New carboalkoxybis(triphenylphosphine)palladium(II) cationic complexes: Synthesis, characterization, reactivity and role in the catalytic hydrocarboalkoxylation of ethene. X-ray structure of trans-[Pd(COOMe)(TsO)(PPh3)2]·2CHCl3

Article abstract of DOI:10.1016/j.molcata.2008.10.002

The cationic complexes trans-[Pd(COOR)(H₂O)(PPh₃)₂](TsO) have been synthesised by reacting cis-[Pd(H₂O)₂(PPh₃)₂](TsO)₂·2H₂O with CO in ROH (R = Me and Et), practically under room conditions, or by methathetical exchange of trans-[Pd(COOMe)Cl(PPh₃)₂] with Ag(TsO) (R = n-Pr, iso-Pr, n-Bu, iso-Bu, sec-Bu). They have been characterised by IR, ¹H NMR and ³¹P NMR spectroscopies. The X-ray investigation of trans-[Pd(COOMe)(TsO)(PPh₃)₂] reveals that the palladium center is surrounded in a virtually square planar environment realized by two PPh₃ trans to each other, the carbon atom of the carbomethoxy ligand and an oxygen atom of the p-toluensulfonate anion, with two crystallization molecules of CHCl₃. The Pd-O-S angle, 151.9 (3)°, is very wide, probably due to the interaction of one CHCl₃ molecule with the complex inner core. The carbomethoxy derivatives react with R′OH yielding the corresponding R′ carboalkoxy derivative (R′ = Et, n-Pr and iso-Pr); ethene does not insert into the Pd-COOMe bond; decarbomethoxylation occurs when treated with TsOH/H₂O in MeOH at 50 °C. All the carboalkoxy are precursors for the catalytic carboalkoxylation of ethene if used in combination of PPh₃ and TsOH, better in the presence of some water. Experimental evidences are more in favor of the so-called "hydride" mechanism rather than the "carbomethoxy" mechanism.

Full text of DOI:10.1016/j.molcata.2008.10.002

Journal of Molecular Catalysis A: Chemical 298 (2009) 103–110
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
New carboalkoxybis(triphenylphosphine)palladium(II) cationic complexes:
Synthesis, characterization, reactivity and role in the catalytic
hydrocarboalkoxylation of ethene. X-ray structure of
trans-[Pd(COOMe)(TsO)(PPh3)2]·2CHCl3
E. Amadioa, G. Cavinatob, A. Dolmellac, L. Ronchina, L. Tonioloa,∗, A. Vavasoria
a Department of Chemistry, University of Venice, Dorsoduro 2137, 30123 Venice, Italy
b Department of Chemistry Sciences, University of Padua, via Marzolo 1, 35100 Padua, Italy
c Department of Pharmaceutical Sciences, University of Padua, via Marzolo 5, 35131 Padua, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 July 2008
Received in revised form 1 October 2008
Accepted 1 October 2008
Available online 14 October 2008
The cationic complexes trans-[Pd(COOR)(H2O)(PPh3)2](TsO) have been synthesised by reacting cis-
[Pd(H2O)2(PPh3)2](TsO)2·2H2O with CO in ROH (R = Me and Et), practically under room conditions, or
by methathetical exchange of trans-[Pd(COOMe)Cl(PPh3)2] with Ag(TsO) (R = n-Pr, iso-Pr, n-Bu, iso-Bu,
sec-Bu). They have been characterised by IR, 1H NMR and 31 P NMR spectroscopies.
The X-ray investigation of trans-[Pd(COOMe)(TsO)(PPh3)2] reveals that the palladium center is sur-
rounded in a virtually square planar environment realized by two PPh3 trans to each other, the carbon
atom of the carbomethoxy ligand and an oxygen atom of the p-toluensulfonate anion, with two crystal-
lization molecules of CHCl3. The Pd–O–S angle, 151.9 (3)◦, is very wide, probably due to the interaction of
one CHCl3 molecule with the complex inner core. The carbomethoxy derivatives react with RꢀOH yield-
ing the corresponding Rꢀ carboalkoxy derivative (Rꢀ = Et, n-Pr and iso-Pr); ethene does not insert into the
Pd–COOMe bond; decarbomethoxylation occurs when treated with TsOH/H2O in MeOH at 50 ◦C.
All the carboalkoxy are precursors for the catalytic carboalkoxylation of ethene if used in combination
of PPh3 and TsOH, better in the presence of some water. Experimental evidences are more in favor of the
so-called “hydride” mechanism rather than the “carbomethoxy” mechanism.
Keywords:
Palladium catalyst
Carbon monoxide
Alkanol
Ethene
Hydrocarboalkoxylation
© 2008 Elsevier B.V. All rights reserved.
1. Introduction
plexes of the type trans-[Pd(COOR)X(PPh3)2] (X = Cl and CN) with R
bulkier than Me (up to cyclohexyl) has been also reported [19–24].
Pd(II)-carboalkoxy complexes are involved as key intermediates
in several catalytic carbonylation reactions, such for example the
oxidative carbonylation of alkanols to carbonates and oxalates, or
of olefins and alkyns to diesters or unsaturated esters [1–3], the
double carbonylation of organic halides [4,5], the carbonylation of
olefins to polyketones [6,7]. They have been proposed as key inter-
mediates also in the hydroesterification of an olefin to monoesters
[8–11].
Methods of synthesis are based on the reaction of Pd(II) com-
plexes with CO and an alkanol, eventually in the presence of a
tertiary amine, or an alkoxide [12–17] or on the oxidative addition
of chloro or cyano formate or phenylcarbonate to Pd(0) complexes
[18–22]. Most of the syntheses reported up to now are relevant
to carbomethoxy derivatives. The synthesis of neutral Pd(II) com-
The cationic carbomethoxy complex trans-[Pd(COOMe)(H2O)
(PPh3)2](TsO) has been synthesised from the neutral trans-
[Pd(COOMe)Cl(PPh3)2] by methathetical exchange with Ag(TsO).
Its role in the catalytic hydrocarbomethylation of ethene has been
investigated [25].
In the present paper we extend the study to the synthesis
and characterization of the new cationic carboalkoxy complexes
trans-Pd(COOR)(H2O)(PPh3)2](TsO), with R = Et, n-Pr, iso-Pr, n-Bu,
iso-Bu, sec-Bu, together with further studies on the reactivity of the
methyl analogue in relation to the catalytic hydrocarbomethyla-
tion of ethene. We also reported the X-ray diffraction structure of
trans-[Pd(COOMe)(TsO)(PPh3)2]·2CHCl3.
