Structural characterisation of solution species implicated in the palladium-catalysed Heck reaction by Pd K-edge X-ray absorption spectroscopy: Palladium acetate as a catalyst precursor

Article abstract of DOI:10.1039/b200617k

Energy dispersive EXAFS (EDE), Quick EXAFS (QEXAFS), ¹³C NMR and X-ray crystallography have been used to probe the co-ordination sphere of palladium in the course of the phosphine free Heck reaction. [Pd₂I₆][NBu₃H]₂ has been isolated from the precatalytic solution and its crystal structure determined. EDE and QEXAFS spectra of the complexes Pd(OAc)₂, Pd(PPh₃)₄ and [Pd₂I₆][NEt₃H]₂ illustrated the value of the technique in structure elucidation. EXAFS of the precatalytic solution detects [Pd₂I₆]^2- and no co-ordinated carbon. EXAFS of the catalytic solution shows a first co-ordination sphere of 2 carbon atoms and a second of 2-2.5 iodines. A scheme involving an equilibrium between the oxidative addition product and the olefin co-ordination species, has been proposed to explain these results.

Full text of DOI:10.1039/b200617k

View Article Online / Journal Homepage / Table of Contents for this issue
Structural characterisation of solution species implicated in the
palladium-catalysed Heck reaction by Pd K-edge X-ray absorption
spectroscopy: palladium acetate as a catalyst precursor
John Evans,a Lynn O’Neill,a Vijaya L. Kambhampati,a Graham Rayner,a Sandra Turin,a
Anthony Genge,a Andrew J. Dentb and Thomas Neisiusc
a Department of Chemistry, University of Southampton, Southampton, UK SO17 1BJ
b CLRC Daresbury Laboratory, Warrington, UK, WA4 4AD
c ESRF, PO Box 220, Grenoble cedex, F-38043, France
Received 17th January 2002, Accepted 12th March 2002
First published as an Advance Article on the web 17th April 2002
Energy dispersive EXAFS (EDE), Quick EXAFS (QEXAFS), 13C NMR and X-ray crystallography have been used
to probe the co-ordination sphere of palladium in the course of the phosphine free Heck reaction. [Pd2I6][NBu3H]2
has been isolated from the precatalytic solution and its crystal structure determined. EDE and QEXAFS spectra
of the complexes Pd(OAc)2, Pd(PPh3)4 and [Pd2I6][NEt3H]2 illustrated the value of the technique in structure
elucidation. EXAFS of the precatalytic solution detects [Pd2I6]2Ϫ and no co-ordinated carbon. EXAFS of the
catalytic solution shows a first co-ordination sphere of 2 carbon atoms and a second of 2–2.5 iodines. A scheme
involving an equilibrium between the oxidative addition product and the olefin co-ordination species, has been
proposed to explain these results.
Palladium catalysed carbon–carbon bond formation is a well-
established tool in synthetic organic chemistry.1,2 As early as
1968 R. F. Heck demonstrated that palladium acetate or
mixtures of palladium acetate and triphenylphosphine could
catalyse the arylation or vinylation of olefins, in what has
become known as the Heck reaction, Scheme 1.3
Scheme 1 The Heck reaction.
Since the late 1980s there has been a revival of interest in the
Heck reaction as a synthetic technique, its attractiveness stem-
ming from its favourable stereo and chemoselectivity, high
yields and relatively mild reaction conditions. Sequential reac-
tions, synthesis of heterocycles and cyclic alkenes, natural
product and asymmetric synthesis are just a few examples of
areas where it has application.4
Until recently, reports concerning investigations of the
mechanistic pathway were scarce and the catalysis has been
assumed to proceed by an initial reduction to zero-valent
palladium, followed by oxidative addition of the aryl halide.5
Insertion of the olefin into the metal–aryl bond follows and the
product is generated by reductive elimination. The base
regenerates the active palladium species, Scheme 2.
Scheme 2 Mechanism of the Heck reaction.
