Palladium (II) complexes with the unsymmetrical H-spirophosphorane ligand HP(OCMe2CMe2O)(OCH2CMe2NH): Synthesis, structural and catalytic studies

Article abstract of DOI:10.1016/j.jorganchem.2011.05.008

We have investigated the reactivity of the unsymmetrical H-spirophosphorane (HSP) ligand HP(OCMe₂CMe₂O)(OCH₂CMe ₂NH) 1 towards different palladium(II) precursors and synthesised the mononuclear complexes [PdCl

Full text of DOI:10.1016/j.jorganchem.2011.05.008

Journal of Organometallic Chemistry 696 (2011) 2985e2992
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Palladium (II) complexes with the unsymmetrical H-spirophosphorane ligand
HP(OCMe2CMe2O)(OCH2CMe2NH): Synthesis, structural and catalytic studies
*
_ ꢀ
Anna Skarzynska , Andrzej Gniewek
Faculty of Chemistry, University of Wrocław, ul. F. Joliot-Curie 14, 50-383 Wrocław, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
We have investigated the reactivity of the unsymmetrical H-spirophosphorane (HSP) ligand HP(OC-
Me2CMe2O)(OCH2CMe2NH) 1 towards different palladium(II) precursors and synthesised the mono-
Received 10 February 2011
Received in revised form
27 April 2011
nuclear complexes [PdCl2{P(OCMe2CMe2O)OCH2CMe2NH2}]
2
and [PdCl(C3H5){P(OCMe2CMe2O)
OCH2CMe2NH2}] 3. The structural features of the compounds are characterised by spectroscopic methods
as well as single crystal X-ray diffraction studies. The complexes are shown to be remarkably active
precatalysts for the Heck and Hiyama cross-coupling reactions. The products of the CeC bond formation
reactions were obtained with high conversion and stereoselectivity. Mechanistic studies of the Heck
reaction reveal that, besides homogenous precatalysts, also heterogeneous Pd(0) nanoparticles are
involved in the catalytic process.
Accepted 13 May 2011
Keywords:
Hydrophosphoranes
Palladium
CeC bond formation
Heck reaction
Hiyama reaction
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
not only with n-butyl acrylate but also with substituted styrenes.
Moreover, the catalyst turns out to be highly effective in the Hiyama
Heterofunctionalised, bidentate P,E ligands (E ¼ N, O) constitute
an important, widely utilised group of ligands in transition metal
chemistry. The vast majority of these compounds comprise heter-
oditopic phosphines, where the phosphorus is the soft donor, and
the hard donor is either a nitrogen or oxygen atom. Less attention
has been paid to mixed bidentate phosphites. Albeit phosphite
ligands in particular P,N heterodonors, have demonstrated to be an
effective ligands for asymmetric catalytic reactions such as allylic
alkylation and Heck cross-coupling reactions [1]. Therefore we
believe that construction of new mixed P,N ligands that can be
easily synthesised from starting material is of great importance. For
this purpose H-spirophosphoranes, capable of P,N- or P,O-coordi-
nation seem to be particularly useful because of their easy
designing. Recently palladium complexes incorporating H-spi-
rophosphorane (HSP) ligands have emerged as efficient catalyst
precursors for Heck cross-coupling reactions [2]. The reported
results show that the H-spirophosphorane palladium complex
coupling reaction of functionalised styrylsilanes with bromo-
benzene [3]. The outstanding features of H-spirophosphoranes,
such as facile synthetic accessibility [4,5] and ambidentate elec-
tronic properties (p-acceptor ability of the phosphorus atom along
with -donor ability of the nitrogen atom), make them attractive
s
for the construction of new palladium precursors suitable for
catalytic processes [6]. Given our interest in the coordination
abilities of H-spirophosphoranes, we were curious to investigate
how fine-tuning inside the ligand molecule could influence cata-
lytic activity.
As a part of our continuing research on the chemistry of H-
spirophosphorane ligands, we report on the syntheses of unsym-
metrical spirophosphorane palladium complexes, an exploration of
their reactivity and application as precatalysts in the Heck and
Hiyama coupling reactions.
2. Results and discussion
[PdCl(m-Cl){P(OCMe2CMe2O)OCMe2CMe2OH}]2 has proved to be
a very good catalyst for the reaction of aryl bromide with n-butyl
acrylate, while the complex [PdCl2{P(OCH2CMe2NH)OCH2C-
Me2NH2}] shows high activity in the cross-coupling of aryl bromide
2.1. Synthesis and spectroscopic characterisation
The unsymmetrical H-spirophosphorane ligand HP(OCMe2C-
Me2O)(OCH2CMe2NH) 1 was prepared by one-pot synthesis in two
stages (eqs. (1) and (2)). It was obtained from equimolar amounts of
pinacol and 2-amino-2-methyl-1-propanol in the presence of one
equivalent of hexamethylphosphorous triamide P(NMe2)3. Pinacol
* Corresponding author. Tel.: þ48 713757254.
_
ꢀ
E-mail address: zag@wchuwr.pl (A. Skarzynska).
0022-328X/$ e see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2011.05.008

_
ꢀ
2986
A. Skarzynska, A. Gniewek / Journal of Organometallic Chemistry 696 (2011) 2985e2992
H2N
O
P
Cl
Cl
NH
O
Cl
O
O
i
ii
Pd
P
H
Pd
O
P
NH2
O
O
P
O
O
NH2
O
O
1
O
2
3
iii
Cl
O
OH
Pd
NH2
P
O
O
4
Scheme 1. (i) [PdCl2(cod)], dichloromethane; (ii) [Pd(m-Cl)(C3H5)]2, toluene; (iii) dichloromethane, þH2O, ꢀC3H6.
was the precursor of choice for the first stage, which proceeded
smoothly to give (OCMe2CMe2O)PNMe2, whereas the use of amino-
alcohol in the second step prevented the undesirable symmetrical
product HP(OCH2CMe2NH)2 from arising in the system. Although
some spectroscopic data for 1 are already known (31P NMR) [7],
there has been no information about the molecular and crystal
structure determination. Therefore, we present the stereochem-
istry of 1 and, additionally, a detailed instrumental analysis.
groups found reflection in infrared spectra. Absorption bands
for appropriate stretching frequencies are present at
n
(PeH) ¼ 2362 cmꢀ1 and
n
(NeH) ¼ 3350 cmꢀ1 respectively.
