Apparent allyl rotation and Pd-N bond rupture in allylpalladium complexes with N-donor ligands - Evidence of an associative mechanism

Article abstract of DOI:10.1002/ejic.200400412

The new didentate N-donor ligands 2-(4-methyl-1H-pyrazol-1-yl)pyrimidine (4Me-pzpm, 1) and 2-(4-bromo-1H-pyrazol-1-yl)pyrimidine (4Br-pzpm, 2) have been synthesised and used to obtain the allylpalladium derivatives [Pd(η³-2Me-C₃H₄)(NN′)]X [X = BAr′₄^-, NN′ = 1 (3), NN′ = 2 (4); X = CF₃SO₃^-, NN′ = 1 (5), NN′ = 2 (6)]. In complexes 3-6 two types of fluxional process have been found: apparent allyl rotation that is observed as H_syn-H_syn, H _anti-H_anti interconversions and H⁴-H ⁶ interchange of the pyrimidine protons that must involve Pd-N(pm) bond rupture. The influence of different factors on both processes - such as the nature of the N-donor ligand, counterion, solvent, complex concentration and addition of water - has been studied. It has been concluded that the apparent allyl rotation has a lower free energy of activation and in both cases the presence of a species with coordinating ability favours the process. Negative entropies of activation have been found. Associative mechanisms involving participation of five-coordinate intermediates have been proposed. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2005.

Full text of DOI:10.1002/ejic.200400412

FULL PAPER
Apparent Allyl Rotation and Pd؊N Bond Rupture in Allylpalladium
Complexes with N-Donor Ligands ؊ Evidence of an Associative Mechanism
[a]
[a]
[a]
´
´
´
Felix A. Jalon, Blanca R. Manzano,* and Belen Moreno-Lara
Keywords: Allyl ligands / Fluxionality / N ligands / Palladium
The new didentate N-donor ligands 2-(4-methyl-1H-pyrazol-
1-yl)pyrimidine (4Me-pzpm, 1) and 2-(4-bromo-1H-pyrazol-
the N-donor ligand, counterion, solvent, complex concentra-
tion and addition of water — has been studied. It has been
1-yl)pyrimidine (4Br-pzpm, 2) have been synthesised and concluded that the apparent allyl rotation has a lower free
used to obtain the allylpalladium derivatives [Pd(η3-2Me-
C3H4)(NNЈ)]X [X = BArЈ4−, NNЈ = 1 (3), NNЈ = 2 (4); X =
energy of activation and in both cases the presence of a spe-
cies with coordinating ability favours the process. Negative
CF3SO3−, NNЈ = 1 (5), NNЈ = 2 (6)]. In complexes 3−6 two entropies of activation have been found. Associative mech-
types of fluxional process have been found: apparent allyl
anisms involving participation of five-coordinate interme-
rotation that is observed as Hsyn−Hsyn, Hanti−Hanti intercon- diates have been proposed.
versions and H4−H6 interchange of the pyrimidine protons
that must involve Pd−N(pm) bond rupture. The influence of
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
different factors on both processes — such as the nature of
Germany, 2005)
Introduction
molecular symmetry. Two main mechanisms have been pro-
posed for the apparent allyl rotation: (i) associative mecha-
nisms[20,28,29,32,35] that involve five-coordinate intermediates
(coordination of the solvent, the anion or other molecules)
that can undergo allyl pseudorotation, and (ii) dissociative
mechanisms[18,19,21Ϫ23,25,26,31,33] with formation of T-shaped
three-coordinate intermediates after dissociation of mono-
dentate or didentate (partial dissociation) ligands. Conse-
quently, one question that is open to debate is whether the
apparent allyl rotation in derivatives with N-donor ligands
involves PdϪN bond breaking or not. In the case of diden-
tate ligands this question is interesting because it would in-
volve the dissociation of one arm of a chelate, a subject of
potential interest, especially in relation to catalytic pro-
cesses. In some examples it has been reported that a ligand
of higher basicity makes the apparent rotation more diffi-
cult,[21] while increased steric hindrance of the N-donor li-
gand can facilitate[20,29] or slow down[21] the process. Nega-
tive entropies of activation have been found in some
cases[29,32] and it has been reported that the process takes
place more easily with coordinating anions[19,29,31] or sol-
vents such as DMSO.[20b] It has also been reported that the
addition of ligand or water has a negligible effect on the
process in a number of examples;[18,21] other examples have
been described where the effect of adding ligand or chlo-
rides is accelerative.[19,28,29] The effect of the anion has also
been analysed.[29,31a] As far as the effect of the complex
concentration is concerned, positive effects,[28,20b] at least to
a limiting value, have been described but in some cases it
has been found that this factor has no influence.[31a,35] In
recent years we have been interested in the synthesis, struc-
tural characterisation and dynamic behaviour of allylpal-
The chemistry of (η3-allyl)palladium complexes has re-
ceived a great deal of attention because these derivatives
can act as precursors or intermediates in different catalytic
processes.[1] The fluxional behaviour of these systems, either
directly related to the allyl group or to the ancillary ligands,
has also been widely studied.[2] Besides the fundamental
interest in these studies, the fluxional behaviour of allylpal-
ladium derivatives has important implications in a variety
of catalytic processes, especially those involving nucleo-
philic attack on (η3-allyl)palladium intermediates in enantio-
selective synthesis.[3] One process frequently encountered
in allylpalladium complexes is a mutual exchange of syn
and anti groups.[4Ϫ12] This process is believed to occur
through an η3-η1-η3 pathway that in some cases is
selective[6a,9a,10,13Ϫ16] due to steric or electronic factors. On
the basis of theoretical calculations concerning the mecha-
nism of the η3-η1-η3 isomerisation in (η3-allyl)palladium
complexes, it has been proposed that the process involves
tetracoordinate (η1-allyl)palladium intermediates with co-
ordination of a solvent molecule or an ancillary ligand.[17]
A second dynamic process that is frequently observed in
complexes with N-donor ligands is the apparent rotation
of the allyl group.[18Ϫ28] In certain examples both of the
aforementioned processes have been observed.[16,29Ϫ35] The
apparent rotation is observed as a syn-syn, anti-anti ex-
change and/or an isomerisation process, depending on the
[a]
´
´
´
´
Departamento de Quımica Inorganica, Organica y Bioquımica,
Universidad de Castilla-La Mancha,
Avda. C. J. Cela, 10, 13071 Ciudad Real, Spain
E-mail: Blanca.Manzano@uclm.es
100
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200400412
Eur. J. Inorg. Chem. 2005, 100Ϫ109

Allyl Rotation and PdϪN Bond Rupture in Allylpalladium Complexes with N-Donor Ligands
FULL PAPER
ladium(ii) complexes with N-donor ligands containing pyr- are limited to the substituent in the 4-position of the
azolyl groups.[23Ϫ28,35] We have described the synthesis and pyrazole, and so it is foreseeable that, if any influence exists,
fluxional behaviour of the complexes [Pd(η3-2Me- it will be only electronic in nature. The pKa values for the
C3H4)(NNЈ)]CF3SO3 [NNЈ ϭ 2-(1H-pyrazol-1-yl)pyrimid- three pyrazole heterocycles are as follows: 4-methylpyrazole
ine ϭ pzpm (I); NNЈ ϭ 2-(1H-pyrazol-1-yl)pyridine ϭ pzpy 3.04; pyrazole 2.48; 4-bromopyrazole 0.63.[36] The new
(II)][23] (see Scheme 1) where the syn-syn, anti-anti intercon- ligands were prepared by reaction of the deprotonated
version was detected along with an interchange process be- pyrazole rings with 2-chloropyrimidine (see Scheme 2 and
tween the H4 and H6 protons of the pyrimidine ring in com- Exp. Sect.).
plex I. This last interchange is only possible through
PdϪN(pm) bond rupture. Moreover, an NOE between the
allylic Hsyn and Hanti protons and H5Ј of the pyrazole ring
was also detected in the case of complex II and this can
only be explained through PdϪN(pz) bond rupture (see
Scheme 1). Considering these facts, and given the similarit-
ies between our work and other results reported pre-
viously,[19] we proposed that it was possible that the appar-
ent allyl rotation could also take place through an initial
Scheme 2
PdϪN bond rupture. In the work described here we have
We synthesised allyl derivatives with two types of
attempted to gain a further insight into these processes by
counterion: triflate (OTf), which has coordinative ability,
looking for more data concerning the free energies of acti-
and the voluminous tetrakis(3,5-trifluoromethylphenyl)-
vation. More precisely, we were interested in determining
whether or not the apparent allyl rotation processes and the
borate (BArЈ4Ϫ), which is non-coordinating. In both cases
the starting material was [Pd(η3-2Me-C3H4)Cl]2 and this
was allowed to react in one step with a stoichiometric
amount of the nitrogen ligands and TlBArЈ4 or, alterna-
tively, in two steps with prior elimination of the chlorides
with AgCF3SO3 and addition of the nitrogen ligands after
the AgCl had been filtered off (see Scheme 3).