Results and discussion
The reaction chosen for study involved the coupling of iodo-
benzene with 2-methylprop-2-en-1-ol to give 1-phenyl-2-methyl-
propanal, Scheme 3.1 The reaction conditions were chosen for
Mechanistic investigations have largely concentrated upon
the role of the phosphine within the process and the initial
oxidative addition step.6,7 To our knowledge, there has been no
similar investigation of the phosphine free catalytic cycle. We
have previously shown that extended X-ray absorption fine
structure spectroscopy (EXAFS) and energy dispersive EXAFS
(EDE) can be used to probe directly the co-ordination sphere of
the metal species within homogeneous catalysis.8 A K-edge Pd
EXAFS study of the phosphine-free Heck reaction was thus
initiated, to investigate further its mechanism.
Scheme 3 The Heck reaction under investigation.
the suitability of reaction conditions to study by time resolved
XAFS spectroscopy. The palladium concentration was 50 mM
and catalyst to reagent ratio was 1 : 20. These concentrations
DOI: 10.1039/b200617k
J. Chem. Soc., Dalton Trans., 2002, 2207–2212
This journal is © The Royal Society of Chemistry 2002
2207

View Article Online
Fig. 1 View of the structure of [Pd2I6][NBu3H]2 with numbering scheme adopted. Ellipsoids are shown at 40% probability and atoms marked with
an asterisk are related by a crystallographic inversion centre at (0.5, 1, 1).
Table 1 Crystallographic data for [Pd2I6][NBu3H]2
Table 2 Selected bond lengths and angles for [Pd2I6][NBu3H]2. Atoms
marked with an asterisk are at equivalent position (1 Ϫ x, 2 Ϫ y, 2 Ϫ z)
Formula
M
Colour, morphology
Crystal dimensions/mm
Crystal system
Space group
a/Å
b/Å
C12H56I6Pd2N
1346.95
Pd(1)–I(1)
Pd(1)–I(2)
Pd(1)–I(3)*
Pd(1)–I(3)
2.611(2)
2.5931(8)
2.604(2)
2.6092(8)
I(1)–Pd(1)–I(2)
I(1)–Pd(1)–I(3)
I(1)–Pd(1)–I(3)*
I(2)–Pd(1)–I(3)
I(2)–Pd(1)–I(3)*
Pd(1)–I(3)–Pd(1)*
I(3)–Pd(1)–I(3)*
92.10(4)
91.50(4)
176.57(3)
176.23(3)
91.30(4)
94.89(4)
85.11(4)
Dark red, plate
0.45 × 0.13 × 0.05
Triclinic
¯
P1 (#2)
11.779(8)
11.787(8)
8.690(5)
110.98(5)
100.57(6)
61.67(4)
991(1)
1
c/Å
α/Њ
β/Њ
Table 2. The structure consists of an association of two tributyl
ammonium units with a centrosymmetric di-anionic Pd2I6 unit.
The palladium atoms are bridged by two iodine atoms.
γ/Њ
U/Å3
Z
Bond lengths and angles are within expected values when
compared to related complexes.9 The Pd–I bond distance for
terminal iodides (2.611(2) and 2.593(8) Å) and bridging
iodides (2.609(8) Å) are similar indicating no lengthening of
this bond through bridging. As expected the I(3)–Pd–I(3)*
angle (85.11(4)Њ) is smaller than the I(1)–Pd–I(2) angle
(92.10(4)Њ), a result of the constraint imposed by the bridging
group. Structures of this anionic complex with other cations
have been reported,10 with the nearest analogue being for the
NBu4ϩ counter ion; little substantial change in the structure of
[Pd2I6]2Ϫ is evident.
The above solution, with triethylamine, was heated to 75 ЊC,
resulting in a colour change from deep red to brown–yellow.
The catalysis proceeded over one hour at this temperature. All
attempts to isolate the active catalytic species resulted in a
conversion of the palladium complex to [Pd2I6][NEt3H]2.
[Pd2I6]2Ϫ is a well known counter-ion 10–12 and is clearly the most
thermodynamically favourable product under these reaction
conditions. Upon completion of the catalysis the solution
reverts to a deep red colour from which [Pd2I6][NEt3H]2 was
isolated as the only palladium-containing species.
