About the different reactivity of dinuclear palladium and platinum compounds with trispyrrolylphosphine: Synthesis and X-ray crystallographic results of new palladium complexes containing P-pyrrolyl bonds

Article abstract of DOI:10.1016/j.jorganchem.2007.05.038

The reaction of [Pt(μ-Cl)(κ,η²-COE-MeO)]₂ (2) (COE-MeO = 2-methoxy-5-cycloocten-1-yl) with trispyrrolylphosphine yields [PtCl{P(pyrl)₃}(κ,η²-COE-MeO)] (3), in contrast to the crown-cycle complex [Pd(μ-Cl) {P(pyrl)₃}]₈ (1) obtained with the analogous palladium complex. The different reactivity has been explained in terms of the higher lability of the carbon-carbon double bond in the starting palladium compound, as compared with the related platinum derivative. The reaction of 1 with the ligand 2-[(pyridin-2-ylmethylene)aminophenol], (HNN′O), in the presence of thalium salt and NEt₃, yields the complex [Pd(κ³N,N′,O-py-CH{double bond, long}N-C₆H₄O)(P(O)(pyrl)₂)] (5), which contains, for the first time, a di(N-pyrrolyl)phosphonato-P ligand. Treatment of [PdCl(κ³N,N′,O-py-CH{double bond, long}N-C₆H₄O)] with P(pyrl)₃ gives the amido derivative [PdCl{κ³P,N,N-(P(pyrl)₂-O-C₆H₄-N-CH(CH₂-CO-CH₃)-py)}] (7), which displays a N-Pd-P-O-C₂ six-membered metallacycle. In addition, the latter compound has been able to insert an acetone molecule into its framework. The crystal structures of 5 and 7 were solved by X-ray diffraction analysis.

Full text of DOI:10.1016/j.jorganchem.2007.05.038

Journal of Organometallic Chemistry 692 (2007) 3882–3891
www.elsevier.com/locate/jorganchem
About the different reactivity of dinuclear palladium
and platinum compounds with trispyrrolylphosphine:
Synthesis and X-ray crystallographic results
of new palladium complexes containing
P–pyrrolyl bonds
a
a
a
´
Inma Angurell , Isabel Martınez-Ruiz , Oriol Rossell ,
a,
b
*
´
Miquel Seco , Pilar Gomez-Sal ,
Avelino Martın , Merce Font-Bardia , Xavier Solans
b
c
c
´
´
a
` `
´
Departament de Quı´mica Inorganica, Universitat de Barcelona, Martı i Franques 1-11, E-08028 Barcelona, Spain
Departamento de Quı´mica Inorga´nica, Universidad de Alcala´, E-28071 Alcala´ de Henares, Spain
b
c
`
Departament de CristalÆlografia, Mineralogia i Diposits Minerals, Universitat de Barcelona,
`
Martı´ i Franques s/n, 08028 Barcelona, Spain
Received 4 May 2007; received in revised form 21 May 2007; accepted 21 May 2007
Available online 2 June 2007
Abstract
The reaction of [Pt(l-Cl)(j,g2-COE-MeO)]2 (2) (COE-MeO = 2-methoxy-5-cycloocten-1-yl) with trispyrrolylphosphine yields
[PtCl{P(pyrl)3}(j,g2-COE-MeO)] (3), in contrast to the crown-cycle complex [Pd(l-Cl) {P(pyrl)3}]8 (1) obtained with the analogous pal-
ladium complex. The different reactivity has been explained in terms of the higher lability of the carbon–carbon double bond in the starting
palladium compound, as compared with the related platinum derivative. The reaction of 1 with the ligand 2-[(pyridin-2-ylmethylene)ami-
nophenol], (HNN0O), in the presence of thalium salt and NEt3, yields the complex [Pd(j3N,N0,O-py-CH@N-C6H4O)(P(O)(pyrl)2)] (5),
which contains, for the first time, a di(N-pyrrolyl)phosphonato-P ligand. Treatment of [PdCl(j3N,N0,O-py-CH@N-C6H4O)] with P(pyrl)3
gives the amido derivative [PdCl{j3P,N,N-(P(pyrl)2-O-C6H4-N-CH(CH2-CO-CH3)-py)}] (7), which displays a N–Pd–P–O–C2 six-mem-
bered metallacycle. In addition, the latter compound has been able to insert an acetone molecule into its framework. The crystal structures
of 5 and 7 were solved by X-ray diffraction analysis.
ꢀ 2007 Elsevier B.V. All rights reserved.
Keywords: Palladium complexes; Trispyrrolylphosphine; Imine; Phosphonate; Insertion
1. Introduction
P(pyrl)3 (pyrl = N-pyrrolyl) is provided, for example, by
the complete substitution of CO in the electron-rich anion
[Rh(CO)4]ꢀ to give [Rh{P(pyrl)3}4]ꢀ [3]. Nevertheless, only
a few examples of coordination compounds of transition
metals stabilized with P(pyrl)3 are known [1,4]. In particu-
lar, the only palladium example described so far has been
obtained in our group by making the dinuclear r/p palla-
dium compound [Pd(l-Cl)(j,g2-COE-MeO)]2 react with
P(pyrl)3 (Eq. (1)) [5].
There is ongoing interest in the study of N-pyrrolyl
phosphines due to their strong electron withdrawing char-
acter, and the catalytic implications of this behaviour [1,2].
Evidence for the exceptional p-accepting properties of
*
Corresponding author.
E-mail address: miquel.seco@qi.ub.es (M. Seco).
0022-328X/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.05.038

I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
3883
ð1Þ
The molecular structure of 1 consists of an eight-palla-
dium(I) crown-cycle containing unsupported Pd–Pd bonds.
This unprecedented geometry prompted us to analyse the
process carefully in order to obtain mechanistic information
about its formation and this work is reported in the present
paper. In this context, the reaction of P(pyrl)3 with the
related platinum complex, [Pt(l-Cl)(j,g2-COE-MeO)]2,
has proved crucial for understanding the formation of 1.
In addition, we have explored the reaction of the octapalla-
dium cluster 1 with the chelating agent 2-[(pyridin-2-ylmeth-
ylene)aminophenol] (HNN0O), which is known to act also
as a tridentate ligand [6], in an attempt to build new supra-
molecular structures. The resulting compound displayed an
unusual P(O)(pyrl)2 ligand. Finally, we describe, for the first
time, the insertion of a P(pyrl)2 moiety into the Pd–O bond
of [PdCl(j3-NN0O)], to form a six-membered palladacycle,
whose structure has been revealed by X-ray crystallography.
