
(dd, 1H, JHH 5.2, 8.0 Hz, H5), 7.09–7.07 (m, 4H, Hα), 6.93 (d,
Synthesis of [PdCl{C(O)Me}(L1-ꢀ2P,N)] 6a
1H, 3JHH 3.5 Hz, H3), 6.57–6.55 (m, 4H, Hβ), 2.18 (d, 3H, 4JHP
3.2 Hz, Me). 13C{1H} NMR (100.5 MHz, CD2Cl2): δ 216.5 (d,
2JCP 12 Hz, CO), 153.3 (d, 2JCP 20 Hz, C8), 143.7 (s, C6), 133.1 (s,
C4), 125.2 (d, 2JCP 7 Hz, C2), 124.0 (d, 2JCP 10 Hz, Cα), 121.7 (d,
Compound 4a (0.078 g, 0.17 mmol) was transferred into a
NMR tube fitted with a Young valve and dissolved in a mini-
mum amount of CD2Cl2. A CO pressure of approximately 10
psi was applied and the NMR tube was shaken vigorously.
After 48 h the solution was transferred into a small Schlenk and
hexane added to precipitate a pale yellow powder. This was
isolated by filtration and dried under reduced pressure. Yield:
0.057 g (69%). Calc. for C21H18ClN2OPPdؒ1/4CH2Cl2: C, 50.2;
H, 3.67; N, 5.51%. Found: C, 50.2; H, 3.64; N, 5.51%. 31P{1H}
3
3JCP 7 Hz, C9), 120.2 (s, C5), 116.4 (d, JCP 8 Hz, Cβ), 112.0 (d,
3JCP 3 Hz, C3), 36.6 (d, 3JCP 46 Hz, Me). 19F{1H} NMR (376.2
MHz, CD2Cl2): δ Ϫ78.2 (s). IR (CD2Cl2, cmϪ1): 1729 [s, ν(CO)].
Synthesis of [Pd{CH2CH2C(O)Me-ꢀ2C,O}(L1-ꢀ2P,N)]OTf 8a
Compound 5a (0.050 g, 0.087 mmol) was transferred to a NMR
tube fitted with a Young valve and dissolved in a minimum
amount of CD2Cl2. A CO pressure of approximately 20 psi was
applied and the NMR tube was shaken vigorously. After 20 min
the NMR tube was immersed in liquid nitrogen and placed
under reduced pressure. After 10 min the Young valve was
closed and the NMR tube was left to reach ambient temper-
ature. This operation was repeated three times in order to elim-
inate unreacted CO. A C2H4 pressure of approximately 10 psi
was applied and the NMR tube was shaken vigorously every
10 min for 1 h. After this time the 31P{1H} NMR spectrum
1
NMR (161.8 MHz, CDCl3): δ 64.9 (s). H NMR (399.8 MHz,
3
3
CDCl3): δ 8.68 (d, 1H, JHH 5.2 Hz, H6), 8.04 (d, 1H, JHH
7.8 Hz, H4), 7.71–7.66 (m, 4H, Hm), 7.61–7.58 (m, 2H, Hp),
7.55–7.50 (m, 4H, Ho), 7.28 (dd, 1H, 3JHH 5.2, 7.8 Hz, H5), 7.17
(dd, 1H, 3JHH 3.6 Hz, 3JHP 2.4 Hz, H2), 6.82 (d, 1H, 3JHH 3.6 Hz,
H3), 2.34 (s, 3H, Me). 13C{1H} NMR (100.5 MHz, CD2Cl2):
2
2
δ 224.3 (d, JCP 6 Hz, CO), 154.8 (d, JCP 21 Hz, C8), 143.2 (s,
C6), 133.5–130.0 (m, Ci, Co, Cm, Cp, C4), 127.0 (d, 2JCP 9 Hz, C2),
121.5 (d, 3JCP 8 Hz, C9), 119.1 (s, C5), 110.9 (s, C3), 39.0 (d, 3JCP
30 Hz, Me). IR (KBr, cmϪ1): 1697 [s, ν(CO)].
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4727

View Article Online
showed complete conversion of 5a to 8a. The contents of the
NMR tube were transferred in a Schlenk, and a small sample
taken to record an IR spectrum. Dichloromethane was added
and the solution was layered with hexane. The two solvents
slowly mixed overnight allowing 8a to precipitate out as crystals
suitable for X-ray crystallography. 31P{1H} NMR (161.8 MHz,
3H, Me). 19F{1H} NMR: δ Ϫ78.8 (s). IR (CD2Cl2, cmϪ1): 1716,
1660, [s, ν(C᎐O)].