Complexation reactions of ligand 1 were carried out with the
use of two palladium precursors: [PdCl2(cod)] (cod ¼ 1,5-cyclo-
octadiene) and [Pd( -Cl)(C3H5)]2 (Scheme 1). An equimolar reac-
m
tion of 1 with appropriate precursors gave rise to the mononuclear
chelating complexes [PdCl2{P(OCMe2CMe2O)OCH2CMe2NH2}] 2 (i)
and [PdCl(C3H5){P(OCMe2CMe2O)OCH2CMe2NH2}] 3 (ii) respec-
tively. Ligand 1 coordinates in its predictable tautomeric form, via
tricoordinated phosphorus and nitrogen atoms. Both complexes
were isolated as yellow or white solids, well soluble in dichloro-
methane and acetonitrile. Prolonged exposure of dichloromethane
solution of 3 to atmospheric moisture causes a new palladium(II)
complex, [PdCl{P(O)(OCMe2CMe2O)}HOCH2CMe2NH2] 4, to form as
yellow crystals suitable for X-ray structural analysis (see
Supplementary material). It is apparently formed as a result of
hydrolytic cleavage of a coordinated H-spirophosphorane ligand
and simultaneous loss of the allyl molecule. The presence of
released propene was confirmed by means of GCeMS in an
experiment performed in a tube sealed with a rubber tap. The
H-spirophosphorane ligand underwent hydrolytic rupture to
form 2-amino-2-methyl-propanol-1 bounded to palladium in
a protonated N,O chelating mode and the phosphonate moiety
P(O)(OCMe2CMe2O) coordinated via the phosphorus atom.
PðNMe2Þ3þ HOCMe2CMe2OH/ðOCMe2CMe2OÞPNMe2
þ 2HNMe2
(1)
ðOCMe2CMe2OÞPNMe2 þHOCH2CMe2NH2 /
HPðOCMe2CMe2OÞðOCH2CMe2NHÞþHNMe2
(2)
The 31P{1H} NMR spectrum of 1 at room temperature exhibits
two resonances, at
d
¼ ꢀ49.21 ppm and ¼ 148.45 ppm, due to the
d
presence of two tautomeric penta- and tri-coordinated forms of
phosphorus atom. The presence of a PeH bond was manifested not
only in the 31P NMR spectrum but also in 1H NMR as a doublet at
7.49 ppm with 1J(PeH) ¼ 772.5 Hz. Besides the resonances assigned
to the proton at the phosphorus atom, the 1H NMR spectrum
exhibited signals for the NH proton: a doublet at
signals assigned to CH2 protons: two multiplets at
d
¼ 2.78 ppm, and
d
¼ 3.54 ppm and
The chemical compositions of 2 and 3 were ascertained by
microanalytical and spectroscopic data. 1H NMR spectra for both
complexes at room temperature show first-order patterns with
d
¼ 3.57 ppm assigned to equatorial and axial protons. Additionally,
six singlets were detected in the aliphatic resonance range, attrib-
uted to six unequal methyl protons. The presence of PeH and NeH
Fig. 2. Hydrogen bonds in ligand 1. Hydrogen atoms not involved in hydrogen inter-
actions are omitted for clarity. Symmetry codes: (i) 0.5 þ x, y, 0.5 ꢀ z; (ii) ꢀ0.5 þ x, y,
0.5 ꢀ z.
Fig. 1. Molecular structure of ligand 1. Displacement ellipsoids are drawn at the 30%
probability level and H atoms are shown as small spheres of arbitrary radii.

_
ꢀ
A. Skarzynska, A. Gniewek / Journal of Organometallic Chemistry 696 (2011) 2985e2992
2987
Table 2
Relevant bond distances (Å) and angles (ꢁ) for HP(OCMe2CMe2O)(OCH2CMe2NH) 1.
Bond lengths
P1eO11
P1eO12
P1eO13
P1eN11
1.6866(9)
1.7146(9)
1.6177(9)
1.638(2)
P2eO21
P2eO22
P2eO23
P2eN21
1.6819(9)
1.7144(9)
1.6198(9)
1.639(2)
Bond angles
O13eP1eO11
O13eP1eO12
O13eP1eN11
O11eP1eO12
N11eP1eO11
N11eP1eO12
91.15(5)
90.72(4)
122.64(5)
177.78(5)
89.67(5)
90.34(5)
O23eP2eO21
O23eP2eO22
O23eP2eN21
O21eP2eO22
N21eP2eO21
N21eP2eO22
91.41(5)
90.94(4)
123.11(5)
177.59(4)
89.78(5)
89.38(5)
Table 3
Hydrogen bond geometry (Å, ꢁ) in ligand 1.
Fig. 3. Molecular structure of complex 2. Displacement ellipsoids are drawn at the 30%
probability level and H atoms are shown as small spheres of arbitrary radii.
DꢀH/A
DꢀH
0.88
0.88
H/A
D/A
DꢀH/A
173
162
N11ꢀH11/O22i
2.15
2.15
3.030 (2)
3.002 (2)
N21ꢀH21/O12ii
three singlets in the range from 1.26 to 1.58 ppm, relevant to six
magnetically unequal methyl protons, as well as a doublets at 4.01
and 3.95 ppm, (for 2 and 3) assignable to CH2 protons. The presence
of wide singlets at 3.93 and 5.01 ppm for both 2 and 3 respectively
corresponds to the NH2 fragment of coordinated H-spi-
rophosphorane ligands. Furthermore, the appearance of NH2
groups is also supported by characteristic vibrations of IR spectra:
Symmetry codes: (i) 0.5 þ x, y, 0.5 ꢀ z; (ii) ꢀ0.5 þ x, y, 0.5 ꢀ z.
and bond angles are given in Tables 1, 2 and 4. The structures of 1
and 2 are comprised of two crystallographically independent
molecules in an asymmetric unit.