F(000)
634.00
2.256
5.593
0.7914–1.0000
3479
2320
154
1.26
0.030
0.038
Dc/g cmϪ3
µ(Mo-Kα)/mmϪ1
Transmission factors (max. and min.)
No. of unique obs. reflections
Unique obs. reflections with [Io > 2.5σ(Io)]
No. of parameters
Goodness of fit
R(Fo)
Rw(Fo)
Max. residual peak/e ÅϪ3
Max. residual trough/e ÅϪ3
1.16
Ϫ0.69
R = Σ (|Fobs|i Ϫ |Fcalc|i)/Σ |Fobs|i. Rw = √[Σ wi(|Fobs|i Ϫ |Fcalc|i)2/Σ wi|Fobs|i2].
GOF = [Σ (|Fobs|i Ϫ |Fcalc|i)/σi]/(n Ϫ m) ≈ 1.
initially posed a problem in terms of catalyst precipitation but
this problem was overcome by choice of NMP as reaction
solvent.
Addition of 2-methylprop-2-en-1-ol to a yellow solution of
palladium acetate, iodobenzene, NMP (N-methyl-2-pyrrol-
idinone) and triethylamine or tributylamine produced an
immediate deep red coloured solution. In the absence of the
2-methylprop-2-en-1-ol, this reaction does not proceed. The
volatile reagents were removed and the remaining red solid
shown to be [Pd2I6][NEt3H]2 or [Pd2I6][NBu3H]2, isolated in
high yield.
Energy dispersive EXAFS (EDE) and QEXAFS studies of
standard complexes
In order to illustrate the value of EXAFS and EDE towards
elucidating the structure of catalytic intermediates, in solution
in donor solvents, the Pd K-edge QEXAFS and EDE spectra
of the standard complexes, Pd(OAc)2, Pd(PPh3)4 and [Pd2I6]-
[NEt3H]2 were obtained. QEXAFS data of palladium acetate
Crystals of [Pd2I6][NBu3H]2 were grown and the crystal struc-
ture determined. The structure is shown in Fig. 1. Crystal data
are given in Table 1 and selected bond lengths and angles in
2208
J. Chem. Soc., Dalton Trans., 2002, 2207–2212

View Article Online
Table 3 Palladium K-edge structural parameters for standard solutions: (a) palladium acetate, NMP, 70mM, QEXAFS, 6 × 5 minutes (averaged);
(b) Pd(PPh3)4, THF, 60 mM, EDE (100 × 2.2 ms); (c) [Pd2I6][NEt3H]2, CH3CN, 70 mM, EDE (100 × 6.6 ms)
c
Sample
Shell
CNa
r/Å
2σ2 b/Å
Af
Ef d (eV)
R(%)
(a)
(b)
(c)
O
P
I
4.37(7)
2.79(20)
3.99(8)
1.998(2)
2.282(8)
2.678(3)
0.0038(3)
0.006(1)
0.006(1)
0.8
1.0
1.0
Ϫ1.1(1)
Ϫ2.4(2.0)
Ϫ7.9(7)
24.6
35.4
13.5
a Co-ordination number. b Debye–Waller factor; σ = root mean-square separation. c Fraction of absorption causing EXAFS d Ef variation in Fermi
level from theory.
in NMP (70 mM) were acquired in a scan time of 30 minutes
(6 × 5 minutes) and the average obtained. Table 3 gives the
structural parameters.
The data are consistent with a first co-ordination sphere of
four oxygen atoms at 1.998 Å. This is in good agreement with
the crystal structure of palladium acetate where each palladium
is surrounded by four bridging acetate ligands with Pd–O dis-
tances in the range of 1.973–2.014 Å.13 However in the crystal
structure palladium acetate exists as a trimer with Pd–Pd non-
bonded distances in the region of 3.105–3.203 Å. A second shell
of two palladium atoms at 3.15 Å was not observed suggesting
that the trimer may be broken down in NMP to a monomer. In
the case of the linear trimer [Ni(acac)2]3 in toluene solution, a
shell could be attributed to an Ni ؒ ؒ ؒ Ni distance,14 so it is
plausible that a shell should be observable for the Pd ؒ ؒ ؒ Pd
atomic pairing in a cyclic trimer.