H4ϪH6 pyrimidine proton interconversion have the same
energy barrier and a common mechanism. Given that the
H4ϪH6 interconversion must involve a PdϪN bond-break-
ing step, comparison of the free energies of activation for
the two processes could provide information about the par-
ticipation of PdϪN bond rupture in the apparent allyl ro-
tation. As stated above, this aspect is now a matter of de-
bate. We also decided to explore the influence on the pro-
cesses of several factors such as the type of ligand, coun-
terion, solvent, complex concentration and addition of
water. These studies may prove useful in elucidating the
type of mechanism (associative or dissociative) operating in
the apparent allyl rotation process.
Scheme 3
Scheme 1
The products are soluble in polar solvents such as
acetone, dichloromethane and THF but are insoluble in
hexane, pentane and diethyl ether.
Results and Discussion
Synthesis of the New Ligands and Complexes
Structural Characterisation
We decided to prepare allylpalladium complexes with two
different ligands containing a pyrimidine ring (in order to
The new ligands and products were characterised by el-
1
study the H4ϪH6 interconversion process) and different 4- emental analysis, and IR, H and 13C NMR spectroscopy.
substituted pyrazole groups: 2-(4-methyl-1H-pyrazol-1-yl)- For the majority of the derivatives, HSQC experiments were
pyrimidine (4Me-pzpm, 1) and 2-(4-bromo-1H-pyrazol-1- also performed and, in some cases, NOE experiments were
yl)pyrimidine (4Br-pzpm, 2; Scheme 2). The results ob- carried out.
tained on these complexes could be compared with those
The IR spectra of the ligands and complexes exhibit the
obtained with pzpm. The differences in the three ligands stretching vibration bands for ν(CϭN) (see Exp. Sect.). The
Eur. J. Inorg. Chem. 2005, 100Ϫ109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
101

´
F. A. Jalon, B. R. Manzano, B. Moreno-Lara
FULL PAPER
1
Table 1. H NMR spectroscopic data for ligands 1 and 2 and complexes 3؊6[a]
Pyrazole
CH3
Pyrimidine
Allyl
Hanti
H5Ј
H3Ј
H5
H4/H6
Hsyn
CH3
1
8.38 (s)
8.73 (s)
8.72 (s)
8.90 (s)
2.14 (s)
Ϫ
2.27 (s)
2.22 (s)
7.62 (s)
7.84 (s)
8.28 (s)
8.33 (s)
7.36 (t), J5,4/5,6 ϭ 4.8
8.78 (d) J4,5/6,5 ϭ 4.8
8.85 (d) J4,5/6,5 ϭ 4.8
9.20 (bs)
Ϫ
Ϫ
4.49 (s)
4.49 (s)
Ϫ
Ϫ
Ϫ
Ϫ
2.25 (s)
2.18 (s)
2
7.48 (t), J5,4/5,6 ϭ 4.8
[b]
3
3.44 (s)
3.46 (s)
3.32 (s)
3.49 (s)
3.49 (s)
3.34 (s)
2.91 (s)
3.05 (s)
3.43 (s)
3.48 (s)
3.34 (s)
3.49 (s)
3[c]
9.25 (bs)
[b]
4
9.12 (s)
9.41(s)
Ϫ
Ϫ
8.53 (s)
8.69 (s)
7.91 (bs)
7.93 (t), J5,4/5,6 ϭ 4.9
9.51 (bs)
9.32 (d)
9.27 (d)
8.56 (d)
8.70 (d)
9.19 (bs)
9.31 (d)
9.25 (d)
9.32 (bs)
9.25 (bs)
9.36 (d)
9.31 (d)
9.15 (bs)
8.99 (bs)
4.57 (s)
4.54 (s)
4.51 (s)
3.93 (s)
4.16 (s)
4.49 (s)
4.50 (s)
2.27 (s)
2.17 (s)
4[c]
4[d]
8.57 (s)
Ϫ
7.93 (s)
7.14 (t), J5,4/5,6 ϭ 5.1
2.05 (s)
5
8.67 (s)
8.88 (s)
2.27 (s)
2.23 (s)
8.22 (s)
8.33 (s)
7.75 (bs)
7.85 (t), J5,4/5,6 ϭ 5.1
2.25 (s)
2.20 (s)
5[c]
6
9.14 (s)
9.45 (s)
8.71 (s)
Ϫ
Ϫ
Ϫ
8.56 (s)
8.73 (s)
8.04 (s)
7.86 (t), J5,4/5,6 ϭ 5.3
7.95 (t), J5,4/5,6 ϭ 5.1
7.72 (bs)
4.58 (s)
2.27 (s)
2.22 (s)
2.17 (s)
6[c]
6[e]
4.59 (s)
4.57 (s)
4.35 (s)
4.29 (s)
3.55 (s)
3.40 (s)
3.39 (s)
3.20 (s)
[a] Unless otherwise specified, room temperature and [D6]acetone as solvent; s: singlet; d: doublet; bs: broad singlet; t: triplet. See Scheme
[b]
Ϫ
[c]
[d]
[e]
2 for the atom numbering.
Resonance under BArЈ4 signals. Ϫ90 °C.
Ϫ60 °C, CDCl3. Ϫ80 °C, CD2Cl2. The resonances for
Ϫ
the BArЈ4 anion are for 3: δ ϭ 7.79 (s, Ho) and 7.68 (Hp) ppm; for 4: δ ϭ 7.78 (s, Ho) and 7.66 (Hp) ppm.
allyl ligand gives rise to a band at 838 cmϪ1 in all the com- involves a syn-syn, anti-anti interconversion, such as an ap-
plexes due to the ν(CϪCH3) vibration. The presence of the parent allyl rotation process, which will be analysed in the
corresponding counterions was confirmed by the presence section on fluxional behaviour. The differentiation between
of the expected bands (see Exp. Sect.).
the two methyl resonances of complex 3 [D6]acetone was
The 1H NMR spectroscopic data for the ligands and achieved with the help of NOE experiments. When H5Ј of
complexes are collected in Table 1. In the cases of deriva- the pyrazole ring was irradiated an NOE was found for the
tives 3 and 4, due to some overlap with the resonances of resonance at δ ϭ 2.27 ppm. An NOE for the resonance at
the counterion, the spectra were recorded both in CDCl3 δ ϭ 2.25 ppm and for the Hanti signal was also found when
and [D6]acetone. The assignment for the nitrogen-ligand the Hsyn resonance was irradiated.
resonances was made on the basis of previous studies with
The corresponding 13C{1H} NMR spectroscopic data are
similar ligands.[23,37] The H3Ј and H5Ј protons of the pyrazo- gathered in Table 2. The assignment for the nitrogen ligands
lyl rings appear as singlets both in the ligands and com- was made on the basis of literature data[23,26,37Ϫ39] and
plexes as a result of the presence of the substituent in the HSQC experiments. In a similar way to the 1H NMR spec-
4-position. In the free ligands the H5 proton of the pyrim- tra, a single signal was observed for the C4 and C6 pyrim-
idine ring appears as a triplet coupled to H4 and H6, which idine carbons (in complex 4 this signal was not observed).
