Bulky triarylarsines are effective ligands for palladium catalysed Heck olefination

Article abstract of DOI:10.1039/b417910b

The four arsines, As{C₆H₃(o-CH₃)(p-Z)} ₃ {Z = H (2a) or OMe (2b)} and As{C₆H₃(o- CHMe₂)(p-Z)}₃ {Z = H (2c) or OMe (2d)} react with [PdCl₂(NCPh)₂] or [PtCl₂(NCBu₁) ₂] to give trans-[MCl₂L₂] or trans-[M ₂Cl₂(μ-Cl)₂L₂]. The crystal structures of trans-[PdCl₂(2a)₂] and [PtCl ₂(2d)₂] have been determined, the latter as its dichloromethane solvate. The structures show that in these complexes, the ligands adopt gga type conformations as do all analogous tri-o-tolyl- and tri-o-isopropylphenylphosphines in square-planar and octahedral complexes. The variable-temperature NMR behaviour of the complexes shows that they are fluxional due to restricted As-C bond rotation. The rate of the fluxionality is more rapid than in the analogous phosphine complexes and this is associated with longer As-C and As-M bonds allowing more free movement. The catalytic activity of the palladium complexes of the arsines and their phosphine analogues for the reaction of 4-bromoacetophenone and n-butyl acrylate has been screened. The results show that the arsines are generally superior to the phosphines as ligands for this catalysis. Tri(o-isopropylphenyl)phosphine and tri(o-isopropylphenyl)arsine are superior to tri-o-tolylphosphine as ligands for this Heck reaction and a p-methoxy substituent improves the arsine catalyst but not the phosphine catalyst. The phosphine catalysts are superior to the arsine catalysts for the reaction of 4-chloroacetophenone and n-butyl acrylate. These observations are discussed in the context of ligand stereoelectronic effects, as measured by the Tolman electronic parameter, v_CO of the [NiL(CO)₃] {L = AsAr₃ or PAr₃}. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b417910b

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Bulky triarylarsines are effective ligands for palladium catalysed
Heck olefination
R. Angharad Baber,a Simon Collard,b Mark Hooper,b A. Guy Orpen,a Paul G. Pringle,*a
Matthew J. Wilkinsona and Richard L. Wingada
a School of Chemistry, University of Bristol, Cantocks Close, Bristol, UK BS8 1TS
b Johnson-Matthey, Royston, Herts, UK
Received 26th November 2004, Accepted 8th February 2005
First published as an Advance Article on the web 10th March 2005
The four arsines, As{C6H3(o-CH3)(p-Z)}3 {Z = H (2a) or OMe (2b)} and As{C6H3(o-CHMe2)(p-Z)}3 {Z = H (2c) or
OMe (2d)} react with [PdCl2(NCPh)2] or [PtCl2(NCBut)2] to give trans-[MCl2L2] or trans-[M2Cl2(l-Cl)2L2]. The
crystal structures of trans-[PdCl2(2a)2] and [PtCl2(2d)2] have been determined, the latter as its dichloromethane
solvate. The structures show that in these complexes, the ligands adopt gga type conformations as do all analogous
tri-o-tolyl- and tri-o-isopropylphenylphosphines in square-planar and octahedral complexes. The variable-
temperature NMR behaviour of the complexes shows that they are fluxional due to restricted As–C bond rotation.
The rate of the fluxionality is more rapid than in the analogous phosphine complexes and this is associated with
longer As–C and As–M bonds allowing more free movement. The catalytic activity of the palladium complexes of the
arsines and their phosphine analogues for the reaction of 4-bromoacetophenone and n-butyl acrylate has been
screened. The results show that the arsines are generally superior to the phosphines as ligands for this catalysis.
Tri(o-isopropylphenyl)phosphine and tri(o-isopropylphenyl)arsine are superior to tri-o-tolylphosphine as ligands for
this Heck reaction and a p-methoxy substituent improves the arsine catalyst but not the phosphine catalyst. The
phosphine catalysts are superior to the arsine catalysts for the reaction of 4-chloroacetophenone and n-butyl acrylate.
These observations are discussed in the context of ligand stereoelectronic effects, as measured by the Tolman
electronic parameter, mCO of the [NiL(CO)3] {L = AsAr3 or PAr3}.
In order to probe ligand stereoelectronic effects in the catalysis
of Heck olefination, we have investigated the catalytic activity
of the palladium complexes of the series of phosphines 1b–
Introduction
Heck olefination1 (eqn. (1)) is an important C–C bond-
forming reaction, made particularly useful by its functional
d. In view of the success of arsine ligands in the catalysis of
group tolerance.2 The palladacycle catalyst derived from tri-o-
another C–C bond-forming reaction, Stille coupling,8 we made
tolylphosphine (1a) was shown by Herrmann and Beller3 to be
the triarylarsines 2a–d and report here rare examples of arsine-
an outstandingly active catalyst precursor and significantly, the
based Heck catalysts.9
turnover rates with chloroarenes were of commercial potential;4
several applications in the fine chemicals industry were recently
reviewed.2 Subsequently, the groups of Milstein,5 Bedford6 and
Results and discussion
Cole-Hamilton7 have shown that the cyclopalladates in Scheme 1
The ligands 1a–d and 2a were made by modifications of
are all excellent catalyst precursors for Heck olefination.
literature methods.10,11 The new arsines 2a–d have been made by
treatment of AsCl3 with the appropriate Grignard reagent (see
Experimental section). Ligands 1b–d and 2a–d were designed to
probe stereoelectronic effects in Heck catalysis.
(1)
Platinum(II) and palladium(II) chemistry
We recently reported11 the coordination chemistry of the phos-
phines 1a–d with platinum(II) and palladium(II). The complexes
of the arsine analogues 2a–d have been characterised principally
by 1H NMR, IR and mass spectrometry. They are more difficult
to characterise than their phosphine analogues because of the
absence of an NMR probe with the simplicity and analytical
1
precision of 31P. From the similarity of the H NMR and IR
spectra of the complexes of 1a–d and 2a–d, it seems reasonable
to conclude that the structures are generally similar but there
are differences and these are presumably associated with the in-
creased M–As and C–As bond lengths and hence smaller cone
angles (see below).