2. Experimental
2.1. Materials
∗
Carbon monoxide and ethene (purity higher then 99%) were
supplied by SIAD Spa (Italy). Methanol was purchased from Baker
Corresponding author. Tel.: +39 041 2348553; fax: +39 041 2348517.
E-mail address: toniolo@unive.it (L. Toniolo).
1381-1169/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2008.10.002

104
E. Amadio et al. / Journal of Molecular Catalysis A: Chemical 298 (2009) 103–110
Table 1
Crystal and refinement data.
(purity > 99.5%, 0.01% of water) and Pd(OAc)2, Ag(TsO), NEt3, PPh3
and p-toluenesulfonic acid were purchased from Aldrich Chem-
icals. NEt3 and the solvents were commercial grade and used
without further purification. [Pd(COOR)Cl(PPh3)2], [PdCl2(PPh3)2],
[Pd(H2O)2(PPh3)2](TsO)2] and [Pd(TsO)2(PPh3)2] complexes were
prepared according to the methods reported in the literature
[24–27].
Complex
Formula
Molecular wt
Color
Crystal system
Space group
a (Å)
C47 H42O5P2SCl6Pd
1099.9
colorless
monoclinic
P21/n (No. 14)
19.558(3)
12.155(2)
21.453(4)
103.87(1)
4951(1)
4
1.476
2232
0.20 × 0.20 × 0.20
3.5/29.6
11938
3892
560
0.058
0.110
0.824
0.763
b (Å)
2.2. General procedure
c (Å)
ˇ (◦)
The IR spectra were recorded in nujol mull on a Nicolet FTIR
instruments mod. Nexus. 1H and 31P NMR spectra were recorded
on a Bruker AMX 300 spectrometer equipped with a BB multinu-
V (Å3)
Z
Dcalc (g cm−3
F(0 0 0)
)
clear probe operating in the FT mode at 300 and 121.5 MHz for 1
H
Crystal dimens (mm)
and 31P, respectively. All the samples were dissolved in deuterated
methanol used also as internal reference for the assignment of the
chemical shifts.
ꢀ limits (◦)
No. of independent data
No. of data with I > 2ꢁ (I)
No. of variables
R(F)a
wR(F2
)
b
2.3. Preparation of trans-[Pd(COOR)(H2O)(PPh3)2](TsO) (R = Me
and Et)
Largest peak in (F (e Å−3
)
GOFc
ꢀ
ꢀ
ꢁꢁ ꢁ ꢁ ꢁꢁ
ꢁ ꢁ
a
R(F) =
Fo
−
Fc
/
Fo
0.1 mmol of [Pd(H2O)2(PPh3)2](TsO)2 was dissolved in 2 ml of
MeOH or EtOH, previously saturated with CO at r.t. The solution
was kept under 2 atm of CO for 5–10 min and then poured into
20 ml of cold water under vigorous stirring. A white precipitated
formed immediately. The suspension was filtered, the solid was
washed with cold water, n-pentane, and dried under vacuum (yield
77%).
ꢂ
ꢅ
1/2
ꢃ
ꢄ
ꢃ
ꢄ
ꢀ
ꢀ
2
2
ꢁ ꢁ ꢁ ꢁ
ꢁ ꢁ
2
2
2
b
wR(F2) =
w
Fo
−
Fc
/
w
Fo
ꢂ
ꢅ
1/2
ꢃ
ꢄ
ꢀ
2
ꢁ ꢁ ꢁ ꢁ
2
2
c
GOF =
w
Fo
−
Fc
/(n − p)
2.6. Hydroesterification of ethene using the carboalkoxy
complexes reported in Table 1 as catalyst precursors in the
relevant ROH
2.4. Preparation of trans-[Pd(COOR)(H2O)(PPh3)2](TsO) with R
bulkier than Et
The precursor was used as such or also in combination with PPh3
and PPh3/TsOH as reported in Section 3. All the experiments were
carried out in a stainless steel autoclave of ca. 250 ml of capacity
following the same procedure already reported, using the solvent
with 800 ppm of water in order to compare the activity with that
previously reported using related precursors [25,28,29].
AgTsO was slowly added to trans-[Pd(COOR)Cl(PPh3)2]
(0.1 mmol) suspended in 2 ml of ROH. The solution was stirred
for few minutes at 15 ◦C till complete precipitation of AgCl and
then quickly filtered using a micro-filter system. The solution was
dropped directly into 20 ml of cold water, under vigorous stirring. A
white solid precipitates which was separated by filtration, washed
with cold water, n-pentane and dried under vacuum (yield 75%).
This procedure gives good results also when R = Me or Et. If the
filtered methanol solution is poured into warm water (50 ◦C), the
complex trans-[Pd(COOR)(TsO)(PPh3)2] separates.
2.7. X-ray data collection, structure solution and refinement
The X-ray data collection was performed at room temper-
ature with a STADI 4 CCD STOE area detector diffractometer
on single-crystal mounted in a thin-walled glass capillary with
graphite-monochromated Mo K␣ radiation (ꢂ = 0.71073 Å). The
crystals were obtained by slow evaporation of a CHCl3/n-hexane
solution of the complex at −10 ◦C. A summary of the X-ray analysis
is listed in Table 1.
The structure was solved by direct methods and refined by
full-matrix least-squares based on F2, where all non-hydrogen
atoms were assigned anisotropic displacement parameters. As
commented in Section 3.2, the solvent molecules suffer from high
thermal motions. The final Fourier difference maps showed no sig-
nificant features, the largest maxima (less than one electron) close
to the chlorine atoms. All calculations were made with programs
of the SHELXTL/PC system and SHELXL93 program [30].
2.5. Reactivity
The reactivity tests under pressure higher than 2 atm were
carried out by dissolving the carboalkoxy complex (0.1 mmol) in
the appropriate alkanol (2 ml) in a 5–10-ml glass bottle placed
in an autoclave of ca. 50 ml volume. The autoclave was first
washed several times with the appropriate gas (CO or ethene),
then pressurised and warmed to the desired pressure and tem-
perature. The solution was stirred with a magnetic bar. After
the desired reaction time was over the autoclave was rapidly
cooled to r.t. (or even to 0 ◦C) and then slowly depressurised.