are equivalent. In the complexes the protons H4 and H6 are A signal shift towards higher frequency with respect to the
inequivalent due to coordination. However, signals for the
two protons are only observed separately at room tempera- Table 2. 13C{1H} NMR spectroscopic data for ligands 1 and 2 and
complexes 3؊6[a]
ture in complex 6. In the other cases these protons give rise
to a single broad signal at this temperature. This pheno-
Pyrazole
Pyrimidine
C5 C4/C6 CH2
η3-Allyl
Cquat CH3
menon clearly indicates that the H4ϪH6 interconversion
process previously detected in [Pd(η3-2Me-C3H4)-
(pzpm)]CF3SO3 (I)[23] is also operating (see section on
C3Ј C4Ј C5Ј CH3 C2
1[b] 145.1 119.5 127.5 9.2 156.1 118.3 158.9
fluxional behaviour). In general, a shift towards higher fre- 1[c] 145.2 119.8 128.2 9.3 157.0 119.7 159.8
2[b] 144.3 97.6 129.4
2[c] 143.9 97.0 129.9
Ϫ
Ϫ
155.3 119.4 159.2
156.0 120.4 159.7
quency is observed for the corresponding resonances of the
complexes with respect to the free ligands due to the ligand-
to-metal σ-donation. As far as the allylic signals are con-
cerned, a single resonance is observed at room temperature
for the Hsyn and also for the Hanti protons. This observation
is not in accordance with a static situation of the complexes,
where two different terminal allylic carbons would be ex-
pected due to the asymmetry of the nitrogen ligands. This
[d]
[d]
[e]
3[b] 148.4
4[b] 147.9
9.0 154.1 120.4 161.3 61.2 (bs) 136.3 23.4
[d]
[e]
131.3
Ϫ
121.3
61.9 (bs) 137.1 23.4
5[c] 148.1 122.5 129.4 8.3 154.8 121.0 162.2 61.2 (bs) 135.5 22.7
6[c] 147.8 99.4 131.7
Ϫ
153.9 121.9 162.3 61.9 (bs) 136.0 22.7
[a]
Unless specified, the signals are singlets; bs ϭ broad singlet. See
[b]
[c]
Scheme 2 for the atom numbering. In CDCl3. In [D6]acetone.
[d]
[e]
Ϫ
Not observed. Resonances under BArЈ4 signals. The reson-
constitutes a clear indication of a dynamic situation that ances of the anions are given in the Exp. Sect.
102 www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2005, 100Ϫ109

Allyl Rotation and PdϪN Bond Rupture in Allylpalladium Complexes with N-Donor Ligands
FULL PAPER
free ligands is also seen. The allylic carbon resonances also mation was insufficient to conclude, with certainty, whether
reflect the apparent allyl rotation process as a single broad the two process have the same energy barrier.
signal appears for the terminal carbons. The expected coup-
In order to analyse the dynamic behaviour of complexes
3Ϫ6 we performed variable temperature H NMR studies
under different conditions in order to obtain the coalesc-
1
ling with boron or fluorine is observed for the Cipso and
CF3 carbons, respectively, of the BArЈ4 anion.[40]
ence temperatures (Tc) of the interconversion processes
H
Fluxional Behaviour
synϪHsyn, HantiϪHanti and H4ϪH6. For the majority of
As stated previously, with these new allyl complexes we the complexes, two Hsyn signals, two Hanti signals and separ-
tried to obtain sufficient information concerning their ate signals for H4 and H6 were observed at low temperature,
fluxional behaviour to draw a conclusion as to whether the as one would expect for a static situation (see Table 1). An
apparent allyl rotation process and the H4ϪH6 interconver- increase in the temperature allowed the corresponding co-
sion have the same energy barrier and follow a common alescence temperatures to be determined. These values are
mechanism. This information and the study of the influence given in Table 3 along with the δν values and the calculated
of different factors could also shed light on the controversy free energies of activation.[41] However, in some cases it was
surrounding mechanisms (associative or dissociative). In the not possible to determine the ∆Gc‡ values. The reasons for
case of derivative I,[23] the corresponding representation of this are that either the signals were not split at the minimum
free energy of activation versus coalescence temperature did experimental temperature or the coalescence was not re-
not allow a conclusion to be drawn as to whether the two ached at the maximum temperature allowed by the solvent
points for the apparent allyl rotation and the one for the (in this case it is indicated that the Tc values are higher than
H4ϪH6 interconversion were in a straight line — the infor- the maximum temperature registered and the ∆Gc‡ values
Table 3. δν, Tc and ∆G‡c data for complexes 3؊6 and I[a]
Entry
Complex
Anion
Solvent
Interchanging
groups
νc (Hz)
Tc (K)
∆G‡c (kJ/mol)
1
3
4
4
4
5
5
6
6
6
6
6
6
I
BArЈ4
BArЈ4
BArЈ4
BArЈ4
OTf
OTf
OTf
OTf
OTf
OTf
OTf
OTf
OTf
OTf
OTf
BArЈ4
BArЈ4
BArЈ4
OTf
OTf
OTf
OTf
OTf
OTf
OTf
OTf
OTf
BArЈ4
BArЈ4
BArЈ4
BArЈ4
BArЈ4
OTf
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
[D6]acetone
CDCl3
CDCl3
CDCl3
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
[D6]acetone
[D6]acetone
[D6]acetone
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
[D6]acetone
[D6]acetone
[D6]acetone
H
H
H
antiϪHanti
antiϪHanti
synϪHsyn
39.8
74.7
2.0
45.3
41.5
14.7
48.1
1.8
213
244
204
303
237
217
246
208
300
250
205
304
230
216
268
252
284
Ͼ331
245
221
311
251
229
Ͼ311
246
209
303
233
263
Ͼ330
226
290
244
205
298
43.6
48.9
46.7
62.6
48.6
46.3
50.4
47.8
63.2
50.3
47.1
62.5
47.3
47.8
55.4
54.2
57.9
Ͼ67.1
49.9
48.9
65.4
50.5
47.6
Ͼ63.8
49.9
48.2
63.8
49.2
53.3
Ͼ66.9
45.6
59.8
49.9
45.2
64.1
2[b]
3[b]
4[b]
5
H4ϪH6
H
antiϪHanti
6
7
8
9
H4ϪH6
H
H
antiϪHanti
synϪHsyn
H4ϪH6
28.5
73.2
1.9
10[b]
11[b]
12[b]
13
H
H
antiϪHanti
synϪHsyn
H4ϪH6
51.2
40.9
10.3
26.6
13.8
60.3
74.1
51.1
18.5
29.7
71.3
29.0
54.8
56.3
1.7
28.5
20.3
63.7
79.1
62.3
46.0
45.7
5.9
H
H
antiϪHanti
synϪHsyn
14
15
16
17
18
19
20
21
I
I
H4ϪH6
4
4
4
6
6
6
6
6
6
6
6
6
4
4
4
4
4
6
6
6
H
antiϪHanti
synϪHsyn
H
H4ϪH6
H
antiϪHanti
synϪHsyn
H
H4ϪH6
22[c]
23[c]
24[c]
25[c]
26[c]
27[c]
28[c]
29[c]
30[c]
31[d]
32[d]
33[e]
34[e]
35[e]
H
antiϪHanti
synϪHsyn
H
H4ϪH6
H
H
antiϪHanti
synϪHsyn
H4ϪH6
H
antiϪHanti
synϪHsyn
H
H4ϪH6
H
synϪHsyn
H4ϪH6
H
H
antiϪHanti
synϪHsyn
OTf
OTf
H4ϪH6
16.3
[a]
[b]
[c]
Unless specified the complex concentration is 1.41 ϫ 10Ϫ2 m and a 300 MHz spectrometer is used.
At 500 MHz.
Complex
concentration of 7.05 ϫ 10Ϫ3 m. [d] With addition of water (5.60 ϫ 10Ϫ2 m in water; H2O:complex ϭ 4:1). [e] With addition of water (3.52
ϫ 10Ϫ2 m in water; H2O:complex ϭ 2.5:1).