The coordination environment of the phosphorus atoms in the
free ligand HP(OCMe2CMe2O)(OCH2CMe2NH) molecule is, as
anticipated, a distorted trigonal bipyramid (TBP), with two oxygen
atoms, O1n and O2n (n ¼ 1, 2), in apical positions. The third oxygen
atom, On3 (n ¼ 1, 2), the nitrogen atom as well as hydrogen occupy
equatorial positions. The angles O11ꢀP1ꢀN11, O13ꢀP1ꢀO12,
O21ꢀP2ꢀN21 and O22ꢀP2ꢀO23 around phosphorus atoms are
almost 90ꢁ. Noticeable distortion from the ideal angle of 120ꢁ is
observed for the angles On3ꢀPnꢀNn1 (n ¼ 1, 2), which are
widened to 122.64(5) and 123.11(5)ꢁ respectively. The axial
distances PnꢀOn3 (n ¼ 1 or 2), 1.6177(9) Å and 1.6198(9) Å, as
stated for the related compounds, are shorter than the two other
equatorial distances, P1ꢀO1n and P2ꢀO2n (n ¼ 1 or 2) (Table 2) [9].
In the crystal lattice, two adjacent molecules are alternately stabi-
lised by intermolecular hydrogen bonds NꢀH/O (Fig. 2, Table 3).
The presence of such bonds has also been determined for analogous
symmetrical phosphorus compounds, e.g. HP(OCH2CH2NH)2,
HP(OC6H4NH)2, HP(OCOCHRNH)2 R ¼ CH3, C3H7 [10e12].
n
(NH2) ¼ 3126e3239 cmꢀ1, which is in accordance with the liter-
ature data published for similar moieties [8]. Complexation of the
ligand, remarkably seen in the IR spectrum (lack of
n
(PꢀH)
stretching vibration), is also manifested in 31P{1H} NMR analysis.
The spectra of 2 and 3 exhibit resonances at 90.70 and 134.85 ppm
respectively.
2.2. X-ray structural determination
Single crystal X-ray diffraction analyses reveal that the stereo-
chemistry of ligand 1 and complex 2 are nearly identical as those
expected from the spectroscopic studies. The molecular structures
of 1 and 2 are shown in Figs. 1 and 3 respectively. Details of the
structural determination as well as a list of selected bond lengths
Table 1
Crystal data for HP(OCMe2CMe2O)(OCH2CMe2NH) 1 and [PdCl2{P(OCMe2CMe2O)
OCH2CMe2NH2}] 2.
The molecules of 2 are specified by phosphorus and nitrogen
atoms of the chelating H-spirophosphorane ligand and two chlo-
rides in cis positions. The coordination sphere of the palladium
atom in [PdCl2{P(OCMe2CMe2O)OCH2CMe2NH2}] is close to square
planar. The phosphorus atoms, Pn (n ¼ 1, 2), subtly deviate from the
plane defined by the Nn1, Cln1, Cln2, (n ¼ 1, 2) atoms by 0.089 and
1
2
Empirical formula
Formula weight
Temperature (K)
C
10H22NO3P
C10H22Cl2NO3PPd
412.56
100(2)
235.26
100(2)
l
(Å)
0.71073
Orthorhombic
Pbca
14.744(3)
18.658(4)
18.807(4)
5174(2)
16
0.71073
Orthorhombic
Pbcn
14.056(4)
16.508(4)
27.388(6)
6355(3)
16
Crystal system
Space group
a (Å)
b (Å)
c (Å)
Table 4
Relevant bond distances (Å) and angles (ꢁ) for [PdCl2{P(OCMe2CMe2O)OCH2C-
Volume (Å3)
Z
Me2NH2}] 2.
Dcalc (g cmꢀ3
)
1.208
1.725
Bond lengths
Pd1eCl11
Pd1eCl12
Pd2eCl21
(mmꢀ1
F(000)
Crystal size (mm)
range(ꢁ)
Number of collected reflection
)
0.20
2048
0.58 ꢂ 0.40 ꢂ 0.12
3.0e27.5
53841
1.60
3328
0.25 ꢂ 0.12 ꢂ 0.10
3.0e27.5
49696
2.3810(9)
2.2866(9)
2.365(2)
Pd1eP1
Pd2eP2
2.1888(9)
2.195(2)
2.070(3)
2.068(3)
m
Pd1eN11
Pd2eN21
Pd2eCl22
2.2805(9)
q
Bond angles
Cl12ePd1eCl11
Cl22ePd2eCl21
P1ePd1eCl12
P2ePd2eCl22
P1ePd1eCl11
P2ePd2eCl21
92.98(3)
91.99(4)
88.87(4)
87.65(4)
178.15(3)
178.47(3)
N11ePd1eP1
N21ePd2eP2
N11ePd1eCl12
N21ePd2eCl22
N11ePd1eCl11
N21ePd2eCl21
93.07(8)
95.06(8)
176.95(7)
177.28(8)
85.08(8)
85.29(8)
Independent reflection (Rint.
)
5939
7301
S
1.02
0.87
R1/wR2 indices
0.033/0.091
0.031/0.064
R1 ¼ SjjFoj ꢀ jFcjj/SjFoj
wR2 ¼ {S[w(Fo ꢀ Fc2)2]/[w(Fo2)2]}1/2
2
2
w ¼ 1/[
s
2(Fo2) þ (aP)2] where P ¼ (Fo þ 2Fc2)/3, a ¼ 0.0602 (1), 0.0316 (2).

_
ꢀ
A. Skarzynska, A. Gniewek / Journal of Organometallic Chemistry 696 (2011) 2985e2992
2988
Table 5
Product distribution in the Heck coupling reactions of bromobenzene and butyl acrylate catalysed by the palladium complexes 2 and 3.