᎐
Synthesis of [Pd(L1-ꢀ2P,N)](OTf)2 10
AgOTf (0.350 g, 1.36 mmol) was added to a dichloromethane
solution of [PdCl2(cod)] (0.165 g, 0.58 mmol) and L1 (0.392 g,
1.30 mmol). The reaction mixture was stirred for 4 h with the
formation of a white precipitate of AgCl. The solid was
removed by filtration and the solvent eliminated under reduced
pressure. The resulting powder was washed with small
amounts of dichloromethane. Crystallisation by slow diffusion
of hexane into a dichloromethane solution gave 10 as colourless
crystals. Yield: 0.520 g (89%). Calc. for C40H30F6N4O6P2PdS2ؒ3/2-
CH2Cl2: C, 43.9; H, 2.93; N, 4.93%. Found: C, 43.7; H, 2.95; N,
4.99%. 31P{1H} NMR (161.8 MHz, CD2Cl2): δ 90.2 (s). 1H
NMR (399.8 MHz, CDCl3): δ 8.82 (m, 2H, H6), 8.28 (dd, 2H,
3JHH 8.0 Hz, 4JHH 1.1 Hz, H4), 7.74–7.68 (m, 8H, Ho), 7.60 (dd,
2H, 3JHH 6.4, 8.0 Hz, H5), 7.55–7.52 (m, 4H, Hp), 7.36–7.30 (m,
8H, Hm), 7.09–7.07 (m, 2H, H2), 7.02 (d, 1H, 3JHH 3.6 Hz, H3).
19F{1H} NMR (376.2 MHz, CD2Cl2): δ Ϫ79.5 (s).
1
CD2Cl2): δ 84.3 (s). H NMR (399.8 MHz, CD2Cl2): δ 8.45 (d,
1H, 3JHH 5.2 Hz, H6), 8.19 (ddd, 1H, 3JHH 8.0 Hz, 4JHH 1.2 Hz,
5JHP 1.2 Hz, H4), 7.69–7.62 (m, 6H, Hp, Ho), 7.60–7.47 (m, 4H,
3
3
Hm), 7.39 (dd, 1H, JHH 5.2, 8.0 Hz, H5), 7.24 (dd, 1H, JHH
3.2 Hz, 3JHP 2.0 Hz, H2), 6.95 (d, 1H, 3JHH 3.2 Hz, H3), 3.16 (t,
2H, 3JHH 6.2 Hz, CH2CO), 2.49 (s, 3H, Me), 1.99 (dt, 2H, 3JHH
6.2 Hz, JHP 1.2 Hz, CH2Pd). 13C{1H} NMR (100.5 MHz,
3
3
2
CD2Cl2): δ 222.4 (d, JCP 8 Hz, CO), 154.5 (d, JCP 14 Hz, C8),
2
143.2 (s, C6), 133.6 (s, C4), 133.1 (d, JCP 20 Hz, Co), 132.8 (s,
Cp), 132.0 (s, Ci), 126.8 (d, 2JCP 7 Hz, C2), 130.0 (d, 3JCP 16 Hz,
Cm), 121.7 (s, C9), 119.1 (s, C5), 111.0 (s, C3), 51.6 (s, CH2CO),
4
2
28.3 (d, JCP 3 Hz, Me), 20.9 (d, JCP 2 Hz, CH2Pd). 19F{1H}
NMR (376.2 MHz, CD2Cl2): δ Ϫ79.1 (s). IR (CD2Cl2, cmϪ1):
1631 [s, ν(CO)].