The reactions of [PtCl2(NCBut)2] or [PdCl2(NCPh)2] in re-
fluxing toluene with 2 equiv. of 2a or 2b gave precipitated
products very slowly, and low yields were obtained unless the
reflux was continued for several days. The IR spectra (in the 200–
400 cm−1 region) were consistent with them being predominantly
the trans-complexes 3–4a,b and the crystal structure of 4a has
been determined (see below). The H NMR spectra for all the
1
Scheme 1
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 5
D a l t o n T r a n s . , 2 0 0 5 , 1 4 9 1 – 1 4 9 8
1 4 9 1
©

View Article Online
◦
mononuclear complexes 3–4a,b at +23 C show broad signals
in the d 2.2–2.4 region for the tolyl-CH3 groups indicating flux-
ionality due to restricted C–As bond rotation (see below). These
signals are less broad than for the corresponding phosphine
analogues (e.g. w1/2 for the CH3 signal for trans-[PtCl2(2b)2] and
trans-[PtCl2(1b)2]11 were 16 and 90 Hz, respectively) consistent
with C–As rotation being more rapid than C–P rotation.
Treatment of [PtCl2(NCBut)2] with 1 equiv. of 2b in refluxing
toluene gave a product assigned to the isomeric binuclear cy-
clometallated species 6b and 6bꢀ on the basis of the characteristic
1H NMR signals for the Pt–CH2 at d 3.47 and 3.43 (ratio 1 : 1)
which have broad 195Pt satellites (2J(PtH) 102 Hz). There are
no previous reports of cyclometallation of tri-o-tolylarsine but
cyclometallated trimesitylarsine complexes are known.12
The reactions of [PtCl2(NCBut)2] or [PdCl2(NCPh)2] with the
bulky arsines 2c and 2d were problematic. Impure products were
obtained contaminated with solvent, metal starting materials
and/or ligand. When [PtCl2(NCBut)2] is treated with 2 equiv. of
2c and 2d the trans mononuclear complexes 3c and 3d are the
1
main products as identified on the basis of H NMR and IR
spectroscopy (see Experimental section). The crystal structure
of 3d has been determined (see below). The reaction of 2c with
[PdCl2(NCPh)2] gave the binuclear species 5c (see Experimental
section for the data). The product of the reaction of 2d with
[PdCl2(NCPh)2] depends on the stoichiometry. Thus the reaction
of [PdCl2(NCPh)2] with 2 equiv. of 2d gave the mononuclear
trans species 4d (m(PdCl) 347 cm−1) whereas with 1 equiv. of 2d,
the IR spectrum (m(PdCl) 352, 293, 257 cm−1) of the product is
consistent with the binuclear species 5d.
X-Ray crystal structures
Molecules of trans-[PdCl2(2a)2] (4a) have crystallographic Ci
symmetry (see Fig. 1 and Table 1 which lists selected bond
The H NMR spectrum of 5c is sharp at −20 ◦C, showing
lengths and angles). The average As–C bond length is 1.94 A,
1
˚
inequivalent CHMe2 peaks and high-frequency aromatic C–
H signals in a similar pattern to its phosphine analog◦ue (see
Experimental section for the data).11 However, at +23 C, the
1H NMR spectrum for 5c shows much coalescence of signals
and is reminiscent of the spectrum of its phosphine analogue
with an average C–As–C bond angle of 104.4◦ and a Pd–A◦s
˚
bond length of 2.43 A. Ligand 2a has a cone angle of 175
in this structure. The average Pd–As–C bond angle is 114.2◦.
As can be seen in Fig. 1, two of the rings are close to a
gauche (g) conformation, the other ring (C1–C7) is in the anti
(a) conformation; the Pd–As(1)–C–C(Me) torsion angles are
−70.0, −54.1 and 173.3◦, i.e. g−g−a conformation. The gga
conformation is observed in all square-planar or octahedral
metal complexes of tri-o-tolylphosphine.11 The rotamer of 4a
observed in the crystal is the meso form. We have shown11 that
there are two rotamers possible in this class of complex (denoted
◦
at 100 C. This indicates that the rotameric fluxionality in the
arsine complex is a significantly lower energy process than for
its phosphine analogue; the estimated DG‡ from coalescence
temperatures is ca. 62 kJ mol−1 for 5c compared to ca. 78 kJ mol−1
in the phosphine analogue.11 The longer M–As and C–As bond
lengths would be expected to allow less restricted rotation.
1 4 9 2
D a l t o n T r a n s . , 2 0 0 5 , 1 4 9 1 – 1 4 9 8

View Article Online
◦
◦
˚
˚
Table 1 Selected bond lengths (A) and angles ( ) for trans-[PdCl2(2a)2]
(4a)
Table 2 Selected bond lengths (A) and angles ( ) for trans-[PtCl2(2d)2]
(3d)
Pd(1)–Cl(1)
Pd(1)–As(1)
As(1)–C(1)
2.3034(11) As(1)–C(8)
1.944(5)
1.947(4)
Pt(1)–Cl(1)
Pt(1)–Cl(2)
Pt(1)–As(2)
Pt(1)–As(1)
As(1)–C(1)
2.292(2)
As(1)–C(21)
As(1)–C(11)
As(2)–C(31)
As(2)–C(41)
As(2)–C(51)
1.962(7)
1.941(7)
1.964(7)
1.967(8)
1.966(7)
2.4303(5)
1.941(4)
As(1)–C(15)
2.3218(19)
2.4262(9)
2.4306(10)
1.949(7)
Cl(1)–Pd(1)–Cl(1A)a 180.00
Cl(1)–Pd(1)–As(1)
Cl(1A)–Pd(1)–As(1)
As(1)–Pd(1)–As(1A) 180.00
C(1)–As(1)–C(8)
C(1)–As(1)–C(15)
C(8)–As(1)–C(15)
C(1)–As(1)–Pd(1)
C(8)–As(1)–Pd(1)
103.92(18)
106.86(19)
116.94(13)
116.35(12)
88.37(3)
91.63(3)
Cl(1)–Pt(1)–Cl(2)
Cl(1)–Pt(1)–As(2)
Cl(2)–Pt(1)–As(2)
Cl(1)–Pt(1)–As(1)
Cl(2)–Pt(1)–As(1)
As(2)–Pt(1)–As(1)
C(1)–As(1)–Pt(1)
C(11)–As(1)–Pt(1)
C(21)–As(1)–Pt(1)
179.52(7)
91.94(5)
87.83(5)
92.20(5)
88.05(5)
175.31(3)
109.06(19)
113.3(2)
121.0(2)
C(31)–As(2)–Pt(1)
C(41)–As(2)–Pt(1)
C(51)–As(2)–Pt(1)
C(11)–As(1)–C(1)
C(11)–As(1)–C(21)
C(1)–As(1)–C(21)
C(31)–As(2)–C(51)
C(31)–As(2)–C(41)
C(51)–As(2)–C(41)
115.2(2)
106.71(19)
120.7(2)
105.9(3)
101.9(3)
104.4(3)
101.4(3)
104.8(3)
106.9(3)
102.32(18) C(15)–As(1)–Pd(1) 109.31(13)
a Inversion symmetry generated atoms (2 − x, −y, −z) are designated
by suffix A.