The white colour of 3 revealed that no compound, analo-
gous to the palladium crown-cycle, had been obtained,
given that the vast majority of platinum compounds con-
taining metal–metal bonds are strongly coloured. Elemental
analyses, NMR spectroscopy and MS MALDI-TOF spec-
trometry confirmed this hypothesis showing that 3 is the
mononuclear species depicted in Eq. (2). Its formation is
easily explained in terms of the rupture of the Pt–Cl bridging
bonds of the dinuclear complex by the incoming phosphine.
The IR spectrum of 3 showed a strong band at 1184 cmꢀ1
,
assigned to mC@C ring breathing pyrrole, confirming the
1
presence of the P(pyrl)3 ligand. In the H NMR spectrum,
the pyrrolyl protons were visible as multiplet resonances
at 6.43 and 7.08 ppm. These chemical shifts appeared at
slightly lower field values than those recorded for the free
P(pyrl)3 (6.35 and 6.78 ppm). The signals of the COE-
MeO group did not undergo significant shifting from those
observed in the starting complex 2. The 13C NMR spectrum
of 3 showed two resonances of the P(pyrl)3 carbon atoms: a
doublet at 113.9 ppm (3JCP = 8.4 Hz) and a second doublet
at 124.8 ppm (2JCP = 8.2 Hz). The most striking feature in
the 31P NMR spectrum is the signal at 68.6 ppm flanked
by 195Pt satellites with a JPPt constant of 6546 Hz. In gen-
eral, Pt(II) complexes containing functionalized N-pyrrolyl
phosphines present JPPt values in the range 3800–4800 Hz,
although there are some derivatives with other phosphine li-
gands that show higher values, examples being
[PtCl2{P(pyrl)2(N-C7H5N)}2] (JPPt = 5409 Hz) [1e] and
[PtCl2{PF2(NMe2)}2] (JPPt = 5690 Hz) [7]. The large P–Pt
coupling constant of 6546 Hz is the highest reported so far
for a Pt(II) complex, suggesting that the phosphine and
the double carbon–carbon bond are both in trans. Complete
assignation of signals in the NMR spectra has been achieved
by NOESY 1H–1H, COSY 1H–1H and HSCQ 1H–13C
experiments (see Supplementary material).
2. Results and discussion
2.1. Palladium and platinum compounds containing P(pyrl)3
Our first goal in this paper was to investigate whether
the platinum complex [Pt(l-Cl)(j,g2-COE-MeO)]2 (2)
reacts with P(pyrl)3 like the analogous palladium com-
pound does. To this end, we first prepared 2 as a white
crystalline powder by reaction of NaMeO with [PtCl2-
1
(COD)] in methanol at room temperature. The H, 13C
NMR and IR spectra confirmed the nature of the product.
Next, 2 was reacted with two equivalents of P(pyrl)3 in ace-
tone at ꢀ10 ꢁC (Eq. (2)). The process was monitored by
31P NMR spectroscopy. The reaction was very clean and
the 31P NMR signal at 78.8 ppm due to free P(pyrl)3 com-
pletely disappeared and a new resonance emerged at
68.6 ppm corresponding to the new compound 3. After
3 h of stirring the volume of the solution was reduced
and a crystalline solid precipitated by addition of hexane.
Compound 3 is rather unstable in solution, precluding
the grow of single crystals for an X-ray crystal structure
determination. The rate of decomposition is increased by
addition of an excess of phosphine. One of the decomposi-
tion products was the complex cis-[PtCl2{P(pyrl)3}2] (4). As
4 had not been previously described we synthesized it fol-
lowing the classical procedure involving the reaction of
[PtCl2(COD)] with P(pyrl)3 in CH2Cl2 at room tempera-
ture. Compound 4 was characterized by elemental analyses
ð2Þ

3884
I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
IR and NMR spectroscopy. Note that in this complex the
JPPt constant is 4801 Hz.
metallic palladium. This result is explained by assuming
that 1 underwent the rupture of the Pd–Cl bridging bonds,
giving unstable Pd(I) complexes which are known to show
disproportionation to metal palladium and Pd(II) com-
plexes. In other words, the complex [PdCl(j,g2-COE-
MeO)]2 firstly reacts with P(pyr)3 to give 1 and then, only
after the addition of an excess of phosphine, undergoes
cleavage of the Pd–Cl bridging bonds. This behaviour con-
trasts with that observed for the platinum complex. In
summary, we conclude that the formation of the octacycl-
opalladium cluster 1 is due to the higher lability of the
coordinated C–C double bond compared with that of the
Pd–Cl bonds in [Pd(l-Cl)(j,g2-COE-MeO)]2.
The formation of 3 from 2 and P(pyrl)3 shed light on the
hypothetical mechanism underlying the formation of 1. It
is clear that the great difference in both reactions is that
the platinum derivative 2 resists the COE-MeO displace-
ment by the phosphine, while the related palladium com-
plex [Pd(l-Cl)(j,g2-COE-MeO)]2 does not. Evidently, the
ease of the COE-MeO labilization depends on the bonding
energy of the metal–carbon bond, it being known that this
is stronger for the platinum–carbon bond. For this reason,
in the process involving the platinum complex 2 and
P(pyrl)3 we only observed cleavage of the platinum–chlo-
ride bond to yield 3. However, in the case of the palladium
complex, the P(pyrl)3 prefers to displace the COE-MeO
ligand, with the central Pd2Cl2 system remaining unaltered.
Under these conditions, the intermediate species that now
contain the one-electron r-COE-MeO ligand was formed.
This species could then undergo the homolytic rupture of
the M–C bond, followed by the formation of four Pd–Pd
bonds to give the cyclo-octanuclear palladium(I) complex
1 (Scheme 1). This behaviour is supported by the observa-
tion of a signal in the ESI mass spectrum of the reaction
solution at m/z = 279 assigned to the [(COE-MeO)2H]+
species.
