Five different fluxional processes in polyfluorophenyl palladium(II) complexes with 2,4,6-tris(3,5-dimethylpyrazol-1-yl)-1,3,5-triazine. The driving effect of the solvent

Article abstract of DOI:10.1021/ic020522a

The polydentate N-donor ligand 2,4,6-tris(3,5-dimethylpyrazol-1-yl)-1,3,5-triazine (Me₂-TpzT) has been used to synthesize the new palladium derivatives Pd(R)₂(Me₂-TpzT), R = C₆F₅, 1; R = m-C₆CIF₄, 2. In the case of complex 2, four different atropisomers have been detected at low temperature. The new complexes exhibit a rich dynamic behavior, including three metallotropic processes (metal-hurdling, 1,4-metallotropic shifts, and an intermolecular process) and two processes involving restricted rotation of aromatic rings (the polyfluorophenyl groups and the uncoordinated pyrazole group adjacent to the metal fragment). The fluxional behavior has been studied by ¹H and ¹⁹F NMR spectroscopy using variable temperature NMR studies and ¹H,¹H and ¹⁹F,¹⁹F EXSY experiments. The study of solutions of 2 in 1,1′,2,2′-tetrachloroethane-d₂ gave the following order for the free energy of activation: pyrazole rotation < 1,4-metallotropic shift < intermolecular exchange < polyfluorophenyl rotation. The process of metal hurdling was not found in this solvent. However, in acetone-d₆ such a process was detected and was found to be of lower energy than the 1,4-metallotropic shift. In dilute acetone-d₆ or 1,1′,2,2′-tetrachloroethane-d₂ solutions, the intermolecular process was not observed. Conclusions concerning the different mechanisms have been deduced from the data obtained.

Full text of DOI:10.1021/ic020522a

Inorg. Chem. 2003, 42, 885−895
Five Different Fluxional Processes in Polyfluorophenyl Palladium(II)
Complexes with 2,4,6-Tris(3,5-dimethylpyrazol-1-yl)-1,3,5-triazine. The
Driving Effect of the Solvent
M. Carmen Carrio´n, Ana Guerrero, Fe´lix A. Jalo´n, Blanca R. Manzano,* and Antonio de la Hoz
Departamento de Qu´ımica Inorga´nica, Orga´nica y Bioqu´ımica, UniVersidad de Castilla-
La Mancha, Facultad de Qu´ımicas, AVda. Camilo J. Cela, 10, E-13071 Ciudad Real, Spain
Rosa M. Claramunt and Viktor Milata†
Departamento de Qu´ımica Orga´nica y Biolog´ıa, Facultad de Ciencias, UNED, Senda del Rey, 9,
E-28040 Madrid, Spain
Jose´ Elguero
Instituto de Qu´ımica Me´dica, CSIC, Juan de la CierVa, 3, E-28006 Madrid, Spain
Received August 26, 2002
The polydentate N-donor ligand 2,4,6-tris(3,5-dimethylpyrazol-1-yl)-1,3,5-triazine (Me -TpzT) has been used to
2
synthesize the new palladium derivatives Pd(R)2(Me2-TpzT), R ) C F , 1; R) m-C ClF , 2. In the case of complex
6
5
6
4
2, four different atropisomers have been detected at low temperature. The new complexes exhibit a rich dynamic
behavior, including three metallotropic processes (metal-hurdling, 1,4-metallotropic shifts, and an intermolecular
process) and two processes involving restricted rotation of aromatic rings (the polyfluorophenyl groups and the
uncoordinated pyrazole group adjacent to the metal fragment). The fluxional behavior has been studied by 1H and
19F NMR spectroscopy using variable temperature NMR studies and 1H,1H and 19F,19F EXSY experiments. The
study of solutions of 2 in 1,1′,2,2′-tetrachloroethane-d2 gave the following order for the free energy of activation:
pyrazole rotation < 1,4-metallotropic shift < intermolecular exchange < polyfluorophenyl rotation. The process of
metal hurdling was not found in this solvent. However, in acetone-d6 such a process was detected and was found
to be of lower energy than the 1,4-metallotropic shift. In dilute acetone-d6 or 1,1′,2,2′-tetrachloroethane-d2 solutions,
the intermolecular process was not observed. Conclusions concerning the different mechanisms have been deduced
from the data obtained.
Introduction
processes1-11 with correlation, in some cases, between the
free energy of activation and the strength of the bond to be
broken.6 In the work reported here we chose the ligand 2,4,6-
tris(3,5-dimethylpyrazol-1-yl)-1,3,5-triazine (Me2-TpzT) (see
Chart 1), mainly because of the rich fluxional behavior that
Fluxional processes are very common in coordination and
organometallic chemistry. Very frequently these processes
are accompanied by rupture of metal-atom bonds and also
by conformational changes in the molecule. We have recently
been interested in fluxional processes in palladium complexes
containing N-donor ligands. In these types of derivatives we
and others have found evidence of Pd-N bond rupture
(1) (a) Gogoll, A.; O¨ rnebro, J.; Grennberg, H.; Ba¨ckvall, J. E. J. Am. Chem.
Soc. 1994, 116, 3631. (b) Gogoll, A.; Grennberg, H.; Axe´n, A.
Organometallics 1997, 16, 1167.
(2) Albinati, A.; Kunz, R. W.; Amman, C. J.; Pregosin, P. S. Organo-
metallics 1991, 10, 1800.
* To whom correspondence should be addressed. Fax: 34-926.29.53.18.
Blanca.Manzano@uclm.es
(3) Delis, J. G. P.; Aubel, P. G.; Vrieze K.; van Leeuwen, P. W. N. M.;
Veldman, N.; Spek, A. L.; van Neer, F. J. R. Organometallics 1997,
16, 2948.
(4) Canovese, L.; Visentin, F.; Chessa, G.; Niero, A.; Uguagliati, P. Inorg.
Chim. Acta 1999, 293, 44.
† On leave from the Department of Organic Chemistry, Faculty of
Chemical and Food Technology, Slovak Technical University, Bratislava,
Slovak Republic SK-812 37.
10.1021/ic020522a CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/16/2003
Inorganic Chemistry, Vol. 42, No. 3, 2003 885

Carrio´n et al.
proposed. Orrell et al.,9 by means of a study with asymmetric
terpyridine ligands coordinated to palladium or platinum
fragments, demonstrated that the mechanism is associatives
implying a quasi-terdentate ligand intermediate and a “tick-
tock” motion of the metal moiety. When the appropriate
coordination sites are available, Orrell also found another
process in palladium or platinum derivatives called metal
hurdling,10 which implies partial dissociation of the metal
fragment by cleavage of a metal-nitrogen bond (M-Ntriazine
in our case) followed by 180° rotation of the metallic group
bonded to the other nitrogen (pyrazole in our case) about
the C-N bond and, finally, recoordination to the new
nitrogen (triazine) atom. In our case such a process would
involve, for example, the change from the A1 to the A3
isomer. Restricted rotation of uncoordinated pyrazole12 or
pyridine8,10,11 rings adjacent to the metal center has been
found to occur at low temperature. In our complexes with
our ligand it is possible for the two aforementioned metallo-
tropic processes to occur. The metal-hurdling process requires
the Pd-N(triazine) bond to be broken and not the Pd-
N(pyrazole) bond, while the 1,4-metallotropic shift involves
cleavage of at least the M-N(pyrazole) bond. For this reason,
we chose a ligand with two heterocycles of different basicity
and, consequently, Pd-N bonds of different strengths in an
attempt to obtain different free energies of activation for the
two processes. When comparing these two processes, we
were interested in evaluating the influence of the coordinating
ability of the solvent, a factor that has not been studied
previously and could give a greater insight into the possible
mechanisms.