A little sample was taken apart for GC analysis. The rest was
quickly poured into water. The precipitate was collected on a fil-
ter, washed with cold water and n-heptane, dried under vacuum.
The nature of the solid was established by IR and NMR spec-
troscopy.
3. Results and discussion
3.1. Synthesis and characterization of the carboalkoxy complexes
reported in Table 2
The reactivity was tested with alkanols, acids (HCl, AcOH and
TsOH), water, water/TsOH and with ethene. In order not to be redun-
dant, the conditions of the tests are reported together with the
results and the discussion in the next section.
Complexes (Ia and b) have been prepared by reacting
[Pd(TsO)2(PPh3)2] or [Pd(H2O)2(PPh3)2](TsO)·2H2O with CO in ROH

E. Amadio et al. / Journal of Molecular Catalysis A: Chemical 298 (2009) 103–110
105
Table 2
Selected IR, 1H NMR and 31 P{ H} NMR data of trans-[Pd(COOR)(H2O)(PPh3)2](TsO) (Ia–g), trans-[Pd(COOMe)(TsO)(PPh3)2] (IIa) and trans-[Pd(COOEt)(TsO)(PPh3)2] (IIb).
1
R
IRa
1Hb
31 Pc
ꢃ
cm−1
ı (ppm)
ı (ppm)
Me (Ia) [REF?]
H2O
3489, 3412
1685
2.58 s: CH3
2.36 s: CH3
19.04 sd
C
O
C–O–C
SO3
1070
1230, 1035, 1013
(IIa)
C
O
1668
1086
1263, 1029, 1000
2.58 s: CH3
2.36 s: CH3
19.04 s
19.01 s
C–O–C
SO3
Et (Ib)
H2O
C
C–O–C
SO3
3671, 3441
1684
1057
1222, 1036, 1011
2.87 q: CH2
2.36 s: CH3
0.49 t: CH3
O
(IIb)
C
O
1663
1088
1268, 1029, 999
2.87 q: CH2
2.36 s: CH3
0.49 t: CH3
19.01 s
19.12 s
C–O–C
SO3
n-Pr (Ic)
H2O
C
C–O–C
SO3
3640, 3485
1671
1083
1257, 1031, 1008
2.81 t: OCH3
2.36 s: CH3
0.88 m: CH2
0.51 t: CH3
O
iso-Pr (Id)
n-Bu (Ie)
H2O
3220
1674
1051
3.82 m: CH
2.36 s: CH3
0.31 d: CH3
19.08 s
19.13 s
19.27 s
19.10 s
C
O
C–O–C
SO3
1223, 1038, 1014
H2O
3640, 3488
1672
1081
1246, 1030, 1007
2.87 t: OCH2
2.36 s: CH3
0.87 m: CH2
0.68 t: CH3
C
O
C–O–C
SO3
iso-Bu (If)
sec-Bu (Ig)
H2O
3142
1669
1068
2.63 d: OCH3
2.36 s: CH3
1.08 m: CH
0.47 d: CH3
C
O
C–O–C
SO3
1218, 1036, 1012
H2O
3144
1671
1059
3.69 m: CH
2.36 s: CH3
0.67 m: CH2
0.41 t: CH3
0.20 d: CH3
C
O
C–O–C
SO3
1231, 1037, 1013
NMR spectra are recorded in CD3OD. Abbreviations: s = singlet; t = triplet; q = quartet; m = multiplet.
a
Nujol mull.
b
ı (1H) values in ppm referenced to CD3OD.
c
ı (31 P) values in ppm from external 85% H3PO4, downfield being taken as positive.
In Ref. [25] it was erroneously reported 16.51 ppm.
d
(2 atm, r.t.):
[Pd(COOR)Cl(PPh3)2] starting from trans-[PdCl2(PPh3)2], in which
case the use of the base is necessary in order to neutralise HCl that
forms during the reaction and that otherwise reverses the reaction
[12,24].
[Pd(TsO)2(PPh3)2] + CO + ROH
H
O
2
−→ trans-[Pd(COOR)(H2O)(PPh3)2](TsO) + TsOH
(1)
Selected IR and NMR data are reported in Table 1. The IR spec-
tra of (Ia–g) show a strong absorption band in the 1684–1670 cm−1
region due to ꢃC of the carboalkoxy ligand. In the correspond-
The reaction with MeOH takes place in a few minutes, whereas
with EtOH is significantly slower as it takes 20–30 min.
O
ing neutral chloride the ꢃC
absorption appears at slight lower
O
frequency [24], which might be due to a stronger ␲-back dona-
tion from the metal to the carboalkoxy moiety in the neutral
complexes.
Reaction (1) with bulkier alkanols gives unsatisfactory results.
An attempt to synthesise complex If failed. The carboalkoxy com-
plexes (Ic–g) with bulkier R (Table 2) have been prepared via
methathetical exchange of the corresponding neutral chloride with
AgTsO, as already experienced for the synthesis of (Ia) [25]:
Absorption bands due to coordinated water are observed in the
3640–3142 cm−1 region [31]. Complexes (Ia–c and Ie) show two
sharp bands of low intensity; complex If shows one band of low
intensity at 3637 cm−1 and an unresolved broad absorption cen-
tered at 3142 cm−1; complexes Id and Ig show a broad unresolved
band centered at 3220 and at 3144 cm−1, respectively. Those at
lower frequency suggest the presence of a hydrogen-bond between
water and TsO−, with consequent lowering of ꢃO–H [31–33]. The
absorptions in the ranges of 1273–1222, and 1014–1007 cm−1 ar−e
identified as some of the characteristic bands of the anionic –SO3
trans-[Pd(COOR)Cl(PPh3)2] + AgTsO
H
O
2
−→ trans-[Pd(COOR)(H2O)(PPh3)2](TsO) + AgCl
(2)
It is noteworthy to point out that the synthesis based on reaction
(1) does not require the presence of a tertiary base, at difference
of the analogous synthesis of the corresponding chloride trans-

106
E. Amadio et al. / Journal of Molecular Catalysis A: Chemical 298 (2009) 103–110
group [34–38]. The band in the 1037–1025 cm−1 is assignable to
the C–O–C stretch of the alkoxycarbonyl group [13,14,24].
Table 3
Relevant bond distances (Å) and angles (◦).