According to this hypothesis, supramolecular structures
containing a different number of {pyrl}3PPd–PdP{pyrl}3
units, for example three or six, cannot be ruled out,
although we have not been able to detect them. It is likely,
as observed for simple supramolecules [8], that the square
molecular species is the one preferred from a thermody-
namic point of view. Finally, the question arises as to the
effect that the addition of an excess of P(pyrl)3 produces
on 1. In fact, solutions of 1 rapidly turned orange from
red to yield [PdCl2(P(pyrl)3)2], which appeared impure with
2.2. Reaction of 1 with 2-[(pyridin-2-ylmethylene)amino-
phenol]
Having described above that an excess of P-donor
ligand produces the decomposition of 1, we became inter-
ested in studying the effect of the singular ligand N-donor
2-[(pyridin-2-ylmethylene)aminophenol] (HNN0O) in order
to determine whether this ligand, able to act as a bis- or tri-
schelate, could stabilize other unusual geometries. In fact,
some palladium complexes containing the ligand as a
bidentate or tridentate mode have recently been described
[6].
Compound 1 was allowed to react with HNN0O, in the
presence of NEt3 to guarantee the anionic form of the
ligand (NN0O) in CH2Cl2, at 0 ꢁC. The solution rapidly
turned purple and metallic palladium precipitated. After
filtration, purple crystals of [Pd(j3N,N0,O-py-CH@N-
C6H4O)(P(O)(pyrl)2)] (5) could be isolated (Eq. (3)). Com-
pound 5 was fully characterized spectroscopically and its
structure was unambiguously confirmed by a single X-ray
crystal structure determination.
Scheme 1.

I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
3885
ð3Þ
The 1H NMR spectrum revealed the protons that corre-
spond to the iminic group (8.38 ppm) and to the HCa-pyr-
idine ligand (8.25 ppm). The signal at 7.30 ppm due to the
OH group of the free ligand was not observed. The 31P
NMR spectrum showed one signal at 41.4 ppm, far from
the expected value for coordinated P(pyrl)3 (in the region
of about 90 ppm) and thus suggesting a significant modifi-
cation on the phosphine ligand. The X-ray structure
revealed that the ligand is actually the di(N-pyrrolyl)phos-
phonato-P group, –P(O)(pyrl)2. The IR spectrum of 5
showed a sharp band at 1186 cmꢀ1 assignable to the
P@O stretch and the mass spectrum (MALDI-TOF)
showed a signal at m/z 483 [M]+ confirming the nature of
the product. This is the first time that the –P(O)(pyrl)2
group appears to be bonded to a metal centre as part of
a coordination compound. The phosphine oxidation is
believed to originate from adventitious water, similar to
that reported in the formation of some diphosphoxane–
rhodium complexes [9]. Since the process described here
supposes the cleavage of Pd–Cl bonds, we repeated the
reaction in the presence of TlBF4, as a halide abstractor
and, under these conditions, the reaction proceeded much
more rapidly. In addition, the absence of chloride ligands
in 5, as well as the precipitation of palladium metal, sug-
gests that the first step of this process would involve the
displacement of the chloride from 1 by the NN0O ligand,
followed by rupture of the metal–metal bond to show
finally disproportionation of Pd(I) to Pd(II) and Pd(0). In
conclusion, treatment of 1 with HNN0O in basic medium
leads to the decomposition of the Pd8 crown-cycle structure
to yield mononuclear square planar palladium(II) com-
plexes, in which three coordination sites are occupied by
the trischelate ligand and the fourth by the di(N-pyrr-
olyl)phosphonato-P group.
Fig. 1. Molecular structure of 5.
Table 1
3
0
˚
Selected bond length (A) and selected bond angles (ꢁ) of [Pd(j N,N ,O-py-
CH@N-C6H4O)(P(O)(pyrl)2)] (5) and [PdCl{j3P,N,N-(P(pyrl)2-O-C6H4-
N-CH(CH2-CO-CH3)-py)}] (7)
5
7
˚
Bond lengths (A)
Pd1–P1
Pd1–O2
Pd1–N4
Pd1–N3
O2–C1
N4–C6
N4–C7
C7–C8
C8–N3
2.2478(7)
2.0278(19)
2.013(2)
2.056(2)
1.334(3)
1.404(3)
1.284(4)
1.457(4)
1.371(3)
1.415(4)
1.4862(19)
1.721(2)
1.719(2)
Pd–Cl
2.3033(7)
2.084(2)
2.0178(18)
2.1701(7)
1.502(4)
1.476(3)
1.387(3)
1.404(3)
1.5963(18)
1.538(4)
1.510(4)
1.477(5)
1.201(4)
Pd–N2
Pd–N1
Pd–P
C11–C7
C7–N1
N1–C6
C1–O1
P–O1
C7–C8
C8–C9
C9–C10
C9–O2
C1–C6
P1–O1
P1–N1
P1–N2
Bond angles (ꢁ)
P1–Pd1–N3
N3–Pd1–N4
N4–Pd1–O2
O2–Pd1–P1
Pd1–N3–C8
Pd1–N4–C7
Pd1–N4–C6
Pd1–O2–C1
C6–N4–C7
100.71(6)
Cl–Pd–P
97.01(3)
96.23(6)
82.82(8)
2.3. Crystal structure of 5
80.53(9)
82.67(8)
96.07(6)
111.10(17)
115.97(18)
112.74(18)
109.86(17)
131.3(2)
Cl–Pd–N2
N2–Pd–N1
N1–Pd–P
Pd–N1–C7
Pd–N1–C6
Pd–P–O1
N3–P–O1
N4–P–O1
N3–P–N4
Pd–P–N3
Pd–P–N4
C6–N1–C7
The molecular structure of compound 5 was solved by
X-ray crystal diffraction methods. Compound 5 crystallizes
with dichloromethane. Thus, in the asymmetric unit of the
unit cell a molecule of compound 5 and half molecule of
dichloromethane are present. Fig. 1 shows the molecular
structure of this compound and in Table 1 a selection of
bond distances and angles are listed. The metal centre
has a distorted square-planar coordination geometry, with
the ligand behaving as NN0O-tridentate, and with the
84.03(6)
107.31(14)
111.44(15)
111.20(7)
103.19(11)
101.80(11)
101.67(12)
121.35(9)
115.23(9)
117.8(2)

3886
I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
˚
di(N-pyrrolyl)phosphonato-P group coordinated trans to
the iminic nitrogen atom N(4).