Fig. 1 Structure of trans-[PdCl2(2a)2] (4a). Hydrogen atoms are omitted
for clarity. Atoms suffixed A are related by symmetry (2 − x, −y, −z).
Fig. 3 Structure of trans-[PtCl2(2d)2] (3d). Hydrogen atoms have been
omitted for clarity.
Fig. 2 Schematic views along the L–M–L axis of the two proposed
meso forms of complexes of the type trans-[MCl2(L)2] where L = 1a–d
or 2a–d; the spheres represent the ortho substituents.11
meso-A and meso-B and illustrated in Fig. 2); the form present
in crystals of 4a is meso-A. The shortest Pd–H distance in the
structure is 3.25 A and involves a methyl C–H on one of the
Fig. 4 Schematic views along the L–M–L axis of the two proposed
rac forms of complexes of the type trans-[MCl2(L)2] where L = 1a–d or
2a–d; the spheres represent the ortho substituents.11
˚
gauche tolyl groups.
Fig. 3 shows the structure of trans-[PtCl2(2d)2] (3d) determined
in crystals of its dichloromethane solvate. Table 2 lists selected
bond lengths and angles in 3d. The two arsine ligands are not
symmetry related in this case and the complex has approximate
P bonds in their phosphine analogues and the cone angles are
correspondingly reduced by ca. 5◦.
Heck catalysis
˚
C2 symmetry. The average As–C bond length is 1.96 A with
an average C–As–C bond angle of 104.2◦, and an average Pt–
As–C bond angle of 114.6◦. The cone angle of 2d is 191◦ in this
structure. The isopropyl groups adopt an orientation that has the
tertiary hydrogen eclipsed to the aromatic ring and towards the
arsenic atom, presumably thereby minimising steric interactions
of the methyl groups. This orientation is the same as that adopted
in all known analogous ortho-isopropylarylphosphine ligands.11
The aryl groups of both arsine ligands in this structure adopt
g+g+a conformations; Pt–As–C–C(Pri) torsion angles are 67.3,
63.8 and 176.1◦ at As(1) and 62.6, 64.9 and 178.4◦ at As(2).
As with the meso form, there are two rotamers possible (rac-
A and rac-B as illustrated in Fig. 4);11 the form present in
crystals of 3d is rac-A. The As–C and As–M distances in 4a
Palladium complexes of 1a–d and 2a–d were screened using high-
throughput apparatus for the Heck reaction shown in eqn. (2).
(2)
The catalysts were either the pre-formed palladium complexes
shown in Scheme 2 or [Pd(OAc)2]/ligand mixtures.
˚
and 3d are approximately 0.12 A longer than the P–C and M–
D a l t o n T r a n s . , 2 0 0 5 , 1 4 9 1 – 1 4 9 8
1 4 9 3

View Article Online
Fig. 5 Conversions in the Heck reaction shown in eqn. (2) with catalysts
derived from phosphines 1 and arsines 2 under the conditions given in
Table 3.
We concluded from these observations that triarylarsines
can be at least as effective as triarylphosphines as ligands
for the Heck reaction. In addition ortho-isopropyl and para-
methoxy substituents were beneficial for the performance of the
triarylarsine catalysts. It was therefore of interest to investigate
whether the effects of these substituents was cumulative and
therefore ligands 1d and 2d were screened.
In a Schlenk tube, a mixture of n-◦butyl acrylate and 4-
bromoacetophenone was heated to 100 C and then 0.5 mol%
of the palladium complexes of 1a, 1d and 2d (i.e. 7a, 8d and
4d, respectively) were added and the reaction (eqn. (2)) was
monitored by GC over 90 min; the results are plotted in Fig. 6.
Scheme 2 Pre-formed palladium complexes screened for catalysis.
The products were analysed by gas chromatography (GC)
which showed that a single alkene product was formed in each
case. Duplicate experiments were carried out for each catalyst
run to ensure reproducibility. The pre-formed catalysts generally
exhibited higher activity than the corresponding in situ catalyst
but the trends in the results were similar. For simplicity, only
the results for the pre-formed catalysts are given in Table 3
and presented graphically in Fig. 5. Herrmann has previously
reported13 that the same catalytic performance was obtained
with palladium acetate or palladium halide complexes and we
have also observed similar results with 1a/Pd(OAc)2 and 7a for
the reaction in eqn. (2).
The following are apparent from Table 3 and Fig. 5:
1. AsPh3 (entry 2) gives a superior catalyst to PPh3 (entry 1).
2. The catalyst derived from 1a gives a more active catalyst
(entry 3) than the arsine analogue 2a (entry 4).
3. The presence of the para-methoxy group in ligands 1b
and 2b has the effect of reducing the activity of the phosphine
catalyst 7b but of improving the activity of the arsine catalyst 4b
significantly (entries 5 and 6).
4. Replacement of the ortho-methyl group with an ortho-
isopropyl group improves the activity of the catalysts derived
from phosphine 1c and arsine 2c (entries 7 and 8).
Fig. 6 Conversions in the Heck reaction shown in eqn. (2) as a function
of time with catalysts derived from phosphines 1a, 1d and arsine 2d under
the conditions given in Table 3.
Table 3 Catalysis results
Entry
Catalyst
Ligand
mol% Pd
Reaction eqn.