The 1H NMR spectra of (Ia–g) show signals in the range of
0.20–3.69 ppm assigned to the alkoxy group protons [24]. The
singlets observed at 2.36 ppm is assigned to the CH3 protons of
the TsO− anion. Two well separated multiplets for the aromatic
protons are centered at about 7.6 and 7.3 ppm. The 31P NMR
Pd–P(1)
Pd–P(2)
S–O(3)
S–C(3)
2.353(2)
2.342(2)
1.452(4)
1.789(6)
Pd–O(3)
Pd–C(1)
C(1)–O(1)
C(1)–O(2)
O(2)–C(2)
2.160(4)
1.978(6)
1.187(7)
1.336(7)
1.458(7)
P(1)–Pd–P(2)
P(1)–Pd–O(3)
P(1)–Pd–C(1)
Pd–O(3)–S
177.2(1)
98.1(1)
89.2(2)
151.9(3)
124.6(5)
C(1)–Pd–O(3)
P(2)–Pd–O(3)
P(2)–Pd–C(1)
O(3)–S–C(3)
Pd–C(1)–O(2)
O(1)–C(1)–O(2)
170.8(2)
84.3(1)
88.3(2)
104.5(3)
110.0(4)
125.4(6)
spectrum shows
a singlet in the range of 19.01–19.27 ppm,
close to that of the relevant neutral chloride complex
[24].
Pd–C(1)–O(1)
As reported in Section 2, complexes (Ia–g) have been precipi-
tated by pouring their alkanol solution into cold water. When warm
water has been used to precipitate the carbomethoxy complex, a
different derivative has been obtained, (IIa). The IR spectrum of
(IIa) does not show absorbtion bands due to coordinated water.
The ꢃCO and ꢃC–O–C are shifted from 1685−and 1036 cm−1 to 1668
and 1029 cm−1 and the bands of the –SO3 group are shifted from
1230 and 1013 cm−1 to 1263 and 1001 cm−1 and are assignable to
a coordinated anion [34–38]. The 1H and 31P NMR spectra of com-
plexes (Ia) and (IIa) in CD3OD do not differ appreciably, probably
due to the fast exchange of labile water, TsO−, solvent. The X-ray
crystal structure of the CHCl3 clathrated of complex (IIa) shows
that it has a trans geometry, in which TsO− coordinates the metal
(in place of H2O) in (Ia) [24]; the 31P NMR spectrum for this com-
plex in CDCl3 shows a singlet at 18.51 ppm slightly lower than
in CD3OD. On the basis of these data complex (IIa), precipitated
from MeOH/H2O is formulated as trans-[Pd(COOMe)(TsO)(PPh3)2].
A trans geometry is assigned also to all the new cationic complexes
(Ib–g).
P(1)–C(1)–P(2)–O(3), C(1)–C(2)–O(1)–O(2) and O(3)–S–C(3). The
atoms in the set P(1)–C(1)–P(2)–O(3) (basal plane) are virtually
coplanar within 0.04 Å, with the Pd center out by 0.06 Å. The set
C(1)–C(2)–O(1)–O(2) is strictly planar, and is almost orthogonal to
the basal plane, the two planes making a dihedral angle of 88.9◦.
Instead, the O(3)–S–C(3) set is tilted from both the previous plane,
being inclined of 28.6◦ in respect of the basal plane and of 69.8◦ in
respect of the C2O2 set.
Looking at bond distances and angles, it appears worth noting
that the Pd–O(3)–S angle is very wide, 151.9(3)◦; this is the largest
value reported to date for a mononuclear Pd(II) complex. A search in
the Cambridge Structural Database (CCD) [40–42] showed that the
closest values found in (trifluoromethanesulfonato) (3-(diethyl-
amino)-propionyl)(diethylamine)palladium(II) and (2-(benzene-1,
1-diyl)-2-methyl-propyl)-(trifluoromethanesulfonato)trimethyl-
phosphine-palladium(II) [43,44], were 150.5◦ and 146.0◦, respec-
tively. In [43], the abnormal widening of the Pd–O–S angle was
attributed to van der Waals contacts between the other oxygen
atoms of the sulfone with hydrogen atoms of a nearby ligands.
In the present case, the large Pd–O(3)–S angle might be due to
the very efficient interaction involving O(4) and the hydrogen
atom of a CHCl3 molecule, as indicated in Fig. 2 (the contact
distance is 1.99 Å, and the pertinent O(4)· · ·H–C(11) angle is 164◦),
and only to a limited extent to steric interactions with the PPh3
group.
The IR spectrum of the solid precipitated by addition of cold
water to the ethanol solution of the carboethoxy derivative shows
double absorptions in the ꢃCO, ꢃC–O–C and ꢃSO regions, suggesting
3
that this solid is a mixture of trans-[Pd(COOMe)(H2O)(PPh3)2](TsO),
(Ib) and trans-[Pd(COOMe)(TsO)(PPh3)2], (IIb).
3.2. X-ray structure analysis of
[Pd(COOMe)(TsO)(PPh3)2]·2CHCl3, (IIa)
The ORTEP [39] drawing of the complex is shown in Fig. 1,
together with the numbering scheme used, while relevant bond
lengths and angles are reported in Table 3. The solid state investi-
gation revealed three sets of atoms lying in well defined planes:
The CCD exploration returned about 20 Pd mononuclear com-
plexes with the metal showing a Pd–O–S–C moiety and a Pd
environment similar to the one found in our compound [45–55].
Fig. 1. A drawing of the complex with the selected numbering scheme. Ellipsoids are at the 50% level; hydrogen atoms are represented by spheres of arbitrary size. The
phenyl rings and the CHCl3 molecules have been omitted for clarity.

E. Amadio et al. / Journal of Molecular Catalysis A: Chemical 298 (2009) 103–110
107
Fig. 2. The interaction between the sulfonato moiety and one of the CHCl3 solvent molecules described in text.
The Pd–O(3) (2.160(4) Å) and O(3)–S (1.452(4) Å) distances agree
reasonably with the average values found in such complexes
(2.175 and 1.461 Å, respectively) [43–55]. Similar considerations
also apply to other geometrical parameters involving the sulfur
atom of the p-toluenesulfonato ligand. In fact, in the present com-
2.33(3) Å found in 227 square planar Pd(II) complexes bearing two
PPh3 groups [40–42].
