Kinetico-mechanistic studies of C-H bond activation on new Pd complexes containing N,N′-chelating ligands

Article abstract of DOI:10.1039/b415613g

The hybrid imine/amine palladium(II) coordination complexes [PdX ₂(κ²-N_imino,N_amino)] (X = Cl, AcO; κ²-N_imino,N_amino = 4ClC ₆H₄CHNCH₂(CH₂)_nN(CH ₃)₂, n = 1, 2) have been prepared in different isomeric forms which include E/Z arrangement around the C=N bond of the hybrid ligand and {Pd(κ²-N_imino,N_amino)} ring conformation. The crystal structures of four of them, E-1AcO, Z-1AcO, E-2AcO and E-2Cl, have been determined and the solution behaviour in acetic acid, the common cyclometallating solvent, for all these systems studied. The complexes in acetic acid solution are shown to maintain the structure determined by X-ray crystallography, as they do in deuterated chloroform. Nevertheless, a partial opening equilibrium of the {Pd(κ²-N_imino,N _amino)} ring is observed by NMR experiments. When the complexes are held in solution for longer periods the corresponding cyclometallated derivatives, 1AcO-CM, 2AcO-CM, 1Cl-CM and 2Cl-CM, containing the {Pd(κ²-C,N_imino)} palladacycle are obtained, as characterized by ¹H NMR spectroscopy. In these compounds the total opening of the N_amino moiety of the ligand has occurred. The C-H bond activation process has been studied kinetico-mechanistically at different temperatures, pressures and acid concentrations; the results agree with the need of an opening of the chelate ring in [PdX₂(κ ²-N_imino,N_amino] prior to the proper cyclometallation reaction. The values of the enthalpies of activation are higher than those observed for known N-monodentated cyclometallating ligands, as should correspond to the contribution of a ligand dechelation pre-equilibrium. The entropies and volumes of activation are also indicative of this predissociation that include an important amount of contractive ordering. The presence of small amounts of triflic acid in the reaction medium accelerates the reaction to the value observed for N _imino-monodentate systems, indicating that the full opening of the chelate ring has taken place. For the badly oriented isomeric forms of the ligand in the chelated complex (Z), the cyclometallation process is even more slow and corresponds directly to the reorganization of the ligand to its cyclopalladation-active (E) conformation.

Full text of DOI:10.1039/b415613g

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Kinetico–mechanistic studies of C–H bond activation on new Pd
complexes containing N,Nꢀ-chelating ligands†
Isabelle Favier,a Montserrat Go´mez,a Jaume Granell,a Manuel Mart´ınez,a Xavier Solansb and
Merce` Font-Bard´ıab
a Departament de Qu´ımica Inorga`nica, Universitat de Barcelona, Mart´ı i Franque`s 1-11,
E-08028, Barcelona, Spain
b Departament de Mineralogia, Cristal·lografia i Dipo`sits Minerals, Universitat de Barcelona,
Mart´ı i Franque`s s/n, E-08028, Barcelona, Spain
Received 10th October 2004, Accepted 5th November 2004
First published as an Advance Article on the web 25th November 2004
The hybrid imine/amine palladium(II) coordination complexes [PdX2(j2-Nimino,Namino)] (X = Cl, AcO;
j2-Nimino,Namino = 4ClC6H4CHNCH2(CH2)nN(CH3)2, n = 1, 2) have been prepared in different isomeric forms which
2
=
include E/Z arrangement around the C N bond of the hybrid ligand and {Pd(j -Nimino,Namino)} ring conformation.
The crystal structures of four of them, E-1AcO, Z-1AcO, E-2AcO and E-2Cl, have been determined and the solution
behaviour in acetic acid, the common cyclometallating solvent, for all these systems studied. The complexes in acetic
acid solution are shown to maintain the structure determined by X-ray crystallography, as they do in deuterated
chloroform. Nevertheless, a partial opening equilibrium of the {Pd(j2-Nimino,Namino)} ring is observed by NMR
experiments. When the complexes are held in solution for longer periods the corresponding cyclometallated
derivatives, 1AcO-CM, 2AcO-CM, 1Cl-CM and 2Cl-CM, containing the {Pd(j2-C,Nimino)} palladacycle are
obtained, as characterized by 1H NMR spectroscopy. In these compounds the total opening of the Namino moiety of
the ligand has occurred. The C–H bond activation process has been studied kinetico–mechanistically at different
temperatures, pressures and acid concentrations; the results agree with the need of an opening of the chelate ring in
[PdX2(j2-Nimino,Namino)] prior to the proper cyclometallation reaction. The values of the enthalpies of activation are
higher than those observed for known N-monodentated cyclometallating ligands, as should correspond to the
contribution of a ligand dechelation pre-equilibrium. The entropies and volumes of activation are also indicative of
this predissociation that include an important amount of contractive ordering. The presence of small amounts of
triflic acid in the reaction medium accelerates the reaction to the value observed for Nimino-monodentate systems,
indicating that the full opening of the chelate ring has taken place. For the badly oriented isomeric forms of the
ligand in the chelated complex (Z), the cyclometallation process is even more slow and corresponds directly to the
reorganization of the ligand to its cyclopalladation-active (E) conformation.
Thorough and comprehensive kinetico–mechanistic studies of
Introduction
the reaction from these [MX2(j2-B,Bꢀ)] complexes to produce
C–H Bond activation via oxidative addition or electrophilic
substitution reactions to produce cyclometallated complexes are
the cyclometallated compounds is even more scarce, and they
have been only described for platinum derivatives.22
well known. These processes are very common with transition
In this paper we present the preparation of the new [PdX2(j2-
metals of group 10 elements containing N-donor ligands.1–4 As a
N
imino,Namino)] coordination complexes, indicated in Chart 1,
general rule, the preparative procedure for these cyclometallated
species is the reaction of a variety of metal salts (i.e. acetate,
chloride, etc.) in presence of the stoichiometric amounts of
the desired ligand which generally contains the heteroatom, B
(B = N, P, S or O).5–10 The process to obtain the final j2-C,B
metallacycles involves the precoordination of B to the metal
centre and further ligand topological arrangement allows for
the C–H bond activation.11,12 The mechanistic study of these
reactions has been carried out both in innocent and protic
solvents, and a good general picture of the reaction mechanism
is available.1,13
Not many of these processes are known with ligands contain-
ing two donor, B,Bꢀ, atoms that permit the preliminary isolation
of the corresponding non-organometallic complexes.14–20 This
is especially true for metals with a d8 configuration, which
favours the formation of four-coordinated planar complexes.