H(21) with [C(21)ꢁ ꢁ ꢁO(1)] = 3.143(4) A (angle C(21)–
H(21)–O(1) = 149.4(3)ꢁ). Another weak hydrogen bond
connects the O(1) of one ligand with the hydrogen atom
H(11) of the neighbouring molecule ([C(11)ꢁ ꢁ ꢁO(1)] =
The whole complex molecule, with the obvious exceptions
of the pyrrolyl rings, is planar (maximum displacement less
˚
˚
than 0.04 A) and two five-membered chelate rings are
3.288(3) A and C(11)–H(11)–O(1) = 136.25(2)ꢁ).
formed upon coordination. The bite angles involved in the
chelation are typical for these tridentate systems. Thus, for
example, the N(3)–Pd(1)–N(4) and the N(4)–Pd(1)–O(2)
angles appear at 80.53(9)ꢁ and 82.67(8)ꢁ, respectively. The
2.4. Reaction of [PdCl(j3-NN0O)] with P(pyrl)3
Having shown that the P(pyrl)3 displaces the double
C@C bond more easily than the bridging chloride in
[Pd(l-Cl)(j,g2-COE-MeO)]2, we were interested to deter-
mine the behaviour of those derivatives containing simulta-
neously two or more sites potentially vulnerable to attack
by the P(pyrl)3. For this purpose, the complex [PdCl-
(j3-NN0O)] (6), having simultaneously Pd–Cl, Pd–N and
Pd–O bonds, was thought to be a good target. Given that
6 had only been described in solution [6], we improved its
synthesis by making [PdCl2(COD)] react with NN0OH in
THF to give [PdCl2(j2-NN0OH)] (see Section 4). After
treatment with n-BuLi in THF, a deep blue solution was
obtained, from which purple crystals of 6 were isolated.
The nature of 6 was determined by NMR spectroscopy
and elemental analyses. The 1H NMR spectrum of 6
showed a signal at 8.34 ppm due to a-HC pyridine protons,
which was shifted at high fields with respect to that
observed in the j2-NN0 coordinated species. The iminic
proton also shifted from 8.70 to 8.44 ppm. Moreover, the
complete disappearance of the OH group confirmed the
three coordination mode for the NN0O ligand.
˚
Pd(1)–N(4) bond distance, 2.013(2) A, is practically identical
to that found in [Pd(j3-NN0O)(Me)] [6], suggesting strong
trans influence of the –P(O)(pyrl)2 ligand. The Pd(1)–O(2)
˚
distance, 2.0278(19) A, is in agreement with those found in
related hydrazone complexes [6] and the C(7)–N(4) distance
˚
of 1.284(4) A is characteristic of a –C@N– iminic system. On
the other hand, the palladium–di(N-pyrrolyl)phosphoryl
˚
bond length [Pd(1)–P(1), 2.2478(7) A] is significantly shorter
and falls towards the lower end of the wide range of Pd–PR3
bond distances. In contrast, the P(1)–N(1) distance,
˚
˚
1.721(2) A and P(1)–N(2) 1.719(2) A, is rather large. The
˚
P(1)–O(1) distance is 1.4862(19) A and corresponds to a
P@O double bond. No other species with this di(N-pyrr-
olyl)phosphonato-P ligand has been previously described,
and only three compounds with the P(O)N2 group bound
to a transition metal have been structurally described,
although in one of them the phosphoryl oxygen [10] is also
coordinated to another metal, and thus their structural
parameters cannot be compared. In the other two examples,
involving metals belonging to the first transition series (Fe,
˚
Co), the M–P bond lengths (3.081 and 2.265 A respectively)
When a solution of 6 in acetone was treated with
P(pyrl)3 a dramatic change in colour was observed and,
after 12 h of stirring, a deep red solution was obtained.
The work-up rendered the unexpected complex [PdCl-
{j3P,N,N-(P(pyrl)2-O-C6H4-N-CH(CH2-CO-CH3)-py)}]
(7), (Eq. (4)), which was identified spectroscopically and by
X-ray structure determination.
˚
[11,12] are greater than the Pd–P distance (2.2478(7) A)
found in 5. This is consistent with the idea that for
LnM–PZx systems, such as the M–P bond distance decreases
(because of increased M–P back bonding), the P–Z lengths
increase due to partial loss of back donation into P–Z r*
orbitals.
It is interesting to note that the crystal structure of this
compound is also maintained for weak interactions due to
both the dichloromethane solvent molecules and (N-pyrr-
olyl)phosphonato ligands (Fig. 2). The solvent molecules
occupy a special position in a binary axis, and bridge two
molecules of 5 through a hydrogen bond O(1)ꢁ ꢁ ꢁC(21)–
ð4Þ
The 31P NMR spectrum indicated the existence of one
signal for the phosphorus nucleus at low field (d = 111.8
ppm) in comparison with the free P(pyrl)3 ligand, but far
from the average values observed for Pd–P(pyrl)3 com-
1
plexes. Furthermore, the H NMR spectrum showed the
shifting of the iminic proton from 8.44 to 5.38 ppm, con-
firming the absence of a C@N double bond. The appear-
ance of new signals at 1.94 and 3.04 ppm, assignable to
the acetone group, was also observed. The IR spectrum
showed bands at 1723 and 1020 cmꢀ1 due to the mC@O
and mP–O stretching, respectively. As revealed by the
Fig. 2. Intermolecular hydrogenbonds in 5.

I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
3887
X-ray crystal structure determination (see below) the most
relevant features of 7 are the formal insertion of the
P(pyrl)2 moiety into the Pd–O bond, to give a six-mem-
bered cycle, and the formation of a C–C bond, through
the incorporation of an acetone molecule.
The formation of 7 can be rationalized in the following
terms. The first step would consist of the replacement of the
chloride ligand by the phosphine. The second would
involve the rupture of a pyrrolyl unit, followed by insertion
of the P(pyrl)2 moiety into the Pd–O bond (Scheme 2). The
formation of a strong P–O bond would be the driving force
of this process. In addition, the basic pyrrolyl group, which
is a good leaving group, could promote the deprotonation
of the acetone molecule and the attack by this species on
the N@C double bond would thus yield the amido ligand
displayed in 7.
In conclusion, the formation of 7 could be the result of
the preferential attack of the P(pyrl)3 on the Pd–Cl against
the Pd–O or Pd–N bonds.
2.5. Crystal structure of 7
Fig. 3. Molecular structure of 7.