Eur. J. Inorg. Chem. 2005, 100Ϫ109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
103

´
F. A. Jalon, B. R. Manzano, B. Moreno-Lara
FULL PAPER
are higher than those calculated for these temperatures). In activation between the two processes will also be clear in
the following discussion we will detail our conclusions con- other plots presented in this paper aimed at evaluating the
cerning the influence of different factors on the fluxional influence of other factors (see below).
behaviour of our allyl complexes.
Effect of the Ligand
Comparison of the data for complexes 3 and 4 (Entries
1Ϫ4) on the one hand and for 5 and 6 on the other, as well
as those for complex I[23] (Entries 5Ϫ9 and 13Ϫ15), gives
an indication of the influence of the ligand. Unfortunately,
in the case of complexes with the 4-Mepzpm ligand it was
not possible to determine all the coalescence temperatures
and obtain the corresponding data. In the case of complex
3 it was only possible to detect the HantiϪHanti interconver-
Figure 1. Linear plot of ∆Gc‡ (kJ/mol) versus Tc (K) for complex 6
sion process while for 5 the splitting of the Hsyn signals was
not achieved. As far as the allylic interconversions are con-
NMR spectrometers (see Table 3): allylic interchange (diamonds,
cerned, and although it would be desirable to have more
in [D6]acetone (1.41 ϫ 10Ϫ2 m) using data from 300 and 500 MHz
ϪϪ); (R2 ϭ 0.9985); H4ϪH6 interchange (squares)
data, it seems that the type of ligand does not have a clear
influence, especially concerning the type of pyrazole ring.
With respect to the H4ϪH6 interchange, three sets of data
Effect of the Counteranion
are available in the case of compounds containing the trifl-
A comparison of Entries 1, 5 and 6 for the complexes
ate anion. In these compounds there is an increase in the
with the 4-Mepzpm ligand and Entries 2Ϫ4 and 7Ϫ12 for
coalescence temperature and in the free energy of activation
those with the 4-Brpzpm ligand should give an indication
on changing from 4Me-pzpm to pzpm to 4Br-pzpm, but
of the influence of the counteranion on the processes in
the three points fit reasonably well (R2 ϭ 0.9924) to a
question. However, the lack of data for the 4-Mepzpm com-
straight line (with positive slope), which might again indi-
plexes precludes a clear conclusion from being drawn. How-
cate the lack of influence of the type of pyrazole in the
ever, with the 4-Brpzpm complexes (see Figure 2) it is pos-
ligand on the process. It is possible that the electronic differ-
sible to conclude that, at least in [D6]acetone solution, there
ences between the three ligands are not sufficient to clearly
is a negligible influence of the anion on both the apparent
influence the processes under investigation. In any case, it
allyl rotation and the H4ϪH6 interconversion. The data rep-
is necessary to bear in mind that for the H4ϪH6 interchange
resented in Figure 2 also confirm the previous conclusion
process the bond that must be broken is between the pal-
that the apparent allyl rotation has a lower free energy of
ladium and the pyrimidine ring, not the pyrazole heterocy-
activation than the H4ϪH6 interconversion.
cle.
Type of Interchange
Before analyzing the influence of other factors, it is im-
portant to consider carefully the data for complex 6. Al-
though the δν value for the HantiϪHanti interconversion
(48.1 Hz, Entry 7) is clearly higher than that for the H4ϪH6
interchange (28.5 Hz, Entry 9), the Tc and ∆Gc‡ values are
markedly higher in the latter case. This clearly indicates that
the H4ϪH6 interchange has a higher energy barrier and will
also have a different mechanism than the apparent allyl ro-
tation. In order to obtain more data for this complex, we
decided to perform the same study with an NMR spec-
trometer of different magnetic field (500 MHz as opposed
to 300 MHz). In this way, we could obtain four data points
for the allylic interconversions (Entries 7, 8, 10 and 11) and
two for the H4ϪH6 interchange (Entries 9 and 12). The cor-
responding data are represented in Figure 1. The points for
Figure 2. Linear plot of ∆G‡c (kJ/mol) versus Tc (K) for complexes
4 and 6 in [D6]acetone (1.41 ϫ 10Ϫ2 m; see Table 3): 4 (squares); 6
(circles); allylic interchange (ϪϪ); R2 ϭ 0.9706
Effect of the Solvent
The studies described below are centred on complexes 4
the allylic interconversions fit very well to a straight line and 6, which contain the ligand 4-Brpzpm, because more
(R2 ϭ 0.9985) while the other two values have free energies coalescence points can usually be measured than for the
of activation that are clearly higher (7.9Ϫ8.9 kJ·molϪ1). 4-Mepzpm derivatives. For complex 4, the data previously
This trend is consistent with our previous conclusion that obtained in [D6]acetone solution (Entries 2Ϫ4) were com-
the H4ϪH6 interchange has a higher energy barrier than pared with those obtained in CDCl3 at the same concen-
the apparent allyl rotation. The difference in free energy of tration (Entries 16Ϫ18). These data are represented in Fig-
104
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 100Ϫ109

Allyl Rotation and PdϪN Bond Rupture in Allylpalladium Complexes with N-Donor Ligands
FULL PAPER
ure 3. The coalescence temperature for the H4ϪH6 inter- tone (Entries 7Ϫ12 and 25Ϫ27) and in CD2Cl2 solutions
change in CDCl3 was not reached and this is indicated by (Entries 19Ϫ24) and for complex 4 in CDCl3 solutions (En-
an arrow in the figure to reflect the fact that the actual tries 16Ϫ18 and 28Ϫ30). The coalescence of the H4 and H6
value is higher than the one indicated. As far as the appar- resonances was not reached up to the maximum tempera-
ent allyl rotation process is concerned, higher ∆Gc‡ values ture allowed by the solvent for complex 6 in CD2Cl2 or
are obtained when the solvent is CDCl3 and the same seems complex 4 in CDCl3. When the results at the two concen-
to apply to the H4ϪH6 interchange. Consequently, for this trations are compared, it is possible to conclude that in the
complex, which contains a counteranion (BArЈ4Ϫ) that does case of 6 in [D6]acetone the three pairs of data are practi-
not have coordinating ability, the presence of a solvent that cally identical — the concentration does not influence
it is able to coordinate has a positive effect on the processes. either of the two processes. In the other two studies, there
is insufficient information about the H4ϪH6 interchange
and for the apparent allyl rotation the influence, if it exists,
is very small (at least in the range of concentrations used).
Effect of the Addition of Water
The study of the influence of the addition of water was
performed with complex 4 in CDCl3 (Entries 16Ϫ18 and
31Ϫ32) in order to have a counterion and a solvent without
coordinating ability. The data were obtained for a water-
free CDCl3 solution and after the addition of four mol of
water per mol of product in this solution. The addition of
Figure 3. Linear plot of ∆Gc‡ (kJ/mol) versus Tc (K) for complex 4
water induces a general lowering of the coalescence tem-
(1.41 ϫ 10Ϫ2 m; see Table 3): CDCl3 (squares); [D6]acetone (dia-
peratures, and the splitting of the Hanti resonances is not
monds); allylic interchange (ϪϪ)
observed even down to Ϫ60 °C. However, the coalescence
of the H4ϪH6 resonances, which was not achieved in the
water-free solution, was observed at 17 °C after the addition
of water. When the data for the allylic interconversions are
taken into amount, it is concluded that the value obtained
after the addition of water is 4.9 kJ·molϪ1 below the
straight line obtained with the data for the water-free solu-
tion. Consequently, at least in the situation where the coun-
teranion and solvent are non-coordinating, the addition of
water favours the apparent allyl rotation. We also studied
the effect of the addition of water when there are species of
coordinating ability in the reaction medium, for example
for complex 6 in [D6]acetone solution (comparison of En-
tries 7Ϫ9 with 33Ϫ35). In this case, there is no appreciable
effect observed in the values of free energy of activation
after the addition of water.