Time/h
Temp/◦C
Conv. (%)
TON
13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Pd(OAc)2/PPh3
Pd(OAc)2/AsPh3
PPh3
AsPh3
1a
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.0001
0.0001
0.0001
0.1
0.1
0.1
0.1
0.1
0.1
0.1
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
16
16
16
16
16
16
16
16
24
24
24
21
21
21
21
21
21
21
100
100
100
100
100
100
100
100
100
100
100
140
140
140
140
140
140
140
7
45
67
34
13
98
94
99
1
89
130
68
7a
4a
7b
4b
8c
5c
7a
8d
4d
7a
4a
4b
8c
5c
8d
4d
2a
1b
26
2b
196
188
200
12200
64200
431000
100
8
1c
2c
1a
1d
6
2d
43
10
1
4
12
9
1a
2a
2b
40
125
85
133
68
1c
2c
1d
13
7
2d
1 4 9 4
D a l t o n T r a n s . , 2 0 0 5 , 1 4 9 1 – 1 4 9 8

View Article Online
Table 4 IR and 31P NMR data for the [Ni(CO)3L] complexes in CH2Cl2
Under these conditions, complexes derived from 1a and 1d
performed similarly but the arsine system derived from 2d greatly
outperformed the phosphines. In another experiment, much
lower levels (10−3 mol%) of the catalysts were added and the
reactions run for 24 h to determine turnover numbers. Under
these conditions (see Table 3, entries 9–11), the arsine system
2d achieves a TON of 431000 which is over 6 times that of the
phosphine analogue 1d and 35 times that of 1a.
L
mCO
dP
PPh3
1a
2068.9
2066.4
2063.9
2064.8
2062.7
42.9
29.4
49.4
31.9
76.6
1b
1c
1d
One great advantage of the 1a/Pd system is that it catal-
yses the Heck olefination of chloroarenes.4 We therefore
screened the phosphine and arsine complexes for catalysis of 4-
chlorobenzophenone with n-butyl acrylate (eqn. (3)). The results
are given in Table 3 (entries 12–18) and shown in Fig. 7. It can be
seen that, in general the phosphine systems are superior to the
analogous arsine systems. The ortho-isopropyl group improves
the performance of both phosphine and arsine-based catalysts.
AsPh3
2a
2071.9
2069.6
2067.7
2064.9
2064.4
2b
2c
2d
(3)
Fig. 8 Conversions in the Heck reaction shown in eqn. (2) with catalysts
derived from phosphines 1 and arsines 2 as a function of Tolman
electronic parameter.
Herrmann et al.13 have suggested that the Heck catalysts
derived from 1a involve cyclopalladates. In an attempt to shed
light on the Pd species that may be present under the conditions
we used for the catalysis, the reactions of dipalladium complex
8c under the catalysis conditions were investigated. Complex 8c
dissolved in hot N,N-dimethylacetamide to give a clear orange
solution. This was then treated with 10 equiv. of Na2CO3 and
heated further to give a cloudy brown solution. The 31P NMR
spectrum of this solution showed several species were present
including 8c and dissociated ligand 1c; the most intense (70%)
signal at d −26.7 corresponded to a new species which we have
also observed upon treatment of 8c with various bases (Cs2CO3,
NEt3, KOBut) in THF or CDCl3 and upon treatment of 8c with
AgBF4 in MeCN followed by NEt3. It is tempting to assign
this signal to the cyclopalladated complex 9c, the platinum
analogue of which we have previously identified.11 However
several attempts to isolate this species have been unsuccessful
because it decomposes upon concentration of its solutions.
Fig. 7 Conversions in the Heck reaction shown in eqn. (3) with catalysts
derived from phosphines 1 and arsines 2 under the conditions given in
Table 3.
The paucity of catalysis data on arsine complexes is partly
due to the perceived disadvantage of arsines in terms of toxicity.
Moreover, early hydrogenation14 and hydroformylation15 studies
indicated that for these reactions, triphenylarsine–rhodium
catalysts were inferior to triphenylphosphine-rhodium cata-
lysts. However there are several examples of C–C coupling
reactions where arsine complexes give more active or selective
catalysts.8,9,16 We have demonstrated here for the Heck reaction
that triarylarsine catalysts can perform as well as or better than
the analogous, very active triarylphosphine catalysts.
To compare the bonding properties of 1a–d with 2a–d, the
ligands were reacted with an excess of [Ni(CO)4] and the IR
spectra of the resulting [NiL(CO)3] complexes were obtained
to determine the frequency of the A1 stretching mode (i.e. the
Tolman electronic parameter17). The results are given in Table 4.
It is clear that the m(CO) values for the arsines are consistently
higher than for the analogous phosphines, consistent with
arsines being poorer r-donors/better p-acceptors.
The catalyst performance in terms of yields from the
chloroarene Heck reaction (eqn. (3)) is plotted against the
Tolman electronic parameter in Fig. 8. There appears to be a
general correlation: the better the r-donor, the higher the yield
which is consistent with oxidative addition of the chloroarene
being rate determining. The results for the bromoarene reaction
(eqn. (2)) show no simple correlation which implies that more
complex ligand stereoelectronic effects determine the rate.
The mechanism of Heck olefination catalysed by palladacy-
cles has been widely discussed. Pd(0)/Pd(II) (with or without
rupture of the Pd–C bond18) and Pd(II)/Pd(IV)19 cycles have
been presented with the suggestion that the high activity of the
metallacycles can be rationalised in terms of the electron richness
of the intermediates leading to rapid oxidative addition of the
ArX. It has also been postulated20 that colloidal or nanocluster
Pd particles are involved and the metallacycle provides a source
of palladium particles at the temperatures of the reactions.
Others have shown21 that when metallic palladium is used as
D a l t o n T r a n s . , 2 0 0 5 , 1 4 9 1 – 1 4 9 8
1 4 9 5

View Article Online
a catalyst, it is leached during Heck reactions to give soluble
palladium(II) species in solution.
J = 2.9, 8.3 Hz), 3.68 (3H, s), 2.27 (3H, s); 13C NMR (CD2Cl2):
d 160.6 (s), 144.3 (s), 134.8 (s), 129.1 (s), 116.1 (s), 112.0 (s), 55.4
(s), 22.2 (s).
The beneficial effect of the o-isopropyl group on the phosphine
and arsine catalysts may be a result of the steric protection
offered by the o-isopropyl substituents thermally stabilising the
catalysts or leading to production of more active palladium
nanoparticles.
Tri(o-isopropylphenyl)arsine (2c)
1-Bromo-2-isopropylbenzene (12.50 g, 63 mmol) was added to
Mg turnings (1.53 g, 63 mmol) in THF (60 cm3) over 5 min.