The EDE spectra of Pd(PPh3)4 in THF solution, 60 mM,
were obtained using 100 scans of 2.2 ms integration time. Fig. 2
Fig. 3 The palladium K-edge k3-weighted EDE data and Fourier
transform, (——, experimental; -------, spherical wave theory), of
[Pd2I6][NEt3H]2 in NMP, 70 mM. 100 scans, integration time: 6 ms.
EDE data and the Fourier transform for the solution and struc-
tural parameters are given in Table 3. Results show a first co-
ordination sphere of four iodines at distances of 2.676 Å from
palladium. Compared with the crystal structure of [Pd2I6]-
[NBu3H]2 the Pd–I distances from the EDE results are some-
what longer than those obtained from crystallography, pos-
sibly due to difficulties in calibration for this particular
spectrum. There is excellent agreement between theoretical and
experimental for the data and the R factor is low, 13.5%, for
time-resolved data.
EDE and QEXAFS of the Heck reaction (Scheme 3)
The previous results indicate the degree of agreement that can
be obtained when investigating the structures of complexes in
solution using QEXAFS and EDE. The same techniques can be
applied to elucidating the structures under catalytic conditions.
The steps from the addition of the precursor, through the onset
of catalysis to its conclusion may be monitored by increasing
the temperature of the charged solution and then holding the
temperature through to the completion of the catalysed reac-
tion. It must, however, be noted that EXAFS data may be a
result of the average of a number of species in solution. Data
may be of poorer quality due to lower metal concentrations or
possible inhomogeneity in the mixture. Despite this, valuable
information can be obtained by the application of this tech-
nique to in situ catalytic reactions.8 Of particular value is the
energy dispersive EXAFS (EDE) technique, where the spectra
Fig. 2 The palladium K-edge k3-weighted EDE data and Fourier
transform, (——, experimental; -------, spherical wave theory), of
Pd(PPh3)4 in THF, 60mM. 100 scans, integration time: 2.2 ms.
shows EDE data and Fourier transform for the solution and
Table 3 gives structural parameters. Results indicate a first co-
ordination shell of three phosphorus atoms at 2.28 Å. This is in
excellent agreement with experimental studies of this solution
where it has been well established that Pd(PPh3)4 exists in solu-
tion primarily as Pd(PPh3)3 with a small amount of Pd(PPh3)2
present.15
EDE spectra of [Pd2I6][NEt3H]2 in acetonitrile, 70 mM, were
obtained with 100 scans 6 ms integration times. Fig. 3 shows
J. Chem. Soc., Dalton Trans., 2002, 2207–2212
2209

View Article Online
Table 4 Palladium K-edge structural parameters for precatalytic solution at room temperature; (a) QEXAFS (4 × 10 minutes, averaged); (b) EDE
(100 scans, integration time 6.5 ms)
c
d
Method
Shell
CNa
r/Å
2σ2 b/Å2
Af
Ef /eV
R (%)
(a)
(b)
I
I
3.2(2)
3.2(16)
2.62(1)
2.63(1)
0.017(0)
0.0105(5)
1.0
1.0
Ϫ5.1(10)
Ϫ5.9(12)
39.3
27.7
a Co-ordination number. b Debye–Waller factor; σ = root mean-square separation. c Fraction of absorption causing EXAFS. d Ef variation in Fermi
level from theory.
in this report were all obtained with an elapsed time of under
1 minute. This allows a snapshot of the solution to be taken at
any point of the catalysis. Quick EXAFS (QEXAFS) required a
five minute scan for useful data. Typically six scans of 5 minute
duration were averaged. For such spectra to be valid, there
should be no change in the predominant palladium species over
this period. Both EDE and QEXAFS experiments were utilised
to study the coupling reaction described in Scheme 3.
As noted above, when all the reagents {Pd(OAc)2/PhI/
2-methylprop-2-en-1-ol/NEt3 in NMP} are charged together a
deep red colour is produced from which [Pd2I6][NEt3H]2 was
isolated. Investigations of the precatalytic solution were
required to see if this complex could be observed in situ. Its
production may have simply been a result of the removal of
volatiles. At this stage oxidative addition of iodobenzene to
zerovalent palladium may have been expected to take place. At
the end of the catalysis, when all the NEt3 has reacted, the
NMR resonances of [Pd2I6][NEt3H]2 are clearly observed.