The molecular structure of 7 is shown in Fig. 3 while the
most relevant geometric features are listed in Table 1. The
metal distorted square-planar coordination observed com-
prises the ligand [NN0O] modified by the formal insertion
of a P(pyrl)2 fragment between the oxygen atom and the
metal centre. This conforms a six-membered ring contain-
ing four heteroatoms (N, P, O, Pd) and a five-membered
[NN0] chelate ring containing the pyridinic group. Both
N-pyridinic and P atoms are in trans, while the amido
N(1) share the two fused rings and occupies the third coor-
dination site. A chloride trans to amido N(1) completes the
coordination around the metal.
envelop form with the N(1) being the out-of-plane atom.
It is noteworthy that the Pd–N(1) distance of
2.0178(18) A is shorter than those reported for previously
˚
described amidopalladium complexes [13]. The C(7)–N(1)
˚
distance of 1.476(3) A corresponds to a C–N single bond
in contrast to that observed in 5 for the C@N iminic bond
˚
(1.284(4) A). Although the C(6)–N(1)–C(7) angle is
117.8(2)ꢁ, the sum of the three angles at N(1) is 336.4ꢁ,
close to a tetrahedral geometry, and this suggests that the
lone pair is localized on the sp3 nitrogen atom. This con-
trasts with the values found in structure 5 and in other pal-
ladium amido complexes, which show a roughly planar
geometry [13]. The P(pyrl)2 fragment is linked to the Pd
The N(1)–Pd–N(2) angle of 82.82(8)ꢁ is similar to those
found in some five-membered palladacycles, shaping an
Scheme 2.

3888
I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
˚
and to O(1) atoms with a P–O(1) distance of 1.5963(18) A
corresponding to a P–O single bond. The pyrrolyl groups
are tilted to the oxygen atom (N(3)–P–Pd and N(4)–P–Pd
angles are 121.35(9)ꢁ and 115.23(9)ꢁ, respectively) and the
˚
N(1)–C(6) distance of 1.387(3) A is a little shorter than
the average values found for N–Caryl bonds. This might
be due to the high strain associated with the generation
of a five- and six-membered ring fused together and sharing
a palladium atom with square-planar geometry. Finally,
one of the most relevant features in 7 is the incorporation
of an acetone molecule into the system through the C(7)
iminic carbon atom.
Scheme 3.
4.2. Synthesis of [PtCl(j,g2-COE-OMe)]2 (2)
3. Conclusion
A solution 0.07 M of NaOMe in methanol (23 mL, 1.61
mmol) was added to [PtCl2(COD)]2 (0.600 g, 1.60 mmol)
and the mixture was stirred at room temperature for 2 h.
The solvent was removed under vacuum, and the residue
was re-dissolved in CH2Cl2 and filtered off. The solvent
was evaporated to dryness and the solid washed with
diethyl ether to obtain [PtCl(j,g2-COE-OMe)]2 as a col-
ourless solid. Yield: 0.509 g, 85%. dH (250 MHz; CDCl3):
The reaction of [PdCl(j,g2-COE-OMe)]2 with P(pyrl)3
affords the eight-palladium(I) crown-cycle compound 1,
while its platinum congener 2 gives the mononuclear com-
plex 3, as a result of the easy cleavage of the platinum–chlo-
ride bond. The different behaviour in both cases has been
ascribed to the higher lability of the C@C double bond in
[PdCl(j,g2-COE-OMe)]2. Compound 1 reacts with the pyr-
idine-aminophenol, (HNN0O), to yield the mononuclear
palladium complex [Pd(j3-NN0O)(P(O)(pyrl)2)] (5), which
contains for the first time a di(N-pyrrolyl)phosphonato-P
group as ligand. On the other hand, [PdCl(j3-NN0O)] (6)
reacts with P(pyrl)3 in acetone to give the six-membered
2
4.73 (2H, s, JHPt = 76 Hz, H5, H6); 3.49 (1H, m, H2);
3.22 (3H, s, H9); 3.0–1.0 (9H, m). dC (62.9 MHz; CDCl3):
101.9 (s, C5, C6); 82.8 (s, C2); 58.1 (s, C9); 35.7 (s, C3);
1
32.8 (s, C4); 31.1 (s, JCPt = 486 Hz, C1); 30.0 (s, C7); 29.7
(s, C8). IR: mmax/cmꢀ1 1087vs, 1073s (C–O). Anal. Calc.
for (C18H30Cl2O2Pt2): C, 29.23; H, 4.09. Found: C, 28.87;
H, 2.55%.
metallacycle
[PdCl{j3-P,N,N-(P(pyrl)2-O-C6H4-N-CH-
(CH2-CO-CH3)-py)}] (7), in which an acetone molecule
has been inserted in the iminic carbon atom. The molecular
structure of 7 confirms the higher lability of the Pd–Cl bond
in 6 in comparison with the Pd–O or Pd–N ones.
4.3. Synthesis of [PtCl(j,g2-COE-OMe)(P(pyrl)3)] (3)
A solution of P(pyrl)3 (0.156 g, 0.68 mmol) in 2 mL of
dichloromethane was added to a stirred solution of
[PtCl(COE-OMe)]2 (0.250 g, 0.34 mmol) in 20 mL of cool
acetone (ꢀ10 ꢁC). The mixture was stirred for a further
2 h at ꢀ10 ꢁC and the resulting yellowish solution was
evaporated to dryness. The residue was re-crystallized in
acetone/n-hexane. The desired product was isolated as a
colourless solid by filtration, washed with n-hexane and
dried under vacuum. Yield: 0.324 g, 80%. dP (101 MHz;
CD2Cl2): 68.6 ppm (1JPPt = 6546 Hz). dH (400 MHz;
CD2Cl2): 7.08 (6H, m, H10); 6.54 (2H, m, H5, H6); 6.43
(6H, m, H11); 3.13 (1H, m, H2); 2.88 (4H, m, H8, H9);
2.64 (1H, m, H8); 2.31 (2H, m, H4); 2.09 (1H, m, H3);
1.99 (3H, m, H7, H3); 1.75 (1H, m, H1). dC (100 MHz;
4. Experimental
4.1. General
All reactions were carried out under an atmosphere of
dry nitrogen using standard Schlenk techniques. All sol-
vents were distilled from appropriate drying agents under
1
N2 prior to use. H, 13C{1H}, 31P{1H} and 195Pt NMR
spectra were recorded at 25 ꢁC on Bruker 250 and Mercury
400 spectrometers. Chemical shifts are reported in ppm rel-
1
ative to external standards (SiMe4 for H and 13C, 85%
H3PO4 for 31P) and coupling constants are given in Hz.