Br
+
R
[Pd]
R
R
+
n
DMF, NaHCO3, [ Bu4N]Br
2
a
b
R
Catalyst 2
Catalyst 3
Yield % of (a)
Yield % of (b)
Yield % of (a)
Yield % of (b)
C(O)OBu
70
30
72
28
[Pd-complex] 1.35$10ꢀ5 mol; NaHCO3 4.4$10ꢀ3 mol; [nBu4N]Br 2.3$10ꢀ3 mol; PhBr (4.36$10ꢀ3 mol); CH2]CHC(O)OBu 1.9$10ꢀ3 mol; 140 ꢁC, 4 h; mesitylene as internal
standard.
Table 6
Product distribution in the Heck coupling reactions of bromobenzene and substituted styrenes catalysed by complexes 2 and 3.
Br
[Pd]
+
n
DMF, [ Bu4N]Br, Cs2CO3
R
R
R = Cl, Me, OMe
Entry
RC6H4CH]CH2
Catalyst 2
Catalyst 3
Conversion
of PhBr (%)
Yield (%) of
trans-stilbene
Conversion
of PhBr (%)
Yield (%) of
trans-stilbene
1
2
3
4
5
3-MeC6H4CH]CH2
4-MeC6H4CH]CH2
2-ClC6H4CH]CH2
4-ClC6H4CH]CH2
77
69
68
91
69
74
66
64
88
66
91
83
82
98
99
88
81
81
94
95
4-OMeC6H4CH]CH2
[Pd-complex] 1.35$10ꢀ5 mol; base Cs2CO3 4.4$10ꢀ3 mol; [nBu4N]Br 2.3$10ꢀ3 mol; PhBr 1.89$10ꢀ3 mol; RC6H4CH]CH2 2.12$10ꢀ3 mol; 140 ꢁC; 4 h; mesitylene as internal
standard.
0.051 Å respectively. More remarkable deviations from the ideal
square planar values, 90 and 180ꢁ, are exhibited by the bond angles
at the metal centre (see Table 4). The six-membered metal chelate
cycles are fixed to half-chair conformations, while the five-
membered phosphorus cycles exhibit envelope conformations.
The nature of the donor atoms: phosphorus or nitrogen, seems to
have a profound effect on the PdꢀCl bonds. The PdꢀCln1 bond
lengths of 2.3810(9) and 2.365 (2) Å are slightly longer than those of
PdꢀCln2, 2.2866(9) and 2.2805(9) Å (n ¼ 1, 2), due to a greater trans
influence of the phosphorus atom with respect to the nitrogen
atom. The PdꢀNn1 bond lengths of 2.070(3) and 2.068(3) Å are
similar to those found for complexes of the formula PdCl2(P w N),
as they are associated with sp3-hybridised amino nitrogen atoms
[13e15]. Other bonding parameters in 2 do not differ significantly
from the expected values reported for the related complexes.
2.3. Catalytic application of complexes 2 and 3
As previously outlined, palladium(II) complexes incorporating
hydrospirophosphorane ligands appeared to be excellent
100
80
60
40
Hg(0) added at t = 20 min.
20
Hg(0) added at t = 0 min.
0
0
50
100
150
200
250
time (minutes)
Fig. 4. Catalyst recycling for the Heck coupling of bromobenzene with 4-Cl-styrene
and the catalyst precursors [PdCl2{P(OCMe2CMe2O)OCH2CMe2NH2}] 2 and [PdCl(C3H5)
{P(OCMe2CMe2O)OCMe2CMe2NH2}] 3.
Fig. 5. Reaction profile study of the Heck cross-coupling of 4-Cl-styrene and bromo-
benzene using 2 as a precatalyst (reaction conditions as stated above) and a poisoning
tests with Hg(0) as an inhibitor.

_
ꢀ
A. Skarzynska, A. Gniewek / Journal of Organometallic Chemistry 696 (2011) 2985e2992
2989
a
30
d
= 7.8 nm
m
25
20
15
10
5
FWHM = 2.8 nm
0
2
1
1
3
4
5
6
7
8
9
10 11 12 13 14 15
diameter [nm]
60
50
40
30
20
10
0
b
d
= 3.1 nm
m
FWHM = 1.5 nm
2
3
4
5
6
7
8
9
10
diameter
c
60
50
40
30
20
10
0
d
= 5.1 nm
m
FWHM = 1.8 nm
2
3
4
5
6
7
8
9
10
diameter
Fig. 6. TEM images and corresponding size histograms (100 nanoparticles measured) of (a) post-catalytic solution using complex 2 as the precatalyst (b) nanoparticles [Pd2]
synthesised from complex 2 under hydrogen atmosphere in THF (c) post-catalytic solution using [Pd2] as a precatalyst.
precatalysts for the Heck and Hiyama cross-coupling reactions
[2,3]. This observation directed our interest to the catalytic prop-
erties of complexes 2 and 3. The methodology of the Heck process
was utilised as optimised before. The coupling reactions were
efficiently performed at an elevated temperature of 140 ꢁC, over
a relatively short reaction time of 4 h, in the presence of the polar
aprotic solvent DMF, which remains the favoured medium for the
Heck reaction. Since a positive effect of an ionic liquid has been
reported in the literature, [nNBu4]Br was used as an additive,
leading to remarkable rate enhancements. Different terminal
alkenes, n-butyl acrylate and substituted styrenes, were chosen as
substrates for the development of the bromobenzene coupling. The
obtained results showed that both H-spirophosphorane complexes,
2 and 3, are very active in the Heck cross-coupling. As seen from

_
ꢀ
A. Skarzynska, A. Gniewek / Journal of Organometallic Chemistry 696 (2011) 2985e2992
2990
Table 7
Product distribution in the Hiyama coupling reactions of bromobenzene and substituted styrylsilanes catalysed by complexes 2 and 3.