NMR data have not provided conclusive evidence of the struc-
ture of the palladium species in the precatalytic solution.
QEXAFS spectra of the solution were acquired over a scan
time of 40 minutes (4 × 10 minutes) and the average obtained.
Typically EDE spectra were obtained by 100 scans and an inte-
gration time of 6.5 ms. Fig. 4 shows EDE data and Fourier
iodines at 2.62 Å. In both experiments a shoulder on this shell is
observed at 2 Å. Attempts to fit this shell to carbon, oxygen and
even phosphorus resulted in a deterioration of the fit; the
Debye–Waller factor for the carbon shell was in the region of
0.058 Å2. So although this shell may be indicating the presence
of a lighter element, there is no definitive information about its
nature. It can be concluded that in this red solution oxidative
addition of the iodobenzene has probably occurred and the
palladium species at this point may be an average of bridged
and solvated structures, Scheme 4.
Scheme 4 Proposed intermediates for precatalytic solution.
When the red precatalytic solution is heated to 75 ЊC there is
a colour change to brown–yellow and catalysis proceeds over an
hour. In situ 13C NMR of this did not allow observation of the
co-ordinated carbon. EDE and QEXAFS experiments were
utilised for evidence of the palladium species present. QEXAFS
spectra were obtained over 50 minutes (10 × 5 minutes) and the
average obtained. EDE spectra were obtained by 100 scans and
11 ms integration time. Fig. 5 shows QEXAFS data and Fourier
transform for the solution and Table 5 gives structural param-
eters for spectra obtained by EDE and QEXAFS. Both experi-
ments agree well and show a first co-ordination sphere of
approximately two carbons and second co-ordination sphere
of ∼ 2 iodines. Possible structures to explain the EXAFS data
are given in Scheme 5. This shows an equilibrium between a
Scheme 5 Proposed catalytic intermediates for catalytic solution.
dimeric oxidative addition product and the olefin co-ordination
complex.
There is no direct evidence to suggest that the oxidative
addition product is dimeric. Inclusion of a palladium shell in
the EDE or QEXAFS, which would indicate the presence of
a Pd ؒ ؒ ؒ Pd shell, neither improves nor worsens the fit. How-
ever, palladium complexes do have a tendency to bridge. It is
expected that the low frequency bending vibrations of the
Pd2(µ-I)2 unit will provide a high dynamic Debye–Waller factor
rendering the Pd ؒ ؒ ؒ Pd shell difficult to detect at elevated tem-
peratures. The opening of this bridge leaves a free co-ordination
site for co-ordination by the olefin before insertion into the
Pd–C σ bond, Scheme 5.
Fig. 4 The palladium K-edge k3-weighted EDE data and Fourier
transform, (——, experimental; -------, spherical wave theory), of
precatalytic solution at room temperature. 100 scans, integration time:
6.5 ms.
transform for the solution and Table 4 gives structural param-
eters for spectra obtained by EDE and QEXAFS. Both experi-
ments agree well and show a first co-ordination sphere of 3.2
2210
J. Chem. Soc., Dalton Trans., 2002, 2207–2212

View Article Online
Table 5 Palladium K-edge structural parameters for catalytic solution at 75 ЊC; (a) QEXAFS (10 × 5 minutes, averaged); (b) EDE (100 scans,
integration time 11 ms)
CNa
r/Å
2σ2 b/Å2
Af
Ef /eV
R (%)
c
d
Method
Shell
(a)
(b)
C
I
C
I
1.5(3)
2.3(3)
1.6(6)
1.8(3)
2.06(3)
2.63(1)
2.13(4)
2.61(1)
0.028(3)
0.028(1)
0.012(8)
0.014(2)
1.0
1.0
Ϫ5.0(12)
Ϫ3.6(26)
36.2
53
a Co-ordination number. b Debye–Waller factor; σ = root mean-square separation. c Fraction of absorption causing EXAFS. d Ef variation in Fermi
level from theory.
identifed by its absorption band at 1721 cmϪ1. Proton and 13C
NMR spectra were recorded on a Bruker AC 300 or AM 360
spectrometer.