Synthesis of [Pd{CH2CH2C(O)Me-ꢀ2C,O}(L2-ꢀ2P,N)]OTf 8b
Synthesis of [Pt(L1-ꢀ2P,N)2](OTf)2 11
As for 8a using compound 5b (0.050 g, 0.091 mmol). 31P{1H}
AgOTf (0.133 g, 0.52 mmol) was added to a dichloromethane
solution of 2a (0.150 g, 0.26 mmol) and L1 (0.080 g, 0.26
mmol). The reaction mixture was stirred for 4 h with the form-
ation of a white precipitate of AgCl. The solid was removed by
filtration and hexane added to precipitate a white powder which
was separated by filtration, washed with hexane and dried
under reduced pressure. Yield: 0.260g (90%). Calc. for C40H30-
F6N4O6P2PtS2ؒ1/4CH2Cl2: C, 43.2; H, 2.75; N, 5.01%. Found: C,
43.2; H, 2.83; N, 5.01%. 31P{1H} NMR (161.8 MHz, CD2Cl2):
δ 62.8 (s, 1JPPt 3601 Hz). 1H NMR (399.8 MHz, CDCl3): δ 8.96
NMR (161.8 MHz, CD2Cl2): δ 91.1 (s). 1H NMR (399.8 MHz,
3
4
CD2Cl2): δ 8.45 (dd, 1H, JHH 5.2 Hz, JHH 1.2 Hz, H6), 8.23
3
5
4
(ddd, 1H, JHH 8.0 Hz, JHP 2.4 Hz, JHH 1.2 Hz, H4), 7.48
3
3
(dd, 1H, JHH 5.2, 8.0 Hz, H5), 7.40 (ps. t, 1H, JHH 3.9 Hz,
3JHP 3.9 Hz, H2), 7.05 (ps. quin, 4H, Hα) 7.02 (d, 1H, JHH
3
3.9 Hz, H3), 6.58–6.56 (m, 4H, Hβ), 3.29–3.25 (m, 2H, CH2-
CO), 2.55 (s, 3H, Me), 2.21–2.17 (m, 2H, CH2Pd). 13C{1H}
2
NMR (75.5 MHz, CD2Cl2): δ 237.9 (s, CO), 154.9 (d, JCP
15 Hz, C8), 144.7 (s, C6), 134.8 (s, C4), 126.3 (d, 2JCP 6 Hz, C2),
125.1 (d, 2JCP 10 Hz, Cα), 122.7 (d, 3JCP 7 Hz, C9), 121.6 (s, C5),
3
(br, 2H, H6), 8.36 (d, 2H, JHH 8.0 Hz, H4), 7.69–7.61 (m, 8H,
3
3
117.9 (d, JCP 9 Hz, Cβ), 114.0 (d, JCP 4 Hz, C3), 52.8 (s,
Ho), 7.54–7.45 (m, 6H, H5, Hp), 7.35–7.31 (m, 8H, Hm), 7.21–
7.17 (m, 2H, H2), 7.04–7.02 (m, 2H, H3). 19F{1H} NMR (376.2
MHz, CD2Cl2): δ Ϫ79.4 (s).
CH2CO), 29.3 (d, JCP 3 Hz, Me), 22.9 (s, CH2Pd). 19F{1H}
4
NMR (376.2 MHz, CD2Cl2): δ Ϫ79.3 (s). IR (CD2Cl2, cmϪ1):
1628 [s, ν(CO)].
Catalytic experiments
Synthesis of [Pd{C(O)CH2CH2C(O)Me-ꢀ2C,O}(L1-ꢀ2P,N)]-
The co-polymerisation reactions were carried out in a 500 cm3
Buchi glass autoclave equipped with a magnetic stirring bar.