Br
[Si]
[Pd]
n
+
DMF, [ Bu4N]F
R
R
[Si] = SiMe2OSiMe3 or Si(OEt)3
R = Cl, Me, OMe
Entry
RC6H4CH]CH[Si]
Catalyst 2
Catalyst 3
Conversion
of PhBr (%)
Yield (%) of
trans-stilbene
Conversion
of PhBr (%)
Yield (%) of
trans-stilbene
1
2
3
4
5
3-MeC6H4CH]CHSi(OEt)3
98
99
99
100
98
96
98
93
97
97
98
99
97
100
99
96
97
91
98
96
4-MeC6H4CH]CHSiMe2OSiMe3
2-ClC6H4CH]CHSiMe2OSiMe3
4-ClC6H4CH]CHSiMe2OSiMe3
4-OMeC6H4CH]CHSiMe2OSiMe3
[Pd-complex] 1.35$10ꢀ5 mol; PhBr 1.31 ꢂ 10ꢀ3 mol, ArCH ¼ CH[Si] 1.44 ꢂ 10ꢀ3 mol; activator [nBu4N]F 2.14$10ꢀ3 mol; 140 ꢁC; 4 h; mesitylene as internal standard.
Table 5, the arylation of n-butyl acrylate performed in the presence
of NaHCO3, in the case of both 2 and 3, proceeds smoothly with
100% conversion to produce mainly a monosubstituted product, i.e.
butyl cinnamate a, in comparable amounts for both precatalysts.
We found Cs2CO3 to be a superior base over NaHCO3 for the
reaction of bromobenzene and substituted styrenes as coupling
partners. The relevant results showing a high degree of selectivity
are presented in Table 6. Incorporation of an allyl group into the
molecule of 3 instead of one chloride ligand had a beneficial effect.
[PdCl(C3H5){P(OCMe2CMe2O)OCH2CMe2NH2}] was found to be
a more effective precatalyst compared to 2, for which lower yields
of trans-stilbenes were produced. The reaction tolerates a variety of
functionalities. The best effects for precatalyst 3 were obtained for
surface, is evidence of heterogeneous catalysis [16c] as well as the
presence of Pd(0) species in the system [16b]. In our case, the
addition of an excess of mercury (260 equivalents) before the
reaction and after 20 min (at this time the yield is 54%) causes the
suppression of catalytic activity (Fig. 5). This outcome shows that,
besides homogenous species, also Pd(0) species might be involved
in this catalytic process. Actually, both the formation of a dark
reaction solution and the presence of small Pd(0) nanoparticles
observed in TEM micrographs in the post-catalysis liquid phase
(Fig. 6a) confirmed this conjecture. The mean particle diameter
and size distribution were calculated by counting 100 particles
from the enlarged micrographs. The size distribution plot was
fitted by means of
a log-normal curve approximation. The
a
styrene bearing both electron-withdrawing and electron-
obtained nanoparticle size distribution is practically symmetric
with a well-defined maximum at 7.8 nm and a relatively narrow
FWHM of 2.8 nm. Most of the Pd(0) particles are round-shaped;
however, some of them form quite irregular aggregates. The
nanoparticles obtained in situ during the Heck reaction from
complex 2 might be additionally stabilised with phosphorous
ligands. Energy dispersive X-ray microanalysis (EDX) showed that
the Pd:P ratio is about 20:1 in the case of the particles presented
in Fig. 6a. To further explore the nature of the phosphorus
moieties, complex 2 and other Heck reaction components in
a 1:10 ratio were placed in an NMR tube. The 31P NMR spectra
were recorded in deuterated DMF at an elevated temperature
(373 K) for 2 h. During that time only one signal was observed at
donating substituents in para position, 4-ClC6H4 and 4-
OMeC6H4 respectively (entry 4 and 5).
a
Recycling experiments were carried out in reactions of bromo-
benzene and 4-chlorostyrene (Fig. 4). They were conducted using
two methods: extraction of organic products with ethyl ether and
loading a new portion of substrates, [nBu4N]Br, and the base Cs2CO3
or filtration of solid components and product extraction from the
liquid phase. Similar results were obtained in both cases: after the
first run, the catalyst activity degraded and the yield decreased
noticeably. This finding (as stated below) might suggest either the
aggregation of nanoparticles formed in situ in the system or the
metal leaching from the NPs.
To get a better understanding of the mechanism of the reac-
tion, we made an attempt to define the nature of the true active
species. A variety of experiments or tests have been used to
determine the essence of the active species in palladium catalysed
reactions [16]. There is no single clear-cut experiment to distin-
guish between homogeneous catalysts and heterogeneous ones.
Therefore, we carried out some mechanistic studies. The reaction
profile curve of product yield against time (Fig. 5) indicates that
the process starts immediately, and over 50% yield of the appro-
priate stilbene is achieved within 20 min. The lack of an induction
period as well as sigmoidal kinetics, characteristic of heteroge-
neous catalysts formed in situ from monometallic precatalysts,
point to the contribution of homogenous catalyst. However, the
induction interval may be too short to be detected in some cases
for heterogeneous catalysis. The widely used test to identify the
nature of the catalyst is the poisoning mercury test, however it is
not universally applicable [16c,17]. The inhibition of catalysis by
Hg(0), which amalgamates the metal catalyst or adsorbs on its
d
¼ 10.61 ppm, which might suggest the formation of a new non-
innocent phosphorous ligand involved in the catalytic process. In
order to understand the effect of the formation of PdNPs on
catalytic activity we have synthesised nanoparticles [Pd2] starting
from complex [PdCl2{P(OCMe2CMe2O)OCH2CMe2NH2}] 2 under
a hydrogen atmosphere in THF (1 atm, r.t., 56 h) adopting the
procedure previously described [17b]. The transmission electron
microscopy analysis of [Pd2] shows the presence of regularly
dispersed NPs of spherical shape with an average size of 3.1 nm,
exhibiting the tendency to partial organisation (Fig. 6b). [Pd2]
was efficiently used as a catalytic precursor in cross-coupling
reaction between bromobenzene and 4-Cl-styrene (140 ꢁC, 4 h,
2.5 mol% Pd) to give 4-Cl-stilbene in 86% yield. The TEM analysis
of post-catalytic solution (Fig. 6c) reveals that the NPs remained
still dispersed but the mean size increases to 5.1 nm. Performed
experiments might suggest involvement of PdNPs in catalytic
process which could serve as a reservoir of soluble Pd(0) active
species [18].