A similar solvent effect is not observed in complex 6
where the anion (OTfϪ) has coordinating ability. In this
case the comparison is made between the [D6]acetone (En-
tries 7Ϫ12) and CD2Cl2 (Entries 19Ϫ21) data. It can clearly
be seen from Figure 4 that the solvent does not have any
influence on the apparent allyl rotation, while for the
H4ϪH6 interchange the conclusion is not as straightfor-
ward. Once again, this figure demonstrates the difference in
free energy of activation between the apparent allyl rotation
and the H4ϪH6 interchange.
When the different plots for the apparent allyl rotation
process are considered, a positive slope in the straight line
obtained is observed. Although we do not believe that it is
possible to give accurate data for the entropy of activation,
it is clear that this parameter is negative.
It must be emphasised that our results indicate that care
must be taken when studying these types of process in sol-
vents that do not have coordinating ability. In solvents such
as acetone, as observed previously,[35] the presence of small
amounts of water does not influence the energy barrier of
the process.
Figure 4. Linear plot of ∆Gc‡ (kJ/mol) versus Tc (K) for complex
6 (1.41 ϫ 10Ϫ2 m; see Table 3): CD2Cl2 (squares); [D6]acetone
(circles); allylic interchange (ϪϪ); R2 ϭ 0.9914
1
Conclusions of the H NMR Variable Temperature Studies
Effect of the Complex Concentration
In order to obtain information about the possible partici-
We deduced that there is a lower energy barrier for the
pation of bimolecular species in these processes, we carried apparent allyl rotation process than for the H4ϪH6 in-
out several variable temperature 1H NMR studies with two terconversion and, consequently, the two processes follow
different concentrations (1.41 ϫ 10Ϫ2 and 7.05 ϫ 10Ϫ3 m). different mechanisms. In the case of the apparent allyl ro-
For complex 6, we performed the studies both in [D6]ace- tation process, and concerning the effect of different factors
Eur. J. Inorg. Chem. 2005, 100Ϫ109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
105

´
F. A. Jalon, B. R. Manzano, B. Moreno-Lara
FULL PAPER
on this phenomenon, we can state the following: (i) a clear by other authors[20b] but, depending on the temperature and
influence of the type of pyrazole present in the nitrogen- the concentration range, the effect can be rather small be-
containing ligand is not found; (ii) the results are the same cause of the existence of a limiting constant value.
when, in [D6]acetone solutions, the counterion is changed
It is also important to consider that the energy barrier
Ϫ
(cf. BArЈ4 versus OTfϪ) or when, for the complexes with for the apparent allyl rotation is lower than for the H4ϪH6
the OTfϪ anion, two different solvents are used ([D6]ace- interconversion process. This latter process necessarily in-
Ϫ
tone and CD2Cl2); (iii) in the complex with the BArЈ4
volves a PdϪN(pm) bond rupture, probably as the limiting
anion, the free energy of activation values are higher in a step. Consequently, the apparent allyl rotation process
non-coordinating solvent such as CDCl3 than in [D6]ace- should occur through a path that does not involve
tone; (iv) the complex concentration does not affect the pro- PdϪN(pm) bond rupture and has a lower energy of acti-
cess in the range studied; (v) when the complex contains a vation. The breaking of the PdϪN(pyrazole) bond is less
non-coordinating anion and the study is carried out in likely because of the higher basicity of this heterocycle (ex-
CDCl3, the addition of water decreases the free energy of cept for the case of 4-Br-pyrazole) with respect to the py-
activation. However, the addition of water does not affect rimidine ring [pKa(pyrimidine) ϭ 1.1]. This again makes a
the dynamic behaviour in [D6]acetone when the counterion dissociative mechanism less likely and points to an associat-
present is OTfϪ; (vi) the entropy of activation is negative.
ive pathway. The partial dissociation of the didentate N-
In the case of the H4ϪH6 interconversion process it is donor ligand in the five-coordinate intermediate has some-
more difficult to draw conclusions because only one ∆Gc‡ times been considered.[20b] In our case, however, this pos-
data point could be obtained for each study. Despite this, sibility can be excluded. Consequently, the associative
we believe that the following general points can be made: mechanism[20,28,29,32,35] we propose would have the follow-
(i) the type of anion, ligand or complex concentration does ing steps (see Scheme 4): (i) formation of a five-coordinate
not have any influence when the solvent is [D6]acetone; (ii) intermediate through coordination of the anion, solvent or
when both the solvent (CDCl3) and anion (BArЈ4Ϫ) are water (L); (ii) pseudorotation of the allyl group in the inter-
non-coordinating, the addition of water reduces the free en- mediate formed, which interchanges the two allylic ter-
ergy of activation but this addition does not affect the val- minals; and (iii) decoordination of the L group, leading to
ues of ∆Gc‡ in [D6]acetone when the counterion is OTfϪ; the final product in which the observed HsynϪHsyn
,
(iii) the entropy of activation is negative.
HantiϪHanti interchange has taken place.
During the studies described here, broadening of the Hsyn
or Hanti resonances was not observed in any case after co-
alescence was achieved. This fact indicates that if the
H
synϪHanti interconversion process (usually proposed to oc-
cur through an η3-η1-η3 mechanism) exists, it must be of
higher energy and is therefore not observed under the con-
ditions used here.
Mechanistic Proposal
Apparent Allyl Rotation
The negative sign of the entropy of activation points to
an associative process with the formation of five-coordinate
intermediates. On the basis of some of the conclusions out-
lined above (ii, iii and v), it is possible to state that a species
with coordinating ability (counteranion, solvent or water)
clearly favours the process. We must consider that when the
counterion is BArЈ4Ϫ, the solvents are CD2Cl2 or CDCl3
and water is not added, the two processes, although with
Scheme 4
H4؊H6 Interconversion
The negative entropy of activation, the positive effect of
higher energies of activation, do take place. It is possible the addition of water when both the solvent and the anion
that in these cases traces of water or ClϪ in the solvents are non-coordinating, and the absence of other effects when
(decomposition of CD2Cl2 or CDCl3), or the low coordin- the solvent is [D6]acetone all indicate that, in this case, an
ating ability of the solvents, allows the process to occur. associative mechanism is in operation. The mechanism
Conclusion iv indicates that bimolecular species are not should involve the following steps (see Scheme 5): (i) forma-
formed and that the free nitrogen of the pyrimidine ring tion of the five-coordinate intermediate through coordi-
does not participate in the formation of the five-coordinate nation of the solvent, anion or water (L); (ii) PdϪN(pm)
intermediates. It is necessary to consider that the formation bond rupture; (iii) internal rotation of the ligand[19] through
of an ‘‘ion pair’’ between the cation and the anion should the CϪN bond that links the two heterocycles in the tetra-
be favoured in concentrated solutions. This would be par- coordinate intermediate formed; (iv) recoordination of the
ticularly true in the case of a non-coordinating solvent, such pyrimidine ring through the other nitrogen atom; and (v)
as complex 6 in CD2Cl2. This phenomenon has been found decoordination of the L group, leading to the final species
106
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 100Ϫ109

Allyl Rotation and PdϪN Bond Rupture in Allylpalladium Complexes with N-Donor Ligands
FULL PAPER
in which the interchange between the positions of the H4
and H6 pyrimidine protons has taken place.
an Innova 500 spectrometer using CDCl3, CD2Cl2 and (CD3)2CO
as solvents; the chemical shifts are given in ppm. Standard ex-
perimental conditions were employed.[25,26] The NOE difference
spectra were recorded with the following acquisition parameters:
spectral width 5000 Hz, acquisition time 3.27 s, pulse width 18 ms,
relaxation delay 4 s, irradiation power 5Ϫ10 dB, number of scans
240. For 1H-13C g-HSQC spectra the standard Varian pulse
sequences were used (V NMR 6.1 C software). The spectra were
acquired using 7996 Hz (1H) and 30154 Hz (13C) spectral widths;
16 transients of 2048 data points were collected for each 256 in-
crements. For variable-temperature spectra the probe temperature
(Ϯ1 K) was controlled by a standard unit calibrated with a meth-
anol reference. Free energies of activation (kJ·molϪ1) were calcu-
lated[41] from the coalescence temperature (Tc) and the frequency
difference between the coalescing signals (extrapolated at the co-
alescence temperature) with the formula ∆G‡c ϭ a·T·[9.972 ϩ log(T/
δν)], where a ϭ 1.914 ϫ 10Ϫ2. The estimated error in the calculated
free energies of activation is Ϯ1.1 kJ·molϪ1. o ϭ ortho, m ϭ meta,
p ϭ para.