In this way the formation of bis(N-monodentated) coordina-
tion complexes, that do not lead to C–H bond activation, is
avoided.21 In this respect, only few metallation studies have been
reported from previously prepared coordination complexes.22,23
where j2-Nimino,Namino corresponds to the hybrid imine/amine
ligands indicated by 1 and 2. The crystal structures have been
determined for E-1AcO, Z-1AcO, E-2AcO and E-2Cl.
The solution behaviour of these complexes has also been
studied, both structurally and reactivity-wise, by NMR and
UV-Vis spectrometries in acetic acid, a common metallating
medium for this type of process,14 and the cyclometallation
process established. The flexibility of the {Pd(j2-Nimino,Namino)}
rings of the non-cyclometallated complexes enabled for the
existence of two isomeric forms in the starting E-2AcO and
E-2Cl compounds, while for the complexes derived from ligand
E-1 are isomerically unique. The final cyclometallated complexes
in acetic acid solution correspond in all cases to species that have
the j2-C,Nimino coordination skeleton. The amino nitrogen acts,
in some cases, in bridging form to another metal centre.14
The kinetico–mechanistic study of the reactions has been
carried out at variable temperature, pressure and acidic con-
ditions, and the corresponding relevant kinetic and activation
parameters have been determined. In all cases the processes
are one order of magnitude slower than those observed for
N-monodentate cyclometallating systems,13 and a dramatic
acceleration of the process in the presence of triflic acid is
detected. This acceleration has been ascribed to a preliminary
dechelating reaction producing N-monodentate complexes very
† Electronic supplementary information (ESI) available: Tables S1 and
S2: values of kobs for the systems studied. Figs. S1–3: Plots of kobs values.
See http://www.rsc.org/suppdata/dt/b4/b415613g/
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2
1 2 3
©

View Article Online
Scheme 2 Synthesis of Z-1AcO.
contrast, the compounds derived from ligand E-2 always exhibit
=
an E arrangement around the C N bond.
The serendipitous exchange of acetato by chloro ligands in
complexes Z-1AcO and E-2AcO occurred in heterogeneous
medium.24,25 When an ethyl acetate–methanol (4 : 1) solution
of the complexes were stirred with commercial acid alumina (ca.
0.2% of chloride), compounds Z-1Cl and E-2Cl were obtained
in quantitative yields (eqn. (1)).
(1)
Chart 1 Imino–amine ligands, E-1 and E-2, and the palladium
coordination complexes prepared.
Alternatively these compounds can also be obtained by column
chromatography over alumina of the corresponding acetato
derivatives (see Experimental section).
Formation of very small amounts of the corresponding
cyclometallated complexes was also observed during the process
(less than 10% for E-2Cl and less than 2% for Z-1Cl). For com-
pound E-1AcO, the presence of alumina leads to decomposition,
and the corresponding chloro complex, the cyclopalladated
derivative, and free p-chlorobenzaldehyde were observed in the
reaction medium. From these results, it seems clear that for
ligand 1, coordination complexes adopting Z conformation are
more robust than the analogous derivatives of type-E.
According to the analytical data, the compounds of stoi-
chiometry corresponding to 1Cl, 2Cl, 1AcO and 2AcO were ob-
tained as monometallic complexes of general formula [PdX2(L)]
(L = 1, 2; X = AcO, Cl), where L acts as a j2-Nimino,Namino
bidentated ligand. The complexes were fully characterized by the
usual techniques. IR spectra showed signals at 1530–1590 cm−1
similar to those already studied. The values determined for the
activation enthalpies, entropies and volumes agree extremely
well with a transition state involving this partial chelate opening
due to the protonation of the amino group in the presence of
acids.
Results
[PdX2(j2-Nimino,Namino)] complexes
Preparation. Ligands E-1 and E-2 were prepared accord-
ing to procedures previously described by condensation of
the aldehyde, 4-ClC6H4CHO, and the corresponding diamine,
H2NCH2(CH2)nNMe2 (n = 1, 2).14 Only the isomer having an E
conformation was observed in solution (as determined by NOE
contacts between imine and methylene protons), analogously
to other similar ligands.22 Neutral acetato (1AcO, 2AcO) and
chloro (1Cl, 2Cl) palladium complexes containing E-1 and E-2
imine/amines, were obtained in toluene solution from palladium
acetate and bis(benzonitrile)dichloropalladium, respectively, as
starting materials (Scheme 1). All these complexes show an E
conformation of the ligand, as observed for the free imine/amine
ligands.
=
assigned to stretching of the C N bond. Structural studies were
carried out, both in solid state (by X-ray diffraction) and in
solution (by NMR spectrometry).
X-Ray structures. Suitable single crystals for X-ray diffrac-
tion measurements were obtained by slow diffusion of diethyl
ether into dichloromethane solutions of E-1AcO, Z-1AcO, E-
2AcO and E-2Cl. Figs. 1 and 2 and Table 1 collect the structures
and relevant information of these complexes.
In these structures, the palladium atom has a distorted square-
planar coordination formed by two cis-oriented nitrogen and
two oxygen (E-1AcO, Z-1AcO, E-2AcO) or chlorine atoms (E-
2Cl). The ring of the complexes are nearly coplanar (torsion
However, at temperatures higher than 10 ◦C, the acetato
complex 1AcO shows important amounts of the species having
the ligand in a Z conformation; Z-1AcO can then be obtained
as the major component at room temperature (Scheme 2). In
Scheme 1 Synthesis of type-E palladium complexes: E-1AcO, E-1Cl, E-2AcO and E-2Cl.