The mixture was heated at reflux for 1 h until the Mg turnings
had disappeared. The grey solution was then allowed to cool
to room temperature before the dropwise addition of AsCl3
(3.80 g, 21 mmol) in THF (50 cm3) at −78 ◦C over 30 min. The
reaction was stirred overnight and then distilled water (50 cm3)
and saturated NH4Cl solution (30 cm3) were added and the
mixture stirred for 30 min. The mixture was then extracted with
Et2O (2 × 50 cm3) and the ethereal extracts combined and dried
over Na2SO4. The solvents were removed in vacuo to yield a
viscous brown oil, which was digested with methanol (60 cm3)
for 48 h to leave a white precipitate. The supernatant methanol
was then removed with a cannula before the powder was washed
with cold methanol (2 × 20 cm3) and the product dried in vacuo,
CH2Cl2 (2 × 50 cm3) was added to precipitate the MgClBr salts,
which were filtered off, The solvent was then evaporated in vacuo
from the flitrate to give the product 2c as a fine white powder
Conclusion
The coordination chemistry of the bulky arsines 2a–d has been
compared with that of the analogous phosphines 1a–d. From the
NMR studies, the greater As–C and As–M bond lengths make
As–C/M–As rotation more rapid than the P–C/M–P rotation
in analogous compounds. The X-ray crystallography and IR
studies reveal that the arsines are less bulky and poorer r-donors
or better p-acceptors than the analogous phosphines. For the
Heck olefinations involving bromoarene substrates, the catalysts
derived from the palladium–arsine complexes were generally
more active than the analogous phosphine derived catalyst;
the bulky arsine 2d yielded a particularly efficient catalyst and
illustrated once again the activating effect an ortho-isopropyl
group can have on catalyst performance.11
1
(2.25 g, 5.20 mmol, 25%). MS (CI) m/z: 432 (M+); H NMR
Experimental
(400 MHz, CD2Cl2): d 7.25 (2H, m), 6.95 (1H, m), 6.75 (1H,
d), 3.36 (1H, septet), 1.07 (6H, d); 13C NMR (CD2Cl2): d 153.5
(s), 138.1 (s), 134.5 (s), 129.3 (s), 126.5 (s), 125.7 (s), 33.2 (s),
24.1 (s).
Unless otherwise stated, all work was carried out under a
dry nitrogen atmosphere, using standard Schlenk-line tech-
niques. Dry N2-saturated solvents were collected from
a
Grubbs system22 in flame and vacuum dried glass-
ware. Dry dimethylacetamide (DMA) for the Heck olefi-
nation reactions was purchased from Aldrich. 4-Bromo-3-
isopropylanisole,23 triarylphosphines 1b–d,11 tri(o-tolyl)arsine
(2a),10 [PtCl2(NCtBu)2]24 and [PdCl2(NCPh)2]25 were prepared
by literature methods. Tri(o-tolyl)phosphine (1a) was purchased
from Strem and all other starting materials were purchased from
Aldrich. NMR spectra were measured on a Jeol Eclipse 300,
Tri(o-isopropyl-p-methoxy)arsine (2d)
4-Bromo-3-isopropylanisole (9.01 g, 39 mmol) was added to
Mg turnings (1.00 g, 41 mmol) in THF (5 cm3) over 5 min.
The mixture was heated at reflux for 2 h. The black solution
was then allowed to cool to room temperature and filtered
to remove excess Mg, before the dropwise addition of AsCl3
(1.87 g, 10.3 mmol) in THF (20 cm3). The reaction was stirred
overnight and then an aqueous NH4Cl solution (2 M, 100 cm3)
was added and the mixture stirred for 30 min. The mixture was
then extracted with Et2O (3 × 30 cm3) and the ethereal extracts
combined and dried over Na2SO4. The solvents were removed
in vacuo to yield a brown solid, which was digested with methanol
(50 cm3) to leave a white precipitate. The supernatant methanol
was then removed with a cannula before the powder was washed
with cold methanol (2 × 20 cm3) and the product dried in vacuo
to give the product 2d as a white powder (2.21 g, 4.23 mmol,
41%). Elemental analysis (%): found (calc.): C, 68.53 (68.95);
Jeol Eclipse 400 or Jeol GX 400. Unless otherwise stated 1H, 13
C
and 31P NMR spectra were recorded at 300, 100 and 121 MHz,
respectively, at room temperature. Mass spectra were recorded
on a MD800 or a Fisons VG Quattro positive electrospray mass
spectrometer. Infrared spectra were recorded on a Perkin Elmer
Spectrum 1 Spectrometer as Nujol mulls between polythene
plates for the Pt and Pd complexes and as CH2Cl2 solutions
for the [NiL(CO)3] species. Elemental analyses were carried out
by the Microanalytical Laboratory of the School of Chemistry,
University of Bristol. GC analysis was done with the help of
Graham Henderson at Johnson Matthey using a Perkin Elemer
Autosystem XL GC with a CP-SIL-5 column.
H, 7.77 (7.52); MS (CI) m/z: 522 (M+); H NMR (400 MHz,
1
CDCl3): d 6.81 (1H, d), 6.73 (1H, d), 6.55 (1H, dd), 3.83 (s, 3H),
3.35 (septet, 1H), 1.05 (d, 6H); 13C NMR (CDCl3): d 160.3 (s),
154.6 (s), 135.4 (s), 127.9 (s), 111.5 (s), 111.1 (s), 55.0 (s), 32.8
(s), 24.0 (s).
Tri(p-methoxy-o-tolyl)arsine (2b)
4-Bromo-3-methylanisole (17.69 g, 88 mmol) was added to Mg
turnings (3.20 g, 132 mmol) in THF (50 cm3) over 5 min. The
mixture was heated at reflux for 3 h until the Mg turnings had
disappeared. The grey solution was cooled in an acetone/ice
bath before the dropwise addition of AsCl3 (5.27 g, 29 mmol)
over 10 min. The reaction was stirred overnight and then distilled
water (95 cm3) was added and the mixture stirred for 30 min.
The mixture was then extracted with Et2O (3 × 50 cm3) and the
ethereal extracts combined and dried over Na2SO4. The solution
was filtered and the solvents were removed in vacuo to yield a
brown solid, which was digested with methanol (2 × 50 cm3).
The solvent was removed in vacuo and CH2Cl2 (2 × 50 cm3) was
added to precipitate the MgClBr salts, which were filtered off.
The solvent was then evaporated in vacuo from the filtrate to
give a white solid, which was digested with methanol to give the
product 2b as a fine white powder (8.35 g, 19.00 mmol, 65%).