The catalysts (5a,b) were transferred via cannula under an
atmosphere of nitrogen. Dried methanol or TfOH were added
using common syringe techniques. The autoclave was pressur-
ised first with C2H4 and then with CO. For the reactions at
50 ЊC, the autoclave was heated in an oil bath. After the
appropriate reaction times the pressure was released and the
polymer removed by filtration, washed thoroughly with di-
chloromethane and dried under reduced pressure. Analysis of
PK: mp 240–260 ЊC. 1H NMR (399.8 MHz, HFIA/C6D6):
δ 2.69 (br, CH2CH2). 13C{1H} NMR (100.5 MHz, HFIA/C6D6):
OTf 9a
A CD2Cl2 solution of 8a (0.055 g, 0.087 mmol) was transferred
into a NMR tube fitted with a Young valve and a CO pressure
of approximately 10 psi was applied. After 24 h at ambient
temperature 31P{1H} NMR spectroscopy showed complete
conversion of 8a into 9a. 31P{1H} NMR (161.8 MHz, CD2Cl2):
δ 69.0 (s). 1H NMR (399.8 MHz, CD2Cl2): δ 8.42 (bd, 1H, 3JHH
5.2 Hz, H6), 8.15 (dt, 1H, 3JHH 8.0 Hz, 4JHH 1.6 Hz, 5JHP 1.6 Hz,
H4), 7.72–7.64 (m, 6H, Hp, Ho), 7.58–7.53 (m, 4H, Hm), 7.38 (dd,
3
3
3
1H, JHH 5.2, 8.0 Hz, H5), 7.16 (dd, 1H, JHH 3.6 Hz, JHP 2.0
Hz, H2), 6.87 (dd, 1H, 3JHH 3.6 Hz, 4JHP 0.6 Hz, H3), 2.86–2.75
(m, 4H, CH2), 2.46 (s, 3H, Me). 13C{1H} NMR (100.5 MHz,
CDCl3): δ 219.1 (s, CO), 217.3 (s, CO), 154.1 (d, 2JCP 14 Hz, C8),
142.6 (s, C6), 133.6 (d, 4JCP 3 Hz, C4), 133.3 (d, 2JCP 15 Hz, Co),
132.9 (s, Cp), 129.7 (d, 3JCP 13 Hz, Cm), 127.1 (d, 1JCP 7 Hz, Ci),
δ 213.1 (s, CO), 36.1 (s, CH2). IR (KBr, cmϪ1): 1695 [s, ν(C᎐O)].
᎐
Crystallography
2
3
Crystallographic data for compounds 1a,b, 3a,b, 7a,b, 8a,b and
10 are summarised in Table 7. All data were collected using a
Nonius KappaCCD diffractometer with the exception of 1a
which were collected on an Enraf-Nonius CAD4. Full matrix
anisotropic refinement was implemented in the final least-
squares cycles for all structures with the specific exceptions
described below. All data were corrected for Lorentz and polar-
isation. Extinction corrections were applied in all cases except
for 8b.
Absorption corrections (multiscan) were applied on the basis
of their individual merits to data for 3a, 3b and 10 (maximum,
minimum transmission factors were 1.04, 0.79; 0.80, 0.70; and
1.09, 0.89, respectively). Hydrogen atoms were included at
calculated positions throughout with the exceptions described
below.
126.7 (d, JCP 6 Hz, C2), 121.6 (d, JCP 4 Hz, C9), 119.1 (s, C5),
4
110.7 (s, C3), 40.1 (d, JCP 25 Hz, Me), 38.2 (s, PdC(O)CH2),
31.2 (s, COCH2). 19F{1H} NMR (376.2 MHz, CD2Cl2): δ Ϫ79.2
(s). IR (CD2Cl2, cmϪ1): 1711, 1622 [s, ν(C᎐O)].