Scheme 5
2-(4-Methyl-1H-pyrazol-1-yl)pyrimidine (1): A mixture of 4-methyl-
pyrazole (0.248 mL, 3 mmol), potassium tert-butoxide (600 mg,
6 mmol) and aliquat 336 (80 mg, 10%) was stirred at 120 °C for
30 min. 2-Chloropyrimidine (515 mg, 4.5 mmol) was then added at
120 °C and the reaction mixture was stirred at this temperature
for 1 h. Extraction with dichloromethane (50 mL), removal of the
solvent from the extract and flash chromatography of the residue
on silica gel using ethyl acetate as the eluent allowed the isolation
of 1. Yield: 62% (298 mg). IR: ν˜ ϭ 1573 cmϪ1 [ν(CϪN)]. C8H8N4
Conclusions
New allylpalladium derivatives with N-donor ligands
have been obtained and their fluxional behaviour studied.
Two processes of apparent allyl rotation (observed as
H
synϪHsyn, HantiϪHanti interconversions) and also H4ϪH6
interconversion of the pyrimidine protons have been ob-
served. This last process must involve, probably as the limit-
ing step, PdϪN(pyrimidine) bond rupture. It has been con- (160.18): calcd. C 59.99, H 5.03, N 34.98; found C 59.70, H 5.29,
N 34.62.
cluded that the apparent allyl rotation process has a lower
free energy of activation. Consequently, this allyl rotation
must not involve PdϪN bond rupture. The influence of fac-
tors such as counterion, solvent and addition of water has
been studied and it is possible to conclude that the presence
of a species with coordinating ability clearly favours both
processes. A bimolecular mechanism can be excluded in
both cases because the complex concentration does not af-
2-(4-Bromo-1H-pyrazol-1-yl)pyrimidine (2): A similar procedure to
that used for 1 was applied for 2. Amounts were as follows: 4-
bromopyrazole (440.8 mg, 3 mmol), potassium tert-butoxide
(600 mg, 6 mmol), aliquat 336 (80 mg, 10%) and 2-chloropyrimid-
ine (515 mg, 4.5 mmol). Yield 60% (405.1 mg). IR: ν˜ ϭ 1570 cmϪ1
[ν(CϪN)]. C7H5BrN4 (225.05): calcd. C 37.36, H 24.89, N 2.24;
found C 37.25, H 24.99, N 2.18.
fect the free energy of activation. Negative entropies of acti- [Pd(η3-C4H7)(4-Mepzpm)]BArЈ4
(3):
4-Mepzpm
(24.3 mg,
Ϫ
0.152 mmol) and TlBArЈ4 (162.3 mg, 0.152 mmol) were added to
a solution of [Pd(η3-C4H7)(µ-Cl)]2 (30 mg, 0.076 mmol) in 10 mL
of toluene. The mixture was stirred at room temperature for 1 h.
The resulting suspension was filtered off. The solid was then ex-
tracted with dichloromethane (40 mL) and the solution was evapo-
rated to dryness. The light-brown solid was washed with hexane
and dried. Yield: 66% (119.1 mg). IR: ν˜ ϭ 1598 and 1566
[ν(CϪN)], 1275, 890 (BArЈ4Ϫ), 839 cmϪ1 ν[(CϪCH3), allyl].
C44H27BF24N4Pd (1184.9): calcd. C 44.60, H 2.30, N 4.73; found
C 44.52, H 2.44, N 4.69. 13C{1H} NMR (125 MHz, CDCl3, 298 K;
vation were obtained. Associative mechanisms with partici-
pation of five-coordinate intermediates have been proposed
for both processes, with a step involving PdϪN(pyrimidine)
bond rupture in the case of the H4ϪH6 interconversion.
On the basis of our studies it is possible to conclude that,
for these types of process, the presence of traces of water in
coordinating solvents like acetone does not affect the free
energy of activation. However, this is not the case for non-
coordinating solvents and great care must be taken in stud-
ies involving such solvents.
Ϫ
1
BArЈ4 resonances): δ ϭ 161.9 (q, JB,C ϭ 50.0 Hz, 4 C, Cipso),
1
135.0 (s, 8 C, Co), 129.2 (m, 8 C, Cm), 124.7 (q, JC,F ϭ 272.6 Hz,
8 C, CF3), 117.7 (s, 4 C, Cp) ppm.
[Pd(η3-C4H7)(4-Brpzpm)]BArЈ4 (4): A similar procedure to that
used for 3 was applied for 4. Amounts were as follows: [Pd(η3-
C4H7)(µ-Cl)]2 (20 mg, 0.051 mmol), 4-Brpzpm (22.9 mg,
0.102 mmol) and TlBArЈ4 (108.9 mg, 0.102 mmol). The solid was
pale brown in colour. Yield: 66% (83.7 mg). IR: ν˜ ϭ 1599 and 1568
[ν(CϪN)], 1279, 890 (BArЈ4Ϫ), 838 cmϪ1 ν[(CϪCH3), allyl].
C43H24BBrF24N4Pd (1249.8): calcd. C 41.32, H 1.94, N 4.48; found
C 41.67, H 2.19, N 4.49. 13C{1H} NMR (75 MHz, CDCl3, 298 K;
Experimental Section
General Comments: All manipulations were carried out under dry,
oxygen-free nitrogen using standard Schlenk techniques. Solvents
were distilled from the appropriate drying agents and degassed be-
fore use. The derivative [Pd(η3-C4H7)(µ-Cl)]2 was synthesised ac-
cording to a literature procedure.[42] Elemental analyses were per-
formed with a Carlo Erba Instruments EA 1108 CHNS/O micro-
Ϫ
1
analyser (University of La Corun˜a). IR spectra were recorded as BArЈ4 resonances): δ ϭ 161.9 (q, JB,C ϭ 49.3 Hz, 4 C, Cipso),
1
1
nujol mulls with a PerkinϪElmer PE 883 IR spectrometer. H and
134.9 (s, 8 C, Co), 129.8 (m, 8 C, Cm), 124.7 (q, JC,F ϭ 272.2 Hz,
13C{1H} NMR spectra were recorded with a Varian Unity 300 and 8 C, CF3), 117.7 (s, 4 C, Cp) ppm.
Eur. J. Inorg. Chem. 2005, 100Ϫ109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
107

´
F. A. Jalon, B. R. Manzano, B. Moreno-Lara
FULL PAPER
[Pd(η3-C4H7)(4-Mepzpm)]OTf
[3] [3a]
B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996, 96,
(5):
AgCF3SO3
(104.3 mg,
[3b]
0.406 mmol) was added to a solution of [Pd(η3-C4H7)(µ-Cl)]2
(80 mg, 0.203 mmol) in THF (20 mL). The mixture was stirred for
1 h and protected from light. The solution was filtered through
Celite to eliminate AgCl. The ligand 4-Mepzpm (65 mg,
0.406 mmol) was then added and the mixture was stirred for 1 h.
The solvent was evaporated to dryness and the pale-yellow residue
was washed with diethyl ether. Yield 70% (133.7 mg). IR: ν˜ ϭ 1596
and 1565 [ν(CϪN)], 1266, 1154, 1032, 784 and 639 (OTfϪ), 839
cmϪ1 ν[(CϪCH3), allyl]. C13H15F3N4O3PdS (470.75): calcd. C
33.17, H 3.21, N 11.90; found C 33.46, H 3.11, N 12.23. 13C{1H}
395Ϫ422.
779Ϫ784.
B. M. Trost, Pure Appl. Chem. 1996, 68,
[3c]
˚
J. D. Oslob, B. Akermark, P. Helquist, P. O.
Norrby, Organometallics 1997, 16, 3015Ϫ3021. [3d] C. Larksarp,
H. Alper, J. Am. Chem. Soc. 1997, 119, 3709Ϫ3715. [3e] H. Hag-
˚
elin, B. Akermark, P.-O. Norrby, Organometallics 1999, 18,
[3f]
2884Ϫ2895.