_
ꢀ
A. Skarzynska, A. Gniewek / Journal of Organometallic Chemistry 696 (2011) 2985e2992
2991
The catalytic activity of the complexes 2 and 3 was compared
4.2. Synthesis of the ligand HP(OCMe2CMe2O)(OCH2CMe2NH) 1
in the Hiyama coupling of bromobenzene and functionalised
styrylsilanes. Like in the Heck procedure, the experiments were
performed under aerobic conditions at 140 ꢁC in DMF over 4 h,
but [nBu4N]F was employed as an activator. The presence of both
electron-withdrawing and electron-donating groups was toler-
ated in the coupling under these conditions. The reaction gave
almost exclusively substituted trans-stilbenes with excellent
yields exceeding 96%, except when an electron-withdrawing
substituent was present in the ortho position in an aromatic
ring (Table 7, entry 3).
The ligand precursor HP(OCMe2CMe2O)(OCH2CMe2NH)
2,2,3,3,8,8-hexamethyl-1,4,6-trioxa-9-aza-5
5-phosphaspiro[4.4]
l
nonane 1 was obtained using the general procedure described in
the literature [22,7]. A mixture of P(NMe2)3 (4.36 cm3, 24 mmol)
and pinacol (Me2COH)2 (2.81 g, 24 mmol) was placed in a round-
bottomed flask equipped with a magnetic stirrer and a reflux
condenser and heated gradually to 140 ꢁC over 3 h with gentle
nitrogen purging. The progress of the reaction was monitored by
GCeMS analyses. The lack of pinacol in the system indicated the
end of the first stage, and next 2-amino-2-methyl-1-propanol
(2.29 cm3, 24 mmol) was added to the cooled mixture. Further
stirring with purging over 2 h at 140 ꢁC led to a final white solid.
The crude product was purified by recrystallisation from hexane
in a freezer providing crystals suitable for X-ray analysis. Yield:
2.5 g (44%).
3. Conclusions
The unsymmetrical H-spirophosphorane ligand HP(OCMe2C-
Me2O)(OCH2CMe2NH) 1, readily obtained from reaction of hex-
amethylphosphoramide with 2-amino-2-methyl-propanol-1 and
pinacol, leads to palladium(II) mononuclear complexes with
a chelating P,N coordination mode. In contrast to the stable
complex [PdCl2{P(OCMe2CMe2O)OCH2CMe2NH2}] 2, the allyl
Anal. Calc. for C10H22NO3P: C, 51.05; H, 9.43; N, 5.95 Found: C,
50.82; H, 9.78; N, 5.84%; 1H NMR (C6D6)
d 0.95, 0.98, 1.14, 1.15, 1.16,
1.25 (18H, s’s, CH3), 2.78 (1H, d, 2J(P, H) ¼ 17.5 Hz, NH), 3.54 (1H, q,
2J(Ha, He) ¼ 8.4 Hz, 3J(P, He) ¼ 51.6 Hz, CH2), 3.57 (1H, q, 2J(Ha,
He) ¼ 8.4 Hz, 3J(P, Ha) ¼ 41.3 Hz, CH2), 7.49 (1H, d 1J(P, H) ¼ 772.5 Hz,
analogue
[PdCl(C3H5){P(OCMe2CMe2O)OCH2CMe2NH2}]
3
undergoes hydrolytic cleavage connected with the rupture of
coordinated spirophosphorane and the formation of the phospho-
nate compound [PdCl{P(O)(OCMe2CMe2O)}HOCH2CMe2NH2] 4.
The catalytic potential of complexes 2 and 3 was tested in CeC
coupling reactions. The presented results show that under similar
reaction conditions the complexes are relatively high-yielding and
stereoselective catalysts for Heck and Hiyama cross-couplings. Very
small disparity was observed in catalytic performance on changing
from unhindered to sterically hindered styrenes or styrylsilanes
utilised as substrates. Little effect was also noticed when using
electron-withdrawing or electron-donating substrates. Finally,
differences in the yield of trans-stilbenes observed in the Heck
coupling imply to some extent that subtle changes of the pre-
catalyst structure may influence the reaction progress. The mech-
anistic investigations indicate that, besides homogenous
precatalysts, Pd(0) nanoparticles formed in situ in the system might
also be involved in the catalytic process.
PH); 31P{1H} NMR (CD2Cl2)
d
ꢀ49.21, 148.45 ppm; IR nmax(nujol)/
cmꢀ1 716vs, 924vs, 977vs, 1047s, 1160vs
n
(CeOeP), 2362s
n
(PꢀH),
3350s
n
(NꢀH); ESI-MS: m/z calcd for [M þ H]þ 236.14 found 236.14.
4.3. Synthesis of the complexes
4.3.1. [PdCl2{P(OCMe2CMe2O)OCMe2CMe2NH2}] 2
[PdCl2(cod)] (0.19 g, 0.67 mmol) was added to a solution of
HP(OCMe2CMe2O)(OCH2CMe2NH) (0.24 g, 1.02 mmol) in dichloro-
methane (4 cm3). The yellow suspension was stirred for 20 min to
produce a yellow solution. The crude product was precipitated with
ethyl ether. The obtained precipitate was filtered, washed with
ethyl ether and dried in vacuo. Yield: 0.25 g (91%) Single crystals
suitable for X-ray analysis were obtained from acetone solution by
slow evaporation.
Anal. Calc. for C10H22Cl2NO3PPd: C, 29.11; H, 5.37; N 3.39 Found:
C, 29.84; H, 5.43; N, 3.38%; 1H NMR (CD2Cl2)
d 1.34, 1.54, 1.58 (18H,
s’s, CH3), 3.93 (2H, br s, NH2), 4.01 (2H, d, 3J(P, H) ¼ 18.7 Hz, CH2); 31
P
NMR (CD2Cl2)
d
90.70 ppm; IR nmax(nujol)/cmꢀ1 279vs, 289vs
4. Experimental section
n
(PdꢀCl), 916s, 958vs, 1031s, 1136s
n
(CꢀOꢀP), 3126m, 3198m
n
(NꢀH2); ESI-MS: m/z calcd for [M þ Na]þ 433.96 found 433.96.