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2
1 2 4

View Article Online
◦
˚
Table 1 Selected bond lengths (A) and angles ( ) for E-1AcO, Z-1AcO,
E-2AcO and E-2Cl (esds in parentheses)
E-1AcO
Z-1AcO
E-2AcO
E-2Cl
Pd–N(1)
Pd–N(2)
Pd–O(1)
Pd–O(3)
Pd–Cl(2)
Pd–Cl(3)
2.044(4)
2.006(2)
2.010(2)
2.006(2)
—
2.050(2)
2.007(2)
2.0142(19)
2.029(2)
—
1.991(2)
2.0683(19)
2.0224(17)
2.0235(19)
—
2.010(3)
2.099(4)
—
—
2.3021(14)
2.2922(16)
—
—
—
N(1)–Pd–N(2)
O(1)–Pd–O(3)
Cl(2)–Pd–Cl(3)
83.77(13) 84.06(9)
89.75(11) 91.26(9)
87.53(8)
84.21(8)
—
87.30(15)
—
88.66(6)
—
—
C(8)–N(1) 2.8◦, for E-1AcO and Z-1AcO, respectively). The
substituent on the iminic nitrogen atom occupies a pseudo-
equatorial position relative to the methyl groups of the dimethy-
lamino moiety for both structures. Another relevant feature is
=
the non-coplanarity between the imino (C N) and the aryl
groups (torsion angle N(2)–C(3)–C(4)–C(5) 32.2◦ and N(2)–
C(7)–C(6)–C(5) 32.8◦ for E-1AcO and Z-1AcO, respectively).
For complexes containing ligand 2, E-2AcO and E-2Cl, the
six-membered metallacycle shows a chair arrangement, with the
substituent on the iminic nitrogen atom occupies a pseudo-axial
position. In contrast to E-1AcO, the aryl and iminic groups are
nearly coplanar (torsion angles N(1)–C(7)–C(5)–C(6) 6.2◦ for
E-2AcO and N(1)–C(7)–C(4)–C(5) 5.6◦ for E-2Cl).
Fig. 1 View of the molecular structures of E-1AcO and Z-1AcO.
Hydrogen atoms (except iminic hydrogen) have been omitted for clarity.
Solution behaviour. The complexes containing ligand E-1
(E-1AcO, Z-1AcO, E-1Cl and Z-1Cl) exhibited only one isomer
in chloroform solution as observed by H NMR spectroscopy.
1
However, compounds with ligand E-2 (E-2AcO and E-2Cl)
present two isomeric forms in chloroform solution. The two
isomers detected for complexes E-2AcO and E-2Cl appear in an
isomeric ratio of ca. (4–6) : 1, constant within the temperature
range studied (323–273 K in CDCl3). NOE contacts revealed
that all the isomers possess an E conformation, consequently
they have to be probably based on different metallacycle
conformations having an exchange energy high enough to be
distinguished in the NMR time scale. The major component of
the E-2AcO and E-2Cl mixture (from now on E-2AcO(M) and
E-2Cl(M)) could be obtained as the sole compound when the
mixture of isomers (E-2AcO(M) + E-2AcO(m) or E-2Cl(M) +
E-2Cl(m)) was purified by flash column chromatography over
acid silica or alumina, for the acetato or chloro complexes,
respectively, using ethyl acetate–methanol (4 : 1) as eluent. Under
these conditions, the minor isomer is not stable as such, and the
corresponding cyclometallated product is formed (see below in
the kinetics section).
Metallation reaction. Coordination complexes [PdX2(j2-
imino,Namino)] containing ligands 1 and 2 gave the corresponding
N
cyclometallated compounds when heated in acetic acid solutions
at 50–60 ◦C for ca. 4–6 h (Scheme 3). The results are analogous
Fig. 2 View of the molecular structures of E-2AcO and E-2Cl.
Hydrogen atoms (except iminic hydrogen) have been omitted for clarity.
angles O(1)–O(3)–N(2)–N(1) = 2.6, 4.0 and 2.4◦ for E-1AcO, Z-
1AcO and E-2AcO, respectively, and Cl(2)–Cl(3)–N(2)–N(1) =
1.2◦ for E-2Cl). The imine ligand exhibits the E conformation
for the two complexes containing ligand 2 (Fig. 2) and for those
with ligand 1, E and Z structures were determined (Fig. 1).
For type-1 complexes, the five-membered metallacycle has
a pseudo-chair conformation, with the palladium, the two
nitrogen and one of the methylenic carbon atoms being nearly
coplanar (torsion angle Pd–N(1)–C(1)–N(2) 1.2◦ and Pd–N(2)–
Scheme 3 Preparation of cyclometallated compounds 1AcO-CM,
1Cl-CM, 2AcO-CM and 2Cl-CM. Reagents and conditions: (i) AcOH,
60 ◦C, 5 h; (ii) solvent evaporation and precipitation with diethyl ether.
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2
1 2 5

View Article Online
from the preparation already described using palladium acetate
and free ligand as starting materials.14 1H NMR spectra in
CDCl3 of these complexes indicate a metallated structure where
the ligand acts as a j2-C,Nimino bidentated ligand. From the
data available,14 we suggest that the Namino donor functions can
act, in some cases, as a bridge to a second palladium atom.
The greater trans influence of the phenyl group when compared
with the anionic ligands in the starting material,11 justifies this
observation.
low due to the low solubility of the complexes in neat acetic acid,
no dependence of the value of the observed rate constant on the
palladium complex concentration was observed. Interestingly,
the reactions carried out with complexes Z-1AcO and Z-1Cl
showed observed rate constants dramatically smaller than those
found for their E counterparts (Chart 1). Furthermore, for Z-
1Cl, decomposition competes with the C–H bond activation
reaction; this fact prevented the determination of the activation
volume for the process.
When 1H NMR spectra of the acetato derivatives, 1AcO-CM
and 2AcO-CM, were recorded in deuterated acetic acid, three
different metallated species in solution are evident in relative
equilibrium ratios 1 : 1.7 : 4 and 1 : 2 : 5.7, respectively.
Phase sensitive 2D NOESY experiments show positive cross-
peaks connecting the methinic protons of these species, thereby
indicating that these compounds exchange in solution. The
methylamine signal of one of these species appears at d values
similar to those of the free ligand indicating so that the amine
group is not coordinated to palladium;15,26 a dinuclear structure
with bridging acetato ligand can be proposed for this compound.