Infrared spectra were recorded on a FT-IR 520 Nicolet
spectrophotometer in the 4000–400 cmꢀ1 range as KBr pel-
lets. MS (ESI) spectra were recorded on a Fisons VG Qua-
tro spectrometer. MALDI-TOF spectra were recorded on a
Voyager DE-RP (Perspective Biosystems) time-of-flight
(TOF) spectrometer.
2
2
CD2Cl2): 124.8 (d, JCP = 8.2 Hz, C10); 120.3 (d, JCP
=
1
2
13.8 Hz, JCPt = 81 Hz, C5 or6); 119.4 (d, JCP = 17.3 Hz,
1JCPt = 79 Hz, C5 or 6); 113.9 (d, JCP = 8.4 Hz, C11); 82.9
3
1
(s, C2); 56.1 (s, C9); 35.2 (s, C3); 31.3 (s, JCPt = 601 Hz,
C1); 29.9 (sbr, C7, C8); 26.8 (s, C4). MS (MALDI DTH-
THF): m/z 566.3 [MꢀOMe]+, 563.3 [MꢀCl]+. IR: mmax
/
The starting materials [PtCl2(COD)]2 [14], [PdCl2-
(COD)]2 [15], P(pyrl)3 [3] and 2-[(pyridin-2-ylmethyl-
ene)aminophenol] (HNN0O) [6] were prepared following
published procedures. Other reagents were purchased from
commercial suppliers. Atom numbering in the NMR data
are indicated in Scheme 3.
cmꢀ1 3126w (CH), 2966–2818w (CH), 1468m (C@C),
1184s (C–N), 1091–1058s (P–N–C), 742s (CH). Anal. Calc.
for (C21H27ClN3OPPt): C, 42.11; H, 4.54; N, 7.01. Found:
C, 42.04; H, 4.45; N, 6.87%. 1H–13C HSQC, 1H–1H COSY,
1H–1H NOESY: see supporting data in Supplementary
material.

I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
3889
4.4. Synthesis of [PtCl2{P(pyrl)3}2] (4)
68%. dH (250 MHz; DMSO-d6): 8.44 (1H, br s, CH@N);
8.34 (1H, d, J = 4.4 Hz, H1); 8.13 (1H, t, J = 7.5 Hz, H3);
7.75 (1H, d, J ꢂ 7.5 Hz, H4); 7.59 (1H, m, H2); 7.37 (1H,
d, J = 7.9 Hz, H5); 7.03 (1H, t, J ꢂ 7.2 Hz, H6); 6.45 (2H,
m, H7, H8). IR: mmax/cmꢀ1 3020w (CH), 1589m (C@N),
1475, 1451m (C@C), 740s (CH). Anal. Calc. for
(C12H9ClN2OPd): C, 42.51; H, 2.67; N, 8.25. Found: C,
42.59; H, 2.60; N, 8.09%.
To a stirred solution of P(pyrl)3 (0.260 g, 1.13 mmol) in
10 mL of dichloromethane at room temperature was added
[PtCl2(COD)] (0.207 g, 0.55 mmol). The solution was stirred
for 30 min and the addition of n-hexane gave the desired
product. Compound 4 was isolated by filtration as a colour-
less powder. Yield: 0.270 g, 67%. dP (101 MHz; CDCl3): 51.4
(1JPPt = 4801 Hz). dH (250 MHz; CDCl3): 6.81 (12H, m),
2
6.28 (12H, m). dC (62.9 MHz; CDCl3): 124.8 (t, JCP
=
4.8. Synthesis of [PdCl{j3-P,N,N-(P(pyrl)2-O-C6H4-N-
CH(CH2–CO–CH3)-py)}] (7)
3
4.1 Hz), 114.6 (t, JCP = 4.7 Hz). dPt (53.8 MHz; CDCl3):
1
ꢀ4258.7 (t, JPPt = 4800 Hz). IR: mmax/cmꢀ1 3137w (CH),
1457m (C@C), 1184vs (C–N), 1069vs (P–N–C), 726vs
(CH). Anal. Calc. for (C24H24Cl2N6P2Pt): C, 39.79; H,
3.34; N, 11.59. Found C, 39.83; H, 3.40; N, 11.37%.
A solution of P(pyrl)3 (0.036 g, 0.16 mmol) in 2 mL of
acetone was added to a solution of compound [PdCl(g3-
NN0O)] (0.053 g, 0.16 mmol) in 10 mL of the same solvent
and the mixture was stirred over night. The solvent was
evaporated to dryness and the crude was extracted 3 times
with 15 mL of diethyl ether. Removal of the solvent fol-
lowed by drying under vacuum gave 7 as a red solid. Yield:
0.038 g, 43%. Crystals suitable for X-ray analyses were
obtained by slow diffusion of n-hexane in a CH2Cl2 solu-
tion of compound 7. dP (101 MHz; CDCl3): 111.8. dH
(250 MHz; CDCl3): 8.95 (1H, m, H1); 7.87 (1H, t, H3);
7.56 (1H, d, H2); 7.36 (3H, br s, H4, pyrl); 7.14 (2H, m,
pyrl); 6.98 (2H, m, H5, H6); 6.83 (1H, d, H8); 6.65 (1H, t,
H7); 6.47 (2H, m, pyrl); 6.39 (2H, m, pyrl); 5.38 (1H, m,
4.5. Synthesis of [Pd(j3-NN0O)(P(O)(pyrl)2)] (5)
A solution of 1 (0.063 g, 0.212 mmol) in 15 mL of
CH2Cl2 was added to a solution of HNN0O (0.051 g,
0.257 mmol) and NEt3 (6 mL, 43 mmol) in 15 mL of
CH2Cl2 at 0 ꢁC. The mixture was stirred for 1 h, and TlBF4
(0.050 g, 0.626 mmol) was added. The solution that resulted
was filtered off and evaporated to dryness. The purple resi-
due was washed with diethyl ether and re-crystallized in
CH2Cl2/hexane to obtain 5 as a purple solid. Yield:
0.025 g, 60%. dP (101 MHz; CDCl3): 41.4. dH (250 MHz;
CDCl3): 8.38 (s, CH@N), 8.25 (d, H1), 8.20–6.5 (m, H2–8),
7.33 (m, pyrl), 6.27 (m, pyrl). IR: mmax/cmꢀ1 3100–3033w
(CH), 1581m (C@N), 1488, 1454m (C@C), 1186s (P@O),
1078–1038s (P–N–C), 754s (CH). MS (MALDI DTH-
THF): m/z 483.3 [M]+. Anal. Calc. for (C20H17N4O2PPd):
C, 49.76; H, 3.55; N, 11.60. Found C, 49.83; H, 3.40; N,
11.47%.