᎐
Synthesis of [Pd{C(O)CH2CH2C(O)Me-ꢀ2C,O}(L2-ꢀ2P,N)]-
OTf 9b
As for 9a using compound 8b (0.055 g, 0.091 mmol). 31P{1H}
NMR (161.8 MHz, CD2Cl2): δ 78.9 (s). 1H NMR (399.8 MHz,
3
4
CD2Cl2): δ 8.45 (dd, 1H, JHH 5.2 Hz, JHH 1.2 Hz, H6), 8.23
(ddd, 1H, 3JHH 8.0 Hz, 5JHP 2.4 Hz, 4JHH 1.2 Hz, H4), 7.47 (dd,
3
3
5
1H, JHH 5.2, 8.0 Hz, H5), 7.42 (ps. t, 1H, JHH 3.9 Hz, JHP
3
3.9 Hz, H2), 7.05 (ps. quin, 4H, Hα), 7.02 (d, 1H, JHH 3.9 Hz,
H3), 6.57–6.56 (m, 4H, Hβ), 2.75–2.74 (m, 4H, CH2), 2.20 (s,
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4728

Table 7 Crystallographic data for complexes 1a,b, 3a,b, 7a,b, 8a,b and 10
Compound
1a
1b
3a
3b
7a
7b
8a
8b
10
Formula
M
T/K
C19H15Cl2N2PPd
479.60
170(2)
C15H13Cl2N4PPd
457.56
150(2)
C20H15ClN2OPRh C16H13ClN4OPRh C22H18F3N2O4PPdS C18H16F3N4O4PPdS C24H22F3N2O4PPdS C20H20F3N4O4PPdS C40H33F6N4O7.50P2PdS2
468.67
446.63
150(2)
600.81
150(2)
578.78
150(2)
628.87
150(2)
606.83
150(2)
1036.16
150(2)
293(2)
Crystal system
Space group
a/Å
b/Å
c/Å
α/Њ
β/Њ
Monoclinic
P21/c
9.849(2)
15.522(2)
12.376(2)
90
111.190(10)
90
1764.1(5)
4
Monoclinic
P21/c
8.7400(1)
10.8700(1)
17.4600(2)
90
99.0630(4)
90
1638.06(3)
4
Triclinic
PϪ1
Monoclinic
P21/c
8.7130(1)
11.0150(2)
17.8380(2)
90
99.2540(10)
90
1689.70(4)
4
Triclinic
PϪ1
Triclinic
PϪ1
Triclinic
PϪ1
Triclinic
PϪ1
Monoclinic
P21/n
13.4210(2)
20.4340(3)
15.6820(2)
90
90.3160(10)
90
4300.64(11)
4
9.0020(2)
13.8360(4)
15.0090(4)
85.7660(11)
82.9660(11)
87.5270(14)
1849.16(8)
4
9.2280(1)
9.3830(2)
14.8820(2)
77.7580(8)
77.8980(8)
75.6470(9)
1203.06(3)
2
11.4130(1)
13.4650(2)
21.9600(3)
88.8130(5)
77.8690(5)
81.8480(6)
3265.89(7)
6
8.2910(1)
11.8690(2)
13.8810(3)
71.4250(9)
77.4640(8)
89.6770(13)
1260.97(4)
2
8.3530(2)
11.2230(4)
13.7950(5)
107.832(2)
103.059(2)
94.394(2)
1184.34(7)
2
γ/Њ
U/Å3
Z
dcalc/g cmϪ3
µ/mmϪ1
Crystal size/mm
Reflections
collected
Independent
reflections
R(int)
1.806
1.855
1.559
1.683
1.166
1.756
1.274
1.659
0.979
1.766
1.081
1.656
0.938
1.702
0.998
1.600
0.684
1.450
0.25 × 0.25 × 0.25 0.30 × 0.30 × 0.30 0.15 × 0.10 × 0.10 0.40 × 0.33 × 0.18 0.20 × 0.15 × 0.08 0.25 × 0.20 × 0.20 0.15 × 0.08 × 0.05 0.05 × 0.05 × 0.03 0.20 × 0.20 × 0.10
3359
33639
21848
29040
22342
29069
20472
15723
41124
3104
4771
8370
3794
5517
9023
4408
4172
7567
D
0.0272
0.0334
0.0222, 0.0531
0.0233, 0.0537
0.0682
0.0404, 0.0959
0.0526, 0.1033
0.0323
0.0217, 0.0608
0.0225, 0.0615
0.0511
0.0308, 0.0767
0.0379, 0.0808
0.0684
0.0384, 0.1024
0.0461, 0.1102
0.0529
0.0348, 0.0969
0.0434, 0.1052
0.0893
0.0497, 0.0902
0.0934, 0.1038
0.0510
0.0462, 0.1159
0.0520, 0.1190
R1, wR2 [I > 2σ(I )] 0.0236, 0.0626
R indices (all data) 0.0262, 0.0654

View Article Online
15 R. Jackstell, H. Klein, M. Beller, K.-D. Wiese and D. Röttger,
Eur. J. Org. Chem., 2001, 3871.