EXAFS studies
QEXAFS spectra were recorded at Station 9.3 of the Synchro-
tron Radiation Source at the Daresbury Laboratory, UK, oper-
ating at 2 GeV (ca. 3.20 × 10Ϫ10 J) and an average current of
140 mA. Data were acquired in fluorescence mode using a
Canberra 13-element germanium solid state detector. EDE
spectra (transmission) were acquired at ID24 of the ESRF,16
Grenoble, operating at 6 GeV and an average current of 190
mA. Data were acquired in energy dispersive mode using an
asymmetric cut (3Њ) Si(111) Laue monochromator. Energy cali-
bration was carried out using a spectrum of a palladium foil.
A masked, Peltier cooled Princeton CCD camera was used to
collect the EDE data; 18 stripes of 64 pixel width, 10 of which
were summed to form a single spectrum, were collected before
readout. The data collection rate was limited by the readout
time of the CCD chip (ca. 250 ms). So, 100 scans of 11 ms
duration (as used for an in situ experiment) would involve
ca. 11 × 18 × 100 ms, of which 11 × 10 × 100 ms (11 s) exposure
would be processed. An additional 100 × 0.3 s was required for
reading out the data, giving an elapsed time of ca. 48 s. Follow-
ing this the sample was moved out of the line of the beam so
that the air path was used to acquire the background (Io) spec-
trum. Background subtracted EXAFS data were obtained
using the program PAXAS.17 Removal of the pre-edge back-
ground was achieved using a polynomial of order 2 and the
post-edge background was subtracted using a polynomial of
order 6, 7 or 8. Curve fitting analyses, by least squares refine-
ment of the non-Fourier filtered k3-weighted EXAFS data were
carried out within EXCURV-92 and -98,18 using spherical wave
methods with ab initio phase shifts and back scattering factors
calculated in the usual manner from relativistic HF–SCF
derived atomic-charge densities. The numbers of independent
parameters used in the fits are within the guideline
Fig. 5 The palladium K-edge k3-weighted QEXAFS data and Fourier
transform, (——, experimental; -------, spherical wave theory), of
catalytic solution at 75 ЊC, averaged over 50 minutes.
Conclusion
QEXAFS and EDE results of the precatalytic solution, at room
temperature, support the presence of [Pd2I6][NEt3H]2. There is
only weak evidence of the oxidative addition steps, or co-
ordination by carbon, at this point. Upon heating to 75 ЊC the
catalysis begins and the species detected by QEXAFS and EDE
is a co-ordination site of 2 carbon atoms and 2 iodine atoms.
This is consistent with an equilibrium of a series of structures
such as the oxidative addition product and the olefin co-
ordinated complex, Scheme 5. This is the step prior to the inser-
tion of the olefin into the Pd–C σ bond in the conventional
catalytic cycle.
Npts = 2(kmax Ϫ kmin)(Rmax Ϫ Rmin)/π.
The R factors are defined as (f|χT Ϫ χE|k3dk/f|χE|k3dk) ×
100%.
Standard deviations are quoted as derived for curve fitting
procedures; expectation errors on interatomic distances are
between 1.0 and 1.5%.19 Errors on co-ordination numbers are
ca. 15%.
Experimental
Solutions for study were prepared in a three neck round
bottomed flask and transferred by canula to a d = 0.5 cm (EDE)
or d = 1 cm (fluorescence QEXAFS) stainless steel solution
cell, with Kapton windows and heater cartridges attached.
All manipulations were carried out under a nitrogen or argon
atmosphere using standard Schlenk techniques. Anhydrous
NMP, palladium acetate and 2-methylprop-2-en-1-ol were
obtained from Aldrich chemicals and used directly. Iodo-
benzene and triethylamine were dried over calcium hydride and
distilled prior to use. The progress of the catalysed reactions
were monitored by extracting samples and using IR spectro-
scopy; these spectra were recorded on a Perkin Elmer 1600
FT-IR spectrometer, in solution using a 1.0 mm NaCl cell.