CHN); 3.04 (2H, m, CH2); 1.94 (3H, s, CH3). IR mmax
/
cmꢀ1: 3120 w (CH), 2966m (CH), 1723m (C@O), 1602m
(C@N), 1481, 1461m (C@C), 1259s (C–N), 1065vs (P–N–
C), 1020m (P–O), 803s (CH). MS ESI(+): m/z = 523.0
[MꢀCl]+. Anal. Calc. for (C23H22ClN4O2PPd): C, 49.39;
H, 3.96; N, 10.11. Found C, 49.20; H, 3.87; N, 9.82%.
4.9. X-ray structure determination of 5
4.6. Synthesis of [PdCl2(j2-NN0OH)]
Crystallographic and experimental details of the crystal
structure determinations are given in Table 2. A suitable
crystal of complex 5 was covered with mineral oil and
mounted in the N2 stream of a Bruker-Nonius Kappa
CCD diffractometer and data were collected using graph-
To a stirred solution of [PdCl2(COD)] (0.040 g, 0.14
mmol) in 50 mL of tetrahydrofuran at room temperature
was added the ligand HNN0O (0.028 g, 0.14 mmol) and
the mixture was stirred for 2 h. The brown solid obtained
was filtered off and dried under vacuum. Yield: 0.050 g,
95%. dH (250 MHz; DMSO-d6): 10.03 (1H, br s, OH);
9.04 (1H, d, J = 5.6 Hz, H1); 8.70 (1H, s, CH@N); 8.38
(1H, t, J = 7.7 Hz, H3); 8.18 (1H, d, J = 7.6 Hz, H4); 7.93
(1H, d, J ꢂ 6.8 Hz, H2); 7.15 (2H, m, H5, H6); 6.89 (1H,
˚
ite-monochromated Mo Ka radiation (k = 0.71073 A).
Data collections were performed at low temperature
150 K with an exposure time of 6 s per frame (13 sets;
1274 frames). Raw data were corrected for Lorenz and
polarization effects. The structure was solved by direct
methods, completed by the subsequent difference Fourier
techniques and refined by full-matrix least-squares on F2
(SHELXL-97) [16]. Anisotropic thermal parameters were used
in the last cycles of refinement for the non-hydrogen atoms.
The hydrogen atoms were included from geometrical calcu-
lations and refined using a riding model. All the calcula-
tions were made using the WinGX system [17].
d, J = 8.5 Hz, H8); 6.81 (1H, t, J = 7.6 Hz, H7). IR: mmax
/
cmꢀ1 3262vs (O–H), 3050w (CH), 1594m (C@N), 1505,
1456m (C@C), 764s (CH).
4.7. Synthesis of [PdCl(j3-NN0O)] (6)
To a solution of compound [PdCl2(g2-NN0OH)] (0.147
g, 0.39 mmol) in 30 mL of tetrahydrofuran was added
250 lL of n-BuLi 1.6 M (0.40 mmol) and the mixture was
stirred 1 h at room temperature. The purple solid obtained
was filtered off and dried under vacuum. Yield: 0.090 g,
4.10. X-ray structure determination of 7
A prismatic crystal of 7 was selected and mounted
on a MAR345 diffractometer with image plate detector.

3890
I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
Table 2
(Generalitat de Catalunya) (Project 2001SGR 00054).
I.A. is indebted to the Ministerio de Ciencia y Tecnologia
for a scholarship.
Crystal data and structure refinement for 5 and 7
Compound
Empirical formula C20H17ClN4O2PPd Æ 0.5CH2Cl2 C23H22ClN4O2PPd
5
7
Formula weight
(g/mol)
525.21
559.27
Appendix A. Supplementary material
Colour
Purple
Red
CCDC 629341 and 628858 contain the supplementary
crystallographic data for 5 and 7. These data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/
conts/retrieving.html, or from the Cambridge Crystallo-
graphic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: (+44) 1223-336-033; or e-mail: deposit@
ccdc.cam.ac.uk. Supplementary data associated with this
article can be found, in the online version, at
doi:10.1016/j.jorganchem.2007.05.038.
Crystal size (mm) 0.3 · 0.28 · 0.2
Crystal system
0.1 · 0.1 · 0.2
Orthorhombic
Pbca
Monoclinic
C12/c1
4105.9(9)
8
Space group
3
˚
V (A )
4620.4(6)
8
Z
Unit cell dimensions
˚
a (A)
20.598(2)
8.2548(14)
24.1512(12)
90
9.0060(10)
17.2090(10)
29.8120(10)
90
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
90.922(12)
90
c (ꢁ)
Index ranges
90
90
ꢀ26 6 h 6 26, ꢀ10 6 k 6 10,
0 6 h 6 12,
0 6 k 6 25,
0 6 l 6 43
Mo Ka
1.608
References
ꢀ31 6 l 6 31
´
[1] (a) V. Rodrıguez, B. Donnadieu, S. Sabo-Etienne, B. Chaudret,
Radiation type
Mo Ka
1.699
1.138
Organometallics 17 (1998) 3809–3814;
Dcalc (Mg/m3)
`
(b) J. Castro, A. Moyano, M.A. Pericas, A. Riera, M.A. Maestro, J.
l (mmꢀ1
)
1.016
´
Mahıa, Organometallics 19 (2000) 1704–1712;
h Range (ꢁ)
k (A)
F(000)
T (K)
Goodness-of-fit
Data collected
Unique observed
data [I > 2r(I)]
3.14–27.51
0.71073
2104
150(2)
1.048
2.46–31.61
0.71073
2256
293(2)
1.242
(c) C.D. Andrews, A.D. Burrows, J.M. Lynam, M.F. Mahon, M.T.