4.1. General procedure
4.3.2. [PdCl(C3H5){P(OCMe2CMe2O)OCMe2CMe2NH2}] 3
[Pd(
Chemicals and deuterated solvents were purchased from Sig-
maeAldrich and Fluka and used as received. All preparations were
performed in an atmosphere of dry, oxygen-free nitrogen, using
conventional Schlenk techniques. Solvents were carefully dried and
m
ꢀCl)(C3H5)]2 (0.13 g, 0.35 mmol) was added to a solution of
HP(OCMe2CMe2O)(OCH2CMe2NH) (0.21 g, 0.89 mmol) in toluene
(4 cm3). The yellow solution was stirred for 30 min at room
temperature, and during that time a white precipitate started to
form in the system. The obtained product was filtered, washed with
ethyl ether and dried in vacuo. Yield: 0.25 g (84%) Anal. Calc. for
C13H27ClNO3PPd: C, 37.34; H, 6.51; N 3.35 Found: C, 37,56; H, 6,55;
deoxygenated by standard methods [19]. [PdCl2(cod)] and [Pd(
Cl)(C3H5)]2 were synthesised as previously reported [20,21].
m-
IR and FIR measurements were performed in KBr or Nujol with
a Bruker 113V FTIR. 1H, 31P{1H} NMR spectra were obtained on
Bruker Avance 500 MHz spectrometer (500 MHz for 1H NMR). The
N, 3.36%; 1H NMR (CDCl3)
br s, CH2), 3.95 (4H, br d, CH2), 5.01 (br s NH2), 5.65 (1H, m, CH); 31
d 1.26, 1.34, 1.42 (18H, s’s, CH3), 3.12 (2H,
P
chemical shifts (
(
d
) are given in ppm towards TMS (1H) and H3PO4
NMR (CDCl3)
460m, 469m,
d
134.85 ppm; IR nmax(nujol)/cmꢀ1 290vs
n
(PdꢀCl),
31P) using deuterated solvents as lock and reference (1H) respec-
n
(PdꢀC), 925vs, 951s, 1136m
n
(CꢀOꢀP), 3163m,
tively. Elemental analyses were performed on a 2400 CHNS Vario EL
III apparatus. Analytical gas chromatographic (GC) analyses were
performed on a Hewlett Packard 5890. Conversion of the substrates
was calculated using the internal standard method. MS spectra
were measured on an ESI Finnigan Mat TSQ 700 instrument. TEM
measurements were carried out using a FEI Tecnai G2 20 X-TWIN
electron microscope (TEM) operating at 200 kV. Specimens for TEM
studies were prepared by putting a droplet of a suspension on
a copper microscope grid covered with a perforated carbon film.
3239m
382.08.
n
(NꢀH2); ESI-MS: m/z calcd for [M ꢀ Cl]þ m/z 382.08; found
4.4. Crystal structure determinations
The data were collected on a KM4CCD diffractometer and cor-
rected for Lorentz and polarisation effects. Data reduction was
carried out using the Oxford Diffraction (Poland) programs. The
structures were solved by direct method using SHELXS and refined

_
ꢀ
A. Skarzynska, A. Gniewek / Journal of Organometallic Chemistry 696 (2011) 2985e2992
2992
by the full-matrix least-squares method on F2 using the SHELXL
software [23]. Non-hydrogen atoms were refined with anisotropic
thermal parameters. The hydrogen atoms were set in calculated
positions and refined using the riding model with a common fixed
isotropic thermal parameter.
Appendix. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jorganchem.2011.05.008.
References
4.5. Representative Heck reaction procedure
[1] (a) P.W.N.M. van Leeuwen, P.C.J. Kamer, C. Claver, O. Pámies, M. Diéguez,
Chem. Rev. 111 (2011) 2077e2118;
The Heck reactions were carried out under a nitrogen atmo-
sphere in a 50 cm3 Schlenk tube equipped with a magnetic stirrer.
In a typical experiment, the flask was charged with the reagents:
the catalyst (1.35 ꢂ 10ꢀ5 mol), bromobenzene PhBr 0.46 cm3
(4.36 ꢂ 10ꢀ3 mol) and n-butyl acrylate CH2]CHC(O)OBu 0.27 cm3
(1.9 ꢂ 10ꢀ3 mol) in DMF as a solvent (3 cm3) containing mesitylene
as an internal standard. [nBu4N]Br 0.75 g (2.3 ꢂ 10ꢀ3 mol) and the
base NaHCO3 0.37 g (4.4 ꢂ 10ꢀ3 mol) were used as additives. The
mixture was heated to 140 ꢁC and stirred for 4 h. After that time, the
reaction mixture was cooled and organic products were separated
by extraction with diethyl ether (3 times with 7 cm3), washed with
water, dried over MgSO4 and analysed by GCeMS. The structures of
the synthesised (E)-Stilbenes were confirmed by GCeMS and NMR
spectroscopy, matching data reported in the literature [24].
(b) A. Marson, J.E. Ernsting, M. Lutz, A.L. Spek, P.W.N.M. van Leeuwen,
P.C.J. Kamer, Dalton Trans. (2009) 621e633;
(c) Y. Mata, M. Diéguez, O. Pámies, C. Claver, Org. Lett. 7 (2005) 5597e5599;
(d) K.N. Gavrilov, O.G. Bondarev, A.V. Korostylev, A.I. Polosukhin, V.N. Tsarev,
N.E. Kadilnikov, S.E. Lyubimov, A.A. Shiryaev, S.V. Zheglov, H.-J. Gais,
V.A. Davankov, Chirality 15 (2003) S97eS103;
(e) K.N. Gavrilov, O.G. Bondarev, A.I. Polosukhin, Russ. Chem. Rev. 73 (2004)
671e699.
_
ꢀ
[2] A. Skarzynska, A.M. Trzeciak, M. Siczek, Inorg. Chim. Acta 365 (2011)
204e210.
_
ꢀ
[3] A. Skarzynska, A.M. Trzeciak, Synthesis submitted.
[4] R. Burgada, R. Setton, F.R. Hartley (Eds.), The Chemistry of Organophosphorus
Compounds, 3, Willey Interscience, New York, 1993, p. 185.