The loss of iodobenzene was monitored using an absorption at
1572 cmϪ1 with the formation of 1-phenyl-2-methylprop-3-one
[Pd2I6][NEt3H]2. The compound (0.38 g) was dissolved in
CH3CN (5 cm3) and transferred directly to a 1 cm solution cell.
Precatalytic species. 2-Methylprop-2-en-1-ol (1 g) was added
to palladium acetate (0.1 g), iodobenzene (2.4 g), triethyl-
amine (1.4 g) and NMP (4 cm3). The mixture was stirred for
J. Chem. Soc., Dalton Trans., 2002, 2207–2212
2211

View Article Online
10 minutes and then transferred to the solution cell. QEXAFS
or EDE data were recorded over a period of 30 minutes.
See http://www.rsc.org/suppdata/dt/b2/b200617k/ for crystal-
lographic data in CIF or other electronic format.
Catalytic species. The above reaction mixture was heated to
50 ЊC and transferred to a preheated solution cell, 50 ЊC. The
solution was heated to 75 ЊC and spectra recorded at this
temperature over one hour. IR spectroscopy indicated that the
reaction was substantially complete after that time.
Acknowledgements
We wish to thank the University of Southampton, EPSRC
and BP Chemicals for financial support and the Directors and
staff of the Daresbury Laboratory and ESRF for access to the
facilities and for their assistance.
In situ 13C NMR studies
All spectra were recorded on a Bruker AM 360 spectrometer.
The samples were run unlocked and referenced against a stand-
ard sample of NMP.
References
1 R. F. Heck, Acc. Chem. Res., 1979, 12, 146.
2 R. F. Heck, Palladium Reagents in Organic Synthesis, Academic
Press, London, 1985; J. Tsuji, Organic Synthesis with Palladium
Compounds, Springer, Berlin, 1980.
Precatalytic species. The solution used for EXAFS studies
was prepared and Cr(acac)3 (10 mg cmϪ3) added. The mixture
was transferred to a NMR tube under an inert atmosphere. The
solution was cooled to Ϫ10 ЊC and 13C spectrum accumulated
over 2 hours at this temperature.
3 R. F. Heck, Acc. Chem. Res., 1979, 12, 146.
4 A. de Mejere and F. E. Meyer, Angew. Chem., Int. Ed. Engl., 1994,
33, 2379 and references therein.
5 W. Cabri and I. Candiani, Acc. Chem. Res., 1995, 28, 2.
6 C. Amatore, E. Carre, A. Jutand, M. A. M’Barki and G. Meyer,
Organometallics, 1995, 14, 5605; C. Amatore, E. Carre, A. Jutand
and M. A. M’Barki, Organometallics, 1995, 14, 1818; C. Amatore,
A. Jutand and M. A. M’Barki, Organometallics, 1992, 11, 3009.
7 T. Mandai, T. Matsumoto and J. Tsuji, Tetrahedron Lett., 1993, 34,
2513; F. Ozawa, A. Kubo and T. Hayashi, Chem. Lett., 1992, 2177.
8 D. Bogg, M. Conyngham, J. M. Corker, A. J. Dent, J. Evans,
R. C. Farrow, V. L. Kambhampati, A. F. Masters, D. N. Macleod,
C. M. Ramsdale and G. Salvini, J. Chem. Soc., Chem. Commun.,
1996, 647; P. Andrews and J. Evans, J. Chem. Soc., Chem. Commun.,
1993, 1246; J. M. Corker and J. Evans, J. Chem. Soc., Chem.
Commun., 1991, 1104; P. Andrews, J. M. Corker, J. Evans and
M. Webster, J. Chem. Soc., Dalton Trans., 1994, 1337.
9 N. W. Alcock, W. L. Wilson and J. H. Nelson, Inorg. Chem., 1993,
32, 3193.
10 F. Maassarani, M. Pfeffer and G. Le Borgne, J. Chem. Soc., Chem.
Commun., 1987, 565; S. Chan, S.-M. Lee, Z. Lin and W.-T. Wong,
J. Organomet. Chem., 1996, 510, 219; F. Neve, A. Crispini and
O. Franscangelli, Inorg. Chem., 2000, 39, 1187.
11 R. Uson, J. Forries, M. A. Uson, J. F. Yaguee, P. G. Jones and
K. Meyer-Baese, J. Chem. Soc., Dalton Trans., 1986, 947.
12 J. L. Graff and M. G. Romanelli, J. Chem. Soc., Chem. Commun,
1987, 337.