The same signals of the other two equilibrium species are shifted
to higher frequencies than for the free ligand, indicating that
the amine nitrogen is probably coordinated to palladium. The
lability of Pd–acetate bonds in protic medium can explain the
equilibrium here reported.13,21
The cyclometallation reactions monitored from the crude
mixtures of E-2AcO and E-2Cl (see Metallation section),
indicated the existence of a fast process with very small spectral
amplitude variations. These changes are the typical for a
cyclometallation reaction, they occur at much shorter times,
and that had to be measured independently. The low solubility
of the complexes did not allow the evaluation of the observed
rate constants under the same experimental conditions than
those used for the main process; in these cases the methodology
indicated in the Experimental section, referring to the use of
concentrated stock solutions in CH2Cl2, had to be used. The
procedure used, together with the high pressure range needed
(producing the solidification of acetic acid at temperatures
below 25 ◦C) did not allow the determination of the activation
volume for the reaction. From the two consecutive processes
observed, the second one produces the same values of kobs
determined before, and was associated the metallation of the
major component, while the first one was associated with
the metallation of the minor component of the mixture. The
amplitude of the spectral changes agrees very well with the
mixture (M/m) ratio determined by 1H NMR spectrometry.
The proton NMR spectra in deuterated (d4) acetic acid ef-
fectively shows that E-2AcO(M)/E-2AcO(m) or E-2Cl(M)/E-
2Cl(m) compound ratio increases with time, due to the faster
cyclometallation reaction of the minor isomers, E-2AcO(m)
and E-2Cl(m). Table S1, ESI† collects all the values of kobs as
a function of the cyclometallating complex, temperature and
pressure. The concentration of acetic acid proved to be an
important factor for the value of the observed rate constant as
shown in Fig. S1, ESI†as already shown for similar palladium
cyclometallating systems studied.13,21,27
In contrast, the 1H NMR spectrum of 2Cl-CM in deuterated
acetic acid shows only one species. The spectral data indicate
that it corresponds to the same species than that seen in
deuterochloroform, no changes take place in the presence of
acid. It is also worth mentioning that there is a definitive absence
of palladium-coordinated acetato groups.
Kinetics. The reactions of all the non-metallated complexes
of ligands E-1 and E-2, kinetically monitored by UV-Vis
spectroscopy in neat acetic acid solution, proved to be the real
1
cyclometallation reactions as checked by H NMR evaluation
of the reaction mixture under the same time and concentration
conditions. Despite the fact that for similar reactions, carried
out on platinum complexes, a preliminary chloride substitution
process by acetate is observed, the chloro ligand remains in
the coordination sphere of the PdII, even after the C–H bond
activation process has occurred (see above). As already indi-
cated, the final reaction mixture corresponds to an equilibrium
mixture of complexes when the acetato compounds are used
as starting material, while only one species is detected in the
case of the chloro derivatives. In all cases the NMR pattern
of the aromatic region is indicative of the formation of a five-
membered cyclometallated ring. The reactions are very clean
under the conditions used in the study as can be seen in Fig. 3(a),
where sharp isosbestic points are envisaged. Although the
concentration of the palladium complexes had to be kept very
In the same context, and given the fact that the presence of
triflic acid produced an important acceleration in cyclometal-
lation reactions on {RhII}2 carboxylato complexes,5,28 the C–H
bond activation reactions of some of the complexes have also
been monitored in the presence of varying concentrations of
triflic acid in acetic acid solutions. For complexes E-1AcO,
E-1Cl and E-2AcO(M) the spectral changes correspond to
those expected for cyclometallation reactions, although the
starting spectrum does not correspond to the original palladium
complex. This is expected, taking into account that protonation
must occur (Fig. 3(b)). In these cases a dramatic acceleration of
Fig. 3 (a) UV-Vis spectral changes of the cyclometallation reaction of complex E-2AcO(M) in neat acetic acid solution at 40 ◦C an◦d [Pd] = 1 ×
10−4 M. (b) UV-Vis spectral changes of the cyclometallation reaction of complex E-2AcO(M) in triflic acid solutions in acetic acid at 40 C [HTrifl] =
0.012 M and [Pd] = 1 × 10−4 M.
1 2 6
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2

View Article Online
Table 2 Values of kinetic (calculated at 25 ◦C from Eyring plots) and activation parameters for the systems studied, as a function of the starting
coordination compound, in neat acetic acid
Complex
104kobs/s−1
DH=/kJ mol−1
DS=/J K−1 mol−1
DV=/cm3 mol−1
E-1AcO
E-1Cl
Z-1AcO
Z-1Cl
E-2AcO(M)
E-2Cl(M)
E-2AcO(m)
E-2Cl(m)
0.30
0.18
0.0026
0.0021
0.16
0.10
9.0
94
84
104
92
109
91
40
5
4
3
5
4
5
9
3
−16 15
−54 13
−12 2 (324 K)
−10 1(327 K)
−0 (348 K)
−22
9
−64 14
Not measureda
−21 2 (323 K)
−15 2 (329 K)
Not measuredb
Not measuredb
29 12
−35 15
−169 30
−179 11
2.7
40
a Decomposition is too fast under cyclometallation conditions. b The procedure needed, indicated in the text, does not allow for its measurement.
the reaction rate is observed, although further increase of the
acid concentration produces an asymptotic decrease (Fig S2,
ESI†). The reactions under these conditions showed a final drift
due to decomposition of the complexes under the high acidity
conditions, as monitored by 1H NMR. When this study is carried
out on the chloro complex E-2Cl, the spectral changes observed
are complex and no reliable rate constants could be determined;
probably for these complexes a preliminary reaction takes place
that enhances the decomposition of the complex. The same type
of decomposition process was observed for type-Z complexes of
ligand 1 (Chart 1) under these conditions. Table S2, ESI,†collects
all the values of kobs for the systems studied as a function of
palladium complex, acetic and triflic acid concentrations and
temperature.
all the C–H activation reaction processes of the 4-ClC6H4CHNR
groups (Fig. 5) indicates that the mechanism for the process is
the same for all the complexes. The transition state proposed
for such reactions is that indicated in Scheme 4, and it should
applicable to all the systems studied.
The fitting of the variation of the values of the rate constants
obtained at different temperatures and pressures to the standard
Eyring and lnk vs. P equations, produced the values of the
thermal and pressure activation parameters indicated in Table 2;
the values of the kinetic parameters calculated at 60 ◦C are also
collected for comparison. A representative sample of the fittings
of the determined values of kobs to the variations in temperature
and pressure is shown in Fig. S3 (ESI†) and Fig. 4.
Fig. 5 Isokinetic relationship for the C–H bond activation on PdII of
4ClC6H4CHNR containing 1 and 2 ligands in neat acetic acid solvent:
ꢂ corresponds to the acetato derivatives E-1AcO and E-2AcO, ꢁto the
chloro derivatives, E-1Cl and E-2Cl, and ꢀ to that of palladium acetate
with the equivalent monodentate imine ligand, 4ClC6H4CHNCH2Ph,
of ref. 13.
The acid concentration dependence found for the systems
studied and shown in Figs. S1 and S2 (ESI†) is revealing. A clear
indication of the possibility of a proton assistance facilitating
the exit of the fully protonated anionic (acetato or chloro) group
from the transition state, as observed in many other PdII and RhII
cyclometallating compounds, is envisaged.5,13,28 The increase of
the reaction rate constant in neat acetic acid is a scarce three-
fold difference from the value in dilute solutions in toluene, the
system being de facto buffered.5 The dependence observed does
not correspond with that observed in other systems (limiting
kinetics behaviour) and expected from a neat pre-equilibrium
protonation of the bound anionic X-ligand. When triflic acid
is added to the reaction system the acceleration is enormous,
and the initial UV-Vis spectrum for the C–H bond activation
reaction does not coincide with the observed in neat acetic acid
solution (Fig. 3). If we speculate that the addition of small
concentrations of triflic acid provokes the opening of the chelate
ring, via protonation of the Namino donor, the metallation process
should take place from the monodentate mode of the ligand
indicated in Scheme 4.
Fig. 4 Plot for the variation with pressure of the rate constants for the
cyclometallation reaction of compounds E-1AcO (ꢀ) and E-2AcO(M)
(᭹) in acetic acid solution.
Discussion
For the simple cyclometallation process of complexes with
the imine ligand in an E conformation, E-1AcO, E-1Cl, E-
2AcO(M), E-2AcO(m), E-2Cl(M) and E-2Cl(m), the values
determined for the thermal and pressure activation parameters
(Table 2) are within the margin expected, and similar to those
obtained for the cyclometallation reaction of a series of N-
monodentate imines in protic solvents.13,27 Nevertheless, the
processes are, except for the minor components of the crude E-
2AcO and E-2Cl mixtures, much slower than for the previously
published systems.13 The isokinetic relationship existing between
The rate constants obtained for the reaction at low concen-
trations of triflic acid for E-2AcO(M) (Table S2, ESI†) are
the same within experimental error than those observed for
the E-2AcO(m) isomer. These observations agree with a very
facile complete dechelation of the ligand with acid buffered
solutions (neat acetic acid) for E-2AcO(m), and a more difficult
one for compound E-2AcO(M), only taking place in strong
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2
1 2 7

View Article Online
Scheme 4 Proposed general mechanism for the cyclometallation process studied.
acidic solutions (triflic acid solutions). 1H NMR monitoring of
the reaction of E-2AcO(M) effectively indicates that, although
decomposition occurs, the initial complex undergoes a much
faster cyclometallation reaction in the presence of very small
quantities of triflic acid. The further decrease observed in
the rate constant with acid concentration, shown in Fig. S2
(ESI†), is more difficult to explain, and probably relates with
a decomposition process. In fact, as indicated in the Kinetics
section the reactions of E-2Cl(M) in triflic acid solutions lead
to decomposition. In this respect, the effect of added triflic acid
on the known cyclometallation reactions of palladium acetate
by the C6H5CHNCH2Ph imine was checked;13 in all cases the
spectral changes indicated decomposition of the ligand.
fact. The organizational increase on going to the transition state
has to be larger for the chloro complexes, given the absence
of possible charge delocalization in the ligand. This factor
further agrees with the non-involvement in the full process of the
acetato group occupying the dechelated position (Scheme 4) (see
below).
Contrarily to the entropies, the volumes of activation for the
systems studied do not show significant differences from the
values for the Nimino-monodentate system. The most negative
values are those for complexes with the less negative activation
entropy, indicating that, as indicated above, changes in solvent-
assisted hydrogen-bond interactions are not dominant in the
system.29–31 The charge separation produced in the intermediate
dechelated species (Scheme 4) must directly electrostrict enough
amounts of acetic acid molecules to compensate for the expected
dechelating positive contribution that would produce less nega-
tive values of DV=. If these molecules were previously ordered
by hydrogen bonding, the electrostrictive contraction would not
imply a negative entropic contribution, as indicated above. The
effect should be larger for the acetato complex, E-2AcO, more
prone to hydrogen bonding interactions with the solvent, than
for the equivalent chloro complex E-2Cl. Although this is the
trend observed, further studies are needed for a more thorough
explanation of the facts.
With respect to the cyclometallation reaction occurring
on complexes E-2AcO(m) and E-2Cl(m), the values of the
parameters indicated in Table 2 are extremely similar to those
for the cyclometallation of the related 4-ClC6H4CHNCH2Ph
monodentate imine. These values determined in neat acetic
acid are: DH= = 59 kJ mol−1, DS= = −99 J K−1mol−1 and
DV= = −17 cm3mol−1,13 practically the same than those for the
cyclometallation reaction of compound E-2AcO(m) (Fig. 5). It
is reasonable to conclude that the minor components of the
crude E-2AcO and E-2Cl mixtures are completely converted, in
acetic acid solution, to complexes with the hybrid ligand in a
The 1H NMR spectrum in deuterated (d4) acetic acid of E-2Cl
is even more revealing. Apart from the two inequivalent methyl
signals of the palladium coordinated amino moiety, a singlet in-
tegrating ca. 5–7% of the total aminomethyl groups is observed.
This signal is assigned to the methyl groups of protonated free
amino group, by comparison of the proton NMR spectrum
of the parent non-condensed amine, H2N(CH2)3N(CH3)2, in the
same medium. Being this so, the metallation process clearly must
take place via the mechanism proposed in Scheme 4. The j2-
N
imino,Namino complex is in a very little displaced pre-equilibrium
with the Nimino-monodentate complex, that further undergoes
the proper C–H activation reaction. This equilibrium is driven
completely to the non-chelated situation by strong acids, as indi-
cated above. As a result, the observed rate constants correspond
de facto to a product of equilibrium and kinetic constants, kobs
=
Kprotk. Thermal and pressure activation parameters correspond
to this product, and include the values of DH◦, DS◦ and DV◦
for the protonation/dechelation equilibrium constant.
The values of the activation entropy for the reaction of
complexes E-1AcO, E-1Cl, E-2AcO(M) and E-2Cl(M) are
less negative than that observed for the Nimino-monodentate 4-
ClC6H4CHNCH2Ph imine system, indicating that the biden-
tate systems must undergo a lesser organization on going to
the transition state. Taking into account the pre-equilibrium
involved in the process the results agree with the operation
of an ordered transition state (Scheme 4) with a positive DS◦
contribution from the dissociation of the amino group of the
ligand. Despite the creation of charges, neither electrostriction,
nor or solvent-assisted hydrogen bonding, seem to cancel the
entropy increase due to dechelation. Possibly the interaction
of the positive NMe2H+ with the negative Pd-bound acetato
group avoids any one of these ordering. The effect should clearly
involve a larger activation enthalpy, given the energetics involved
on a chelate bond dissociation, DH◦. Effectively, the values
collected in Table 2 are 30–40 kJ mol−1 larger than the obtained
for the Nimino-monodentate system.13
N
imino-monodentate mode. From there on, the cyclometallation
process has to show very similar characteristics to those observed
for the monodentate imine ligands mentioned; Scheme 4 shows
this process. If this assumption is true, the ratio observed for
the kobs(m)/kobs(M) ratio for both complexes is revealing. The
value indicates that the rate of C–H bond activation of the
complex having the ligand in the j2-Nimino,Namino form represents
only a 5–8% of that observed for the monodentated form of
the hybrid ligand. These values agree extremely well with the
NMR-estimated j1-Nimino/j2-Nimino,Namino ratio for E-2Cl, which
is a clear indicative of the different equilibrium composition
of the complexes in acetic acid medium. The same conclusions
could be drawn if the minor isomers of the E-2AcO and E-
2Cl crude mixtures correspond directly to monodentate systems
with a dangling Namino group, or to dinuclear systems with the
The differences between the values found for DS= between the
acetato and chloro complexes of the same imine ligand merit
some attention, given the fact the DS◦ contributions should be
equivalent for the two species. For complexes derived from both
ligands E-1 and E-2, the acetato compounds have more positive
activation entropies. The different nature of the X aniono ligand
in the transition state indicated in Scheme 4, easily explains this
N
imino,Namino donor ligand acting as a bridge between two Pd
1
centres. H NMR spectrometry does not allow for any further
distinction given the nature of the minor component and the
rapid cyclometallation reaction observed.
For the complexes where ligand 1 is in a Z arrangement
(Chart 1) within the non-metallated complex, an isomerization
1 2 8
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2

View Article Online
process to the E conformation has to take place before the C–H
bond activation occurs. Given the fact that the observed rate con-
stants collected in Table 2 are much smaller than those obtained
for the E counterparts, it is clear that the process measured
de facto as cyclometallation, corresponds to the simple Z to
E isomerization reaction, the C–H bond activation to produce
the cyclometallated complex not being the rate limiting process.
Although the differences in the thermal activation parameters
indicated in Table 2 are not very important, they depend on the
presence of acetato or chloro ligands in the coordination sphere
of PdII. Furthermore, the value determined for DV= definitively
differs from the values expected for a cyclometallation reaction.
These differences were not expected, but they have already been
found for similar platinum cyclometallation processes, where the
50 MHz) spectrometers in CDCl3 and with SiMe4 standard,
unless otherwise cited. IR spectra were recorded on a Nicolet
520 FT-IR, Nicolet 510 FT-IR and FT-IR Nicolet Impact
400 spectrometers. FAB mass chromatograms were obtained
on a Fisons V6-Quattro instrument. Elemental analyses were
carried out by the Servei de Recursos Cient´ıfics i Te`cnics de la
Universitat Rovira i Virgili with a Carlo Erba EA1108-elemental
analyzer.
Complexes
(E)-Bis(acetato){[Nꢀ -(4-chlorobenzylidene)-N,Nꢀ -dimethyl-
ethane-1,3-diamine]-N,Nꢀ}palladium(II), E-1AcO. 0.543
g
(2.4 × 10−3 mol) of Pd(AcO)2 were dissolved in 20◦cm3 of
toluene, the mixture was stirred during 30 min at −10 C, and
then 0.509 g (2.4 × 10−3 mol) of E-1 were added. A yellow
precipitate is immediately formed, the mixture was stirred for
a further 2 h and then filtered. The solid obtained was washed
with diethyl ether and dried under reduced pressure. Purification
was carried out by recrystallization in a mixture of CH2Cl2–
Et2O. Yield: 0.460 g (35%). Anal. Calc. for C15H21ClO4N2Pd
(435.2): C, 41.39; H, 4.86; N, 6.44%. Found: C, 41.05; H, 4.10;
N, 6.25%. MS (FAB) m/z 376.9 ([M − AcO]+). IR (KBr): 1630
involvement of the acetato ligand favouring the rotation of the
22
=
=
N CH bond has been established. For the rotation of C N
bond to take place, the loss of the double bond character for the
C(+)–N(−) tautomeric form has to be stabilized. The existence
of a well oriented dangling acetato-oxygen in the transition
state, capable of interaction with the positive charge situated
on the iminic carbon, should facilitate better than a palladium-
bonded chloro ligand the reorganization of the double bond.
The mentioned pseudo-equatorial position of the iminic carbon,
as determined in the solid state structure, agrees with this easy
interaction. This trend is the one observed in Table 2, where
the effect is seen to be mainly entropic, given the important
difference from −24 to −63 J K−1 mol−1 on going from the
acetato (Z-1AcO) to the chloro derivatives (Z-1Cl). Evidently
if the above mentioned effect is true it would be more feasible
in non-acidic media, and further studies in this field are being
carried out.
−1
1
=
(CH N); 1567 and 1412 (AcO) cm . H NMR (500 MHz) d
8.28 (d, 8.5 Hz, 2H), 8.03 (s, 1H), 7.47 (d, 8.5 Hz, 2H), 4.17 (m,
2H), 2.78 (2, 6H), 2.53 (m, 2H), 1.96 (s, 3H), 1.21 (s, 3H) ppm.
13C NMR (100.6 MHz) d 178.7 (C), 178.3 (C), 169.1 (CH),
139.9 (C), 132.5 (CH), 128.8 (CH), 128.0 (C), 65.4 (CH2), 63.9
(CH2), 51.3 (CH3), 48.2 (CH3), 23.6 (CH3), 22.1 (CH3) ppm.
(E)-Dichloro{[Nꢀ -(4-chlorobenzylidene)-N,Nꢀ -dimethyl-
ethane-1,3-diamine]-N,Nꢀ}palladium(II), E-1Cl. 0.078 g (2 ×
10−4 mol) of [PdCl2(PhCN)2] was dissolved in 5 cm3 of toluene
and 0.043 g (2 × 10−4 mol) of E-1 was then added. A yellow
precipitate was immediately observed from the brown solution.
It was filtered, washed with toluene and dried under vacuum.
The product was purified by dissolving it in dichloromethane,
filtered over Celite and the solvent removed to dryness under
vacuum. Yield 0.048 g (62%). Anal. Calc. for C11H15Cl3N2Pd
(388.0): C, 34.05; H, 3.90; N, 7.22%. Found: C, 35.51; H,
Conclusions
The cyclometallation processes occurring on the series of
compounds having chelate j2-Nimino,Namino ligands, depicted in
Chart 1, have been studied from a kinetico–mechanistic perspec-
tive in neat acetic acid solutions. The processes are seen to take
place with the intervention of a tetracentered transition state as
that established in previous studies. Nevertheless, the reaction of
this bidentate ligands is one order of magnitude slower than that
for their Nimino-monodentate counterparts, and the differences
are assigned to the chelate j2-Nimino,Namino nature of the ligand.
A pre-equilibrium dechelation process has been established as
needed for the C–H bond activation to occur. The dramatic
acceleration observed when strong acid is present in the medium
is related with the total opening of the j2-Nimino,Namino chelate.
The resulting species undergoes the C–H bond activation at
the same rate than that observed previously for monodentate
systems; activation parameters corroborate this fact.
Isolation of the Z badly oriented analogues of some of the
non-metallated compounds has also been achieved, and their
cyclometallation mechanism studied. The reaction is found to
be more than one order of magnitude slower, and it is associated
with the Z to E isomerization of the ligand geometry previous to
the C–H activation. Both the crystal structure determined and
the thermal and baric activation parameters prove the previously
suggested implication of the spectator ligand in the transition
=
4.94; N, 7.02%. IR (KBr): 1643 (CH N); 1098 and 1025
(amine) cm−1. 1H NMR (500 MHz) d 9.13 (d, 8.5 Hz, 2H), 8.09
(s, 1H), 7.60 (d, 8.5 Hz, 2H), 4.85 (m, 1H), 3.74 (m, 1H), 2.77
(s, 3H), 2.56 (s, 3H), 2.40 (m, 1H), 1.95 (m, 1H) ppm. 13C NMR
(100.6 MHz) d 170.0 (CH), 141.1 (C), 133.2 (CH), 129.8 (CH),
128.4 (C), 62.0 (CH2), 61.0 (CH2), 51.8 (CH3) ppm.
(Z)-Bis(acetato){[Nꢀ -(4-chlorobenzylidene)-N,Nꢀ -dimethyl-
ethane-1,3-diamine]-N,Nꢀ}palladium(II), Z-1AcO. 0.530
g
(2.4 × 10−3 mol) of Pd(AcO)2 were dissolved in 20 cm3 of
toluene and 3 cm3 of a 0.9 M solution of E-1 in toluene (2.7 ×
10−3 mol) were added at room temperature. The mixture was
stirred overnight and then filtered. The pale yellow solid was
washed with diethyl ether and dried under reduced pressure.
Yield: 0.400 g (39%). Anal. Calc. for C15H21ClO4N2Pd·1.7H2O
(465.6): C, 38.66; H, 5.24; N, 6.01%. Found: C, 38.65; H, 5.72;
=
N, 6.27%. IR (KBr): 1637 (CH N); 1604, 1394 (AcO); 1104
1
and 1019 (amine) cm−1. H NMR (500 MHz) d 8.09 (bs, 1H),
7.46 (d, 8.5 Hz, 2H), 7.37 (d, 8.5 Hz, 2H), 3.84 (dd, 6.0 Hz,
2.0 Hz, 2H), 2.75 (s, 6H), 2.61 (t, 6.0 Hz, 2H), 1.98 (s, 3H), 1.94
(s, 3H) ppm. 13C NMR (100.6 MHz) d 178.7 (C), 178.4 (C),
169.2 (CH), 139.2 (C), 130.9 (CH), 129.8 (CH), 128.6 (C), 64.7
(CH2), 54.7 (CH2), 50.7 (CH3), 23.6 (CH3), 23.5 (CH3) ppm.
=
state for the C N bond rotation.
Experimental
General
(Z)-Dichloro{[Nꢀ -(4-chlorobenzylidene)-N,Nꢀ -dimethyl-
ethane-1,3-diamine]-N,Nꢀ}palladium(II), Z-1Cl. Complex Z-
1Cl was obtained by column chromatography of Z-1AcO
(0.086 g) using acid alumina (% Cl− < 0.2) and a mixture of
AcOEt–MeOH = 4 : 1 as eluent. Yield: 0.071 g (92%). Anal.
Calc. for C11H15Cl3N2Pd (388.0): C, 34.05; H, 3.90; N, 7.22%.
Alumina was purchased from Merck. [PdCl2(PhCN)2] was
prepared as previously described.32 Ligands E-1 and E-2
were prepared by the methodology previously described.14 The
solvents were purified by standard procedures and distilled
under nitrogen. NMR spectra were recorded on Varian XL-
500 (1H), Bruker DRX-500 (1H), Bruker DRX-250 (1H), Varian
Mercury-400 (1H, 13C 100.6 MHz) and Varian Gemini (13C,
=
Found: C, 33.61; H, 4.61; N, 8.06%. IR (KBr): 1637 (CH N);
1
and 1104, 1019 (amine) cm−1. H NMR (500 MHz) d 9.03 (bs,
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2
1 2 9

View Article Online
1H), 7.40 (m, 4H), 3.97 (m, 2H), 2.96 (s, 6H), 2.63 (t, 6.0 Hz,
2H) ppm. 13C NMR (100.6 MHz) d 170.5 (CH), 139.0 (C),
131.1 (CH), 129.7 (CH), 129.1 (C), 65.7 (CH2), 55.5 (CH2), 51.9
(CH3) ppm.
plate detector (E-1AcO, Z-1AcO, E-2AcO) and on a Enraf-
Nonius CAD4 four-circle (E-2Cl) diffractometers. Intensities
were collected with graphite monochromatized Mo-Ka ra-
˚
diation (k = 0.71073 A), operating at room temperature.
Lorentz-polarization but no absorption corrections were made.
The structures were solved by direct methods using SHELXS
computer program33 and refined by full-matrix least-squares
method on with SHELX97 computer program.34 Crystal data
are summarized in Table 3.
(E)-Bis(acetato)-{[Nꢀ -(4-chlorobenzylidene)-N,Nꢀ -dimethyl-
propane-1,3-diamine]-N,Nꢀ}palladium(II), E-2AcO. 0.264
g
(1.3 × 10−3 mol) of Pd(AcO)2 were dissolved in 20 cm3 of
toluene and 0.264 g (1.3 × 10−3 mol) of E-2 were then added
at room temperature. The mixture was stirred for 3 h and
then filtered. The pale yellow solid was washed with diethyl
ether and dried under reduced pressure. Yield: 0.326 g (62%).
Anal. Calc. for C16H23ClO4N2Pd (449.2): C, 42.78; H, 5.16; N,
E-1AcO. The structure was determined using 5901 reflec-
tions. The function minimized was Rw[(|Fo| − |Fc| )2] where
2
2
2
w = [r2(I) + (0.0514P)2 + 1.7165P]−1 and P = (|Fo|
+
2|Fc| )/3; f , f ꢀꢀ and f ꢀꢀ were taken from International Tables
of X ray Crystallography.35 Five hydrogen atoms were located
from a difference synthesis and refined with an overall isotropic
temperature factor. 21 hydrogen atoms were computed and
refined using a riding model, with an isotropic temperature
factor equal to 1.2 times the equivalent temperature factor of
the atom which are linked.
6.24%. Found: C, 42.13; H, 5.36; N, 6.92%. MS (FAB) m/z
2
+
=
391.0 ([M–AcO] ). IR (KBr): 1620 (CH N); 1560 and 1405
(AcO) cm−1. 1H NMR Major isomer (80%): (500 MHz) d 9.42
(d, 8.5 Hz, 2H), 7.88 (s, 1H), 7.65 (d, 8.5 Hz, 2H), 5.21 (m, 1H),
3.56 (m, 2H), 2.44 (s, 3H), 2.34 (s, 3H), 2.20 (m, 2H), 1.95 (s,
3H), 1.70 (m, 1H), 1.54 (s, 3H) ppm. 13C NMR (100.6 MHz)
d 178.8 (C), 178.2 (C), 168.6 (CH), 140.5 (C), 132.6 (CH),
129.7 (CH), 129.1 (C), 62.9 (CH2), 62.8 (CH2), 52.7 (CH3), 49.7
(CH3), 27.2 (CH2), 23.3 (CH3), 22.5 (CH3) ppm. Minor isomer
(20%): (500 MHz, selected signals) d 9.32 (d, 8.5 Hz, 2H), 8.12
(s, 1H), 7.63 (d, 8.5 Hz, 2H), 4.85 (m, 1H), 3.76 (m, 1H), 3.56
(m), 2.52 (s, 3H), 2.35 (s, 3H), 1.98 (s, 3H) ppm.
Z-1AcO. The structure was determined using 6183 reflec-
2
2
tions. The function minimized was Rw[(|Fo| − |Fc| )2] where
2
2
w = [r2(I) + (0.0944P)2]−1 and P = (|Fo| + 2|Fc| )/3; f , f ꢀꢀ and f ꢀꢀ
were taken from International Tables of X ray Crystallography.35
Three hydrogen atoms were located from a difference synthesis
and refined with an overall isotropic temperature factor. 20
hydrogen atoms were computed and refined using a riding
model, with an isotropic temperature factor equal to 1.2 times
the equivalent temperature factor of the atom which are linked.
(E)-Dichloro{[Nꢀ-(4-chlorobenzylidene)-N,Nꢀ-dimethylpropa-
ne-1,3-diamine]-N,Nꢀ}palladium(II), E-2Cl. 0.5
g
(1.3
×
10−3 mol) of [PdCl2(PhCN)2] was dissolved in 20 cm3 of toluene
and 0.250 g (1.1 × 10−3 mol) of E-2 was added. A yellow
precipitate was immediately formed from the brown solution. It
was filtered, washed with toluene and the dried under vacuum.
The product was purified by dissolving it in dichloromethane,
filtering over Celite and drying under vacuum. Yield: 0.468 g
(100%). The compound consists in an isomer mixture as that
indicated above for E-2AcO. Selected 1H NMR signals for the
minor isomer, E-2Cl(m), are: (500 MHz) d 9.0 (d, 8.5 Hz, 2H),
8.22 (s, 1H), 7.65 (d, 8.5 Hz, 2H) ppm.
E-2AcO. The structure was determined using 5288 reflec-
2
2
tions. The function minimized was Rw[(|Fo| − |Fc| )2] where
2
2
w = [r2(I) + (0.0574P)2 + 0.289P]−1 and P = (|Fo| + 2|Fc| )/3;
f , f ꢀꢀ and f ꢀꢀ were taken from International Tables of X ray
Crystallography.35 Seven hydrogen atoms were located from
a difference synthesis and refined with an overall isotropic
temperature factor. 17 hydrogen atoms were computed and
refined using a riding model, with an isotropic temperature
factor equal to 1.2 times the equivalent temperature factor of
the atom which are linked.
The major isomer component for the E-2Cl isomer mixture
was obtained by column chromatography of E-2AcO (0.150 g)
using acid alumina (%Cl− < 0.2) and a mixture of AcOEt–
MeOH = 4 : 1 as eluent. Yield: 0.113 g (93%). Anal. Calc. for
C12H17Cl3N2Pd (402.03): C, 35.85; H, 4.26; N, 6.97%. