Cationic (fluoromesityl)palladium(II) complexes

Article abstract of DOI:10.1016/j.jorganchem.2006.05.039

Halide abstraction from [Pd(μ-Cl)(Fmes)(NCMe)]₂ (Fmes = 2,4,6-tris(trifluoromethyl)phenyl or nonafluoromesityl) with TlBF₄ in CH₂Cl₂/MeCN gives [Pd(Fmes)(NCMe)₃]BF₄, which reacts with monodentate ligands to give the monosubstituted products trans-[Pd(Fmes)L(NCMe)₂]BF₄ (L = PPh₃, P(o-Tol)₃, 3,5-lut, 2,4-lut, 2,6-lut; lut = dimethylpyridine), the disubstituted products trans-[Pd(Fmes)(NCMe)(PPh₃)₂]BF₄, cis-[Pd(Fmes)(3,5-lut)₂(NCMe)]BF₄, or the trisubstituted products [Pd(Fmes)L₃]BF₄ (L = CN^tBu, PHPh₂, 3,5-lut, 2,4-lut). Similar reactions using bidentate chelating ligands give [Pd(Fmes)(L-L)(NCMe)]BF₄ (L-L = bipy, tmeda, dppe, OPPhPy₂-N,N′, (OH)(CH₃)CPy₂-N,N′). The complexes trans-[Pd(Fmes)L₂(NCMe)]BF₄ (L = PPh₃, tht) (tht = tetrahydrothiophene) and [Pd(Fmes)(L-L)(NCMe)]BF₄ (L-L = bipy, tmeda) were obtained by halide extraction with TlBF₄ in CH₂Cl₂/MeCN from the corresponding neutral halogeno complexes trans-[Pd(Fmes)ClL₂] or [Pd(Fmes)Cl(L-L)]. The aqua complex trans-[Pd(Fmes)(OH₂)(tht)₂]BF₄ was isolated from the corresponding acetonitrile complex. Overall, the experimental results on these substitution reactions involving bulky ligands suggest that thermodynamic and kinetic steric effects can prevail affording products or intermediates different from those expected on purely electronic considerations. Thus,water, whether added on purpose or adventitious in the solvent, frequently replaces in part other better donor ligands, suggesting that the smaller congestion with water compensates for the smaller M-OH₂ bond energy.

Full text of DOI:10.1016/j.jorganchem.2006.05.039

Journal of Organometallic Chemistry 691 (2006) 3862–3873
www.elsevier.com/locate/jorganchem
Cationic (fluoromesityl)palladium(II) complexes
*
´
´
Camino Bartolome, Raquel de Blas, Pablo Espinet , Jose Miguel Martın-Alvarez,
Fernando Villafane
˜
Quı´mica Inorga´nica, Facultad de Ciencias, Universidad de Valladolid, 47005 Valladolid, Spain
Received 20 April 2006; received in revised form 18 May 2006; accepted 18 May 2006
Available online 3 June 2006
Warmly dedicated to Professor Antonio Abad on occasion of his premature retirement.
Abstract
Halide abstraction from [Pd(l-Cl)(Fmes)(NCMe)]2 (Fmes = 2,4,6-tris(trifluoromethyl)phenyl or nonafluoromesityl) with TlBF4 in
CH2Cl2/MeCN gives [Pd(Fmes)(NCMe)3]BF4, which reacts with monodentate ligands to give the monosubstituted products trans-
[Pd(Fmes)L(NCMe)2]BF4 (L = PPh3, P(o-Tol)3, 3,5-lut, 2,4-lut, 2,6-lut; lut = dimethylpyridine), the disubstituted products trans-
[Pd(Fmes)(NCMe)(PPh3)2]BF4, cis-[Pd(Fmes)(3,5-lut)2(NCMe)]BF4, or the trisubstituted products [Pd(Fmes)L3]BF4 (L = CNtBu,
PHPh2, 3,5-lut, 2,4-lut). Similar reactions using bidentate chelating ligands give [Pd(Fmes)(L–L)(NCMe)]BF4 (L–L = bipy, tmeda, dppe,
OPPhPy2-N,N0, (OH)(CH3)CPy2-N,N0). The complexes trans-[Pd(Fmes)L2(NCMe)]BF4 (L = PPh3, tht) (tht = tetrahydrothiophene) and
[Pd(Fmes)(L–L)(NCMe)]BF4 (L–L = bipy, tmeda) were obtained by halide extraction with TlBF4 in CH2Cl2/MeCN from the corre-
sponding neutral halogeno complexes trans-[Pd(Fmes)ClL2] or [Pd(Fmes)Cl(L–L)]. The aqua complex trans-[Pd(Fmes)(OH2)(tht)2]BF4
was isolated from the corresponding acetonitrile complex. Overall, the experimental results on these substitution reactions involving
bulky ligands suggest that thermodynamic and kinetic steric effects can prevail affording products or intermediates different from those
expected on purely electronic considerations. Thus,water, whether added on purpose or adventitious in the solvent, frequently replaces in
part other better donor ligands, suggesting that the smaller congestion with water compensates for the smaller M–OH2 bond energy.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Palladium; Fluoromesityl; Steric effects; Kinetic; Aqua; Cationic
1. Introduction
plex with three labile acetonitrile ligands, [Pd(Fmes)(NC-
Me)3]BF4, and its reactivity towards mono and bidentate
ligands. The chemistry of cationic complexes containing
labile ligands is a very active field, mainly due to the catalytic
properties of many of these complexes in olefin polymeriza-
tion processes [4]. Cationic palladium complexes containing
the fluorinated aryl ring, C6F5 have also been used as inser-
tion-triggered radical polymerization of norbornene and
copolymerization of norbornene with simple olefins [5].
The bulky ligand 2,4,6-tris(trifluoromethyl)phenyl (nona-
fluoromesityl or Fmes) gives rise to interesting structural
features when coordinated to main group [1], or to transition
metals [2], due to its high steric requirements combined with
a certain electron withdrawing character. Its coordination to
palladium has led to some unusual complexes, such as an
unprecedented self-assembled pyramidal tetrametallic com-
plex with a halide in the apex in complexes [Pd4(Fmes)4X5]ꢀ
(X = Cl, Br, I), a number of neutral aqua complexes, or the
formation of a new halocarbon ligand acting as halo-donor
chelate [3]. Herein we report the synthesis of a cationic com-
2. Results and discussion
2.1. Synthesis and characterization of
[Pd(Fmes)(NCMe)3]BF4
*
The reaction of [Pd(l-Cl)(Fmes)(NCMe)]2 [3e] with
Corresponding author.
TlBF4 in a dichloromethane/acetonitrile mixture leads to
E-mail address: espinet@qi.uva.es (P. Espinet).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.05.039

C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
3863
the formation of [Pd(Fmes)(NCMe)3]BF4 (1a), which was
isolated as a white solid in 92% yield (Scheme 1).
tive processes are preferred in mono(fluoromesityl)palla-
dium(II) complexes [3e].
The m(C „ N) IR absorptions of MeCN in 1a appear at
2343 and 2326 cmꢀ1. The 19F NMR spectrum of 1a in
CDCl3 shows one broad signal at ꢀ59.66 ppm for the
two ortho-CF3, groups and a singlet at ꢀ63.29 ppm for
the para-CF3. Addition of MeCN causes the sharpening
of the ortho-CF3 signal, giving rise to a fine singlet, as in
2.2. Reactivity of [Pd(Fmes)(NCMe)3]BF4 towards
monodentate ligands
The monodentate ligands chosen to study the reactivity
of 1a were tBuNC, different phenylphosphines as P-donors,
and isomeric lutidines as N-donors. The latter offer the
possibility of changing the ligand steric hindrance while
keeping to a minimum the variation of donor ability.
According to the electronic trans effect and trans influ-
ence, the ligand exchange should operate as follows: The
associative substitution should lead the incoming ligand
to coordinate trans to Fmes, which is the group with the
higher trans effect in 1a. However, all the ligands used have
a higher trans influence than MeCN (and lower than
Fmes), which makes the initial kinetic product thermody-
namically unstable, due to the highly destabilizing mutual
trans influence of the incoming ligand and the Fmes group
[8]. Therefore, a subsequent isomerization can be expected.
If only electronic factors were considered and all reactions
were fast, the successive substitution and isomerization
reactions should follow the path proposed in Scheme 3.
The experimental results found (Scheme 4) indicate that
other factors must also be taken into account: (a) steric fac-
tors due to repulsion between ligands may be thermody-
namically significant and may prevail over electronic
factors in determining which isomer is the most stable
1
the typical pattern for Fmes complexes [3]. The H NMR
spectrum of 1a in CDCl3 at room temperature shows two
sharp singlets (relative intensity 2:9) arising from the Fmes
hydrogen atoms and the hydrogens of the three coordi-
nated NCMe ligands. At 253 K the signal of the latter splits
into two singlets (relative intensity 3:6). At this temperature
the 19F NMR spectrum also shows a sharpening of the
ortho-CF3 signal. Moreover, both in the 1H NMR and
19F NMR low temperature spectra a small amount of a
minor complex, likely to be [Pd(Fmes)(NCMe)2(OH2)]BF4
(5a) (see data in Section 4), is detected. These features sug-
gest that the three acetonitrile ligands become equivalent
through a associative displacement and exchange that is
catalyzed by small amounts of water or, more efficiently,
by added MeCN. With just traces of water in the medium
1
the process is fast in the H NMR time scale and in the
exchange limit for the 19F NMR time scale. The exchange
1
can be frozen in the H NMR time scale upon cooling at
253 K, and catalytically accelerated in the 19F NMR time
scale by free MeCN. Scheme 2 shows only the equilibrium
between 1a and 5a where the MeCN trans to Fmes is
replaced by an aqua ligand from traces of water present
in CDCl3. For the equivalences observed the other two
acetonitriles must also enter in the exchange process (for
instance by Berry pseudo-rotation in the pentacoordinated
intermediate produced during the associative substitution
process). The lability of the MeCN ligand, and the easy for-
mation of organometallic [6], and coordination palladium
aqua complexes [7] is well documented.
All the attempts to isolate 5a were unsuccessful, but
other cationic complexes described below are also in equi-
librium with their corresponding aqua complexes, and one
of them could be fully characterized.
Since MeCN may be easily replaced by other ligands,
complex 1a is a good precursor for other cationic (fluorom-
esityl)palladium(II) complexes. We have shown before
that, although dissociative substitutions seem to operate
in bis(fluoromesityl)palladium(II) complexes [3f], associa-
Scheme 3. Ideal sequential substitution/isomerization mechanism under
electronic control in [Pd(Fmes)(NCMe)3]BF4 (1a).
Scheme 1. Synthesis of [Pd(Fmes)(NCMe)3]BF4 (1a).
Scheme 2. Process detected in the solutions of [Pd(Fmes)(NCMe)3]BF4 (1a).