J. C. Anderson, R. J. Cubbon, J. D. Harling,
[3g]
Tetrahedron: Asymmetry 2001, 12, 923Ϫ935.
P. J. Harring-
ton, in Comprehensive Organometallic Chemistry (Eds.: E. W.
Abel, F. G. A. Stone, G. Wilkinson, L. S. Hegedus), Pergamon,
Oxford, 1995, vol. 12, p. 882Ϫ894 and references cited therein.
[4a] P. W. N. M. van Leeuwen, P. W. Praat, J. Chem. Soc., Chem.
[4]
[4b]
1
Commun. 1970, 365Ϫ366.
F. G. A. Stone, N. M. Boag, M.
NMR (125 MHz, [D6]acetone, 298 K): δ ϭ 121.7 (q, JC,F ϭ 322.2
Hz, 1 C, CF3) ppm.
Green, J. L. Spencer, J. Organomet. Chem. 1977, 127,
[4c]
C51ϪC54.
R. E. Rülke, V. E. Raasjager, P. Wehman, C. J.
[Pd(η3-C4H7)(4-Brpzpm)]OTf (6): A similar procedure to that used
for 5 was applied for 6. Amounts were as follows: Ag CF3SO3
(78.1 mg, 0.304 mmol), [Pd(η3-C4H7)(µ-Cl)]2 (60 mg, 0.152 mmol),
4-Brpzpm (68.4 mg, 0.304 mmol). The solid was white in colour.
Yield: 80% (126.6 mg). IR: ν˜ ϭ 1597 and 1566 [ν(CϪN)], 1269,
1147, 1034, 785 and 639 (OTfϪ), 838 cmϪ1 ν[(CϪCH3), allyl].
C12H12BrF3N4O3PdS (535.62): calcd. C 26.91, H 2.26, N 10.46;
found C 27.01, H 2.11, N 10.27. 13C{1H} NMR (125 MHz, [D6]-
Elsevier, P. W. N. M. van Leeuwen, K. Vrieze, J. Fraanje, K.
Boubitz, A. L. Spek, Organometallics 1996, 15, 3022Ϫ3031.
[5] [5a]
J. M. Brown, D. I. Hulmes, P. J. Guiry, Tetrahedron 1994,
50, 4493Ϫ4506. [5b] T. D. W. Claridge, J. M. Long, J. M. Brown,
D. Hibbs, M. B. Hursthouse, Tetrahedron 1997, 53, 4035Ϫ4050.
[6] [6a]
P. Barbaro, P. S. Pregosin, R. Salzmann, A. Albinati, R. W.
Kunz, Organometallics 1995, 14, 5160Ϫ5170. [6b] T. Hosokawa,
Y. Wakabayashi, K. Hosokawa, T. Tsuji, S. I. Murahashi,
[6c]
Chem. Commun. 1996, 859Ϫ860.
S. Ramdeehul, L. Barloy,
1
acetone, 298 K): δ ϭ 121.8 (q, JC,F ϭ 322.2 Hz, 1 C, CF3) ppm.
J. A. Osborn, A. De Cian, J. Fischer, Organometallics 1996, 15,
[6d]
5442Ϫ5444.
H. A. Ankersmit, N. Veldman, A. L. Spek, K.
Preparation of NMR Spectroscopic Samples: Samples with a con-
centration of 1.41 ϫ 10Ϫ2 m were prepared by dissolving the corre-
sponding amount of the complex in 0.5 mL of the deuterated sol-
vent. Samples with a concentration of 7.05 ϫ 10Ϫ3 m were prepared
by adding 0.5 mL of the corresponding deuterated solvent to the
previous sample. Samples for the studies of Entries 31 and 32
(Table 3) were prepared by adding 0.510 µL (2.8 ϫ 10Ϫ2 mmol) of
water to the original solution. Samples for the studies of Entries
33Ϫ35 were prepared by adding 0.320 µL (1.76 ϫ 10Ϫ2 mmol) of
water to the original sample.
Vrieze, G. Van Koten, Inorg. Chim. Acta 1996, 252, 339Ϫ354.
[6e]
L. Barloy, S. Ramdeehul, J. A. Osborn, C. Carlotti, F. Tau-
lelle, A. De Cian, J. Fisher, Eur. J. Inorg. Chem. 2000, 12,
2523Ϫ2532.
[7]
L. Canovese, F. Visentin, P. Uguagliati, V. Lucchini, G. Ban-
doli, Inorg. Chim. Acta 1998, 277, 247Ϫ252.
[8] [8a]
I. Schwarz, J. Rust, C. W. Lehmann, M. Braun, J. Or-
ganomet. Chem. 2000, 605, 109Ϫ116. [8b] F. Gomez-de la Torre,
Y. Gutierrez, F. A. Jalon, B. R. Manzano, A. Rodrıguez, Mon-
´
´
´
´
atsh. Chem. 2000, 131, 1267Ϫ1280.
[9] [9a]
J. Krause, R. Goddard, R. Mynott, K.-R. Pörschke, Or-
ganometallics 2001, 20, 1992Ϫ1999. [9b] P. Braunstein, F. Naud,
A. Dedieu, M. M. Rohmer, A. De Cian, S. J. Rettig, Organome-
tallics 2001, 20, 2966Ϫ2981.
Acknowledgments
We are grateful to the Spanish DGES/MCyT (Project N. BQU-
[10]
[11]
[12]
M. Kollmar, B. Goldfuss, M. Reggelin, F. Rominger, G.
Helmchen, Chem. Eur. J. 2001, 7, 4913Ϫ4927.
L. Canovese, G. Chessa, C. Santo, F. Visentin, P. Uguagliati,
Organometallics 2002, 21, 4342Ϫ4349.
´
´
2002-00286) and the Consejerıa de Ciencia y Tecnologıa de la Junta
de Comunidades de Castilla-La Mancha (PBI-02-002) for finan-
cial support.
M. Ogasawara, K. Takizawa, T. Hayashi, Organometallics
2002, 21, 4853Ϫ4861.
[1]
[1a] J. Tsuji, Palladium Reagents and Catalysts, Innovation in Or-
[13] [13a]
C. Breutel, P. S. Pregosin, R. Salzmann, A. Togni, J. Am.
[1b]
Chem. Soc. 1994, 116, 4067Ϫ4068. [13b] P. S. Pregosin, R. Salz-
mann, A. Togni, Organometallics 1995, 14, 842Ϫ847.
ganic Synthesis, J. Wiley & Sons, Chichester, 1996.
S. A.
[13c]
Godleski, in Comprehensive Organic Synthesis (Eds.: B. M.
J.
Trost, I. Fleming), Pergamon Press: New York, 1991, vol. 4,
Herrmann, P. S. Pregosin, R. Salzmann, A. Albinati, Or-
ganometallics 1995, 14, 3311Ϫ3318.
[1c]
chapter 3.3.
P. J. Harrington, in Comprehensive Organomet-
[14]
allic Chemistry II (Eds.: E. W. Abel, F. G. A. Stone, G. Wilkin-
son, R. J. Puddephatt), Elsevier, New York, 1995, vol. 12,
p.797. [1d] J.-E. Bäckvall, Metal-Catalysed Cross Coupling Reac-
tions, VCH, Weinheim, Germany, 1998, p. 339. [1e] B. M. Trost,
´
´
´
´
R. Fernandez-Galan, F. Jalon, B. R. Manzano, J. Rodrıguez de
la Fuente, M. Vrahami, B. Jedlicka, W. Weissensteiner, Or-
ganometallics 1997, 16, 3758Ϫ3768.
[15]
[16]
P. Dotta, P. G. A. Kumar, P. S. Pregosin, Magn. Reson. Chem.
2002, 40, 653Ϫ658.
B. Crociani, S. Antonaroli, G. Bandoli, L. Canovese, F. Visen-
tin, P. Uguagliati, Organometallics 1999, 18, 1137Ϫ1147.