3.3. Reactivity
plex the S–C(3) distance and the O(4)
S
O(5), O(3)–S–C(3) angles
The reactivity has been tested with both the carbomethoxy com-
plexes (Ia) and (IIa). They behave in the same way, therefore for the
sake of simplicity here below we report the tests done with (IIa)
only, if not otherwise indicated.
are 1.789(6) Å, 115.7(4) and 104.5(3)◦, whereas the corresponding
average parameters found in about 30 Pd square planar complexes
showing at least 1 sulfonato ligand are 1.809 Å, 116.4◦ and 102.1◦,
respectively [40–42].
3.3.1. Reactivity with alkanols
In the CCD there are also about 15 structures in which a
mononuclear tetracoordinated Pd atom is bound by a –C( O)–O–C
moiety [14,16,17,22,25,43,56–61], most of them also showing
the metal coordinated by two phosphine ligands. The aver-
age Pd–C distance and the Pd–C–( O), Pd–C–O angles in the
reported structures are 1.979 Å, 126.6◦ and 112.9◦, respectively.
These values compare quite well with 1.978(6) Å, 124.6(5)◦ and
110.0(4)◦ in the complex described here. Among known struc-
tures, the complexes most closely resembling these values are
trans-aqua-carbomethoxy-bis(triphenylphosphine)-palladium
(1.975 Å, 124.5◦ and 111.4◦) [25], trans-carbomethoxy-chloro-
bis(triphenylphosphine)palladium (1.972 Å, 125.3◦ and 111.9◦)
[56] and chloro-(3-hydroxypropoxycarbonyl-C)-(2-(pyridin-2-
yl)ethyldiphenylphosphine-N, P)-palladium(II) (1.964 Å, 124.5◦
and 112.7◦) [17].
At 80 ◦C, under 45 atm of CO, the carbomethoxy complex reacts
with RꢀOH (Rꢀ = n-Pr, iso-Pr), used also as a solvent, giving com-
plexes (Ic–d). These complexes react with MeOH under the same
conditions giving the carbomethoxy derivative.
Under the above conditions and in the presence of a base such
as triethylamine, complex (IIa) in MeOH is reduced to Pd(0) com-
plexes, [Pd3(CO)3(PPh3)3] and [Pd(CO)(PPh3)3], the latter forms
prevalently when the reaction is carried out also in the presence
of 1 mol of PPh3 per Pd atom. The reduction occurs with concomi-
tant formation of dimethyl carbonate (DMC) and oxalate (DMO), as
here below schematised:
(IIa)MeOH, −CO→and NEt3(CH3) CO + (CH3COO)2 + Pd(0)
(3)
2
−(NEt H)(TsO)
3
Under similar conditions the analogous complexes trans-
[Pd(COOMe)Cl(PPh3)2] and trans-[Pd(OAc)(COOMe)(PPh3)2] give
DMC or DMO, respectively [61–63].
The two CHCl3 molecules in the unit cell, they are animated
by high thermal motions, and only one of them, as said above,
seems to interact with the complex inner core. Instead, the phenyl
ring of the p-toluenesulfonato moiety establishes a rather long-
range (4.22 Å) ␲-interaction with a phenyl ring bound to P(2). Other
structural parameters, like the six P–Cph distances (average value
3.3.2. Reactivity with acids HX (X = Cl, OAc and TsO)
The reactions have been carried in MeOH. At r.t. (IIa) reacts with
1 equiv. of HCl affording trans-[Pd(COOMe)Cl(PPh3)2], whereas
when treated with 2 equiv. of acid the carboalkoxy moiety evolves
with formation of MeOH, CO according to reaction (4) in which step
1.819 Å) do not show any new feature, and the same can be said of
the two Pd–PPh3 bonds, that match quite well the average value of

108
E. Amadio et al. / Journal of Molecular Catalysis A: Chemical 298 (2009) 103–110
(ii) practically reverses the reaction that leads to the synthesis of
trans-[Pd(COOMe)Cl(PPh3)2] from [PdCl2(PPh3)2], CO and MeOH in
the presence of NEt3 [12,24]:
3.3.5. Catalytic properties of complexes (Ia–g) in the
hydrocarboalkoxylation of ethene
The catalytic activity of complex (Ia) in the hydroesterication of
ethene in MeOH has been already reported [25,29]. Under standard
conditions (40 atm, CO/ethene = 1/1, 70–100 ◦C) significant catalytic
activity is shown only in the presence of both PPh3 and TsOH
(Pd/P/TsOH = 1/8/8). After catalysis the starting complex was recov-
ered (65%) as trans-[Pd(COEt)(TsO)(PPh3)2], related to the “hydride”
cycle (see Scheme 1) [29]. This acyl complex reacts with MeOH to
give expected methyl propanoate (MP) in an almost stoichiometric
amount and is a catalyst precursor for the methoxycarbonylation
of a different olefin to give the expected ester and an almost sto-
ichiometric amount of MP. This proves that the acyl complex is
sufficiently stable to be isolated while being reactive enough to
enter the catalytic cycle. Using [Pd(TsO)2(PPh3)2] as catalyst pre-
cursor, it was found that a hydride source, such as H2, H2O, or
TsOH, present a promoting effect on the catalysis [28]. All these
experimental evidences are more in favor of the “hydride” cycle
rather than of the “carbomethoxy” cycle [67]. In addition, others
found that using a closely related system, derived from Pd(AcO)2 in
(IIa)HCl,−–→TsOHtrans-[PdCl(COOMe)(PPh3) ]
2
(i)
−→
HCl, –MeOH, –COtrans-[PdCl2(PPh3) ]
(4)
2
(ii)
At r.t. the carboalkoxy moiety is stable when treated with
AcOH or TsOH, even in excess; with the first acid, in the ratio
Pd/AcOH = 1/1, (IIa) is partially converted to the corresponding
acetate, trans-[Pd(COOMe)(OAc)(PPh3)2], whereas the conversion
is complete using an excess of AcOH.
When treated at 50 ◦C with TsOH (Pd/TsOH = 1/6–1/10), complex
(IIa) is unstable yielding [Pd(TsO)2(PPh3)], analogously to reaction
(4), step(ii).
The complexes trans-[Pd(COOMe)X(PPh3)2] (X = Cl and AcO) are
easily obtained also by adding 1 equiv. of LiX to (IIa) dissolved in
MeOH.
combination with an excess of PPh3 and of TsOH [68], there was
+
formation of phosphine degradation products such as MePPh3
,
3.3.3. Reactivity with water and with water/TsOH
EtPPh3 and EtCOCH2CH2PPh3+, isolated as TsO− salts, thus pro-
viding further evidences for the “hydride” mechanism, the last two
being Pd mediated side products of the “hydride” route [69].