[5] B.A. Arbuzov, N.A. Polezhaeva, Russ. Chem. Rev. 43 (1974) 933e973.
[6] K.N. Gavrilov, A.I. Polosukhin, O.G. Bondarev, S.E. Lyubimov, K.A. Lyssenko,
P.V. Petrovskii, V.A. Davankov, J. Mol. Catal. A: Chem. 196 (2003) 39e53.
[7] R. Burgada, C. Laurenco, J. Organomet. Chem. 66 (1974) 255e270.
[8] A.V. Korostylev, O.G. Bondarev, K.A. Lyssenko, A.Yu. Kovalevsky,
P.V. Petrovskii, G.V. Tcherkaev, I.S. Mikhel, V.A. Davankov, K.N. Gavrilov, Inorg.
Chim. Acta 295 (1999) 164e170.
[9] V.F. Zheltukhin, A.I. Dovbysh, D.N. Sadkova, A.B. Dobrynin, O.N. Kataeva,
I.A. Litvinov, V.A. Alfonsov, Russ. Chem. Bull. Int. Ed. 54 (2005) 1935e1938.
[10] P.F. Meunier, R.O. Day, J.R. Devillers, R.R. Holmes, Inorg. Chem. 17 (1978)
3270e3276.
[11] L. Yu, Z. Liu, H. Fang, Q.L. Zeng, Y.F. Zhao, Amino Acids 28 (2005) 369e372.
[12] L. Yu, Z. Liu, H. Fang, Y.-F. Zhao, Acta Crystallogr. E61 (2005) o261eo262.
[13] N.D. Jones, S.Jo. Ling Foo, B.O. Patrick, B.R. James, Inorg. Chem. 43 (2004)
4056e4063.
[14] K.N. Gavrilov, O.G. Bondarev, R.V. Lebedev, A.A. Shiryaev, S.E. Lyubimov,
A.I. Polosukhin, G.V. Grintselev-Knyazev, K.A. Lyssenko, S.K. Moiseev,
N.S. Ikonnikov, V.N. Kalinin, V.A. Davankov, A.V. Korostylev, H.-J. Gais, Eur. J.
Inorg. Chem. (2002) 1367e1376.
[15] A.V. Korostylev, O.G. Bondarev, S.Y. Lyubimov, A.Yu. Kovalevsky,
P.V. Petrovskii, V.A. Davankov, K.N. Gavrilov, Inorg. Chim. Acta 303 (2000)
1e6.
4.6. Hiyama reaction procedure
The Hiyama reactions were carried out under a nitrogen atmo-
sphere in a 50 cm3 Schlenk tube equipped with a magnetic stirrer.
In a typical experiment, the flask was charged with the reagents:
the catalyst (1.31 ꢂ 10ꢀ5 mol), bromobenzene PhBr 0.14 cm3
(1.31 ꢂ 10ꢀ3 mol), ArCH]CH[Si] 1.44 ꢂ 10ꢀ3 mol in DMF as
a solvent (2 cm3) containing mesitylene as an internal standard.
The activator [nBu4N]F (2.14 ꢂ 10ꢀ3 mol) was used as an additive.
The mixture was heated to 140 ꢁC and stirred for 4 h. After that time
the reaction mixture was cooled and organic products were sepa-
rated by extraction with diethyl ether (3 times with 7 cm3), washed
with water, dried over MgSO4, and analysed by GCeMS.
[16] (a) M. Diéguez, O. Pámies, Y. Mata, E. Teuma, M. Gómez, F. Ribaudo,
P.W.N.M. van Leeuwen, Adv. Synth. Catal. 350 (2008) 2583e2598;
(b) N.T.S. Phan, M. Van Der Sluys, C.W. Jones, Adv. Synth. Catal. 348 (2006)
609e679;
(c) J.A. Widegren, R.G. Finke, J. Mol. Catal. A: Chem. 198 (2003) 317e341;
(d) J. Li, M. Lutz, A.L. Spek, G.P.M. van Klink, G. van Koten, R.J.M. Klein Gebbink,
J. Organomet. Chem. 695 (2010) 2618e2628.
Acknowledgements
Financial support of the Polish Ministry of Science and Higher
Education (Grant No N204 028538) is gratefully acknowledged. The
authors are grateful to Dr Mariusz Majchrzak for preparing styr-
ylsilanes and Ms Marzena Dejerling for testing the Heck and
Hiyama reactions.
[17] (a) G.M. Whitesides, M. Hackett, R.L. Brainard, J.-P.P.M. Lavalleye,
A.F. Sowinski, A.N. Izumi, S.S. Moore, D.W. Brown, E.M. Staudt, Organome-
tallics 4 (1985) 1819e1830;
(b) F. Fernández, B. Cordero, J. Durand, G. Muller, F. Malbosc, Y. Kihn, E. Teuma,
M. Gómez, Dalton Trans. (2007) 5572e5581.
[18] C.C. Cassol, A.P. Umpierre, G. Machado, S.I. Wolke, J. Dupont, J. Am. Chem. Soc.
127 (2005) 3298e3299.
[19] D.D. Perrin, W.L.F. Armarego, Purification of Laboratory Chemicals, third ed.
Pergamon Press, Oxford, 1998.
[20] D. Drew, J.R. Doyle, Inorg. Synth. 13 (1972) 52.
Appendix A. Supplementary material
[21] Y. Tatsuno, T. Yoshida, S. Otsuka, Inorg. Synth. 28 (1990) 342e345.
[22] D. Houalla, M. Sanchez, L. Beslier, R. Wolf, Org. Magn. Reson. 3 (1971) 45e74.
[23] G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112e122.
[24] (a) Y. Leng, F. Yang, K. Wei, Y. Wu, Tetrahedron 66 (2010) 1244e1248;
(b) R. Cella, A. Stefani, Tetrahedron 62 (2006) 5656e5662 and references
therein.
CCDC 810570e810572 contain the supplementary crystallo-
graphic data for compound 1, 2 and 4, respectively. These data can
be obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.