C. Lee, in Catalytic Asymmetric Synthesis, 2nd ed. (Ed.: I.
[1f]
Ojima), Wiley-VCH, New York, 2000, p. 593Ϫ649.
A.
Pfaltz, M. Lautens, in Comprehensive Asymmetric Catalysis
(Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, Hei-
delberg, 1999, p.833Ϫ886.
[17]
[18]
´
N. Solin, K. J. Szabo, Organometallics 2001, 20, 5464Ϫ5471.
A. Albinati, R. W. Kunz, C. J. Ammann, P. S. Pregosin, Or-
ganometallics 1991, 10, 1800Ϫ1806.
[2] [2a]
K. Vrieze, Dynamic Nuclear Magnetic Resonance Spec-
troscopy (Eds.: L. M. Jackman, F. A. Cotton), Academic Press,
New York, 1975, chapter 11, p. 441. [2b] P. S. Pregosin, R. Salz-
mann, Coord. Chem. Rev. 1996, 155, 35Ϫ68. [2c] J. W. Faller, in
[19]
A. Gogoll, J. Örnebro, H. Grennberg, J.-E. Bäckvall, J. Am.
Chem. Soc. 1994, 116, 3631Ϫ3632.
[20] [20a]
B. Crociani, F. Di Bianca, A. Giovenco, T. Boschi, Inorg.
Encyclopedia of Inorganic Chemistry (Ed.: R. B. King), John
Wiley and Sons, Chichester, 1994, vol. 7, p. 3925Ϫ3926.
A. Jalon, B. R. Manzano, F. Gomez-de la Torre, A. M. Guer-
[20b]
[2d]
Chim. Acta 1987, 127, 169Ϫ182.
B. Crociani, S. Antonar-
F.
oli, M. Paci, F. Di Bianca, L. Canovese, Organometallics 1997,
´
´
16, 384Ϫ391.
´
rero, A. M. Rodrıguez, Targets in Heterocyclic Systems, Chem-
[21]
´
istry and Properties (Eds.: O. A. Attanasi, D. Spinelli), Roma,
1999, vol. 3, p. 399Ϫ438.
A. Gogoll, H. Grennberg, A. Axen, Organometallics 1997, 16,
1167Ϫ1178.
108
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 100Ϫ109

Allyl Rotation and PdϪN Bond Rupture in Allylpalladium Complexes with N-Donor Ligands
FULL PAPER
[22]
[23]
[33]
[34]
Y. Takao, T. Takeda, J. Watanabe, J. Setsune, Organometallics
1999, 18, 2936Ϫ2938.
J. W. Faller, H. L. Stokes-Huby, M. A. Albrizzio, Helv. Chim.
Acta 2001, 84, 3031Ϫ3042.
´
`
M. A. Pericas, C. Puigjaner, A. Riera, A. Vidal-Ferran, M.
J. Elguero, A. Fruchier, A. de la Hoz, F. A. Jalon, B. R. Man-
´
´
´
zano, A. Otero, F. Gomez-de la Torre, Chem. Ber. 1996, 129,
Gomez, F. Jimenez, G. Muller, M. Rocamora, Chem. Eur. J.
2002, 8, 4164Ϫ4178.
589Ϫ594.
[24]
[25]
[35]
´
´
´
A. M. Guerrero, F. A. Jalon, B. R. Manzano, R. M. Clara-
F. A. Jalon, B. R. Manzano, A. Otero, M. C. Rodrıguez, J.
Organomet. Chem. 1995, 494, 179Ϫ185.
´
´
´
munt, M. D. Santa Marıa, C. Escolastico, J. Elguero, A. M.
´
´
´
´
F. Gomez-de la Torre, A. de la Hoz, F. A. Jalon, B. R. Man-
Rodrıguez, M. A. Maestro, J. Mahıa, Eur. J. Inorg. Chem.
´
´
zano, A. Otero, A. M. Rodrıguez, M. C. Rodrıguez-Perez, A.
2002, 3178Ϫ3189.
[36]
[37]
´
´
Echevarrıa, J. Elguero, Inorg. Chem. 1998, 37, 6606Ϫ6614.
J. Catalan, J.-L. M. Abboud, J. Elguero, Adv. Heterocycl. Chem.
[26]
[27]
[28]
[29]
[30]
´
´
F. Gomez-de la Torre, A. de la Hoz, F. A. Jalon, B. R. Man-
1987, 41 187Ϫ274.
´
´
´
J. Elguero, A. Guerrero, F. Gomez de la Torre, A. de la Hoz,
zano, A. M. Rodrıguez, J. Elguero, M. Martınez-Ripoll, Inorg.
´
´
Chem. 2000, 39, 1152Ϫ1162.
F. A. Jalon, B. R. Manzano, A. Rodrıguez, New. J. Chem. 2001,
´
´
N. Arroyo, F. Gomez-de la Torre, F. A. Jalon, B. R. Manzano,
25, 1050Ϫ1060.
[38] [38a]
´
B. Moreno-Lara, A. M. Rodrıguez, J. Organomet. Chem. 2000,
L. A. Oro, M. Esteban, R. M. Claramunt, J. Elguero, C.
Foces-Foces, F. H. Cano, J. Organomet. Chem. 1984, 276,
603, 174Ϫ184.
[38b]
´
´
´
M. C. Carrion, A. Dıaz, A. Guerrero, F. A. Jalon, B. R. Man-
zano, A. Rodrıguez, R. L. Paul, J. C. Jeffery, J. Organomet.
79Ϫ97.
M. A. Esteruelas, L. A. Oro, M. C. Apreda, C.
´
´
Foces-Foces, F. H. Cano, R. M. Claramunt, C. Lopez, J. El-
Chem. 2002, 650, 210Ϫ222.
guero, M. Begtrup, J. Organomet. Chem. 1988, 344, 93Ϫ108.
[39] [39a]
˚
´ ´
M. Fajardo, A. de la Hoz, E. Dıez-Barra, F. A. Jalon, A.
S. Hansson, P-O. Norrby, M. P. T. Sjögren, B. Akermark, M.
´
E. Cucciolito, F. Giordano, A. Vitagliano, Organometallics
1993, 12, 4940Ϫ4948.
J. Sprinz, M. Kiefer, G. Helmchen, M. Reggelin, G. Huttner,
O. Walter, L. Zsolnai, Tetrahedron Let. 1994, 35, 1523Ϫ1526.
Otero, A. Rodrıguez, J. Tejeda, D. Belletti, M. Lanfranchi, M.
A. Pellinghelli, J. Chem. Soc., Dalton Trans. 1993, 1935Ϫ1939.
[39b]
´ ´
F. A. Jalon, B. R. Manzano, A. Otero, M. C. Rodrıguez-
´
Perez, J. Organomet. Chem. 1995, 494, 179Ϫ185.
M. Brookhart, B. Grant, A. F. Volpe Jr., Organometallics 1992,
11, 3920Ϫ3922.
[31] [31a]
[40]
[41]
[42]
J. G. P. Delis, J. H. Groeen, K. Vrieze, P. W. N. M. van
Leeuwen, N. Veldman, A. L. Spek, Organometallics 1997, 16,
[31b]
551Ϫ562.
L. Canovese, F. Visentin, P. Uguagliati, G.
J. Sandström, Dynamic NMR Spectroscopy, Academic Press,
London, 1982.
W. T. Dent, R. Long, G. Wilkinson, J. Chem. Soc. 1964,
1585Ϫ1588.
Chessa, V. Lucchini, G. Bandoli, Inorg. Chim. Acta 1998,
275/276, 385Ϫ394.
[32]
R. J. van Haaren, K. Goubitz, J. Fraanje, G. P. F. Van
Strijdonck, H. Oevering, B. Coussens, J. N. H. Reek, P. C. J.
Kamer, P. W. N. M. van Leeuwen, Inorg. Chem. 2001, 40,
3363Ϫ3372.
Received May 19, 2004
Early View Article
Published Online November 10, 2004
Eur. J. Inorg. Chem. 2005, 100Ϫ109
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
109