Elemental analysis (%): found (calc.): C, 65.15 (65.75); H, 6.47
(6.20); MS (CI) m/z: 438 (M+); 1H NMR (400 MHz, CD2Cl2): d
6.71 (1H, d,3J = 3.5 Hz), 6.59 (1H, d, 4J = 8.3 Hz), 6.52 (1H, dd,
trans-[PtCl2(2a)2] (3a)
Arsine 2a (0.121 g, 0.347 mmol) and [PtCl2(NCBut)2] (0.075 g,
0.174 mmol) were dissolved in toluene (5 cm3) and heated at
◦
85 C for 15 h to give a green–yellow precipitate. The solution
was allowed to cool to room temperature before removal of
the solvent by cannula. The resulting solid was then dissolved
in CH2Cl2 and filtered through a pad of Florisil to give a
yellow solution. The solvent was reduced and hexane was
added to induce precipitation. The solution was filtered off
and the remaining solid was washed with cold toluene (2 ×
1 cm3) before drying in vacuo to give 3a as a pale yellow
powder (0.050 g, 0.048 mmol, 28%). Elemental analysis (%) for
3a·CH2Cl2 (presence of solvent confirmed by 1H NMR): found
(calc.): C, 48.74 (49.30); H, 4.00 (4.23); MS (FAB) m/z 890
(M+ − 2Cl); IR: m(Pt–Cl) 343 cm−1; 1H NMR (CD2Cl2): d 7.71
(1H, br s), 7.32 (1H, dt), 7.13 (2H, m), 2.25 (3H, br s).
1 4 9 6
D a l t o n T r a n s . , 2 0 0 5 , 1 4 9 1 – 1 4 9 8

View Article Online
trans-[PtCl2(2b)2] (3b)
1.45 (d), 1.25 (d), 1.20 (d), 1.17 (d), 0.98 (d), −0.01 (d), −0.06
(d), −0.11 (d), −0.20 (d).
Complex 3b was prepared as a yellow solid (0.20 g, 0.17 mmol,
61%) in a similar fashion to 3a. Elemental analysis (%) for
1
[PdCl2(2d)]2 (5d)
3b·H2O (presence of solvent confirmed by H NMR): found
(calc.): C, 49.63 (49.67); H, 4.94 (4.86); MS (FAB) m/z 1142
(M+), 1105 (M+ − Cl), 1070 (M+ − 2Cl); IR: m(Pt–Cl) 342 cm−1;
1H NMR (400 MHz, CD2Cl2): d 7.58 (1H, br s), 6.68 (2H, m),
3.72 (3H, s), 2.22 (3H, br s); 13C NMR (CD2Cl2): d 161.7 (s),
144.8 (s), 136.0 (br s), 120.3 (s), 117.3 (s), 111.5 (s), 54.1 (s),
23.4 (s).
Complex 5d was prepared as an orange solid (0.396 g, 0.25 mmol,
74%) in a similar fashion to 4a except that 1 equiv. of 2d was
used rather than 2 equivs. Elemental analysis (%) for 5d·2CH2Cl2
1
(presence of solvent confirmed by H NMR): found (calc.): C,
47.51 (47.44); H, 5.22 (5.27); MS (ES) m/z 1362 ([Pd2Cl3(2d)2]+),
665 ([Pd2Cl2(2d)2]2+); IR: m(Pd–Cl) 352, 293, 257 cm−1; 1H NMR
(CDCl3): d 8.88 (br), 7.19 (m), 6.98 (br), 6.73 (br), 3.81 (br, m),
3.37 (br), 2.65 (br), 2.48 (br), 2.37 (br), 2.03 (br), 1.56 (br), 1.35
(br), 1.07 (br), 0.09 (br).
trans-[PtCl2(2c)2] (3c)
Complex 3c was prepared as a pale yellow solid in a similar
fashion to 3a but satisfactory elemental analyses were not
obtained. IR: m(Pt–Cl) 346 cm−1; 1H NMR (400 MHz, CD2Cl2):
d 8.65 (br), 7.65 (br), 7.53 (br), 7.30 (br), 7.14 (br), 7.08 (br), 3.50
(br), 3.20 (br), 2.37 (br), 1.80 (br), 1.45 (br), 1.18 (d), 1.07 (br),
0.92 (br), −0.15 (br), −0.20 (br).
[PtCl(2b − H)]2 (6b)
Arsine 2b (0.156 g, 0.356 mmol) and [PtCl2(NCBut)2] (0.154 g,
0.356 mmol) were dissolved in toluene (10 cm3) and heated at
reflux for 22 h to give a dark brown solution. The volatiles
were removed in vacuo and the resulting residue was re-dissolved
in CH2Cl2 (20 cm3). The solution was filtered through a 2 cm
thick plug of Florisil which was washed through with a further
60 cm3 of CH2Cl2 to give a pale yellow solution. The solvent
was removed in vacuo to give 6b as a pale yellow solid (0.150 g,
0.11 mmol, 63%). Elemental analysis (%): found (calc.): C, 42.75
(43.13); H, 3.86 (3.92); MS (ES) m/z 1298 ([Pt2Cl(2b − H)2]+);
IR: m(Pt–Cl) 281, 248 cm−1; 1H NMR (CDCl3): d 7.17 (m), 6.92
trans-[PtCl2(2d)2] (3d)
Complex 3d was prepared as an orange solid in a similar fashion
to 3a but satisfactory elemental analyses were not obtained. IR:
m(Pt–Cl) 347 cm−1; 1H NMR (CDCl3): d 8.64 (br), 7.38 (br), 7.18
(br), 6.68 (br), 3.88 (br), 3.77 (br), 3.61 (br), 3.32 (br), 2.52 (br),
1.83 (br), 1.29 (br), 1.00 (br), −0.02 (br).
2
(m), 6.79 (d), 6.76 (d), 6.64 (m), 3.76 (s), 3.47 (s, J(Pt–H) =
trans-[PdCl2(2a)2] (4a)
2
102 Hz), 3.43 (s, J(Pt–H) = 102 Hz), 2.67 (s), 2.53 (s); 13C
NMR (75 MHz, CDCl3): d 161.88 (s), 161.85 (s), 161.56 (s),
161.49 (s), 157.79 (s), 157.73 (s), 144.04 (s), 144.00 (s), 133.73
(br s), 131.87 (s), 131.65 (s), 128.06 (s), 12.96 (s), 121.16 (v br),
117.45 (s), 117.43 (s), 112.97 (s), 112.72 (s), 111.52 (s), 111.29
(s), 111.03 (s), 110.93 (s), 55.41 (s), 55.32 (s), 23.18 (s), 23.12 (s),
14.83 (s), 14.43 (s).