16 A. D. Burrows, M. F. Mahon and M. Varrone, unpublished results.
17 P. S. Pregosin and R. W. Kunz, 31P and 13C NMR of Transition
Metal Phosphine Complexes, Springer-Verlag, 1979.
18 S. M. Aucott, A. M. Z. Slawin and J. D. Woollins, J. Chem. Soc.,
Dalton Trans., 2000, 2559.
19 K. R. Reddy, W.-W. Tsai, K. Surekha, G.-H. Lee, S.-M. Peng,
J.-T. Chen and S.-T. Liu, J. Chem. Soc., Dalton Trans., 2002, 1776.
20 R. G. Pearson, Inorg. Chem., 1973, 12, 712.
21 G. P. C. M. Dekker, C. J. Elsevier, K. Vrieze and P. W. N. M. van
Leeuwen, Organometallics, 1992, 11, 1598; Y. Kayaki, H. Tsukamoto,
M. Kaneko, I. Shimizu, A. Yamamoto, M. Tachikawa and
T. Nakajima, J. Organomet. Chem., 2001, 622, 199.
22 G. A. Luinstra and P. H. P. Brinkmann, Organometallics, 1998, 17,
5160; P. H. P. Brinkmann and G. A. Luinstra, J. Organomet. Chem.,
1999, 572, 193.
As stated earlier, the asymmetric units in compounds 3a and
7b, were seen to contain, respectively, 2 and 3 molecules, while
that in 10 comprises one cation, two triflate anions (one severely
disordered) and 1.5 water molecules. The disordered triflate
region of the electron density map was somewhat of a mine-
field, but optimal convergence was ultimately achieved by mod-
elling the anion over two sites in a 60:40 ratio. One of phenyl
rings in the cation also exhibited positional disorder in site-
occupancy proportions of 55:35. The ADPs associated with the
carbons in this disordered phenyl group were restrained to have
similar values in the final convergence cycles. Hydrogen atoms
in the full water molecule (based on O7) present in the motif of
compound 10 were readily located and refined subject to dis-
tance restraints (0.89 Å from the parent atom and 1.45 Å from
each other). Unfortunately, the hydrogen atoms associated with
the half occupancy water (O8) could not be located and hence
were omitted from the least-squares refinement.
23 E. Drent and P. H. M. Budzelaar, Chem. Rev., 1996, 96, 663.
24 K. R. Reddy, K. Surekha, G.-H. Lee, S.-M. Peng, J.-T. Chen and
S.-T. Liu, Organometallics, 2001, 20, 1292.
CCDC reference numbers 217632–217640.
25 A. Sen, Acc. Chem. Res., 1993, 26, 303.
26 C. Bianchini, A. Meli, G. Müller, W. Oberhauser and E. Passaglia,
Organometallics, 2002, 21, 4965; C. S. Shultz, J. Ledford, J. M.
DeSimone and M. Brookhart, J. Am. Chem. Soc., 2000, 122, 6351;
F. C. Rix, M. Brookhart and P. S. White, J. Am. Chem. Soc., 1996,
118, 4746; J. Ledford, C. S. Shultz, D. P. Gates, P. S. White, J. M.
DeSimone and M. Brookhart, Organometallics, 2001, 20, 5266.
27 B. Milani, A. Scarel, G. Mestroni, S. Gladiali, R. Taras, C. Carfagna
and L. Mosca, Organometallics, 2002, 21, 1323; B. A. Markies,
D. Kruis, M. H. P. Rietveld, K. A. N. Verkerk, J. Boersma,
H. Kooijman, M. T. Lakin, A. L. Spek and G. Van Koten, J. Am.
Chem. Soc., 1995, 117, 5263; S. Stoccoro, G. Minghetti, M. A.
Cinellu, A. Zucca and M. Manassero, Organometallics, 2001, 20,
4111.
28 G. P. C. M. Dekker, A. Buijs, C. J. Elsevier, K. Vrieze, P. W. N. M.
van Leeuwen, W. J. J. Smeets, A. L. Spek, Y. F. Wang and C. H.