In the present case, we found that also complexes (Ib–g), tested
in the relevant alkanol at 100 ◦C, 40–50 atm, do not present any
significant catalytic activity and that extensive decomposition to
palladium metal occurs at 100 ◦C. Only in the presence of both
added PPh3 and TsOH catalytic activity is observed. Qualitative
observations indicate that the activity lowers with increasing bulk-
iness of the alkanol.
+
Complex (IIa) in MeOH/H2O (10/1, v/v) does not react signifi-
cantly in 1 h even at 50 ◦C, as it is recovered unreacted upon adding
cold water to the solution, whereas when to this MeOH/H2O solu-
tion TsOH is added (Pd/TsOH = 1/6) the carbomethoxy moiety is
unstable and (IIa) is converted to [Pd(H2O)2(PPh3)2] (TsO)2].
H
O, TsOH
2
(IIa)
−→
cis − [Pd(H2O)2(PPh3)2](TsO)2
(5)
–CO, –MeOH
Taking into consideration the reactivity of (IIa) with TsOH
reported at Section 3.3.2, it appears that (i) the acid promotes the
demethoxylation of the carbomethoxy moiety, probably through
protonation of the oxygen atom of the methoxy group, and that (ii)
water does not play a significant role in the demethoxylation step.
When the same experiment is carried out under CO pressure
(20 atm), reduction to Pd(0) complexes occurs, with formation of
[Pd3(CO)3(PPh3)3] or [Pd(CO)(PPh3)3], the latter forming when the
reaction is carried out also in the presence of 1 equiv. of PPh3. The
reduction occurs with concomitant formation of CO2, probably via
a reaction closely related to the water gas shift reaction through the
intermediacy of a Pd(II)–(COOH) species [64–66]:
That being said, in addition to the fact that ethene does not
insert into the Pd–COOMe bond under the condition reported in
Section 3.3.4, we present here further evidences that are more in
favor of the “hydride” cycle. In an experiment carried out under
conditions in which complex (Ia) is relatively stable, i.e., 40 ◦C and
in the presence of 2 equiv. of added PPh3, catalysis is not observed
even under 80 atm total pressure (CO/ethene 1/1). At difference,
under these conditions, but also in the presence of 10 equiv. of
TsOH, there is formation of MP (TOF = 10 h−1), together with minor,
though significant, amounts of light CO/ethene co-oligomers hav-
ing only keto–ester end groups (KE in Scheme 1), the most abundant
of them being 4-oxohexanoate (n = 2). Even more significant, no
dimethyl succinate (DMS) or higher diesters co-oloigomers were
detected by GC of the reaction mixture.
+
CO, H
O
−CO
–H
Pdꢁ2+ −→ Pd–COOHꢁ+ −→ Pd–Hꢁ+−→ Pd(0)
(6)
2
2
+
−H
3.3.4. Reactivity with ethene
Both the “carbomethoxy” and the “hydride” cycles lead to the
formation of MP and KE co-oligomers. Though it is well known that
intermediates arising from the insertion of ethene (a), (c) and (f)
are in equilibrium with the so-called ␤-chelates, and those arising
from the insertion of CO (b) and (g) are in equilibrium with the so-
called ␥-chelates, the omission of these chelates do not question the
validity of the reasoning that follows, therefore they are omitted for
the shake of clarity.
After pressuring a methanol solution of (IIa) with ethene
(1–40 atm) at r.t. for 1 h, the complex has been recovered unre-
acted, together with a minor amount of Pd(0) complexes (ca. 10%).
In the MeOH solution no methyl propanoate was detected by GC.
Nor any insertion was observed in CD2Cl2/MeOH from −78 ◦C up to
50 ◦C in a NMR tube pressurised with 6 atm of ethene.
In principle, ethene could insert into the Pd–(COOMe) bond with
formation of a Pd–(CH2–H2–COOMe) moiety, which may be in equi-
librium with a so-called ␤-esterchelate, as it has been found to
occur promptly with cationic cis-chelatediphosphine Pd(II) com-
plex [Pd(COOCH3(CH3CN)(dibpp)](TfO) (dibpp = 1,3-(iBu2P)2C3H6)
in CH2Cl2/MeOH/CH3CN even at −30 ◦C at ambient pressure, as
established by NMR multinuclear spectroscopy [11]. It has been
also found that upon rising the temperature up to 25 ◦C the ␤-
esterchelate reacts with MeOH yielding methyl propanoate and
the starting carbomethoxy complex, thus the “carboalkoxy” cycle
for the hydromethoxycarbonylation of ethene was demonstrated
under the conditions just reported [11].
In the “carbomethoxy” cycle, insertion of ethene into
a
Pd–COOMe bond gives intermediate (a), which upon protonolysis
by MeOH yields MP and a Pd–OMe species, which inserts CO to con-
tinues the catalytic cycle. Protonolysis competes with the further
subsequent insertion of CO and ethene, with formation of inter-
mediates (b) and (c), which, after protonolysis with MeOH, give
the KE co-oligomers and again a Pd–OMe species, that continues
the catalysis. It should be noted that the copolymerization process
stops after the insertion of a few molecules of monomers, i.e., the
termination process competes effectively with the chain growing
process. However, since it has been found that in the copolymeriza-

E. Amadio et al. / Journal of Molecular Catalysis A: Chemical 298 (2009) 103–110
109
Scheme 1. Proposed reaction mechanism.
tion process CO insertion is faster than ethene insertion [6,7], it is
hard to explain why after CO insertion into intermediate (a) (or (c))
there is no formation of dimethyl succinate (or of higher diesters).
Formation of the diesters would occur via methanolysis with for-
mation of a Pd–H species which can start the “hydride” cycle. The
shift from one mechanism to the other has been demonstrated by
multinuclear NMR spectroscopy [11]. For balance requirements of
the two cycles there should be also formation of diethylketone or
of co-oligomers having only keto ending groups for any molecule
having only ester ending groups. However, co-oligomers of this type
did not form nor diethyl ketone.
Therefore is more likely that MP and the ketoesters form though
a catalytic cycle that starts from a Pd–H species and terminates
via attack of MeOH to a Pd–acyl bond of intermediates (e) and (g),
with concomitant reformation of the hydride species. The hydride
would form after decarbomethoxylation of the Pd–COOMe bond
promoted by TsOH/H2O (cf. Section 3.3.3), followed by the inter-
action of H2O with CO (cf. reaction (6): under catalytic conditions
ethene would insert into the Pd–H bond before deprotonation).