Palmer, New J. Chem. 25 (2001) 824–826;
(d) J. Huang, C.M. Haar, S.P. Nolan, W.J. Marshall, K.G. Moloy,
J. Am. Chem. Soc. 120 (1998) 7806–7815;
(e) A.D. Burrows, M.F. Mahon, M. Varrone, Dalton Trans. (2003)
4718–4730.
˚
62592
4720 (0.1231)
44591
7281 (0.0390)
´
[2] (a) A.M. Trzeciak, T. Glowiak, R. Grzybek, J.J. Ziolkowski, J.
Chem. Soc., Dalton Trans. (1997) 1831–1837;
(b) A.M. Trzeciak, T. Glowiak, J.J. Ziolkowski, J. Organomet.
Chem. 552 (1998) 159–164;
(c) A.M. Trzeciak, B. Borak, Z. Ciunik, J.J. Ziolkowski, M.F.C.
Guedes da Silva, A.J.L. Pombeiro, Eur. J. Inorg. Chem. (2004) 1411–
1419.
(Rint
)
´
Absorption
correction
Parameters
refined
Final R indices
[I > 2r(I)]
R indices (all
data)a
None
None
293
´
271
R1 = 0.0313, wR2 = 0.0755
R1 = 0.0400, wR2 = 0.0796
R1 = 0.0458,
wR2 = 0.0885
R1 = 0.0499,
[3] K.G. Moloy, J.L. Petersen, J. Am. Chem. Soc. 117 (1995) 7696–7710.
[4] (a) C.E.F. Rickard, W.R. Roper, S.D. Woodgate, L.J. Wright, J.
Organomet. Chem. 643–644 (2002) 168–173;
wR2 = 0.0899
(b) C. Babij, C.S. Browning, D.H. Farrar, I.O. Koshevoy, I.S.
Podkorytov, A.J. Poe¨, S.P. Tunik, J. Am. Chem. Soc. 124 (2002)
8922–8931;
(c) A.D. Burrows, R.W. Harrington, M.F. Mahon, M.T. Palmer, F.
Senia, M. Varrone, Dalton Trans. (2003) 3717–3726;
(d) A.D. Burrows, M.F. Mahon, M. Varrone, Dalton Trans. (2004)
3321–3330;
(e) G.R. Clark, G.-L. Lu, C.E.F. Rickard, W.R. Roper, L.J. Wright,
J. Organomet. Chem. 690 (2005) 3309–3320;
(f) R.A. Stockland Jr., M.C. Kohler, I.A. Guzei, M.E. Kastner, J.A.
Bawiec III, D.C. Labaree, R.B. Hochberg, Organometallics 25 (2006)
2475–2485.
P
P
P
P
a
R1 = iFoj ꢀ jFci/[ jFoj]; wR2 = {[ w(Fo2 ꢀ Fc2)2]/[ w(Fo2)2]}1/2
.
Crystallographic and experimental details are summarized
in Table 2. Unit-cell parameters were determined from
32825 reflections (3ꢁ < h < 31ꢁ) and refined by least-
squares method. Intensities were collected with graphite-
monochromatized Mo Ka radiation. 44591 reflections
were measured in the range 2.46ꢁ 6 h 6 31.61ꢁ. 7281 of
which were non-equivalent by symmetry (R (on I) =
0.039). 6900 reflections were assumed as observed applying
the condition I > 2r(I). Lorentz-polarization but no
absorption corrections were made. The structure was
solved by direct methods, using the SHELXS computer pro-
gram and refined by full-matrix least-squares method with
the SHELX-97 computer program [16].
´
´
[5] I. Angurell, I. Martınez-Ruiz, O. Rossell, M. Seco, P. Gomez-Sal, A.
´
Martın, Chem. Commun. (2004) 1712–1713.
[6] A. Bacchi, M. Carcelli, C. Pelizzi, G. Pelizzi, P. Pelagatti, S. Ugolotti,
Eur. J. Inorg. Chem. (2002) 2179–2187.
[7] J. Chatt, R. Mason, D.W. Meek, J. Am. Chem. Soc. 97 (1975) 3826–
3827.
[8] (a) M. Ferrer, M. Mounir, O. Rossell, E. Ruiz, M.A. Maestro,
Inorg. Chem. 42 (2003) 5890–5899;
´
(b) M. Ferrer, L. Rodrıguez, O. Rossell, J. Organomet. Chem. 681
(2003) 158–166.
Acknowledgements
[9] A.D. Burrows, M.F. Mahon, M.T. Palmer, M. Varrone, Inorg.
Chem. 41 (2002) 1695–1697.
We acknowledge support from the Ministerio de Ciencia
[10] H. Nakazawa, Y. Kadoi, T. Mizuta, K. Miyoshi, H. Yoneda, J.
Organomet. Chem. 366 (1989) 333–342.
´
y Tecnologıa (Project BQU2003-01131) and the CIRIT

I. Angurell et al. / Journal of Organometallic Chemistry 692 (2007) 3882–3891
3891
[11] M. Nieger, E. Niecke, R. Detsch, [BADMEZ] Cambridge Structural
Database, V5.27, January 2006.
(d) W. Henderson, B.K. Nicholson, A.G. Oliver, Polyhedron 13
(1994) 3099–3104.
[12] A.K.-S. Tse, Ru-ji Wang, T.C.W. Mak, Kin Shing Chan, J. Chem.
Soc., Chem. Commun. (1996) 173–174.
[14] J.X. McDermott, J.F. White, G.M. Whitesides, J. Am. Chem. Soc. 98
(1976) 6521–6528.
[13] (a) M. Yamashita, J.F. Hartwig, J. Am. Chem. Soc. 126 (2004) 5344–
5345;
[15] J. Chatt, L.M. Vallarino, L.M. Venanzi, J. Chem. Soc. (1957) 3413–
3416.
(b) M.W. Hooper, J.F. Hartwig, Organometallics 22 (2003) 3394–
3403;
(c) N. Kuhn, M. Grathwohl, C. Nachtigal, M. Steimann, Z.
Naturforsch. B 56 (2001) 704–710;
[16] G.M. Sheldrick, SHELXL-97: Program for Crystal Structure Analysis,
University of Go¨ttingen, Go¨ttingen, Germany, 1998.
[17] WinGX System: L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837–
838.