Complex 4a was prepared as a yellow powder (0.76 g, 0.83 mmol,
91%) in a similar fashion to 3a replacing [PdCl2(NCPh)2]
for [PtCl2(NCBut)2]. Elemental analysis (%) for 4a·1/3CHCl3
1
(presence of solvent confirmed by H NMR): found (calc.): C,
55.71 (55.65); H, 4.68 (4.67); MS (FAB) m/z 839 (M+ − Cl), 802
(M+ − 2Cl); IR: m(Pd–Cl) 356 cm−1; 1H NMR (CDCl3): d 7.73
(1H, br m) 7.32 (1H, dt), 7.20 (2H, m), 2.37 (3H, br s).
High-throughput Heck olefinations
trans-[PdCl2(2b)2] (4b)
The pre-formed palladium complexes of ligands 1a–d or 2a–
d (0.01 mmol) or ligands 1a–d or 2a–d (0.04 mmol) with
[Pd(OAc)2] (0.02 mmol) were added to vials containing aliquots
of DMA solutions of 4-bromoacetophenone (1.0 cm3, 2.0 M,
2.0 mmol) and mesitylene (0.12 g, 1.0 M, 1.0 mmol). Then
a solution of n-butylacrylate (1.0 cm3, 2.8 M, 2.8 mmol) was
added and finally Na2CO3, (2.2 mmol). The vials were placed in
a Baskerville “lunch-box” and purged with argon ten times to a
pressure of 10 atm, before releasing ◦the pressure to 2 atm. The
“lunchbox” was then heated at 100 C for 17 h with agitation.
The products were analyzed by GC.
Complex 4b was prepared as an orange solid (1.31 g, 1.12 mmol,
82%) in a similar fashion to 4a. Elemental analysis (%) for
1
4b·CHCl3 (presence of solvent confirmed by H NMR): found
(calc.): C, 50.11 (50.15); H, 4.85 (4.72); MS (FAB) m/z 1020
(M+ − Cl), 982 (M+ − 2Cl); IR: m(Pd–Cl) 340 cm−1; H NMR
1
(CDCl3): d 7.62 (1H, br), 6.73 (2H, m), 3.80 (3H, s), 2.33 (3H,
br s); 13C NMR (CD2Cl2): d 164.1 (s), 146.3 (s), 138.0 (br s),
126.8 (s), 119.3 (s), 113.7 (s), 57.0 (s), 27.6 (s).
trans-[PdCl2(2d)2] (4d)
Complex 4d was prepared as an orange solid (0.40 g, 0.23 mmol,
43%) in a similar fashion to 4a. Elemental analysis (%) for 4d·2d
Monitored Heck olefinations
1
Schlenk flasks were filled with NaOAc (0.45 g, 5.5 mmol), 4-
bromoacetophenone (0.925 g, 5 mmol) and n-butyl acrylate
(1.0 cm3, 7.0 mmol) and were degassed via freeze–pump thawing
before heating to 100 ◦C and injection of hot (100 ◦C) DMA
solutions (3.0 cm3) of 7a, 8d or 4d (0.0125 mmol). The flasks
were sampled periodically (see Fig. 4) and products analysed by
either NMR or GC.
(presence of 1 equiv. of free ligand confirmed by H NMR):
found (calc.): C, 62.39 (61.95); H, 6.81 (6.76); IR: m(Pd–Cl)
1
347 cm−1; H NMR (400 MHz, CD2Cl2): d 8.40 (br) 7.22 (br),
6.64 (br), 3.76 (br, m), 3.44 (br), 3.15 (br), 2.39 (br), 1.80 (br),
1.38 (br), 1.20 (br), 0.93 (br), −0.04 (br), −0.12 (br).
[PdCl2(2c)]2 (5c)
Complex 5c was prepared as an orange solid (0.629 g, 0.48 mmol,
53%) in a similar fashion to 4a. Elemental analysis (%) for
5c·C6H5CH3 (presence of solvent confirmed by 1H NMR): found
(calc.): C, 55.51 (55.85); H, 5.91 (5.69); MS (FAB) m/z 1185
(M+ − Cl), 1148 (M+ − 2Cl); IR: m(Pd–Cl) 352, 301, 264 cm−1;
1H NMR (CD2Cl2): d 8.79 (br), 7.59 (br), 7.45 (br), 7.13 (m),
3.66 (br), 3.24 (br), 2.38 (br), 1.95 (br), 1.49 (br), 1.23 (br), 0.99
(br), −0.05 (br), −0.13 (br); 1H NMR (CD2Cl2, −20 ◦C): d 8.83
(m), 8.71 (m), 7.65 (m), 7.44 (m), 7.15 (m), 3.61 (m), 3.37 (m),
3.19 (m), 3.00 (m), 2.53 (d), 2.30 (m), 2.20 (d), 1.95 (d), 1.51 (d),
[NiL(CO)3] complexes
[Ni(CO)4] was added dropwise from an inverted cylinder into a
solution of ca. 0.1 g of ligand (1a–d) or (2a–d) in CH2Cl2 (10 cm3)
in a Schlenk tube. The mixture was then frozen by immersion in
liquid N2 before detaching the reaction vessel from the tubing
connecting it to the cylinder. The solutions were allowed to warm
to room temperature and left to stir for 16 h. The solvent and
remaining [Ni(CO)4] were evaporated off under reduced pressure
into a trap containing bleach. The product was dissolved in
D a l t o n T r a n s . , 2 0 0 5 , 1 4 9 1 – 1 4 9 8
1 4 9 7

View Article Online
Table 5 Crystal and refinement data
trans-4a
References
1 (a) R. F. Heck, J. Am. Chem. Soc., 1968, 90, 5518; (b) R. F. Heck,
J. Am. Chem. Soc., 1971, 93, 6896; (c) J. Tsuji, Acc. Chem. Res., 1969,
2, 151.
2 J. G. deVries, Can. J. Chem., 2001, 79, 1086, and references therein.
3 W. A. Herrmann, C. Brossmer, K. Ofele, C. P. Reisinger, T. Priermeier,
M. Beller and H. Fischer, Angew. Chem., Int. Ed. Engl., 1995, 34,
1844.
4 (a) Applied Homogeneous Catalysis with Organometallic Compounds,
ed. B. Cornils and W. A. Herrmann, VCH, Weinheim, 1996; (b) W. A.