These results do not exclude that, under the conditions in which
catalysis occurs as just reported, ethene insertion into a Pd–COOMe
bond might take place to some extent. As a matter of fact it has
been found that the dibpp-based cationic complex under NMR con-
ditions, both the acyl complex, related to the “hydride”, and the
␤-ester chelate, related to the “carbomethoxy” cycle, easily form at
−60 ◦C and at −30 ◦C, respectively. In addition, it has been found
that the acyl complex undergoes methanolysis at −30 ◦C in a few
minutes to give the ester as required from the “hydride” cycle,
whereas methanolysis of the ␤-ester chelate, required to give MP
from the “carbomethoxy” cycle, is slow on a timescale of days,
at 25 ◦C. It was suggested that slow methanolysis of the ␤-ester
chelate, rather than slow insertion of an alkene into the Pd–COOMe
bond, directs the catalysis towards the “hydride” mechanism
[11].

110
E. Amadio et al. / Journal of Molecular Catalysis A: Chemical 298 (2009) 103–110
[27] G. Cavinato, A. Vavasori, L. Toniolo, A. Dolmella, Inorg. Chim. Acta 357 (2004)
However, it should be underlined that the dibpp- and the
2737.
PPh3-based systems are too different to consent a straightforward
comparison. A part that dibpp acts only as cis-chelating ligand and
that the acyl- and carbomethoxy-complexes isolated after catalysis
using the PPh3-based system have a trans-geometry,1 the two sys-
tems work under different conditions. The second one requires the
use of an excess of PPh3 in order to prevent deactivation to Pd metal.
When PPh3 is added to the dibpp system the insertion of ethene
into the Pd–COOMe bond is inhibited [11]. In addition, the PPh3
system, in order to be catalytically efficient, requires also the use of
an excess of TsOH and H2O, which lead to decarbomethoxylation of
the Pd–COOMe bond.
[28] A. Vavasori, G. Cavinato, L. Toniolo, J. Mol. Catal. A: Chem. 176 (2001) 11.
[29] G. Cavinato, L. Toniolo, A. Vavasori, J. Mol. Catal. A: Chem. 219 (2004) 233.
[30] G.M. Sheldrick, SHELXTL/PC, Version 5.03, Siemens Analytical X-ray Instru-
ments Inc., Madison, WI, 1994;
G.M. Sheldrick, SHELXL-93, Program for the Refinement of Crystal Structures,
University of Göttingen, Germany, 1993.
[31] J. Vicente, A. Arcas, D. Bautista, P.G. Jones, Organometallics 16 (1997) 2127.
[32] G.J. Kubas, C.J. Burns, G.R.K. Khaisa, L.S. Van Der Sluys, G. Kiss, C.D. Hoff,
Organometallics 11 (1992) 3390.
[33] D.M. Branan, N.W. Hoffman, E.A. McElroy, N. Prokopuk, A.B. Salazar, M.J. Rob-
bins, W.E. Hill, T.R. Webb, Inorg. Chem. 30 (1991) 1200.
[34] U. Bohner, G. Zundel, J. Phys. Chem. 89 (1985) 1408.
[35] E. Horn, M.R. Snow, Aust. J. Chem. 37 (1984) 1375.
[36] O.H. Bailey, A. Ludi, Inorg. Chem. 24 (1985) 2582.
Not only, but the evidences here reported for the “hydride”
mechanism, together with those reported in previous studies
[25,27–29,68,69], refer to the actual catalysis conditions.
It is worth mentioning that also neutral trans-
[Pd(COOMe)Cl(PPh3)2] treated at 95 ◦C with 1-hexene does
not insert the olefin into the Pd–COOMe bond [71]. In this case the
much greater coordination capacity of the chloride ligand and the
bulkier 1-hexene might prevent the insertion.
[37] W. Lasser, U. Thewalt, J. Organomet. Chem. 302 (1986) 201.
[38] E.A. Robinson, Can. J. Chem. 391 (1961) 247.
[39] C.K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National Laboratory, Oak
Ridge, TN, 1976.
[40] F.H. Allen, Acta Crystallogr. B 58 (2002) 380.
[41] F.H. Allen, W.D.S. Motherwell, Acta Crystallogr. B 58 (2002) 407.
[42] Cambridge Crystallographic Database (Version 5.29 of November 2007).
[43] O.P. Anderson, A.B. Packard, Inorg. Chem. 18 (1979) 1129.
[44] J. Campora, J.A. Lopez, P. Palma, P. Valerga, E. Spillner, E. Carmona, Angew. Chem.
Int. Ed. Engl. 38 (1999) 147.
[45] R.B. Bedford, C.S.J. Cazin, S.J. Coles, T. Gelbrich, P.N. Horton, M.B. Hursthouse,
M.E. Light, Organometallics 22 (2003) 987.
[46] A.D. Burrows, M.F. Mahon, M. Varrone, J. Chem. Soc., Dalton Trans. (2003) 4718.
[47] M. Yamashita, I. Takamiya, K. Jin, K. Nozaki, Organometallics 25 (2006) 4588.
[48] D.B. Grotjahn, Y. Gong, L. Zakharov, J.A. Golen, A.L. Rheingold, J. Am. Chem. Soc.
28 (2006) 438.
[49] C.J. Chapman, C.G. Frost, M.F. Mahon, J. Chem. Soc., Dalton Trans. (2006) 2251.
[50] T. Kochi, K. Yoshimura, K. Nozaki, J. Chem. Soc., Dalton Trans. (2006) 25.
[51] D.K. Newsham, S. Borkar, A. Sen, D.M. Conner, B.L. Goodall, Organometallics 26
(2007) 3636.
[52] W. Weng, C.-H. Chen, B.M. Foxman, O.V. Ozerov, Organometallics 26 (2007)
3315.
[53] M. Agostinho, P. Braunstein, C. R. Chim. 10 (2007) 666.
[54] M. Agostinho, P. Braunstein, R. Welter, J. Chem. Soc., Dalton Trans. (2007) 759.
[55] M. Agostinho, P. Braunstein, Chem. Commun. (2007) 58.
[56] L.N. Zhir-Lebed, L.G. Kuz’mina, Y.T. Struchkov, O.N. Temkin, V.A. Golodov, Koord.
Khim. 4 (1978) 1046.