Stam, Organometallics, 1992, 11, 1937; K. R. Reddy, C.-L. Chen,
Y.-H. Liu, S.-M. Peng, J.-T. Chen and S.-T. Liu, Organometallics,
1999, 18, 2574; P. Braunstein, M. D. Fryzuk, M. Le Dall, F. Naud,
S. J. Rettig and F. Speiser, J. Chem. Soc., Dalton Trans., 2000, 7,
1067; Y.-C. Chen, C.-L. Chen, J.-T. Chen and S.-T. Liu,
Organometallics, 2001, 20, 1285.
See http://www.rsc.org/suppdata/dt/b3/b309882f/ for crystal-
lographic data in CIF or other electronic format.
Acknowledgements
The EPSRC is thanked for financial support and Johnson-
Matthey is thanked for a generous loan of platinum metals.
References
1 E. K. van den Beuken, W. J. J. Smeets, A. L. Spek and B. L. Feringa,
Chem. Commun., 1998, 223.
2 E. Shirakawa and T. Hiyama, J. Organomet. Chem., 1999, 576, 169.
3 K. R. Reddy, K. Surekha, G.-H. Lee, S.-M. Peng and S.-T. Liu,
Organometallics, 2000, 19, 2637.
4 K. G. Moloy and J. L. Petersen, J. Am. Chem. Soc., 1995, 117, 7696.
5 C. D. Andrews, A. D. Burrows, J. M. Lynam, M. F. Mahon and
M. T. Palmer, New J. Chem., 2001, 25, 824.
6 A. D. Burrows, M. F. Mahon, M. T. Palmer and M. Varrone,
Inorg. Chem., 2002, 41, 1695.
29 A. Sommazzi and F. Garbassi, Prog. Polym. Sci., 1997, 22, 1547.
30 H.-K. Luo and D.-G. Li, J. Mol. Catal. A, 2001, 171, 23.
31 V. V. Grushin, Chem. Rev., 1996, 96, 2011; M. Portnoy and
D. Milstein, Organometallics, 1994, 13, 600.
7 H. Brunner and H. Weber, Chem. Ber., 1985, 118, 3380.
8 H. Brunner, W. Li and H. Weber, J. Organomet. Chem., 1985, 288,
359.
9 P. G. Cozzi, N. Zimmermann, R. Hilgraf, S. Schaffner and A. Pfaltz,
Adv. Synth. Catal., 2001, 343, 450.
10 A. D. Burrows, M. F. Mahon and M. Varrone, Inorg. Chim. Acta,
2003, 350, 152.
11 A. D. Burrows, CrystEngComm, 2001, 3, 217.
12 A. D. Burrows, R. W. Harrington, M. F. Mahon, M. T. Palmer,
F. Senia and M. Varrone, Dalton Trans., 2003, 3717.
13 S. Fischer, L. K. Peterson and J. F. Nixon, Can. J. Chem., 1974, 52,
3981.
14 A. Frenzel, M. Gluth, R. Herbst-Irmer and U. Klingebiel,
J. Organomet. Chem., 1996, 514, 281.
32 I. Tóth and C. J. Elsevier, Organometallics, 1994, 13, 2118.
33 M. Portnoy, F. Frolow and D. Milstein, Organometallics, 1991, 10,
3960.
34 J. Chatt, L. M. Vallarino and L. M. Venanzi, J. Chem. Soc., 1957,
3413.
35 D. Drew and J. R. Doyle, Inorg. Synth., 1972, 13, 47.
36 G. Giordano and R. H. Crabtree, Inorg. Synth., 1990, 28, 88.
37 R. E. Rülke, J. M. Ernsting, A. L. Spek, C. J. Elsevier, P. W. N. M.
van Leeuwen and K. Vrieze, Inorg. Chem., 1993, 32, 5769.
38 A. D. Burrows, M. F. Mahon, S. P. Nolan and M. Varrone,
Inorg. Chem., 2003, 42, 7227.
D a l t o n T r a n s . , 2 0 0 3 , 4 7 1 8 – 4 7 3 0
4730