Catalytic species. The above solution was prepared and
heated to 65 ЊC in a NMR probe. The 13C spectrum was
accumulated over two hours at this temperature with a printout
obtained every 30 min.
Preparative studies
Isolation of [Pd2I6][NEt3H]2 and [Pd2I6][NBu3H]2. 2-Methyl-
prop-2-en-1-ol (1 g) was added to palladium acetate (0.1 g),
iodobenzene (2.4 g), and NMP (4 cm3) and triethylamine (1.4 g)
or tributylamine (2.6 g). The mixture was stirred for 30 minutes
at room temperature and volatile reagents removed in vacuo.
The resulting red solid was washed with diethyl ether to yield
[Pd2I6][NEt3H]2 (0.17 g, 65%) or [Pd2I6][NBu3H]2 (0.21 g, 70%).
[Pd2I6][NEt3H]2. Elemental analysis: C: 11.81 (12.22); H: 2.32
1
(2.71); N: 23.45 (23.77)%. H NMR (CDCl3)(300 MHz): δ 1.4
(t, 3H), 3.1 (q, 2H). 13C NMR (CDCl3) (75 MHz): δ 11.63,
43.27.
13 A. C. Skapski and M. L. Smart, Chem. Commun, 1970, 658.
14 J. M. Corker and J. Evans, J. Chem. Soc., Chem. Commun, 1991,
1104; J. M. Corker, PhD Thesis, University of Southampton, 1990.
15 J. F. Fauvarque, F. Pfluger and M. Troupel, J. Organomet. Chem,
1981, 208, 419; C. Amatore and F. Pfluger, Organometallics, 1990, 9,
2276; C. Amatore, A. Jutand, F. Khalil and L. Mottier,
Organometallics, 1993, 12, 3168.
[Pd2I6][NBu3H]2. Elemental analysis: C: 21.22 (21.40); H:
1
3.88 (4.16); N: 1.92 (2.08)%. H NMR (CDCl3)(300 MHz):
δ 0.72 (t, 3H), 1.015–1.45(m, 6H). 13C NMR (CDCl3) (75
MHz): δ 13.94, 16.01, 28.22, 32.04.
16 M. Hagelstein, A. San Miguel, T. Ressler, A. Fontaine and
J. Goulon, J. de Physique IV, 1997, 1, C2–303.
Crystallographic studies
17 N. Binsted, PAXAS Program for the analysis of X-ray absorption
spectra, University of Southampton, 1988.
18 EXCURV98, CCLRC Daresbury Laboratory Computer Program,
Cheshire, 1998.
19 J. M. Corker, J. Evans, H. Leach and W. Levason, J. Chem. Soc.,
Chem. Commun., 1989, 181.
20 P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman,
S. Garcia-Granda, R. O. Gould, J. M. M. Smits, C. Smyskalla,
PATTY, The DIRDIF Program System, Technical Report of
the Crystallography Laboratory, University of Nijmegen, The
Netherlands, 1992.
Dark red crystals of [Pd2I6][NBu3H]2 were grown from a con-
centrated 2-methylprop-2-en-1-ol solution. Data collection
used a Rigaku AFC7S four circle diffractometer, with graphite-
monochromated Mo-Kα X radiation (λmax = 0.7103 Å),
T = 150 K, ω–2θ scans. The structure was solved by direct
methods using PATTY,20 and then developed by iterative cycles
of full least squares refinement and difference Fourier syntheses
which located all non-H atoms in the asymmetric unit.21 All
non-H atoms were refined anisotropically and H atoms were
placed in fixed calculated positions with d (C–H) = 0.96 Å.
CCDC reference number 168848.
21 TeXsan: Crystal Structure Analysis Package, Molecular Structure
Corporation, Houston, TX, 1992.
2212
J. Chem. Soc., Dalton Trans., 2002, 2207–2212