Herrmann, V. P. W. Bo¨hm and C. P. Reisinger, J. Organomet. Chem.,
1999, 576, 23; (c) A. Zapf and M. Beller, Top. Catal., 2002, 19, 101.
5 M. Ohff, A. Ohff, M. E. van der Boom and D. Milstein, J. Am. Chem.
Soc., 1997, 119, 11687.
6 D. A. Albisson, R. B. Bedford, S. E. Lawrence and P. N. Scully,
Tetrahedron Lett., 1998, 39, 9793.
7 S. Gibson, D. F. Foster, R. P. Tooze and D. J. Cole-Hamilton, Chem.
Commun., 2001, 779.
8 V. Farina and G. P. Roth in Advances in Metal–Organic Chemistry,
ed. L. S. Liebskind, JAI Press, New York, 1996, vol. 5.
9 (a) A. Kojima, C. D. L. Boden and M. Shibasaki, Tetrahedron Lett.,
1997, 38, 3459; (b) S. Y. Cho and M. Shibasaki, Tetrahedron Lett.,
1998, 39, 1773.
trans-3d·3CH2Cl2
Colour, habit
Size/mm
Formula
Yellow plate
0.3 × 0.2 × 0.05
C42H42As2Cl2Pd
873.90
Yellow block
0.2 × 0.2 × 0.2
C63H84As2Cl8O6Pt
1565.83
Monoclinic
P21/n (14)
14.116(3)
24.806(5)
19.594(4)
99.57(3)
6766(2)
¨
Mr
Crystal system
Space group (no.)
Monoclinic
P21/c (14)
10.6786(18)
10.9964(19)
15.825(3)
93.093(3)
1855.6(5)
2
˚
a/A
˚
b/A
˚
c/A
b/◦
3
˚
V/A
Z
4
l/mm−1
2.443
3.408
100
47826
15511
0.0553
0.0654
2.70, −2.37
T/K
173
19130
4250
0.089
0.047
0.65, −0.72
Reflections collected
Unique data
Rint
Final R1 [I > 2r(I)]
Dqmax, min/e A
−3
˚
10 (a) G. J. Burrows and E. E. Turner, J. Chem. Soc., 1920, 1373;
(b) J. A. S. Howell, M. G. Palin, P. C. Yates, P. McCardle, D.
Cunningham, Z. Goldschmidt, H. E. Gottlieb and D. Hezroni-
Langeman, J. Chem. Soc., Perkin Trans. 2, 1992, 1772.
11 R. A. Baber, A. G. Orpen, P. G. Pringle, M. J. Wilkinson and R. L.
Wingad, Dalton Trans., 2005, 659, and references therein.
12 (a) S. A. Dias and E. C. Alyea, Transition Met. Chem., 1979, 4, 205;
(b) E. C. Alyea, G. Ferguson, J. Malito and B. L. Ruhl, Can. J. Chem.,
1988, 66, 3162.
CH2Cl2 and analysed by 31P NMR and IR spectroscopy (see
Table 4).
X-Ray crystal structure analyses of 4a and 3d·3CH2Cl2
X-Ray diffraction experiments on trans-[PdCl2(2a)2] (4a) and
trans-[PtCl2(2d)2]·3CH2Cl2 (3d·3CH2Cl2) were carried out at
−100 ◦C on a Bruker SMART diffractometer using Mo-Ka
13 W. A. Herrmann, C. Brossmer, C. P. Reisinger, T. H. Riermeier, K.
¨
Ofele and M. Beller, Chem. Eur. J., 1997, 3, 1357.
14 (a) J. A. Osborn, F. H. Jardine, J. F. Young and G. Wilkinson, J. Chem.
Soc. A, 1966, 1711; (b) J. T. Mague and G. Wilkinson, J. Chem. Soc.
A, 1966, 1736.
˚
X-radiation, k = 0.71073 A. Crystal data and refinement data
are given in Table 5. Absorption corrections were based on
2
15 J. T. Carlock, Tetrahedron, 1984, 40, 185.
16 L. A. van der Veen, P. K. Keveen, P. C. J. Kamer and P. W. N. M. van
Leeuwen, Chem. Commun., 2000, 333.
17 C. A. Tolman, Coord. Chem. Rev., 1977, 77, 313.
18 V. P. W. Bo¨hm and W. A. Herrmann, Chem. Eur. J., 2001, 7, 4191.
19 (a) B. L. Shaw, New J. Chem., 1998, 77; (b) B. L. Shaw, S. D. Perera
and E. A. Staley, Chem. Commun., 1998, 1361.
20 (a) A. H. M. de Vries, J. M. C. A. Mulders, J. H. M. Mommers,
H. J. W. Henderickx and J. G. de Vries, Org. Lett., 2003, 5, 3285;
(b) M. R. Eberhard, Org. Lett., 2004, 6, 2125.
equivalent reflections and structures refined against all Fo
data with hydrogen atoms riding in calculated positions. In
3d·3CH2Cl2 the solvent is severely disordered and the model
used, although the best that could be obtained, is not completely
satisfactory given the large residual electron density features
close to the solvent molecules and the presence of apparent
3
˚
voids of volume ca. 30 A nearby.
CCDC reference numbers 256982 and 256983
See http://www.rsc.org/suppdata/dt/b4/b417910b/ for cry-
stallographic data in CIF or other electronic format.
21 A. Biffis, M. Zecca and M. Basato, Eur. J. Inorg. Chem., 2001, 1131.
22 A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen and
F. J. Timmers, Organometallics, 1996, 15, 1518.
23 (a) H. Konishi, K. Aritomi, T. Okano and J. Kiji, Bull. Chem. Soc.
Jpn., 1989, 62, 591; (b) N. J. Hales, H. Heeney, J. H. Holinshead and
S. V. Ley, Tetrahedron, 1995, 51, 7741.
24 D. Fraccarollo, R. Bertani, M. Mozzon, U. Belluco and R. A.
Michelin, Inorg. Chim. Acta., 1992, 201, 15.
Acknowledgements
We should like to thank Graham Henderson (Johnson-Matthey)
for help with the catalysis experiments, EPSRC for support,
Johnson Matthey for an Industrial CASE award (to M. W.) and
The Leverhulme Trust for a Research Fellowship (to P. G. P.).
25 J. R. Doyle, P. E. Slade and H. B. Jonassen, Inorg. Synth., 1960, 6,
218.
1 4 9 8
D a l t o n T r a n s . , 2 0 0 5 , 1 4 9 1 – 1 4 9 8