3864
C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
Scheme
6. Rearrangement
upon
heating
trans-[Pd(Fmes)(NC-
Me)2(PPh3)]BF4 (3b).
Scheme 4. Syntheses of cationic complexes with monodentate ligands
from 1a.
with PPh3 illustrate the effect of increasing steric hin-
drance after successive substitutions. The increasing num-
ber of bulkier ligands hinders the formation of more
substituted products and slows down the isomerization
processes, whereas higher temperature and excess of
incoming ligand favors the isomerization to the thermody-
namic products. Thus the substitution degree of the prod-
ucts will depend on the steric requirements of the ligands.
Consistently, the reaction of 1a with P(o-Tol)3 (conic
angle = 194ꢁ for the exo3 conformation, [10]) only affor-
ded the product of monosubstitution trans-[Pd(Fmes)-
(NCMe)2P(o-Tol)3]BF4 (3c). In this case, 3c is stable
and does not isomerize to the cis isomer. Possibly the
cis isomer is less stable than the trans one because of
severe repulsion between the two bulky ligands in the
cis arrangement. Thus, this is a case where steric effects
(favoring the trans isomer) prevail on electronic ones
(trans influence favoring the cis isomer). In contrast,
ligands with lower conic angle, such as PHPh2 (conic
one; (b) steric factors may be kinetically significant, render-
ing some of the proposed steps (i.e. isomerizations) slow;
and (c) a product may be thermodynamically unstable
towards rearrangement in two or more other products.
The reaction of 1a with PPh3, a middle sized ligand
(conic angle 145ꢁ, [9]) with a moderately high trans influ-
ence, in a 1:1 ratio in mild conditions led to the monosub-
stituted trans-[Pd(Fmes)(NCMe)2(PPh3)]BF4 (3b), whereas
excess of ligand and heating at reflux in MeCN allowed us
to isolate trans-[Pd(Fmes)(NCMe)(PPh3)2]BF4 (2b).
Complex 2b may be also obtained from the reaction of
trans-[PdCl(Fmes)(PPh3)2] [3a] with TlBF4 as halogen
extractor in refluxing MeCN (Scheme 5). On the other
hand, heating 3b led to a mixture of 3b, 2b, 1a, and FmesH,
as well as the monosubstituted complex cis-3b (see Section
4). The latter was found to be unstable towards its rear-
rangement to 1a and 2b (Scheme 6).
These results indicate that 1a reacts easily (room tem-
perature, 1:1 ratio) with PPh3 to give the kinetic mono-
substituted product 3b (I in Scheme 3). The associative
isomerization at room temperature is slow. The attack
of a second PPh3 to obtain the disubstituted product
requires higher temperature, which also facilitates the
isomerization. Thus III was not detected, and the thermo-
dynamic product 2b (corresponding to IV in Scheme 3)
was isolated. All attempts to coordinate a third PPh3 were
unsuccessful; either the trisubstituted product is too
unstable or the associative attack is too slow. The results
t
angle 128ꢁ, [9]) or BuNC (conic angle not tabulated but
t
obviously small, as the Bu group is two bonds far from
Pd), rapidly led to the trisubstituted products [Pd(Fmes)-
(PHPh2)3]BF4 (1d) and [Pd(Fmes)(CNtBu)3]BF4 (1e), even
under mild reaction conditions. Furthermore, lower ratios
t
PHPh2 or BuNC to Pd lead to mixtures of 1a and the tri-
substituted product.
A similar discussion may be applied to the N-donor luti-
dines (dimethylpyridines) used, which are electronically similar
but sterically different. Steric hindrance is to be expected in the
perpendicular to the coordination plane of palladium where
the ortho substituents of the aryl and the lutidine rigs will
clash. The reactions of 1a with the less hindered lutidine, 3,5-
lut, proceeded smoothly and lead to the mono-, di-, and tri-
substituted products, trans-[Pd(Fmes)(3,5-lut)(NCMe)2]BF4
(3f), cis-[Pd(Fmes)(3,5-lut)2(NCMe)]BF4 (2f) and [Pd(Fmes)-
(3,5-lut)3]BF4 (1f), respectively, depending on the reactants
ratio. Therefore, the less substituted products could be
obtained in this case, in contrast to what is observed (see
Scheme 5. Syntheses of cationic complexes with monodentate ligands by
halide abstraction.
t
above) for other small ligands, such as BuNC and PHPh2.

C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
3865
It should be noted that 2f is the only disubstituted product with
the incoming ligands coordinated mutually cis (III in Scheme
3) that has been isolated in the solid state (complexes 2b
(L = PPh3) and 2i (L = tht) could only be detected in solu-
tion). The NMR spectra show that 2f is always accompanied
with small amounts of 1f and 3f (Scheme 7), suggesting that
this cis complex is unstable towards rearrangement.
Fmes palladium(II) complexes, P–F coupling is observed
when the Fmes group and the phosphines are coordinated
cis, but not when these groups are mutually trans [3]. The
1H NMR spectrum of 1d could be interpreted only after
a number of incoherent and selective 31P irradiation
experiments.
The kind of phosphine used in 3c can undergo the so-
called ‘‘two-ring-flip mechanism’’, [11] which exchanges
the exo2 and exo3 conformers (two or three ortho-methyl
groups in metal complexes of this kind of phosphine are
oriented toward the metal atom). In the case of 3c the
19F NMR spectrum at 213 K in CDCl3 shows two non-
equivalent ortho-CF3 groups but one singlet in the
31P{1H} NMR spectrum. This suggests that Pd–P rotation
is slow at this temperature. The low temperature 1H NMR
spectrum shows a broad signal at 8.6 ppm, which is
assigned to the H6 endo of the exo2 conformer (in this con-
formation the o-methyl group is oriented away from the
metal) [10]. This supports that the rotation around the P–
C ipso bond, which exchanges the exo2 and exo3 conform-
ers (where two or three ortho-methyl groups are oriented
toward the metal atom), has also been arrested (in the
NMR timescale) at this temperature. One of the conform-
ers is clearly present in higher proportion, as observed also
for neutral complexes [Pd(Fmes)XP(o-Tol)3]2 previously
reported [3e].
The treatment of 1a with the more hindered 2,4-lut in a
1:1 ratio at room temperature led to the monosubstituted
product trans-[Pd(Fmes)(2,4-lut)(NCMe)2]BF4 (3g). When
the reaction was carried out in refluxing CHCl3 with a
4:1 excess of ligand, the trisubstituted product [Pd(Fmes)-
(2,4-lut)3]BF4 (1g) was isolated. The disubstituted species
could not be obtained. The addition of 2,6-lut (the
most hindered lutidine) to 1a only led to the monosubsti-
tuted product trans-[Pd(Fmes)(2,6-lut)(NCMe)2]BF4 (3h).
Apparently, the high steric crowding in the axial positions
of Pd precludes the formation of more substituted products.
The complex trans-[Pd(Fmes)(NCMe)(tht)2]BF4 (2i)
(tht = tetrahydrothiophene) could be synthesized treating
trans-[PdCl(Fmes)(tht)2] [3a] with TlBF4 in MeCN
(Scheme 5). This reaction has to be carried out with excess
of free tht in the reaction mixture, otherwise it leads to mix-
tures of 2i and a new species. The latter could not be
isolated pure, but the 1H NMR spectrum shows two
inequivalent NCMe ligands and one coordinated tht mole-
cule (1:1:1) proving that it is the cis isomer of [Pd(Fmes)-
(NCMe)2(tht)]BF4. The ortho-CF3 groups of 2i display a
very broad signal in the 19F NMR spectrum in CDCl3 at
room temperature. This signal sharpens both in dry CDCl3
and in wet CDCl3. In the latter case, the MeCN signal
appears broad and at chemical shifts near to that of free
MeCN. These data point to an equilibrium between 2i
and the aqua complex trans-[Pd(Fmes)(OH2)(tht)2]BF4
(5i), where a water molecule has replaced the MeCN
ligand. In fact, complex 5i was selectively obtained and iso-
lated by refluxing 2i in a THF/H2O mixture (see Section 4).
The characterization of the lutidine complexes from
their spectroscopic data is straightforward. An X-ray crys-
tal structure analysis was carried out for the 3f. A perspec-
tive view of the structure is given in Fig. 1, and selected
distances and angles are listed in Table 1. The palladium
atom is essentially square planar with minor distortions.
The C–C–C angle at the ipso carbon atom of the Fmes
2.3. Characterization of cationic complexes containing
monodentate ligands
The 19F and 31P NMR spectra of all the complexes with
phosphines are informative about the relative coordination
of the Fmes group and phosphines. As reported for other
Fig. 1. ORTEP diagram of the cation [Pd(Fmes)(3,5-lut)3]+ of 3f showing
the atom numbering scheme; the ellipsoids are drawn at 30% probability
level.
Scheme 7. Solution behavior of cis-[Pd(Fmes)(3,5-lut)2(NCMe)]BF4 (2f).