[57] G. Del Piero, M. Cesari, Acta Crystallogr. B 35 (1979) 2411.
[58] R. Santi, A.M. Romano, R. Garrone, R. Millini, J. Organomet. Chem. 566 (1998)
37.
References
[1] M. Beller, A.M. Tafesh, in: B. Cornils, W.A. Hermann (Eds.), Applied Homoge-
neous Catalysis with Organometallic Compounds, I, VCH, Weinheim, 1996, p.
187.
[2] S. Uchiumi, K. Ataka, T. Matsuzaki, J. Organomet. Chem. 576 (1999) 279.
[3] B. Gabriele, G. Salerno, M. Costa, in: M. Beller (Ed.), Catalytic Carbonylation
Reactions, 18, Springer, 2006, p. 239 (Top. Organomet. Chem.).
[4] J.-T. Chen, A. Sen, J. Am. Chem. Soc. 106 (1984) 1506.
[5] F. Osawa, N. Kawasaky, H. Okamoto, T. Yamamoto, A. Yamamoto,
Organometallics 6 (1987) 1640.
[6] A. Sen, Acc. Chem. Res. 26 (1933) 303.
[7] E. Drent, P.H.M. Budzelaar, Chem. Rev. 96 (1996) 663.
[8] D.M. Fenton, J. Org. Chem. 38 (1973) 3192.
[9] D.E. James, L.F. Hines, J.K. Stille, J. Am. Chem. Soc. 98 (1982) 1806.
[10] D. Milstein, Acc. Chem. Res. 21 (1988) 428.
[11] J. Liu, B.T. Heaton, J.A. Iggo, R. Whyman, J.F. Bickley, A. Steiner, Chem. Eur. J. 12
(2006) 4417.
[12] M. Hidai, M. Kokura, Y. Uchida, J. Organomet. Chem. 52 (1973) 431.
[13] F. Rivetti, U. Romano, J. Organomet. Chem. 154 (1978) 323.
[14] A. Sacco, G. Vasapollo, C.F. Nobile, A. Piergiovanni, M.A. Pellinghelli, M. Lan-
franchi, J. Organomet. Chem. 356 (1988) 397.
[15] G.P.C.M. Dekker, C.J. Elsevier, K. Vrieze, P.W.N.M. van Leeuwen, C.F. Roobeck, J.
Organomet. Chem. 430 (1992) 357.
[16] G.D. Smith, B.E. Hanson, J.S. Merola, F.J. Waller, Organometallics 12 (1993) 568.
[17] P. Giannoccaro, D. Cornacchia, S. Doronzo, E. Mesto, E. Quaranta, M. Aresta,
Organometallics 25 (2006) 2872.
[18] P. Fitton, M.P. Johnson, J.E. McKeon, Chem. Commun. (1968) 6.
[19] S. Otsuka, A. Nakamura, T. Yoshida, M. Naruto, K. Ataka, J. Am. Chem. Soc. 95
(1973) 3180.
[59] E. Gallo, F. Ragaini, S. Cenini, F. Demartin, J. Organomet. Chem. 586 (1999) 190.
[60] J. Ruiz, M.T. Martinez, F. Florenciano, V. Rodriguez, G. Lopez, J. Perez, P.A.
Chaloner, P.B. Hitchcock, Inorg. Chem. 42 (2003) 3650.
[61] Y. Nishihara, Y. Inoue, M. Itazaki, K. Takagi, Org. Lett. 7 (2005) 2639.
[62] F. Rivetti, U. Romano, J. Organomet. Chem. 174 (1979) 221.
[63] G. Cavinato, L. Toniolo, J. Organomet. Chem. 444 (1993) C65.
[64] V.N. Zudin, G.N. Il’inich, V.A. Likholobov, Y.I. Yermakov, J. Chem. Soc. Chem.
Commun. (1984) 545.
[65] V.N. Zudin, V.D. Chinakov, V.M. Nekipelov, V.A. Likholobov, Yu.I. Yermakov, J.
Organomet. Chem. 289 (1985) 425.
[66] V.N. Zudin, V.D. Chinakov, V.M. Nekipelov, V.A. Rogov, V.A. Likholobov, Yu.I.
Yermakov, J. Mol. Catal. 52 (1989) 27.
[20] W. Keim, J. Becker, A.M. Trzeciak, J. Organomet. Chem. 372 (1989) 447.
[21] Y. Nishihara, M. Miyasaka, Y. Inoue, T. Yamaguchi, M. Kojima, K. Takagi,
Organometallics 26 (2007) 4054.
[22] R. Hua, H. Takeda, S. Onozawa, Y. Abe, M. Tanaka, J. Am. Chem. Soc. 123 (2001)
2899.
[23] A. Dervisi, P.G. Edwards, P.D. Newman, R.P. Tooze, S.J. Coles, M.B. Hursthouse, J.
Chem. Soc., Dalton Trans. (1999) 1113.
[24] R. Bertani, G. Cavinato, L. Toniolo, G. Vasapollo, J. Mol. Catal. 84 (1993) 165.
[25] G. Cavinato, A. Vavasori, L. Toniolo, F. Benetollo, Inorg. Chim. Acta 343 (2003)
183.
[67] G. Cavinato, L. Toniolo, A. Vavasori, in: M. Beller (Ed.), Catalytic Carbonylation
Reactions, 18, Springer, 2006, p. 125 (Top. Organomet. Chem.).
[68] W.G. Reman, G.B.J. De boer, S.A.J. Van Langen, A. Nahuijsen (Shell), Eur. Pat., EP
411721 A3 (1991).
[69] R.P. Tooze, K. Whiston, A.P. Malyan, M.J. Taylor, N.W. Wilson, J. Chem. Soc., Dalton
Trans. (2000) 3441.
[70] P.W.N.M. van Leeuwen, M.A. Zuideveld, B.H.G. Swennenhuis, Z. Freixa, P.C.J.
Kamer, K. Goubitz, J. Fraanje, M. Lutz, A.L. Spek, J. Am. Chem. Soc. 125 (2003)
5523.
[71] G. Cavinato, L. Toniolo, J. Organomet. Chem. 398 (1990) 187.
[26] J.M. Jenkins, J.C. Verkade, Inorg. Synth. 11 (1968) 108.
1
This is the preferred geometry, however, in solution it is likely that cis-species
are also present, as it is well known that this geometry favors the insertion reactions
as well as the product-forming step [70].