Chart 1
their metal derivatives could exhibit. This ligand possesses
six bidentate coordination sites, and when the ligand is
coordinated to a single metallic fragment, it is possible that
six identical (degenerate) isomers can exist. If we denote
the three pyrazole rings as A, B, and C, the three nitrogen
atoms of the triazine as 1, 3, and 5, and consider the nitrogen
atoms that are coordinated, we can designate these isomers
as A1, A3, B3, B5, C5, and C1 (see Chart 1, isomer A1 is
represented as an example). Orrell et al. have studied in detail
the fluxional behavior of Re complexes of this ligand12 as
well as different coordination complexes (including pal-
ladium derivatives8-11) and similar ligands such as pyrazolyl-
pyridines,13 pyrazolylbipyridine,14 2,4,6-tris(pyrazolyl)pyrim-
idines,15 2,2′:6′,2′′-terpyridine8,9,11 and 2,4,6-tris(2-pyridyl)-
pyrimidine.10 Processes involving 1,4-metallotropic shifts
were described for these types of systems.8-17
In the case of our ligand this would imply, for example,
changing from the A1 to the C1 isomer. There has been some
controversy regarding the mechanism of this process, and
both associative8,10-15 and dissociative17 pathways have been
The restricted rotation of the uncoordinated pyrazole ring
adjacent to the metal center was also a possible process, and
this would be favored by the presence of the methyl
substituents on pyrazolyl groups. Another aim of our work
was to analyze the possible presence of intermolecular
processes similar to those that we detected in allyl-palladium
derivatives with similar ligands.7 The metallic fragments
chosen were bis(polyfluorophenyl)palladium systems because
they usually give Pd-C bonds of high stability and also
because 19F NMR could give supplementary information
about this system. We also studied other fragments, such as
allylpalladium(II), with these types of ligands, but in these
cases hydrolysis of the ligand occurred.6 In addition, the
reaction of Pd(C6F5)2(cod) (cod ) 1,5-cyclooctadiene) with
2,4,6-tris(4-methylpyrazol-1-yl)-1,3,5-triazine was previously
studied and was found to give complex mixtures.6 For this
reason, we chose the ligand with a dimethylpyrazolyl group
in order to obtain a more basic heterocycle that could stabilize
the final complex. The presence of polyfluorophenyl groups
in the new derivatives also led to the possibility of analyzing
any possible process involving restricted rotation of these
rings, an area that has been previously studied in a wide
variety of systems.8,10,18 Besides the bis(pentafluorophenyl)
derivative, we also synthesized the corresponding complex
with the m-C6ClF4 group. This ring, which contains two
different sides, in the case of restricted rotation could give
rise to the formation of atropisomerssthus providing supple-
mentary information about the fluxional processes.
(5) (a) Elguero, J.; Fruchier, A.; de la Hoz, A.; Jalo´n, F. A.; Manzano, B.
R.; Otero, A.; Go´mez-de la Torre, F. Chem. Ber. 1996, 129, 589. (b)
Go´mez-de la Torre, F.; Jalo´n, F. A.; Lo´pez-Agenjo, A.; Manzano, B.
R.; Rodr´ıguez, A.; Sturm, T.; Weissensteiner, W.; Mart´ınez-Ripoll,
M. Organometallics 1998, 17, 4634. (c) Elguero, J.; Guerrero, A.;
Go´mez-de la Torre, F.; de la Hoz, A.; Jalo´n, F. A.; Manzano, B. R.;
Rodr´ıguez, A. M. New. J. Chem. 2001, 25, 1050.
(6) Go´mez-de la Torre, F.; de la Hoz, A.; Jalo´n, F. A.; Manzano, B. R.;
Otero, A.; Rodr´ıguez, A. M.; Rodr´ıguez-Pe´rez, M. C.; Echevarr´ıa, A.;
Elguero, J. Inorg. Chem. 1998, 37, 6606.
(7) Go´mez-de la Torre, F.; de la Hoz, A.; Jalo´n, F. A.; Manzano, B. R.;
Rodr´ıguez, A. M.; Elguero, J.; Mart´ınez-Ripoll, M. Inorg. Chem. 2000,
39, 1152.
(8) Abel, E. W.; Orrell, K. G.; Osborne, A. G.; Pain, H. M.; Sik, V.;
Hursthouse, B.; Abdul Malik, K. M. J. Chem. Soc., Dalton Trans.
1994, 3441.
(9) Abel, E. W.; Gelling, A.; Orrell, K. G.; Osborne, A. G.; Sik, V. Chem.
Commun. 1996, 2329.
(10) (a) Gelling, A.; Olsen, M. D.; Orrell, K. G.; Osborne, A. G.; Sik, V.
Chem. Commun. 1997, 587. (b) Gelling, A.; Olsen, M. D.; Orrell, K.
G.; Osborne, A. G.; Sik, V. Inorg. Chim. Acta 1997, 264, 257.
(11) Gelling, A.; Orrell, K. G.; Osborne, A. G.; Sik, V. J. Chem. Soc.,
Dalton Trans. 1998, 937.
(12) Gelling, A.; Orrell, K. G.; Osborne, A. G.; Sik, V. J. Chem. Soc.,
Dalton Trans. 1996, 3371.
(13) Garc´ıa, M. A. M.; Gelling, A.; Noble, D. R.; Orrell, K. G.; Osborne,
A. G.; Sik, V. Polyhedron 1996, 15, 371.
(14) Gelling, A.; Orrell, K. G.; Osborne, A. G.; Sik, V. J. Chem. Soc.,
Dalton Trans. 1994, 3545.
(15) Gelling, A.; Noble, D. R.; Orrell, K. G.; Osborne, A. G.; Sik, V. J.
Chem. Soc., Dalton Trans. 1996, 3065.
(16) Civitello, E. R.; Dragovich, P. S.; Karpishin, T. B.; Novick, S. G.;
Bierach, G.; O’Connell, J. F.; Westmoreland, T. D. Inorg. Chem. 1993,
32, 237.
(17) Rotondo, E.; Giordano, G.; Minniti, D. J. Chem. Soc., Dalton Trans.
Lett. 1996, 253.
886 Inorganic Chemistry, Vol. 42, No. 3, 2003

Fluxional Processes in PdR2(Me2-TpzT)
Free energies of activation were calculated22 from the coalescence
temperature (Tc) and the frequency difference between the coalesc-
ing signals (extrapolated at the coalescence temperature) with the
formula ∆Gcq ) aT[9.972 + log(T/δν)], a ) 1.914 × 10-2. The
spectra were recorded each 10° and each 1° in the proximities of
the coalescence temperature. We have only considered the signals
separated with δν high enough to make the calculations accurate.
In the EXSY experiments the formula ∆Gq ) aT[10.319 + log(T/
k)], a ) 1.914 × 10-2, was used. The estimated error in the
calculated free energies of activation is (1.0 kJ‚mol-1. We have
tried to dry the deuterated solvents as much as possible in order to
avoid any effect of water. In any case, the results are reproducible.
Preparation of Complexes. (a) [Pd(C6F5)2(Me2-TpzT)] (1). To
a solution of Me2-TpzT (63.2 mg, 0.17 mmol) in tetrahydrofuran
(THF; 35 mL) was added Pd(C6F5)2(cod) (95.5 mg, 0.17 mmol).
The mixture was stirred for 2 h at room temperature, and the
solution was partially evaporated and hexane added. The resulting
pale yellow solid obtained was separated by filtration and washed
with hexane (2 × 10 mL). Yield: 89.2 mg (65%). Anal. Calcd for
C30H21F10N9Pd: C, 44.82; H, 2.63; N, 15.68. Found: C, 44.65; H,
2.55; N, 15.43. 13C NMR (1,1′,2,2′-tetrachloroethane-d2, room temp)
δ 162.6 (s, Cquat, triazine), 162.4 (s, 2C, Cquat, triazine), 157.0 (bs,
Experimental Section
General Comments. All manipulations were carried out under
an atmosphere of dry oxygen-free nitrogen using standard Schlenk
techniques. Solvents were distilled from the appropriate drying
agents and degassed before use. The preparation of the ligand Me2-
TpzT has been described previously.19 [Pd(C6F5)2(cod)] and [Pd(m-
C6ClF4)2(cod)] were synthesized as described in the literature for
similar complexes.20 Elemental analyses were performed using a
Thermo Quest FlashEA 1112 microanalyzer. IR spectra were
recorded in Nujol mulls on a Perkin-Elmer 883 (200-4000 cm-1
range) spectrophotometer. NMR spectra were recorded on a
VARIAN UNITY 300 spectrometer operating at 299.980 MHz for
1H: bs ) broad singlet; dm ) doublet of multiplets. Spectra were
recorded at the temperature indicated ((0.1 K) with a probe
calibrated with methanol. The standard VARIAN pulse sequence
was used (VNMR 6.1 software, COSY and EXSY pulse sequences).
The NOE difference spectra were recorded with the following
conditions: 4000 Hz; acquisition time, 3.744 s; pulse width, 90°;
relaxation delay, 5 s; irradiation power, 5-10 dB. 19F,19F COSY
spectra were acquired using a 30604 kHz spectral width; 24
transients of 4096 data points were collected for each 240 t1
increments. A 1 s relaxation delay, a 12 µs (45°) pulse width, and
a 0.134 s acquisition time were used. A second 19F,19F COSY
spectrum was acquired using a 2572 kHz spectral width; 8 transients
of 512 data points were collected for each 214 t1 increments. A 1
s relaxation delay, a 12 µs (45°) pulse width, and a 0.19 s acquisition
time were used. The data were processed using zero filling and
sine-bell functions in both dimensions before Fourier transformation.
The 2D exchange spectra (EXSY) were acquired in the phase-
sensitive mode using the States Haberkorn method.21 Typically for
1H,1H EXSY, a 2130 kHz spectral width and 16 transients of 512
data points were collected for each 176 t1 increments. A 1 s
relaxation delay, a 23 µs (90°) pulse width, and a 0.24 s acquisition
time were used. For 19F,19F EXSY experiments, a 2692 kHz spectral
width and 16 transients of 512 data points were collected for each
112 t1 increments. A 1 s relaxation delay, a 20 µs (90°) pulse width,
and a 0.19 s acquisition time were used. The free induction decays
were processed with square cosine-bell filters in both dimen-
sions, and zero filling was applied prior to double Fourier
transformation.
C3), 155.9 (s, C3), 153.7 (bs, C3), 147.0 (dm, 1JCF ) 225 Hz, Cortho
,
C6F5), 146.9 (bs, C5), 144.5 (s, C5), 142.4 (bs, C5), 135.1 (dm, 1JCF
) 240 Hz, Cmeta, C6F5), 114.0 (bs, C4), 114.0 (s, C4), 111.0 (bs,
C4), 15.5 (s), 15.4 (bs), 14.0 (s), 13.0 (s, 2C), 12.1 (bs) (Me3 and
Me5), the Cipso and Cpara of the C6F5 groups have not been observed.
IR (Nujol): 799 and 785 cm-1 (C6F5).
(b) [Pd(m-C6ClF4)2(Me2-TpzT)]‚1/4C6H14 (2‚1/4C6H14). To a
solution of Me2-TpzT (40.0 mg, 0.11 mmol) in acetone (30 mL)
was added Pd(C6ClF4)2(cod) (64.0 mg, 0.11 mmol). The mixture
was stirred for 2 h and the solution evaporated to dryness. The
resulting white solid was washed with pentane (2 × 10 mL).
Yield: 52.0 mg (57%). The solid could be recrystallized from
dichloromethane/hexane. Anal. Calcd for C31.5H24.5Cl2F8N9Pd: C,
44.08; H, 2.88; N, 14.69. Found: C, 44.18; H, 2.71; N, 14.75. The
1
hexane found in the elemental analysis is also observed in the H
NMR spectrum. 13C NMR (1,1′,2,2′-tetrachloroethane-d2, room
temp): δ 162.6 (s, Cquat, triazine), 161.4 (s, 2C, Cquat, triazine),
158.1 (bs, C3), 155.9 (s, C3), 153.8 (bs, C3), 146.8 (bs, C5), 144.4
(s, C5), 142.6 (bs, C5), 114.1 (bs, C4), 114.0 (s, C4), 111.0 (bs, C4),
15.6 (s), 15.4 (bs), 14.0 (s), 13.1 (s, 2C), 12.4 (bs) (Me3 and Me5),
the carbons of the C6ClF4 group have not been observed. IR
(Nujol): 802 and 751 cm-1 (C6ClF4).
Determination of the kinetic parameters required two experiments
with mixing times of 1 s (optimized from 1 to 2 s in order to find
the value that gives the cross-peaks with higher intensity) for the
exchange experiment and 0.02 s for the nonexchange spectra. The
cross-peak/diagonal ratio was determined by integrating the volume
under the peaks.
Preparation of the NMR Samples for the 2D EXSY Experi-
ments. A saturated acetone solution of 2 was prepared by stirring
10 mg of the complex in 0.5 mL of acetone-d6. The solution was
filtered and introduced into an NMR tube and the sample used for
2D EXSY (1H and 19F) experiments. A more dilute solution was
prepared by taking 0.25 mL of the aforementioned solution and
adding an additional acetone-d6 (0.25 mL) to the NMR tube. A
1H,1H EXSY experiment was performed on this sample.
A solution of 2 in 1,1′,2,2′-tetrachloroethane-d2 was prepared in
an NMR tube with 7.5 mg of the complex and 0.5 mL of the
deuterated solvent. The diluted solution was prepared with 3.5 mg
of the complex and 0.5 mL of the solvent. 2D EXSY (1H and 19F)
experiments were performed on these solutions.
(18) See, for example: (a) Casares, J. A.; Coco, S.; Espinet, P.; Lin, Y.-S.
Organometallics 1995, 14, 3058. (b) Albe´niz, A. C.; Cuevas, J. C.;
Espinet, P.; de Mendoza J.; Prados, P. J. Organomet. Chem. 1991,
410, 257. (c) Casas, J. M.; Fornie´s, J.; Mart´ın, A.; Menjo´n, B.; Toma´s,
M. J. Chem. Soc., Dalton Trans. 1995, 2949. (d) Fornie´s, J.; Mart´ınez,
F.; Navarro, R.; Urriolabeitia, E.; Welch, A. J. J. Chem. Soc., Dalton
Trans. 1995, 2805. (e) Ara, I.; Delgado, E.; Fornie´s, J.; Herna´ndez,
E.; Lalinde, E.; Mansilla, N.; Moreno, M. T. J. Chem. Soc., Dalton
Trans. 1996, 3201. (f) Falvello, L. R.; Fornie´s, J.; Fortun˜o, C.; Mart´ın,
A.; Mart´ınez-Sarin˜ena, A. P. Organometallics 1997, 16, 5849. (g)
Fornie´s, J.; Go´mez-Saso, M. A.; Mart´ın, A.; Mart´ınez, F.; Menjo´n
B.; Navarrete, J. Organometallics 1997, 16, 6024.
Results and Discussion
(19) Guerrero, A.; Jalo´n, F. A.; Manzano, B. R.; Claramunt, R. M.; Cabildo,
P.; Infantes, L.; Cano, F. H.; Elguero, J. J. Heterocycl. Chem.,
submitted for publication.
Synthesis and Characterization. The reaction of 2,4,6-
(20) Espinet, P.; Mart´ınez-de Ilarduya, J. M.; Pe´rez-Briso, C.; Casado, A.
L.; Alonso, M. A. J. Organomet. Chem. 1998, 551, 9.
tris(3,5-dimethylpyrazol-1-yl)-1,3,5-triazine, Me2-TpzT, with
(21) States, D. J.; Haberkorn, R. A.; Ruben, D. J. J. Magn. Reson. 1982,
48, 286.
(22) Sandstro¨m, J. Dynamic NMR Spectroscopy; Academic Press: London,
1982.
Inorganic Chemistry, Vol. 42, No. 3, 2003 887

Carrio´n et al.
Chart 2
Table 1. 1H NMR Chemical Shifts (δ, ppm) of Me2-TpzT, [Pd(C6F5)2(Me2-TpzT)] (1) and [Pd(m-C6ClF4)2(Me2-TpzT)] (2)a
pyrazole A
pyrazole B
Me3
pyrazole C
deriv
solvent
T (K)
H4
Me3
Me5
H4
Me5
H4
Me3
Me5
Me2-TpzT
acet-d6
298
6.22 (q)
4JHH ) 1.0
6.66 (bs)
6.62 (s)
6.25 (s)
6.61 (s)
6.24 (s)
2.26 (s)
2.84 (d)
1
1
1
2
2
acet-d6
acet-d6
tetra-d2
acet-d6
tetra-d2
298
253
253
253
253
1.84 (bs)
1.85 (s)
1.80 (s)
1.84 (s)
1.78 (s)
2.96 (bs)
2.98 (s)
2.84 (s)
2.98 (s)
2.83 (s)
6.44 (bs)
6.41 (s)
6.22 (s)
6.40 (s)
6.22 (s)
2.28 (bs)
2.30 (s)
2.33 (s)
2.29 (s)
2.32 (s)
2.76 (bs)
2.77 (s)
2.66 (s)
2.77 (s)
2.66 (s)
5.87 (bs)
5.85 (s)
5.66 (s)
5.82 (s)
5.66 (s)
2.09 (bs)
2.11 (s)
2.16 (s)
2.12 (s)
2.15 (s)
2.51 (bs)
2.55 (s)
2.46 (s)
2.54 (s)
2.46 (s)
a Acet-d6 ) acetone-d6; tetra-d2 )1,1′,2,2′-tetrachloroethane-d2.
Pd(R)2(cod) (R ) C6F5, m-C6ClF4) in a 1:1 ratio led to the
new derivatives Pd(R)2(Me2-TpzT), R ) C6F5, 1, m-C6ClF4,
2 (see Chart 2).
this pyrazole was particularly shielded as a consequence of
the ring current anisotropy of the pentafluorophenyl group25
situated in close proximity. This effect has been previously
described in the literature,11,26 and we have also found it in
several other types of complex.7,27 The differentiation
between pyrazolyl groups B and C was made on the basis
of the higher shielding expected for Me3 (and also for the
Me5 if rotation of the free pyrazolyl groups at room
temperature is assumed) of ring C due to the anisotropy
effects of the proximal pentafluorophenyl ring. Besides, as
one would expect, the resonances of pyrazole B are the most
similar to those in the free ligand (see Table 1). In the
aforementioned NOE experiments, transfer of magnetization
between protons of pyrazoles C and B was observed in
acetone-d6 solution, while the same effect between protons
of A and C pyrazoles was detected in 1,1′,2,2′-tetrachloro-
ethane-d2 for both derivatives. This effect will be explained
below in the section concerning fluxional behavior. Such
interchanges were used to differentiate the Me3 and Me5
resonances of pyrazoles B and C once the signals of pyrazole
A had been assigned. This situation is in accordance with
the expected chemical shifts.
The solids are air-stable in the solid state as well as in
acetone or 1,1′,2,2′-tetrachloroethane solutions and can be
kept for several weeks under a nitrogen atmosphere without
evidence of decomposition. In addition, there is no evidence
of decomposition even after heating the tetrachloroethane
solutions at 373 K. The complexes are soluble in acetone
and in most chlorinated solvents but insoluble in hexane.
The IR spectra show two bands in the X-sensitive region,23,24
which indicates the presence of two different polyfluoro-
phenyl groups.
1
1
(a) H and 13C NMR Spectra. The H NMR spectra
(acetone-d6 or 1,1′,2,2′-tetrachloroethane-d2) of 1 and 2 at
room temperature, although some signals are slightly broad,
show signals consistent with the presence of three types of
pyrazolyl group. This is in accordance with the expected
structure, which contains a cis-PdR2 fragment coordinated
to one pyrazole and one N-triazine donor atom. At 253 K
all signals were narrow (see Table 1). The use of NOE
experiments on H4 allowed the assignment of the Me3 and
Me5 signals of each pyrazole ring. The more deshielded
resonances for H4 and Me5 were assigned to the coordinated
pyrazole ring (A in Chart 2). Note that the Me3 signal of
The 13C NMR data (room temperature, 1,1′,2,2′-tetrachloro-
ethane-d2, see Experimental Section) are consistent with the
presence of three different pyrazolyl groups. The assignment
(23) (a) Uso´n, R.; Fornie´s, J. AdV. Organomet. Chem. 1988, 28, 219. (b)
Uso´n, R.; Fornie´s, J.; Gimeno, J.; Espinet, P.; Navarro, R. J.
Organomet. Chem. 1974, 81, 115. (c) Uso´n, R.; Fornie´s, J.; Navarro,
R.; Garc´ıa, M. P. Inorg. Chim. Acta 1979, 33, 69.
(24) Maslowsky, E., Jr. Vibrational Spectra of Organometallic Compounds;
Wiley: New York, 1977.
(25) Alkorta, I.; Elguero, J. New J. Chem. 1998, 22, 381.
(26) (a) Orellana, G.; Alvarez-Ibarra, C.; Santoro, J. Inorg. Chem. 1988,
27, 1025. (b) Steel, P. J.; Constable, E. C. J. Chem. Soc., Dalton Trans.
1990, 1389.
(27) Carrio´n, M. C.; D´ıaz, A.; Guerrero, A.; Jalo´n, F. A.; Manzano, B. R.;
Rodr´ıguez, A. New J. Chem. 2002, 26, 305.
888 Inorganic Chemistry, Vol. 42, No. 3, 2003

Fluxional Processes in PdR2(Me2-TpzT)
Table 2. 19F NMR Chemical Shifts Data for Complexes 1 and 2a
compd
T (K)
Fo1/Fo1′
Fo2/Fo2′
Fm1/Fm1
-165.6 (m)
′
Fm2/Fm2
′
Fp1
Fp2′
1b
343
-116.2 (m)
-129.5 (t;
3JFF ) 21.3)
-161.4 (t)
-161.1 (t)
1b
1b
293
253
-115.7 (bs)
-115.7 (m)
-116.1 (m)
-115.4 (m)
-116.2 (m)
-119.2 (bs)
-116.8 (m)
-119.8 (m)
-117.1 (m)
-118.9 (m)
-164.8 (m)
-164.5 (m)d
-166.0 (m)
-163.4 (t)
-163.2 (t)
-165.5 (m)
-166.3 (m)
-167.1 (m)
-168.1 (m)
1c
213
-166.0 (m)
-166.4 (m)
-163.1 (t)
3JFF ) 19.8
-163.2 (t)
3JFF ) 19.8
F2
F6
F4
F5
2b
253
-94.2 (m) (1)
-94.5 (m) (2)
-95.0 (s) (3)
-95.1 (m) (4)
-95.9 (bs) (5+6)
-110.4 (dd)
-111.6 (d; J ) 17)
-111.7 (d; J ) 18.3)
-114.6 (d; J ) 36.6)
-114.7 (d; J ) 36.6)
-140.9 (d; J ) 21.4)
-142.9 (d; J ) 21.4)
-164.6 (m)
-165.5 (m)
-166.3 (m)
-98.6 (s) (7)
J ) 31.6, 21.3.
-98.8 (s) (8)
-110.7 (m)
-111.2 (d; J ) 23)
-111.3 (d; J ) 27.5)
a The labeling of ring 1 and 2 for complex 1 could be the reverse. b 1,1′,2,2′-tetrachloroethane-d2. c Acetone-d6. d Signals overlapped.
Chart 3
of the triazine7 or pyrazole carbons was made according to
literature.7,28 The signals corresponding to two pyrazoles are
slightly broad, while those of the other pyrazole are narrow.
This assignment is in accordance with the behavior observed
1
in the variable temperature H NMR studies.
(b) 19F NMR Spectra. The 19F NMR spectrum of 1 in
1,1′,2,2′-tetrachloroethane-d2 at room temperature (see Table
2) confirmed the existence of two different pentafluorophenyl
rings by the appearance of two triplets for the Fpara atoms.
Two broad signals for the Fortho and another two for the Fmeta
atoms were also observed. However, at low temperature (213
K) four different signals were observed for the Fortho atoms
and another four for the Fmeta. The resonances corresponding
to fluorine atoms of the same ring (rings 1 and 2, see Chart
2 and Table 2) were assigned on the basis of 19F,19F COSY
spectra. Besides the intraannular couplings, an interannular
coupling was observed between Fortho1′ and Fortho2. For the
two Fortho or the two Fmeta atoms of the same ring to be
chemically different, not only is a restricted rotation of the
pentafluorophenyl rings necessary but also the absence of
planarity in the ligand. On considering the reported data
concerning similar ligands containing pyridyl8,10,11 or pyr-
azolyl rings,12 we propose that at low temperature there is a
restricted rotation of the uncoordinated pyrazole ring C. In
contrast to pyrazole B, this ring should experience some
degree of steric hindrance due to the adjacent pentafluoro-
phenyl ring. The most plausible conformation for this
compound (and also for 2) is that in which the planes of the
polyfluorophenyl groups and pyrazole C approximate a
mutually parallel arrangement that is orthogonal to the
coordination plane, with pyrazole B parallel to this latter
plane.
29 for a similar system) and by means of a 19F,19F COSY
spectrum. Cross-peaks were observed between the signals
of F5 and those of F4 and F6. Some interannular couplings
were observed between F2-F6 and F2-F2 resonances, and
these will be discussed in more detail below. Although only
two or three resonances appeared for the F4 and F5 atoms,
respectively, the regions containing F2 (seven signals) and
F6 (eight signals) were more informative. This number of
signals indicates the existence of eight different phenyl rings.
Considering that two different tetrafluorophenyl groups are
expected for each species, four isomeric forms must be
present in solution at this temperature, a situation that is in
accordance with the existence of four atropisomers (see Chart
3 for a schematic view of the possible rotamers considering
an averaged perpendicular orientation of the C6ClF4 groups).
To justify the presence of four different atropisomers, and
in a way similar to complex 1, not only must there be a
restricted rotation of the phenyl rings but also the planarity
of the ligand must be brokensa direct consequence of the
restricted rotation of pyrazole C. In the case of a free rotation
of the uncoordinated pyrazole rings, isomers I and II, on
one side, and III and IV, on the other side, would be
enantiomers and would not be distinguishable by NMR
(Chart 3). This problem of restricted rotation of the pyrazole
ring C and the polyfluorophenyl rings will be discussed in
the section concerning fluxional behavior.
The 19F NMR spectrum of 2 at room temperature (1,1′,2,2′-
tetrachloroethane-d2) contains broad signals, whereas the
same spectrum at 253 K shows narrow signals. The assign-
ment of the different types of fluorine atoms was made on
the basis of bibliographic references (see, for example, ref
(28) Oro, L. A.; Esteban, M.; Claramunt, R. M.; Elguero, J.; Foces-Foces,
C.; Cano, F. H. J. Organomet. Chem. 1984, 276, 79. Esteruelas, M.
A.; Oro, L. A.; Apreda, M. C.; Foces-Foces, C.; Cano, F. H.;
Claramunt, R. M.; Lo´pez, C.; Elguero, J.; Begtrup, M. J. Organomet.
Chem. 1988, 344, 93. Fajardo, M.; de la Hoz, A.; D´ıez-Barra, E.; Jalo´n,
F. A.; Otero, A.; Rodr´ıguez, A.; Tejeda, J.; Belletti, D.; Lanfranchi,
M.; Pellinghelli, M. A. J. Chem. Soc., Dalton Trans. 1993, 1935.
(29) Albe´niz, A. C.; Casado, A. L.; Espinet, P. Organometallics 1997, 16,
5416.
Inorganic Chemistry, Vol. 42, No. 3, 2003 889

Carrio´n et al.
Table 3. Calculated Free Energy of Activation for Some Processes in
Complex 1 in 1,1′,2,2′-Tetrachloroethane-d2
coalescence of the signals due to pyrazole A and C was
observed in a way similar to that described for the 1,1′,2,2′-
tetrachloroethane-d2 solution. However, a clear broadening
of the pyrazole B signals was observed when the temperature
was increased. The limit of the boiling point of the acetone
prevented the achievement of signal coalescences. It is
possible to conclude that the process of “metal hurdling” is
only observed when acetone is present, presumably due to
its coordinating ability.
a
a
T
δν
∆Gc
nucleus Pz interchange (K) (Hz) (kJ mol-1
)
process
1H
1H
1H
19F
19F
19F
A h C (H4)
A h C (Me3)
A h C (Me5)
325 164.9
319 108.5
321 126.9
63.9
63.7
63.7
64.2
56.6
56.1
1,4-metallotropic shift
1,4-metallotropic shift
1,4-metallotropic shift
1,4-metallotropic shift
pyrazole rotation
F
F
F
para1 h Fpara2 343 540.1
ortho1 h Fortho1
meta1 h Fmeta1
′
287 130.1
293 267.6
′
pyrazole rotation
a (1 kJ mol-1
.
A variable temperature 19F NMR study was also under-
taken in 1,1′,2,2′-tetrachloroethane-d2 in the case of com-
pound 1. As we have previously reported, four Fortho, four
The detailed assignment of the F2 resonances to the four
atropisomers will be covered in a separate section after a
discussion of the information obtained from the study of the
fluxional behavior.
Fmeta, and two Fpara resonances were observed at low
temperature. When the temperature was increased, the
coalescence between the Fortho and Fmeta resonances of the
same ring was achieved (Fo1-Fo1′, Fo2-Fo2′, Fm1-Fm1′, Fm2-
Fm2′). A further increase in the temperature led to the
coalescence of the signals of ring 1 with those of ring 2.
Consequently, Fortho or Fmeta signals undergo two types of
interchange, while those of Fpara experience only one. Thus,
in the latter case it was possible to calculate the free energy
of activation without interference, while in the former case
only two barriers (Fo1-Fo1′, Fm1-Fm1′, first process) were
calculated; these corresponded to coalescence temperatures
that were low enough that the broadening due to the second
process could be considered negligible. The data for these
studies are given in Table 3. In principle, the possible
processes that lead to interconversion of signals due to atoms
in the same ring that are of lower energy could be the
following: (i) a rotation of pyrazole C, (ii) a rotation of both
pentafluorophenyl groups, or (iii) a combination of both types
of rotation. We propose alternative i on the basis that this
type of process has been found to be of low energy in similar
systems.8,10-12 Besides, from the data obtained for complex
2 (see below), in which similar behavior is expected, it is
possible to conclude that rotation of both polyfluorophenyl
groups with a similar energy barrier does not take place. The
interchange of atoms of different rings indicates that they
interconvert their positions with respect to the pyrazole or
triazine rings, and this is possible via the 1,4-metallo-
tropic shift. Coincidentally, the ∆Gcq value obtained from
the 19F NMR spectra fits well with those obtained for this
process from the 1H NMR spectra (see Figure 1) and,
consequently, we can conclude that in this metallotropic shift
not only does the interchange of pyrazoles A and C take
place but the polyfluorophenyl rings also exchange their
positions.
The effect of the atropisomerism on the chemical shifts
of the pyrazole protons should be very small. The presence
of the four atropisomers is not detected in the H NMR
spectra of the 1,1′,2,2′-tetrachloroethane-d2 solution, probably
due to the similarity of the chemical shifts. However, small
splittings of the Me3 (CDCl3) or H4 signals (acetone-d6) were
observed at low temperature (183-213 K). The δν measured
were too small, and consequently the calculation of free
energies of activation from these data is not considered in
the following section of fluxional behavior.
1
Fluxional Behavior. (a) Complex 1. To gain a deeper
insight into the fluxional behavior of complex 1, several
1
variable temperature H NMR studies in acetone-d6 and
1,1′,2,2′-tetrachloroethane-d2 were undertaken. The study in
acetone-d6 started at 253 K, and in this case all signals were
narrow. However, when the temperature was increased, the
coalescence of signals due to the uncoordinated pyrazolyl
rings [Me3(B)-Me3(C) and Me5(B)-Me5(C)] was observed.
As the temperature was increased further, the corresponding
signals that emerge after the coalescences were observed to
broaden along with the resonances of the coordinated
pyrazole ring (A), but the coalescence of these signals was
not achieved due to the limit of the boiling point of the
solvent. The corresponding free energies of activation were
not calculated because the concomitant broadening of the
signals due to the presence of at least two processes made
these calculations inaccurate. The spectra of the samples in
1,1′,2,2′-tetrachloroethane-d2 at 253 K also contained narrow
signals. An increase in temperature led to the coalescence
of resonances due to pyrazoles A and C (Me3, Me5, and H4),
but the signals of pyrazole B were narrow even at 413 K. In
this case we observed only one process, so the corresponding
free energies of activation were calculated (see Table 3). One
conclusion that can be drawn from these data is that the 1,4-
metallotropic shift process that involves interconversion of
pyrazole rings A and C takes place in 1,1′,2,2′-tetrachloro-
ethane-d2 and, very probably, also in acetone-d6. However,
the interconversion of pyrazoles B and C, which would take
place via a “metal-hurdling” process, is only observed in
acetone-d6.
1
(b) Complex 2. Variable temperature H NMR studies
(acetone-d6 or 1,1′,2,2′-tetrachloroethane-d2 solutions) on
complex 2 showed similar behavior to that found for 1.
However, in the case of the 19F NMR spectra, the large
number of signals made it extremely difficult to determine
which pairs of resonances coalesced at each temperature. At
high temperature only one resonance was observed for each
type of fluorine atom, indicating that all the atropisomers
interconverted and, in addition, that the two tetrafluorophenyl
groups (trans to pyrazole or triazine) interchange their
environments. For this reason we decided to carry out 2D
To evaluate the influence of the coordinating ability of
the solvent, a variable temperature 1H NMR study in acetone-
d6/1,1′,2,2′-tetrachloroethane-d2 (1:1) was performed. The
890 Inorganic Chemistry, Vol. 42, No. 3, 2003

Fluxional Processes in PdR2(Me2-TpzT)
Table 4. Calculated Free Energy of Activation for the Different
Processes Detected in 2 from 2D EXSY Studiesa
Pz
inter-
∆G
b
entry solvent nucleus change (kJ mol-1
)
process
pyrazole rotation
metal hurdling
metal hurdling
1
2
3
4
5
6
7
8
9
tetra-d2
acet-d6
19F
1H
1H
1H
1H
1H
1H
60.5
63.3
63.9
d
66.2
68.1
65.7
65.4
72.0
70.6
d
B h C
B h C
B h C
A h C
A h C
A h C
c
acet-d6
tetra-d2
acet-d6
metal hurdling
1,4-metallotropic shift
1,4-metallotropic shift
1,4-metallotropic shift
1,4-metallotropic shift
intermolecular metal interchange
intermolecular metal interchange
intermolecular metal interchange
intermolecular metal interchange
C6F4Cl rotation
c
acet-d6
tetra-d2
tetra-d2
acet-d6
19F A h C
1H
1H
1H
1H
19F
A h B
A h B
A h B
A h B
10 tetra-d2
q
c
Figure 1. ∆Gc versus Tc for complex 1 for the 1,4-metallotropic shift
11 acet-d6
12 tetra-d2
13 tetra-d2
c
process in 1,1′,2,2′-tetrachloroethane-d2 obtained after the coalescence
d
1
temperatures: (b) from H NMR spectra; (9) from 19F NMR spectra (R2
74.5e
) 0.9579).
a Tetra-d2 ) 1,1′,2,2′-tetrachloroethane-d2; acet-d6 ) acetone-d6. b (1
kJ mol-1 c Half-concentration. d Not detected. e Determined by the coal-
.
escence temperature.
EXSY experiments on this compound. In this method each
single process is separated in the 2D map and interference
between processes are therefore avoided.30 A special feature
of 2D EXSY is that the kinetic constant for each independent
process is deduced from the data in the NMR spectra. More
recently, 1D EXSY techniques have been introduced.31 The
essential feature of a quantitative 2D EXSY experiment is
the relationship between the intensity of a cross-peak and
the rate constants for chemical exchange. Cross-peaks in the
spectrum correspond to nuclei that exchange from one site
to another. The intensities of those cross-peaks do not
correspond directly to the exchange matrix but to its
exponential form.30
spectrum in acetone contained cross-peaks corresponding to
three different processes with three energy barriers: (i)
interchange of pyrazoles B and C with ∆Gq ) 63.3 kJ‚mol-1
(entry 2, Table 4), (ii) interchange of pyrazoles A and C
with ∆Gq ) 66.2 kJ‚mol-1 (entry 5), and (iii) interchange
of pyrazoles A and B with ∆Gq ) 72.0 kJ‚mol-1 (entry 9).
The first interchange must be due to a metal-hurdling process
and the second to a 1,4-metallotropic shift. The metal-
hurdling process has a lower energy barrier than the 1,4-
metallotropic shift in this solvent (entries 2 and 5). The third
process has not been detected in similar systems.10,12 We
considered the possibility that the interchange between
pyrazoles A and B could be due to an intermolecular process.
To verify this hypothesis, we performed a 2D EXSY
experiment in acetone-d6 solution with half-concentration.
Interestingly, the first two processes took place with barriers
similar to those found for the former solution (compare
entries 2 and 3 and also 5 and 6) and were not markedly
affected by the dilution. However, the interchange between
pyrazoles A and B was not observed in the dilute solution
(entries 9 and 11), which is a clear indication of an
intermolecular process.
1H,1H EXSY studies were also performed in 1,1′,2,2′-
tetrachloroethane-d2 solution. In this case, only two processes
were observed and these correspond to (i) interchange of
pyrazoles A and C with an energy barrier of 65.7 kJ‚mol-1
(entry 7) and (ii) interchange of pyrazoles A and B with an
energy barrier of 70.6 kJ‚mol-1 (entry 10). The first process
must be due to a 1,4-metallotropic shift, and the second, if
we take into account the results obtained in acetone, must
be the intermolecular interchange. It is noteworthy that in
this noncoordinating solvent the process of metal-hurdling
is not observed (entry 4), a situation in agreement with the
results obtained from the monodimensional spectra. In this
case, cross-peaks between pyrazoles B and C are not
observed while the intermolecular process is present. There-
fore, this process cannot involve complete dissociation of
the metal fragment from the Me2-TpzT ligand followed by
recombination, because such a process would lead to
interchange of all pyrazolyl groups. Confirming the results
Rate constants can be deduced from the spectrum accord-
ing to the following equation:
R ) -(ln A)/τm ) -X(ln Λ)X-1/τm
where Aij ) Iij/Mj and τm is the mixing time. Iij(τm)/Mj and
X are the square matrix of eigenvectors of Ai such that X-1AX
- Λ ) diag(λi), with λi being the ith eigenvalue of A. Iij can
be deduced by measuring the volume of each peak intensity
directly from the spectrum. Mj is the volume of the diagonal
peak of the spectrum registered with a mixing time close to
0, without any chemical exchange.
An essential feature of the process is that
Kijpi ) Kjipj
where Kij and Kji are the rate constants of processes i f j
and j f I, respectively, and pi is the relative population of
the ith site.
Considering the coalescence of all signals, as detected in
the variable temperature 19F NMR spectra, a maximum of
56 correlations is expected in the 2D map in the well-resolved
F2 region. These experiments were performed in 1,1′,2,2′-
1
tetrachloroethane-d2 solution for H and 19F and also in
1
1
acetone-d6 for H (in all cases at 253 K). The H NMR
(30) Perrin, C. L.; Gipe, R. K. J. Am. Chem. Soc. 1984, 106, 4036. Bremer,
J.; Mendez, G. L.; Moore, W. J. J. Am. Chem. Soc. 1984, 106, 4691.
Perrin, C. L.; Dwyer, T. J. Chem. ReV. 1990, 90, 935.
(31) Perrin, C. L.; Engler, R. E. J. Am. Chem. Soc. 1997, 119, 585.
Inorganic Chemistry, Vol. 42, No. 3, 2003 891

Carrio´n et al.
that the variable temperature 19F NMR study showed, after
several coalescences, only one signal for each type of fluorine
atom. Consequently, at a relatively high temperature, the
fluorine atoms for the two blocks described above experience
an interchange. The corresponding barrier calculated for the
F2 atoms by the coalescence method (Tc ) 348 K) is 74.5
kJ‚mol-1 (entry 13). This process must imply the rotation
of one or two polyfluorophenyl groups. In this case it is
meaningless to consider whether the rotation of the two rings
has the same energy barrier or not, because they are
interconverting by the 1,4-metallotropic shiftsa process that
is of lower energy.
Assignment of Atropisomers. Assignment of the 19F
NMR signals to atropisomers I-IV and the determination
of their relative ratios was undertaken by considering the
long-range through-space correlations in the 19F,19F COSY
spectra and interconversions detected in the 19F,19F EXSY
spectra (in all cases we used the F2 signal).
As discussed above, the 19F NMR spectra for F2 would
be expected to show eight signals, although only seven are
observed due to the overlapping of signals 5 and 6.
Figure 2. 19F,19F EXSY experiment in the region of F2 for complex 2.
Pyrazole rotation interchanges signals 1-3, 2-4, 6-7, and
5-8 and interconverts atropisomers I-II (with the chlorine
atoms in syn dispositions) and III-IV (with the chlorine
atoms in anti dispositions) (Scheme 1). This means that each
of the stated pairs of signals correspond to resonances of
different isomers with the same relative disposition of the
chlorine atoms and with the same heterocycle (triazine or
pyrazole) in a trans arrangement.
obtained in acetone, the intermolecular process is not
observed when a 1,1′,2,2′-tetrachloroethane-d2 solution with
half-concentration is measured (entry 12).
19F,19F EXSY studies were performed for F2 because the
signals of this fluorine atom do not show large couplings
and are simpler than those of F6, which appear as doublets.
If we number the signals from 1 to 8 (5 and 6 overlapped)
in order of increasing field, a first process of lower energy
(∆Gq ) 60.5 kJ‚mol-1) (entry 1, Table 4) with cross-peaks
between the pairs 1-3, 2-4, 5-8, and 6-7 is observed (see
Figure 2). A second process (∆Gq ) 65.4 kJ‚mol-1, entry
8), in which the pair of signals 1 and 3 interchange with the
pair 6 and 7 on one side and also the pair 2 and 4 with the
pair 5 and 8 on the other side, is also observed (see Scheme
1, right). On the basis of this correlation, the eight signals
are classified into two blocks that do not correlate with one
another. The first process, which has no equivalence in any
1,4-Metallotropic shift exchanges the pairs 1-3 with 6-7
and 2-8 with 4-5. Although this process does not inter-
convert atropisomers, both C6F4Cl groups in each compound
interchange their positions. Consequently, it is possible to
define two blocks of signals: 1,3,6,7 and 2,4,5,8swith each
block corresponding to the same orientation for the chlorine
atoms (see Scheme 1). In the 19F,19F COSY spectra (253
K), where the long-distance correlations can be observed, a
cross-peak between signals 2 and 5 of F2 has been found. It
is necessary to consider that the magnitude of this through-
space coupling is correlated with the nonbonding F‚‚‚F
distance,32 and, consequently, it indicates that the resonances
belong to one atropisomer that contains the chlorine atoms
in a syn disposition. The same reasoning should be applicable
to signals 4 and 8. If we take into account the values of the
integrals in the 19F spectrum, it is possible to deduce that
one of these isomers has an abundance of 21% (signals 2
and 5) and the other of 28% (signals 4 and 8). After analyzing
molecular models, it is apparent that a greater degree of steric
hindrance would exist between Me3 and the chlorine atom
in isomer I (see Chart 2) and, consequently, we propose that
this isomer corresponds to that indicated in Table 5 and with
1
interchange observable in the H ,1H EXSY study, should
be the rotation of one ring (we must assume that the
1
atropisomers have not been observed separately in the H
NMR spectra in 1,1′,2,2′-tetrachloroethane-d2 solution). This
process could be (i) the restricted rotation of the pyrazole C
or (ii) the restricted rotation of one of the polyfluorophenyl
rings (the rotation of both C6ClF4 rings would lead to a larger
number of cross-peaks). We propose that this process
corresponds to the first possibility on the basis of systems
described previously8,10-12 and by analogy to the results found
for complex 1. The second process, which has an energy
barrier similar to that found in the 1H NMR study (compare
entries 7 and 8), must correspond to the 1,4-metallotropic
shift. It is unreasonable to believe that a process observed
(32) (a) Ernst, L.; Ibrom, K. Angew. Chem., Int. Ed. Engl. 1995, 34, 1881.
(b) Ernst, L.; Ibrom, K.; Marat, K.; Mitchell, R. H.; Bodwell, G. J.;
Bushnell, G. W. Chem. Ber. 1994, 127, 1119. (c) Lyga, J. W.;
Henrie, R. N.; Meier, G. A.; Creekmore, R. W.; Patera, R. M. Magn.
Reson. Chem. 1993, 31, 323. (d) Mallory, F. B.; Mallory, C. W.;
Ricker, W. M. J. Am. Chem. Soc. 1975, 97, 4770, and references
therein.
by 1H NMR is not present in the corresponding study by 19
NMR.
F
Despite the fact that the 19F,19F EXSY study at 253 K did
not reveal any other exchange process, we must consider
892 Inorganic Chemistry, Vol. 42, No. 3, 2003

Fluxional Processes in PdR2(Me2-TpzT)
Scheme 1. Analogy between the Expected Reduction in the Number of Haloarene Rings (left) and the Resonance Groups Correlated in the 19F,19F
EXSY Experiment of 2 (Right)
Table 5. Assignment of the F2 Resonances of Complex 2 to the Four
Atropisomers and Their Abundance
coordinating solvent. Besides, the process is not accompanied
by the rotation of any of the tetrafluorophenyl rings;
otherwise a unique block of all correlated signals would be
obtained in the 19F,19F EXSY spectrum of complex 2. Given
the fact that we have found the same energy barrier for the
interconversion of pyrazoles A and C and the interchange
of the relative positions of the two polyfluorophenyl rings
with respect to the heterocyclic ligands, this process should
involve the “tick-tock” motion of the PdR2 moiety. All these
data are in accordance with the associative mechanism
proposed by Orrell,9 which involves a pentacoordinate
intermediate containing a terdentate ligand, as shown in
signals 2 and 5. The assignment of signals 4 and 8 to isomer
II is straightforward.
Scheme 2. We must consider that in the first and the last
step of the process it is necessary to rotate one uncoordinated
pyrazole ring. However, considering our data, this rotation
can take place with a lower energy barrier.
If we consider isomers III and IV (block of signals 1, 3,
6, 7), it is envisaged that cross-peaks between the resonances
of F2 and F6 should be observed in the COSY spectrum. In
fact, this situation was found for signals 1 and 6 of F2 with
other resonances of F6 (-111.6 and -110.6, respectively).
These two couplings probably correspond to two different
isomers because one of the two possible couplings between
the Fortho atoms of two rings is usually higher than the other
due to the tilting of the two polyfluorophenyl rings.29 This
led to the conclusion that the resonances 1, 7 on one side
and 3, 6 on the other side belonged to the same isomer.
Although the integral values are not markedly different, it
is possible to conclude that they are in accordance with the
assigned pairs of signals. On the basis of the same argument
applied to isomers I and II, we tentatively assign signals 3
and 6 (with an abundance of 24%) to isomer III.
We have demonstrated for the first time that the coordinat-
ing ability of the solvent has a dramatic influence on the
process of metal hurdling. In fact, this phenomenon is not
observed in tetrachloroethane solutions. The accelerating
effect of species with coordinating ability (solvent, counter-
anions, added ligands, etc.) has been used as an argument
supporting both dissociative33 (formation of threecoordinate
intermediates) or associative34 (formation of pentacoordinate
intermediates) pathways. When kinetic measurements have
been made (for example in isomerization, substitution
reactions, etc.; see, for example, ref 35), it has been found
that pentacoordinate intermediates are normally involved. We
therefore propose that the prior coordination of the solvent
is necessary to give a pentacoordinate intermediate, which
Mechanistic Considerations. The conclusions reached
from the study of the dynamic behavior of complexes 1 and
2 give a greater insight into the mechanisms involved in the
different processes.
It is clear that the 1,4-metallotropic shift is an intra-
molecular process that does not need the presence of a
(33) Gogoll, A.; O¨ rnebro, J.; Grennberg, H.; Ba¨ckvall, J. E. J. Am. Chem.
Soc. 1994, 116, 3631.
(34) Hansson, S.; Norrby, P. O.; Sjo¨gren, M. P. T.; Åkermark, B.;
Cucciolito, M. E.; Giordano, F.; Vitagliano, A. Organometallics 1993,
12, 4940.
(35) Espinet, P. ComprehensiVe in Organometallic Chemistry; Pergamon
Press: Oxford, U.K., 1982; Vol. 6, pp 235 and 412.
Inorganic Chemistry, Vol. 42, No. 3, 2003 893

Carrio´n et al.
Scheme 2. Proposed Mechanism for the 1,4-Metallotropic Shift
Scheme 3. Proposed Mechanism for the Process of Metal Hurdling
subsequently suffers rupture of the Pd-N1(triazine) bond.
This process is then followed by a 180° rotation of the metal-
bound pyrazole group around the C(triazine)-N(pyrazole)
axis, formation of the Pd-N3(triazine) bond, and dissociation
of the solvent (see Scheme 3). An alternative mechanism
implying a three-coordinate intermediate formed after rup-
ture of the Pd-N1(triazine) bond and stabilized with the
solvent cannot be completely ruled out. The fact that the
process of metal hurdling has a lower energy barrier in
acetone solutions than the 1,4-metallotropic shift may be
related with the Pd-N(triazine) bond being weaker than the
Pd-N(pyrazole) bond due to the lower basicity of the triazine
heterocycle.
Conclusions
We have synthesized the new complexes Pd(R)2(Me2-
TpzT), Me2-TpzT ) 2,4,6-tris(3,5-dimethylpyrazol-1-yl)-
1,3,5-triazine, R ) C6F5, 1; R ) C6ClF4, 2. Four different
atropisomers have been detected at low temperature in
complex 2. The fluxional behavior in acetone-d6 or 1,1′,2,2′-
tetrachloroethane-d2 solutions of the new derivatives has been
analyzed in detail by means of H and 19F one- and two-
1
dimensional NMR techniques. Five different processes have
been found, three of which involve metallotropy (namely,
metal hurdling, 1,4-metallotropic shift, and an intermolecular
process) and two of restricted rotation of aromatic rings (the
uncoordinated pyrazole ring adjacent to the metal center and
the polyfluorophenyl groups). The 1,4-metallotropic shift
takes place in both solvents. The coordinating ability of the
solvent has a dramatic effect on the metal-hurdling process
in the sense that it is not observed in tetrachloroethane
solutions while in acetone solutions was present. In this
solvent, it was of lower energy than the 1,4-metallotropic
shift. These two processes (especially the metal hurdling)
are not significantly affected by the concentration, whereas
the intermolecular process is not observed in dilute acetone
solutions. In tetrachloroethane solutions of complex 2 the
following order for the free energy of activation was found:
As far as the intermolecular process that interconverts
pyrazoles A and B is concerned, we have already explained
that this cannot involve complete dissociation of the metallic
fragment. Bearing in mind the dramatic effect of dilution,
we propose that an interaction exists between two molecules
of the complex in such a way that a pentacoordinate
intermediate is also formed, but this time with the participa-
tion of one uncoordinated pyrazole from the other molecule
(the less crowded ring will be pyrazole B). This would lead
to the rupture of the Pd-N(triazine) bond and finally to the
concerted transfer of the metallic fragment from one
molecule to the other.
894 Inorganic Chemistry, Vol. 42, No. 3, 2003

Fluxional Processes in PdR2(Me2-TpzT)
pyrazole rotation < 1,4-metallotropic shift < intermolecular
exchange < polyfluorophenyl rotation. Considering these
obtained data, we propose that the three metallotropic proc-
esses proceed through associative mechanisms via penta-
coordinate intermediates, implying coordination of the
solvent (metal hurdling), the N atom of the adjacent pyrazole
(1,4-metallotropic shift), or the pyrazole N atom of another
molecule (intermolecular process). The data from the 19F,19F
EXSY and COSY experiments allowed complete assignment
of the F2 resonances to the different atropisomers of 2.
Acknowledgment. Thanks are given to the DGI/MCyT
of Spain for financial support (Project No. BQU2002-00286).
IC020522A
Inorganic Chemistry, Vol. 42, No. 3, 2003 895