3866
C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
Table 1
˚
Selected bond lengths (A) and angles (ꢁ) for [Pd(Fmes)(3,5-lut)3]BF4 (3f)
Pd(1)–C(11)
Pd(1)–N(3)
Pd(1)–N(2)
Pd(1)–N(1)
C(11)–Pd(1)–N(3)
C(11)–Pd(1)–N(2)
2.010(7)
2.035(5)
2.039(6)
2.103(6)
90.4(2)
N(3)–Pd(1)–N(1)
N(2)–Pd(1)–N(1)
N(3)–Pd(1)–N(2)
C(11)–Pd(1)–N(1)
C(16)–C(11)–C(12)
91.7(2)
87.7(2)
177.6(2)
177.9(2)
113.8(7)
90.2(2)
group is significantly less than 120ꢁ [113.8(7)ꢁ], due to the
electronic effects of the electropositive metal and electro-
negative CF3 substituents at these positions [12]. The
Pd(1)–N(1) distance is larger than Pd(1)–N(3) or Pd(1)–
N(2), due to the higher trans influence of the Fmes group
compared to lutidine. The Fmes and two lutidine ligands
are tilted in the same sense making angles of 86.5, 79
(N1, trans to Fmes), and 88ꢁ (N3, cis to Fmes) to the coor-
dination plane. The other lutidine (N2, cis to the Fmes) is
tilted also in the same sense making an angle of 65.1ꢁ. This
torsion avoids the contact between one lutidine methyl
group and one Fmes trifluoromethyl group of a neighbor-
ing cation in the crystal.
Low temperature NMR spectra of 1g were needed in
order to obtain additional information. The 19F NMR
room temperature spectrum shows the presence of a major
compound (91%) and two minor isomers [13]. Assuming
that Fmes rotation is hindered due to its high steric require-
ments, restricted rotation of the 2,4-lut ligands around the
Pd–N bonds will give rise to three atropisomers A–C in
Fig. 2. The 1H NMR spectrum of 1g at 243 K clearly indi-
cates that the major isomer observed corresponds to C, as
the aromatic hydrogens and the methyl groups of the three
lutidines are non-equivalent (see Section 4). C is the less
hindered atropisomer, and therefore the most stable.
Scheme 8. Syntheses of cationic complexes with bidentate ligands ((a) and
(b) 4a, 4b; (a) 4c, 4d, 4e).
ent bidentate ligands L–L in a 1:1 ratio lead to the cationic
complexes [Pd(Fmes)(L–L)(NCMe)]BF4 (L–L = bipy, 4a;
tmeda, 4b; dppe, 4c; OPPhPy2 ꢀ N,N0, 4d; MeC(OH)-
Py2 ꢀ N,N0, 4e) in high yields. All the reactions were fast,
and no kinetic intermediates were detected. This is in con-
trast with the behavior observed in the analogous reaction
of [Pd(l-Cl)(Fmes)(NCMe)]2 with dppe, which lead to
kinetic and thermodynamic products, ([PdCl(Fmes)(NC-
Me)]2(l-dppe) and [PdCl(Fmes)(dppe)], respectively) [3e].
The alternative procedure treating the corresponding chlo-
rocomplexes [3a,3e] with TlBF4 in MeCN, was also used to
synthesize 4a and 4b.
The characterization of 4a, 4b, and 4c from their spec-
troscopic data is straightforward. The chemical shift
assigned to H6 in the complexes containing coordinated
pyridyl groups (4a, 4d, and 4e) is clearly higher for the
pyridyl moiety cis to NCMe. The same effect has been
observed for the corresponding neutral chlorocomplexes
containing the same chelating ligands, although the shield-
ing effect of the halogen is higher [3e].
2.4. Synthesis and characterization of cationic complexes
with bidentate ligands
One of the ortho-CF3 groups in the complex
[Pd(Fmes)(OPPhPy2 ꢀ N,N0)(NCMe)]BF4 (4d) appears
very upfield in the 19F NMR spectrum at room tempera-
ture, which can be attributed to the anisotropic shielding
Scheme 8 collects the complexes obtained with chelating
ligands and their syntheses. The reactions of 1a with differ-
Fig. 2. Atropisomers in [Pd(Fmes)(2,4-lut)3]BF4 (1g).

C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
3867
produced by the phenyl group of the chelating ligand, as
observed before for the analogous complexes [PdCl(Fmes)-
(OPPhPy2-N,N0)] [3e] and [PdBr(C6F5)(OPPhPy2-N,N0)]
[14]. Moreover, both ortho-CF3 groups give unusually
broad signals, indicating that some dynamic process is
in bis(fluoromesityl) complexes. Bonding Pdꢁ ꢁ ꢁF interac-
tions can then be discarded, so the crowding in 4e is
responsible for the short non-bonding Pdꢁ ꢁ ꢁF contacts, as
previously reported [3a]. The conformer found in the solid
state has the methyl group of the (OH)(CH3)CPy2 ligand
oriented towards the axial position of the coordination
plane. The hydrogen atom of the OH group was located
in a difference Fourier map and refined [15]. It is involved
in O–Hꢁ ꢁ ꢁF intermolecular contacts with the BF4 counter-
anion, with a rather short H(1)ꢁ ꢁ ꢁF(12) distance of
1
occurring. The H NMR spectrum of 4d is as expected for
the ligand chelated as N–N donor, but the signals of the
two non-equivalent H3 atoms in the pyridyl groups are also
broad. In a variable temperature study the H3 signals are
well resolved pseudotriplets at 333 K and also at 213 K.
These two protons remain non-equivalent at high tempera-
tures, so an exchange process between both pyridyl groups
can be discarded. A similar temperature effect is observed
in the 19F NMR spectra in which at the ortho-CF3 signals
sharpen at 213 K without any appreciable change in their
chemical shifts. The fluxional behavior of 4d can be
explained by the equilibrium depicted in Scheme 9 in which
4d undergoes displacement of the nitrogen trans to Fmes
(the most trans labilizing ligand) by an aqua ligand from
traces of water, to give an aqua complex D in non-detectable
amounts. The exchange rate is fast at high temperature, thus
the signals from both the ortho-CF3 and H3 atoms become
sharp. When the temperature decreases, only the signals
from 4d are observable. The concentration of the putative
aqua complex is very small, what makes the averaged chem-
ical shift in the very uneven equilibrium almost identical to
the chemical shift of very major component 4d, recorded at
low temperature.
Two conformers, (endo-CH3) and (endo-OH), are plausi-
ble for 4e. A single-crystal X-ray analysis was undertaken
to establish which conformer is stable in the solid state.
A perspective view of the cation [Pd(Fmes){(OH)(CH3)-
CPy2-N,N0}(NCMe)]+ is shown in Fig. 3. Selected bond
lengths and angles are listed in Table 2. The square planar
geometry of Pd is somewhat distorted towards tetrahedral,
as previously observed for the neutral [PdCl(Fmes){(OH)-
0
˚
(CH3)CPy2-N,N }] [3e]: N(2) is located 0.195(9) A above
the plane described by the Pd(1), C(11) and N(1) atoms,
˚
and N(3) is 0.209(9) A below. This distortion is not as large
as for cis-bis(fluoromesityl)palladium complexes in which
the square-planar geometry is severely distorted [3f]. As
usual, the C–C–C angle at the ipso carbon atom of the
Fmes group is less than 120ꢁ [ca. 114.2(9)ꢁ] [12]. The Fmes
ligand is nearly perpendicular to the coordination plane.
The Pdꢁ ꢁ ꢁF3C-ortho distances are longer than those found
Fig. 3. Top: ORTEP diagram of the cation [Pd(Fmes){(OH)(CH3)C-
Py2 ꢀ N,N0}(NCMe)]+ of 4e showing the atom numbering scheme; the
ellipsoids are drawn at 30% probability level. Bottom: Intermolecular
Hꢁ ꢁ ꢁF hydrogen bonds observed in 4e.
Scheme 9. Water coordination in [Pd(Fmes)(OPPhPy2-N,N0)(NCMe)]BF4 (4d).

3868
C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
Table 2
4. Experimental
˚
Selected bond lengths (A) and angles (ꢁ) for 4e
Pd(1)–N(1)
Pd(1)–C(11)
Pd(1)–N(2)
Pd(1)–N(3)
N(1)–Pd(1)–C(11)
C(11)–Pd(1)–N(2)
1.973(9)
1.992(9)
2.029(8)
2.089(8)
86.4(4)
N(1)–Pd(1)–N(3)
N(2)–Pd(1)–N(3)
N(1)–Pd(1)–N(2)
C(11)–Pd(1)–N(3)
C(16)–C(11)–C(12)
92.4(3)
88.0(3)
174.5(4)
174.1(4)
114.2(9)
4.1. General comments
All reactions were carried out under an atmosphere of
dry N2, and at room temperature unless otherwise indicated.
Solvents were purified according to standard procedures
[16]. 1,3,5-C6H3(CF3)3 (FmesH) was purchased from Fluo-
rochem, and Li(Fmes) was prepared as described in the lit-
erature [1a,16,17], and used immediately in situ without
further purification. The chlorofluoromesitylcomplexes
used as starting materials were prepared as previously
described [3]. TlBF4 was obtained as reported [18] (Caution.
Tl(I) derivatives are toxic and should be handled with care).
Infrared spectra were recorded in a Perkin–Elmer 883 or
1720X apparatus as Nujol mulls between polystyrene films
from 4000 to 200 cmꢀ1. Only the most significant absorp-
tions are herein indicated for clarity (for the rest of IR
absorptions see Supplementary Material). NMR spectra
were recorded on Bruker AC-300 or ARX-300 instruments
in dry CDCl3 at room temperature unless otherwise stated.
NMR spectra are referred to TMS, CFCl3, or 85% aqueous
H3PO4, coupling constants are measured in Hz, and trans
and cis are referred to Fmes unless otherwise indicated.
93.8(3)
˚
1.682(12) A, and a O(1)–H(1)ꢁ ꢁ ꢁF(12) angle of 158.5(4)ꢁ.
Most Hꢁ ꢁ ꢁF distances found in the Cambridge Data Base
˚
are within the range 1.73–2.5 A, and only few are shorter
˚
than 1.7 A, as in this case.
As observed for 1a and 2i, solutions of 4d and 4e also
show equilibria with their respective aqua com-
plexes [Pd(Fmes)(OH2)(OPPhPy2-N,N0)]BF4 (5d) and
[Pd(Fmes)(OH2){(OH)(CH3)CPy2-N,N0}]BF4 (5e) in wet
deuterated solvents. The addition of water to 4d and 4e
leads to observable amounts of the aqua complexes, which
could not be isolated in the solid state. As for complexes 4,
the signal of H6 in aqua complexes containing coordinated
pyridyl groups (5a, 5d, and 5e) appears clearly downfield
for the pyridyl group cis to the aqua ligand. This shielding
is lower than that observed for the acetonitrile complexes 4,
and much lower than for the corresponding neutral chloro-
complexes containing the same chelating ligands [3e].
Therefore, the coordination-induced shift value for these
ligands follows the sequence: Cl > NCMe > OH2. The
aqua complexes give back the acetonitrile complexes 4d
and 4e upon addition of free MeCN.
19
The signals in the F NMR assigned to 11BF4ꢀ and 10BF4ꢀ
appear within the range ꢀ153 to 154 ppm for all the
complexes described (Supplementary Material). Elemental
analyses were performed on a Perkin–Elmer 2400B micro-
analyzer. Electrical conductivity measurements were
carried out at room temperature with a Crison 522 conduc-
tivimeter on ca. 5 · 10ꢀ4 M solutions; the range of molar
conductivity for 1/1 electrolytes is 135–155 S cm2 molꢀ1
in acetonitrile solutions [19], although values of ca.
80 S cm2 molꢀ1 have been also described [20]. The molar
conductivities of all the complexes herein described are in
the range 94–115 S cm2 molꢀ1 (Supplementary Material).
3. Conclusions
Ligand substitution reactions at [Pd(Fmes)(NCMe)3]-
BF4 (1a) occur under mild conditions. A single Fmes group
in a monoarylated complex can tilt and facilitate the
approach of the incoming ligand, making associative sub-
stitutions a kinetically accessible pathway, at least for the
first substitution. Subsequent substitutions can become
more difficult depending on the ligands. The kinetic prod-
ucts of the initial substitution are consistent with the higher
trans effect of the Fmes group. All the ligands used have
higher trans effect than MeCN, what should make this
kinetic product thermodynamically unstable towards isom-
erization, due to the tendency of Pd(II) to avoid ligands
with high trans influence in mutually trans positions. How-
ever, the subsequent evolution of the reactions is not that
simple, revealing that thermodynamic and kinetic steric
effects also count and can prevail on the purely electronic
predictions.
It is worth noting that in a number of cases H2O,
whether added or as adventitious water in the solvent,
has been detected to replace other ligands in equilibrium,
including chelating pyridyl ligands. This frequent occur-
rence suggests that the smaller size of the water molecule
might also be a contribution to favor this otherwise unu-
sual ligand competition.
4.2. [Pd(Fmes)(NCMe)3]BF4 (1a)
A 100 mL flask was successively charged with [Pd(l-
Cl)(Fmes)(NCMe)]2 [3e] (1.310 g, 1.42 mmol), MeCN
(8 mL), CH2Cl2 (50 mL), and TlBF4 (0.860 g, 2.98 mmol),
and the mixture was stirred for 16 h. The solution was then
filtered on dry Celite, the solvents were pumped off, and the
residue was washed with Et2O (3 · 10 mL), yielding 1.567 g
(92%) of 1a as an off-white solid. IR: 2343 s, 2326 vs, 2298
vs. 19F NMR (253 K) d ꢀ59.60 (s, ortho-CF3, 6F), ꢀ63.10
(s, para-CF3, 3F). 1H NMR (253 K) d 7.86 (s, C6H2(CF3)3,
2H), 2.52 (s, trans-CH3CN, 3H), 2.35 (s, cis-CH3CN, 6H).
Anal. Calc. for C15H11BF13N3Pd: C, 30.16; H, 1.86; N,
7.03. Found: C, 30.15; H, 1.93; N, 6.85%.
4.3. [Pd(Fmes)(PHPh2)3]BF4 (1d)
PHPh2 (0.074 g, 0.4 mmol) was added to a solution of 1a
(0.060 g, 0.1 mmol) in CH2Cl2 (10 mL), and the mixture
was stirred for 1 h. The volatiles were pumped off, and the

C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
3869
residue was washed with Et2O (3 · 10 mL), dried in vacuo,
and stored under nitrogen, yielding 0.056 g (54%) of 1d as a
white solid. 19F NMR d ꢀ59.60 (t, JPF = 8.0, ortho-CF3,
6F), ꢀ63.60 (s, para-CF3, 3F). 31P{1H} NMR d 0.2 (dsept,
2JPP = 34.0 (cis), JPF = 8.0, cis-PHPh2, 2P), ꢀ8.4 (t,
CF3, 3F of atropisomer C), ꢀ63.44 (s, para-CF3, 3F of
minor species). 19F NMR (333 K) d ꢀ57.99 (br, ortho-
CF3, 6F all species), ꢀ 63.63 (s, para-CF3, 3F all species).
1H NMR (243 K, atropisomer C) d 8.21 (d, J = 6.0,
NC5H3(CH3)2, H6, 1H), 8.09 (d, J = 6.0, NC5H3(CH3)2,
H6, 1H), 7.84 (s, C6H2(CF3)3, 2H), 7.85 (d, J = 6.0,
NC5H3(CH3)2, H6, 1H), 7.19 (s, NC5H3(CH3)2 trans, H3,
1H), 7.08 (s, NC5H3(CH3)2 cis, H3, 2H), 7.05 (d, J = 6.0,
NC5H3(CH3)2, H5, 1H), 7.00 (d, J = 6.0, NC5H3(CH3)2,
H5, 1H), 6.89 (d, J = 6.0, NC5H3(CH3)2, H5, 1H), 2.42
(s, NC5H3(CH3)2, 3H), 2.36 (s, NC5H3(CH3)2, 3H), 2.32 (s,
NC5H3(CH3)2, 6H), 2.31 (s, NC5H3(CH3)2, 3H), 2.25 (s,
1
2JPP = 34.0, trans-PHPh2, 1P). H {31P} NMR d 7.20 (m,
C6H5, 30H, and C6H2(CF3)3, 2H), 6.60 (s, trans-PHPh2,
1
1H), 6.15 (s, cis-PHPh2, 2H). H NMR d 7.20 (m, C6H5,
1
3
30H; C6H2(CF3)3, 2H), 6.60 (dt, JPH = 300.0, JPH = 3.0,
1
PHPh2 trans, 1H), 6.15 (AA‘XX’M system, JPH = 377.0,
2JPP = 357 (trans), JPH = 8.0 (cis), JPH = 7.0 (trans),
PHPh2 cis, 2H). Anal. Calc. for C45H35BF13P3Pd: C,
52.33; H, 3.42. Found: C, 52.17; H, 3.45%.
3
3
1
NC5H3(CH3)2, 3H). H NMR (298 K, atropisomer C) d
8.24 (m, NC5H3(CH3)2, H6, 1H), 8.12 (m, NC5H3(CH3)2,
H6, 1H), 7.91 (d, J = 6.0, H6, NC5H3(CH3)2, 1H), 7.83 (s,
C6H2(CF3)3, 2H), 7.20 (s, H3, NC5H3(CH3)2 trans, 1H),
7.08 (m, NC5H3(CH3)2 and NC5H3(CH3)2, 3H), 6.97 (m,
NC5H3(CH3)2, H5, 1H), 6.89 (m, NC5H3(CH3)2, H5, 1H),
2.44 (s, NC5H3(CH3)2, 3H), 2.36 (s, NC5H3(CH3)2, 3H),
2.33 (m, NC5H3(CH3)2, 12H). 1H NMR (333 K, all species)
d 8.20 (s, NC5H3(CH3)2, H6, 2H), 7.99 (d, J 6.0,
NC5H3(CH3)2, H6, 1H), 7.82 (s, C6H2(CF3)3, 2H), 7.20
(s, NC5H3(CH3)2 trans, H3, 1H), 7.13 (d, J = 6.0,
NC5H3(CH3)2, H5, 1H), 7.08 (s, NC5H3(CH3)2 cis, H3,
2H), 6.93 (br, NC5H3(CH3)2, H5, 2H), 2.42 (s,
NC5H3(CH3)2, 6H), 2.38 (s, NC5H3(CH3)2, 6H), 2.33 (s,
NC5H3(CH3)2, 6H). Anal. Calc. for C30H29BF13N3Pd: C,
45.28; H, 3.67; N, 5.28. Found: C, 44.94; H, 3.60; N, 5.27%.
4.4. [Pd(Fmes)(CNtBu)3]BF4 (1e)
CNtBu (0.028 g, 0.33 mmol) was added to a solution of
1a (0.060 g, 0.1 mmol) in CH2Cl2 (10 mL), and the mixture
was stirred for 8 h. Then Et2O was added (ca. 10 mL) and
the solution was cooled to ꢀ20 ꢁC. The microcrystalline
white solid obtained was filtered, washed with Et2O
(3 · 3 mL), and dried in vacuo, yielding 0.054 g (74%). IR
(CH2Cl2): 2306 m, 2235 s. 19F NMR d ꢀ61.30 (s, ortho-
1
CF3, 6F), ꢀ63.28 (s, para-CF3, 3F). H NMR d 8.00 (s,
C6H2(CF3)3, 2H), 1.69 (s, CNC4H9 trans, 9H), 1.41 (s,
CNC4H9 cis, 18H). Anal. Calc. for C24H29BF13N3Pd: C,
39.83; H, 4.04; N, 5.81. Found: C, 39.85; H, 3.88; N, 5.79%.
4.5. cis-[Pd(Fmes)(3,5-lut)3]BF4 (1f)
4.7. trans-[Pd(Fmes)(NCMe)(PPh3)2]BF4 (2b)
A 100 mL flask was successively charged with 1a
(0.060 g, 0.1 mmol), CH2Cl2 (10 mL), and 3,5-lutidine
(0.043 g, 0.4 mmol), and the mixture was stirred for 10 h.
Work up as for 1e yielded 0.061 g (77%). 19F NMR d
Method A. A 100 mL flask was successively charged with
1a (0.060 g, 0.1 mmol), MeCN (10 mL), and PPh3 (0.105 g,
0.4 mmol), and the mixture was refluxed for 16 h. Then the
solution was filtered on dry Celite, and the volatiles were
pumped off. The solid residue was crystallized in CH2Cl2–
hexane at ꢀ20 ꢁC, yielding a microcrystalline white solid,
which was filtered, washed with cold hexane (3 · 3 mL),
dried in vacuo, and stored under nitrogen. Yield: 0.061 g
(59%). Method B. A 100 mL flask was successively charged
with trans-[Pd(Fmes)Cl(PPh3)2] [3a] (0.095 g, 0.1 mmol),
MeCN (10 mL), and TlBF4 (0.032 g, 0.11 mmol), and the
mixture was refluxed for 6 h. The volatiles were pumped
off, CH2Cl2 (10 mL) was added to the residue, which was fil-
tered on dry Celite. Then hexane was added (ca. 15 mL) and
the solution was cooled to ꢀ20 ꢁC, yielding 0.068 g (65%) of
2b. IR: 2322 m, 2294 w. 19F NMR d ꢀ58.59 (t, JPF = 5.0,
ortho-CF3, 6F), ꢀ63.40 (s, para-CF3, 3F). 31P{1H} NMR
1
ꢀ58.10 (s, ortho-CF3, 6F), ꢀ63.34 (s, para-CF3, 3F). H
NMR d 8.09 (s, NC5H3(CH3)2 trans, H2, 2H), 7.99 (s,
NC5H3(CH3)2 cis, H2, 4H), 7.80 (s, C6H2(CF3)3, 2H),
7.53 (s, NC5H3(CH3)2 trans, H4, 1H), 7.41 (s,
NC5H3(CH3)2 cis, H4, 2H), 2.32 (s, NC5H3(CH3)2 trans,
6H), 2.23 (s, NC5H3(CH3)2 cis, 12H). Anal. Calc. for
C30H29BF13N3Pd: C, 45.28; H, 3.67; N, 5.28. Found: C,
45.19; H, 3.62; N, 5.46%.
4.6. [Pd(Fmes)(2,4-lut)3]BF4 (1g)
A 100 mL flask was successively charged with 1a
(0.060 g, 0.1 mmol), CHCl3 (15 mL), and 2,4-lutidine
(0.043 g, 0.4 mmol), and the mixture was refluxed for
22 h. Then the solution was filtered on dry Celite, Et2O
was added (ca. 15 mL) to the filtrate, and the solution
was cooled to ꢀ20 ꢁC. The microcrystalline white solid
obtained was filtered, washed with Et2O (3 · 3 mL), and
dried in vacuo, yielding 0.040 g (51%). 19F NMR (243 K,
atropisomer C) d ꢀ58.37 (s, ortho-CF3, 3F), ꢀ58.42 (s,
ortho-CF3, 3F), ꢀ63.17 (s, para-CF3, 3F) (C: 91%). 19F
NMR (298 K) d ꢀ57.54 (br, ortho-CF3, 6F, minor species),
ꢀ58.23 (s, ortho-CF3, 3F of isomer C), ꢀ58.31 (s, ortho-
1
d 20.5 (sept, JPF = 5.0). H NMR d 7.50 (m, C6H5, 30H),
7.28 (s, C6H2(CF3)3, 2H), 1.58 (t, JPH = 1.0, CH3CN,
3H). Anal. Calc. for C47H35BF13NP2Pd: C, 54.28; H,
3.39; N, 1.35. Found: C, 53.98; H, 3.58; N, 1.21%.
4.8. cis-[Pd(Fmes)(3,5-lut)2(NCMe)]BF4 (2f)
A 100 mL flask was successively charged with 1a
(0.060 g, 0.1 mmol), CH2Cl2 (10 mL), and 3,5-lutidine

3870
C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
(0.021 g, 0.2 mmol), and the mixture was stirred for 30 min.
Work up as for 1e yielded 0.032 g (44%) of 2f as a white
solid. IR: 2334 m, 2307 w. 19F NMR d ꢀ58.77 (s, ortho-
NC5H3(CH3)2 trans, H2, 2H), 7.89 (s, NC5H3(CH3)2 cis,
H2, 2H), 7.86 (s, C6H2(CF3)3, 2H), 7.55 (s, NC5H3(CH3)2
trans, H4, 1H), 7.42 (s, NC5H3(CH3)2 cis, H4, 1H), 2.40
(s, NC5H3(CH3)2 trans, 6H), 2.30 (s, CH3CN, 3H), 2.21
(s, NC5H3(CH3)2 cis, 6H). Anal. Calc. for C25H23BF13-
N3Pd: C, 41.15; H, 3.18; N, 5.76. Found: C, 40.93; H,
3.05; N, 5.58%.
6F), ꢀ63.24 (s, para-CF3, 3F). 31P{1H} NMR d 14.5 (s).
1H NMR d 7.95 (d, JPH = 2.00, C6H2 (CF3)3, 2H), 7.68
(m, C6H4CH3, 3H), 7.56 (m, C6H4CH3, 3H), 7.39 (m,
C6H4CH3, 6H), 2.26 (s, C6H4CH3, 9H), 1.91 (s, CH3CN,
6H). 19F NMR (213 K) d ꢀ58.57 (s, ortho-CF3, 3F),
ꢀ59.78 (s, ortho-CF3, 3F), ꢀ63.20 (s, para-CF3, 3F).
31P{1H} NMR (213 K) d 13.8 (s). 1H NMR (213 K) d
8.60 (br, C6H4CH3, H6, 1H), 7.96 (d, JPH = 2.0,
C6H2(CF3)3, 2H), 7.32 (m, C6H4CH3, 11H), 3.00 (br,
C6H4CH3, CH3 endo, 3H), 1.94 (m, CH3CN and
C6H4CH3, 12H). Anal. Calc. for C34H29BF13N2PPd: C,
47.44; H, 3.40; N, 3.25. Found: C, 47.50; H, 3.50; N,
3.05%.
1
CF3, 6F), ꢀ63.26 (s, para-CF3, 3F). H NMR d 8.15 (s,
4.9. trans-[Pd(Fmes)(NCMe)(tht)2]BF4 (2i)
Tetrahydrothiophene (0.088 g, 1 mmol) and TlBF4
(0.032 g, 0.11 mmol) were successively added to a solution
of [Pd(Fmes)Cl(tht)2] [3a] (0.060 g, 0.1 mmol) in MeCN
(10 mL), and the mixture was refluxed for 15 h. Then the
volatiles were pumped off and the residue was extracted
with CH2Cl2 (20 mL) and filtered on dry Celite. The vola-
tiles were pumped off again, and the solid residue was
washed with Et2O (3 · 10 mL), yielding 0.041 g (59%) of
2i as a pale yellow solid. IR: 2328 m, 2300 m. 19F NMR
4.13. trans-[Pd(Fmes)(3,5-lut)(NCMe)2]BF4 (3f)
A 100 mL flask was successively charged with 1a
(0.060 g, 0.1 mmol), CH2Cl2 (10 mL), and 3,5-lutidine
(0.011 g, 0.1 mmol), and the mixture was stirred for
30 min. Work up as for 1e yielded 0.042 g (63%). IR:
2339 s, 2311 m. 19F NMR d ꢀ59.43 (s, ortho-CF3, 6F),
ꢀ63.23 (s, para-CF3, 3F). 1H NMR
d
8.21 (s,
NC5H3(CH3)2, H2, 2H), 7.89 (s, C6H2(CF3)3, 2H), 7.58
(s, NC5H3(CH3)2, H4, 1H), 2.44 (s, NC5H3(CH3)2, 6H),
2.28 (s, CH3CN, 6H). Anal. Calc. for C20H17BF13N3Pd:
C, 36.20; H, 2.58; N, 6.33. Found: C, 35.98; H, 2.51; N,
6.25%.
1
d ꢀ60.05 (s, ortho-CF3, 6F), ꢀ63.24 (s, para-CF3, 3F). H
NMR d 7.92 (s, C6H2(CF3)3, 2H), 2.96 (m, SC4H8, Ha,
4H), 2.53 (s, CH3CN, 3H), 2.05 (m, SC4H8, Hb, 4H). Anal.
Calc. for C19H21BF13NPdS2: C, 32.99; H, 3.06; N, 2.02.
Found: C, 32.85; H, 3.00; N, 1.73%.
4.14. trans-[Pd(Fmes)(2,4-lut)(NCMe)2]BF4 (3g)
4.10. trans-[Pd(Fmes)(NCMe)2(PPh3)]BF4 (3b)
A 100 mL flask was successively charged with 1a
(0.090 g, 0.15 mmol), CH2Cl2 (15 mL), and 2,4-lutidine
(0.016 g, 0.15 mmol), and the mixture was stirred for 2 h.
Work up as for 1e yielded 0.048 g (48%). IR: 2337 s,
2310 vw. 19F NMR d ꢀ59.30 (s, ortho-CF3, 6F), ꢀ63.23
NC5H3(CH3)2, H6, 1H), 7.91 (s, C6H2(CF3)3, 2H), 7.31
(s, NC5H3(CH3)2, H3, 1H), 7.28 (d, J = 6.0, NC5H3(CH3)2,
H5, 1H), 2.99 (s, NC5H3(CH3)2, 3H), 2.43 (s,
NC5H3(CH3)2, 3H), 2.24 (s, CH3CN, 6H). Anal. Calc.
for C20H17BF13N3Pd: C, 36.20; H, 2.58; N, 6.33. Found:
C, 35.96; H, 2.41; N, 6.04%.
A 100 mL flask was successively charged with 1a
(0.090 g, 0.15 mmol), CH2Cl2 (15 mL), and PPh3 (0.039 g,
0.15 mmol), and the mixture was stirred for 6 h. Work up
as for 1e yielded 0.110 g (89%). IR: 2343 m, 2332 m,
2315 m. 19F NMR d ꢀ59.26 (s, ortho-CF3, 6F), ꢀ63.20
(s, para-CF3, 3F). 1H NMR
d 8.50 (d, J = 6.0,
1
(s, para-CF3, 3F). 31P{1H} NMR d 19.5 (s). H NMR d
7.95 (d, JPH = 2.0, C6H2(CF3)3, 2H), 7.60 (m, C6H5,
15H), 1.91 (d, JPH = 1.0, CH3CN, 6H). Anal. Calc. for
C31H23BF13N2PPd: C, 45.48; H, 2.83; N, 3.42. Found: C,
45.45; H, 2.99; N, 3.35%.
4.11. cis-[Pd(Fmes)(NCMe)2(PPh3)]BF4 (cis-3b)
4.15. trans-[Pd(Fmes)(2,6-lut)(NCMe)2]BF4 (3h)
Assigned signals: 19F NMR d ꢀ59.04 (ortho-CF3, d,
JPF = 3.8 Hz) ꢀ63.40 (para-CF3, s). 31P{1H} NMR d
26.56, m.
A 100 mL flask was successively charged with 1a
(0.090 g, 0.15 mmol), CH2Cl2 (15 mL), and 2,6-lutidine
(0.016 g, 0.15 mmol), and the mixture was stirred for 6 h.
The volatiles were pumped off, and the residue was washed
with Et2O (3 · 10 mL), yielding 0.067 g (67%) of 2e as a
white solid. IR: 2340 s, 2310 s. 19F NMR d ꢀ58.94 (s,
ortho-CF3, 6F), ꢀ63.24 (s, para-CF3, 3F). 1H NMR d
7.94 (s, C6H2(CF3)3, 2H), 7.74 (t, J = 8.0, NC5H3(CH3)2,
H3, 1H), 7.30 (d, J = 8.0, NC5H3(CH3)2, H4, 2H), 3.19 (s,
NC5H3(CH3)2, 6H), 2.24 (s, CH3CN, 6H). Anal. Calc.
for C20H17BF13N3Pd: C, 36.20; H, 2.58; N, 6.33. Found:
C, 36.12; H, 2.56; N, 6.24%.
4.12. trans-[Pd(Fmes)(NCMe)2{P(o-Tol)3}]BF4 (3c)
A 100 mL flask was successively charged with 1a
(0.060 g, 0.1 mmol), CH2Cl2 (10 mL), and P(o-Tol)3
(0.032 g, 0.1 mmol), and the mixture was stirred for 2 h.
The volatiles were pumped off, and the residue was
washed with hexane (3 · 10 mL), yielding 0.074 g (86%)
of 2b as a white solid which was stored under nitrogen.
IR: 2334 m, 2308 m. 19F NMR d ꢀ59.19 (s, ortho-CF3,

C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
3871
4.16. [Pd(Fmes)(bipy)(NCMe)]BF4 (4a)
tated. The mixture was cooled to ꢀ20 ꢁC, and the solid
was filtered, washed with Et2O (3 · 5 mL), and dried in
vacuo. Yield: 0.096 g (70%). IR: 2324 m, 2294 w. 19F
NMR d ꢀ59.47 (t, JPF = 5.0, ortho-CF3, 6F), ꢀ63.33 (s,
para-CF3, 3F). 31P{1H} NMR d 56.6 (dsept, JPP = 21.0,
Method A. 2,20-Bipyridyl (0.017 g, 0.11 mmol) was
added to a solution of 1a (0.060 g, 0.1 mmol) in CH2Cl2
(10 mL), and the mixture was stirred for 30 min. Then
hexane was added (ca. 10 mL) and the solution was con-
centrated in vacuo, and cooled to ꢀ20 ꢁC. The microcrys-
talline white solid obtained was filtered, washed with
hexane (3 · 5 mL), and dried in vacuo. Yield: 0.047 g
(70%). Method B. A 100 mL flask was successively charged
with [Pd(Fmes)Cl(bipy)] [3a] (0.058 g, 0.1 mmol), MeCN
(10 mL), and TlBF4 (0.032 g, 0.11 mmol), and the mixture
was refluxed for 45 min. The volatiles were pumped off,
CH2Cl2 (10 mL) was added to the residue, which was fil-
tered on dry Celite. Work up as before gave 0.037 g
(55%) of 5b. IR: 2333 w, 2307 w. 19F NMR d ꢀ60.03 (s,
ortho-CF3, 6F), ꢀ63.17 (s, para-CF3, 3F). 1H NMR d
8.97 (d, J = 5.0, C5H4N, H6 cis to NCMe, 1H), 8.35 (d,
J = 8.0, C5H4N, 1H), 8.25 (d, J = 8.0, C5H4N, 1H), 8.19
(td, J = 8.0 and 1.0, C5H4N, 1H), 8.12 (td, J = 8.0 and
1.0, C5H4N, 1H), 8.01 (s, C6H2(CF3)3, 2H), 7.95 (t,
J = 6.0, C5H4N, 1H), 7.33 (t, J = 6.0, C5H4N, 1H), 7.23
(t, J = 5.0, C5H4N, 1H), 2.52 (s, CH3CN, 3H). Anal. Calc.
for C21H13BF13N3Pd: C, 37.56; H, 1.95; N, 6.26. Found: C,
37.38; H, 1.98; N, 5.93%.
1
JPF 5.0, P cis, 1P), 48.5 (d, JPP = 21.0, P trans, 1P). H
NMR d 7.65 (m, C6H5, 10H), 7.63 (d, JPH = 3.0,
C6H2(CF3)3, 2H), 7.37 (m, C6H5, 2H), 7.15 (m, C6H5,
8H), 3.00 (m, PCH2, 2H), 2.40 (m, PCH2, 2H), 2.14 (s,
CH3CN, 3H). Anal. Calc. for C37H29BF13NP2Pd: C,
48.63; H, 3.20; N, 1.53. Found: C, 48.37; H, 3.15; N, 1.20%.
4.19. [Pd(Fmes)(NCMe)(OPPhPy2-N,N0)]BF4 (4d)
OPPhPy2 (0.062 g, 0.22 mmol) were added to a solution
of 1a (0.119 g, 0.2 mmol) in CH2Cl2 (15 mL), and the solu-
tion was stirred for 1 h. Work up as for 1e yielded 0.149 g
(94%) of a beige solid. IR: 2334 m, 2315 w. 19F NMR d
ꢀ58.63 (br, ortho-CF3, 3F), ꢀ63.20 (br, ortho-CF3, 3F),
ꢀ63.31 (s, para-CF3, 3F). 19F NMR (213 K) d ꢀ58.69 (s,
ortho-CF3, 3F), ꢀ62.84 (s, ortho-CF3, 3F), ꢀ63.45 (s,
1
para-CF3, 3F). 31P{1H} NMR d 21.5 (s). H NMR d 9.36
(d, J = 5.0, C5H4N, H6 cis to NCMe, 1H), 8.85 (br,
C5H4N, H3, 1H), 8.49 (br, C5H4N, H3, 1H), 8.29 (m,
C5H4N, H4, 1H), 8.18 (m, C5H4N, H5 and H6, 2H), 8.02
(s, C6H2(CF3)3, 1H), 7.73 (m, C5H4N, H4, 6H), 7. 67 (s,
C6H2(CF3)3, 1H), 7.33 (m, C5H4N, H5, 1H), 2.39 (s,
NCCH3, 3H). Anal. Calc. for C27H18BF13N3OPPd: C,
40.76; H, 2.28; N, 5.28. Found: C, 40.52; H, 2.56; N, 4.85%.
4.17. [Pd(Fmes)(NCMe)(tmeda)]BF4 (4b)
Method A. Tetramethylethylenediamine (0.006 g,
0.051 mmol) was added to a solution of 1a (0.030 g,
0.05 mmol) in MeCN (10 mL), and the mixture was stirred
for 3 h. The volatiles were pumped off and the white solid
was washed with Et2O (3 · 5 mL) and dried in vacuo.
Yield: 0.047 g (70%). Method B. A 100 mL flask was suc-
cessively charged with [Pd(Fmes)Cl(tmeda)] [3e] (0.108 g,
0.2 mmol), MeCN (15 mL), and TlBF4 (0.064 g,
0.22 mmol), and the mixture was refluxed for 24 h. The vol-
atiles were pumped off, CH2Cl2 (20 mL) was added to the
residue, which was filtered on dry Celite. Then Et2O was
added (ca. 20 mL) and the solution was cooled to
ꢀ20 ꢁC. The microcrystalline off-white solid obtained was
filtered, washed with Et2O (3 · 5 mL), and dried in vacuo.
Yield: 0.103 g (82%). IR: 2333 m, 2307 w. 19F NMR d
4.20. [Pd(Fmes)(NCMe){(OH)(CH3)CPy2-N,N0}]BF4
(4e)
(OH)(CH3)CPy2 (0.022 g, 0.11 mmol) was added to a
solution of 1a (0.060 g, 0.1 mmol) in CH2Cl2 (10 mL) and
the solution was stirred for 3 h. Then hexane was added
(ca. 10 mL) and the solution was concentrated in vacuo,
and cooled to ꢀ20 ꢁC. The microcrystalline white solid
obtained was filtered, washed with hexane (3 · 5 mL), and
dried in vacuo, yielding 0.052 g of 4e (73%). IR: 2340 m,
2313 m. 19F NMR (Me2CO-d6) d ꢀ57.71 (br, ortho-CF3,
3F), ꢀ59.30 (br, ortho-CF3, 3F), ꢀ62.09 (s, para-CF3, 3F).
1H NMR (Me2CO-d6) d 9.09 (d, J = 5.5, C5H4N, H6 cis to
NCMe, 1H), 8.45 (d, J = 8.0, C5H4N, H3, 1H), 8.32 (d,
J = 4.0, C5H4N, H3, 1H), 8.25 (m, H4 and H5 of C5H4N
and C6H2(CF3)3, 3H), 8.06 (s, C6H2(CF3)3, 1H), 7.88 (d,
J = 5.5, C5H4N, H6, 1H), 7.76 (dd, J = 5.0 and 9.5,
C5H4N, H4, 1H), 7.30 (ddd, J = 7.5, 6.0 and 1.5, C5H4N,
H5, 1H), 6.71 (s, OH, 1H) 3.00 (s, CH3, 3H), 2.57 (s,
NCCH3, 3H). Anal. Calc. for C23H17BF13N3OPd: C,
38.61; H, 2.39; N, 5.87. Found: C, 38.05; H, 2.40; N, 5.45%.
1
ꢀ58.17 (s, ortho-CF3, 6F), ꢀ63.30 (s, para-CF3, 3F). H
NMR d 7.87 (s, C6H2(CF3)3, 2H), 2.94 (m, NCH2, 2H),
2.88 (m, NCH2, 2H), 2.84 (s, NCH3, 6H), 2.37 (s, NCH3,
6H), 2.35 (s, CH3CN, 3H). Anal. Calc. for
C17H21BF13N3Pd: C, 32.33; H, 3.35; N, 6.65. Found: C,
32.02; H, 3.14; N, 6.45%.
4.18. [Pd(Fmes)(NCMe)(dppe)]BF4 (4c)
1,2-Bis(diphenylphosphino)ethane (0.060 g, 0.15 mmol)
was added to a solution of 1a (0.090 g, 0.15 mmol) in
MeCN (10 mL), and the mixture was stirred for 1 h. The
solution was concentrated in vacuo to ca. 1–2 mL, and
Et2O was added to the residue until a white solid precipi-
4.21. [Pd(Fmes)(NCMe)2(OH2)]BF4 (5a)
Assigned signals: 19F NMR (253 K) d ꢀ59.75 (s, ortho-
1
CF3, 6F), ꢀ63.00 (s, para-CF3, 3F). H NMR (253 K) d
2.30 (s, H2O).

3872
C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
4.22. [Pd(Fmes)(OH2)(OPPhPy2-N,N0)]BF4 (5d)
Table 3
Crystal data and structure refinement for 3f and 4e
Assigned signals: 19F NMR d ꢀ58.55 (br, ortho-CF3,
3F), ꢀ62.89 (br, ortho-CF3, 3F), ꢀ63.23 (s, para-CF3,
3f
4e
Empirical formula
Formula weight
Temperature (K)
C30H29BF13N3Pd C23H17BF13N3OPd
1
3F). H NMR d 9.17 (d, J = 5.0, C5H4N, H6 cis to OH2,
795.77
298(2)
0.71073
Monoclinic
P2(1)/n
12.5397(17)
16.121(2)
16.559(2)
90
97.048(4)
90
3322.0(8)
4
715.61
298(2)
0.71073
Triclinic
1H), 7.89 (s, C6H2(CF3)3, 1H), 7.53 (s, C6H2(CF3)3, 1H),
˚
Wavelength (A)
7.33 (m, C5H4N, H5, 1H).
Crystal system
Space group
ꢀ
P1
4.23. [Pd(Fmes)(OH2){(OH)(CH3)CPy2-N,N0}]BF4
(5e)
˚
a (A)
8.459(3)
11.856(4)
13.354(4)
95.595(9)
91.412(9)
94.905(7)
1327.3(7)
2
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
Assigned signals: 19F NMR (Me2CO-d6) d ꢀ57.87 (s,
ortho-CF3, 3F), ꢀ59.11 (s, ortho-CF3, 3F), ꢀ62.09 (s,
1
3
˚
para-CF3, 3F). H NMR (Me2CO-d6) d 8.93 (d, J = 5.5,
V (A )
C5H4N, H6 cis to OH2, 1H), 6.69 (s, COH, 1H), 3.07 (s,
CH3, 3H).
Z
Dcalc (g cmꢀ3
)
1.591
0.657
1592
1.791
0.814
704
Absorption coefficient (mmꢀ1
F(000)
)
4.24. trans-[Pd(Fmes)(OH2)(tht)2]BF4 (5i)
Crystal size (mm)
H Range for data
collection
Reflections collected
Independent reflections
Absorption correction
Maximum and
minimum transmission factor
Data/restraints/parameters
Goodness-of-fit on F2
R1 [I > 2r(I)]
0.09 · 0.08 · 0.08 0.07 · 0.05 · 0.01
1.77–23.25ꢁ
1.53–23.40ꢁ
A solution of 2i (0.035 g, 0.05 mmol) in THF/H2O
(10 mL, 1:1) was refluxed for 6 h. The solvents were then
removed in vacuo and the residue was extracted with
Et2O (3 · 5 mL). The colorless solution was concentrated
in vacuo, and cooled to ꢀ20 ꢁC. The microcrystalline white
solid obtained was filtered, washed with cold hexane
(3 · 3 mL), and dried in vacuo, yielding 0.010 g (30%).
19F NMR d ꢀ60.11 (s, ortho-CF3, 6F), ꢀ63.22 (s, para-
15741
4762
6347
3848
SADABS
SADABS
1.000000 and
0.639593
4762/0/439
1.009
0.0471
0.1300
1.000000 and
0.577003
3848/0/385
1.000
0.0616
0.1501
wR2 (all data)
1
CF3, 3F). H NMR d 7.89 (s, C6H2(CF3)3, 2H), 2.92 (m,
SC4H8, Ha, 8H), 2.24 (br, H2O, 2H), 2.00 (m, SC4H8,
Hb, 8H).
Crystallographic data (excluding structure factors) for the
structures 3f and 4e reported in this paper have been depos-
ited with the Cambridge Crystallographic Data Centre as
Supplementary publication with the following deposition
numbers: CCDC-294981, and CCDC-294982 for complexes
3f, and4e respectively. Copies of the data can be obtained
free of charge on application to the CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK, fax: (internat.) +44 1223 336
033, email: deposit@ccdc.cam.ac.uk.
4.24.1. Experimental procedure for X-ray crystallography
Suitable single crystals of 3f and 4e were grown by slow
diffusion of a concentrated dichloromethane solution of the
complex into diethylether at ꢀ20 ꢁC and mounted in glass
fibers, and diffraction measurements were made using a
Bruker SMART CCD area-detector diffractometer with
˚
Mo Ka radiation (k = 0.71073 A) [21]. Intensities were inte-
grated from several series of exposures, each exposure cov-
ering 0.3ꢁ in x, the total data set being a hemisphere [22].
Absorption corrections were applied, based on multiple
and symmetry-equivalent measurements [23]. The structure
was solved by direct methods and refined by least squares
on weighted F2 values for all reflections (see Table 3) [24].
All non-hydrogen atoms were assigned anisotropic dis-
placement parameters and refined without positional con-
straints. Hydrogen atoms were taken into account at
calculated positions and their positional parameters were
refined. The hydrogen atom of the hydroxyl group was
located in a difference Fourier map and refined. To correct
the systematic shortening of O–H bonds measured by X-ray
diffraction, the H atom position was normalized. An O–H
Acknowledgments
´
Financial support by the Ministerio de Educacion y
Ciencia (Project CTQ2004-07667/BQU) and the Junta de
´
Castilla y Leon (Project No. VA 060/04) is very gratefully
acknowledged.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.jorganchem.
2006.05.039.
References
˚
distance of 0.967 A was imposed, a value that has been
observed in alcohols by neutron diffraction [15]. Refinement
proceeded smoothly to give R1 = 0.0616 for 3f and
R1 = 0.0471 for 4e based on the reflections with I > 2r(I).
Complex neutral-atom scattering factors were used [25].
[1] (a) F.T. Edelmann, Comments Inorg. Chem. 12 (1992) 259, and
references therein;
(b) R.D. Schulter, A.H. Cowley, D.A. Atwood, R.A. Jones, M.R.
Bond, C.J. Carrano, J. Am. Chem. Soc. 115 (1993) 2070;

C. Bartolome´ et al. / Journal of Organometallic Chemistry 691 (2006) 3862–3873
3873
(c) R.D. Schulter, H.S. Isom, A.H. Cowley, D.A. Atwood, R.A.
Jones, F. Olbrich, S. Corbelin, R.J. Lagow, Organometallics 163
(1994) 4058.
[9] C.A. McAuliffe, in: G. Wilkinson, R.D. Gillard, J.A. McCleverty
(Eds.), Comprehensive Coordination Chemistry, vol. 2, Pergamon,
Oxford, 1987, p. 1012.
[2] (a) V.C. Gibson, C. Redshaw, L.J. Sequeira, K.B. Dillon, W.
Clegg, M.R.J. Elsegood, J. Chem. Soc., Chem. Commun. (1996)
2151;
[10] R.A. Widenhoefer, H.A. Zhong, S.L. Buchwald, Organometallics 15
(1996) 2745, and references therein.
[11] (a) A.K. Colter, I.I. Schuster, R.J. Kurland, J. Am. Chem. Soc. 87
(1965) 2278;
(b) A.K. Colter, I.I. Schuster, R.J. Kurland, J. Am. Chem. Soc. 87
(1965) 2279.
[12] (a) A. Domenicano, A. Vaciago, C.A. Coulson, Acta Crystallogr.
Sect. B 31 (1975) 221;
(b) K.B. Dillon, V.C. Gibson, J.A.K. Howard, C. Redshaw, L.
Sequeira, J.W. Yao, J. Organomet. Chem. 528 (1997) 179;
(c) M. Belay, F.T. Edelmann, J. Organomet. Chem. 479 (1994)
C21;
´
(d) P. Espinet, S. Martın-Barrios, F. Villafane, P.G. Jones,
˜
Organometallics 19 (2000) 290.
(b) A. Domenicano, P. Mazzeo, A. Vaciago, Tetrahedron Lett.
(1976) 1029;
´
´
[3] (a) C. Bartolome, P. Espinet, F. Villafane, S. Giesa, A. Martın, A.G.
˜
Orpen, Organometallics 15 (1996) 2019;
(c) A. Domenicano, P. Murray-Rust, A. Vaciago, Acta Crystallogr.
Sect. B 39 (1983) 457, and references therein;
(d) R. Norrestam, L. Schepper, Acta Chem. Scand. Ser. A 35 (1981)
91.
´
´
´
(b) C. Bartolome, R. de Blas, P. Espinet, J.M. Martın-Alvarez, F.
Villafane, Angew. Chem., Int. Ed. 40 (2001) 2521;
˜
´
(c) C. Bartolome, P. Espinet, L. Vicente, F. Villafane, J.P.H.
˜
Charmant, A.G. Orpen, Organometallics 21 (2002) 3536;
[13] We are convinced that the minor signals (6%, 3%) correspond to
the aqua complexes cis and trans-[Pd(Fmes)(OH2)(lut)2] formed
by substitution of 2,4-lut with traces of water present in the
solvent.
´
(d) C. Bartolome´, P. Espinet, J.M. Mart´ın-Alvarez, F. Villafane,
˜
Inorg. Chim. Acta 347 (2003) 49;
´
´
´
(e) C. Bartolome, P. Espinet, J.M. Martın-Alvarez, F. Villafane, Eur.
˜
´
J. Inorg. Chem. (2003) 3127;
[14] J.A. Casares, P. Espinet, J.M. Martınez-Ilarduya, Y.-S. Lin, Organo-
´
´
´
(f) C. Bartolome, P. Espinet, J.M. Martın-Alvarez, F. Villafane, Eur.
˜
metallics 16 (1997) 770.
J. Inorg. Chem. (2004) 2326.
[4] (a) See, for example: A. Sen, T.W. Lai, R.R. Thomas, J. Organomet.
Chem. 358 (1988) 567;
[15] F.H. Allen, O. Kennard, D.G. Watson, L. Brammer, A.G. Orpen,
R. Taylor, J. Chem. Soc., Perkin Trans. 2 (1987) S1.
[16] D.D. Perrin, W.L.F. Armarego, Purification of Laboratory Chemi-
cals, third ed., Pergamon Press, Oxford, 1988.
[17] M. Scholz, H.W. Roesky, D. Stalke, K. Keller, F.T. Edelmann, J.
Organomet. Chem. 366 (1989) 73.
(b) A. Sen, Z. Jiang, Organometallics 12 (1993) 1406;
(c) L.K. Johnson, S. Mecking, M. Brookhart, J. Am. Chem. Soc. 118
(1996) 267;
´
(d) S.A. Svejda, M. Brookhart, Organometallics 18 (1999) 65.
[18] F.J. Arnaiz, J. Chem. Educ. 74 (1997) 1332.
´
´
[5] R. Lopez Fernandez, PhD thesis, University of Valladolid, Spain,
[19] W. Geary, Coord. Chem. Rev. 7 (1971) 81.
[20] (a) T.D. Dubois, D.W. Meek, Inorg. Chem. 8 (1969) 146;
(b) R.L. Dutta, D.W. Meek, D.H. Busch, Inorg. Chem. 9 (1970)
1215.
[21] SMART V5.051 Diffractometer Control Software, Bruker Analytical X-
ray Instruments Inc., Madison, WI, 1998.
[22] SAINT V6.02 Integration Software, Bruker Analytical X-ray Instru-
ments Inc., Madison, WI, 1999.
[23] G.M. Sheldrick, SADABS: A Program for Absorption Correction with
the Siemens SMART System, University of Go¨ttingen, Germany,
1996.
2004.
[6] J. Vicente, A. Arcas, Coord. Chem. Rev. 249 (2005) 1135.
[7] (a) P. Leoni, M. Sommovigo, M. Pasquali, S. Midollini, D. Braga,
P. Sabatino, Organometallics 10 (1991) 1038;
(b) P.J. Stang, D.H. Cao, G.T. Poulter, A.T. Arif, Organometallics
14 (1995) 1110;
(c) M. Akita, T. Miyaji, N. Muroga, C. Mock-Knoblauch, W. Adam,
S. Hikichi, Y. Moro-oka, Inorg. Chem. 39 (2000) 2096;
(d) M. Fuss, H.U. Siehl, B. Olenyuk, P.J. Stang, Organometallics 18
(1999) 758.
[8] For the concepts of ‘‘antisymbiotic effect’’ and ‘‘transphobia’’ see,
respectively: R.G. Pearson, Inorg. Chem. 12 (1973) 712;
[24] SHELXTL Program System Version 5.1; Bruker Analytical X-ray
Instruments Inc., Madison, WI, 1998.
´
J. Vicente, J.A. Abad, A.D. Frankland, M.C. Ramırez de Arellano,
Chem. Eur. J. 5 (1999) 3066.
[25] International Tables for Crystallography, vol. C. Kluwer, Dordrecht,
1992.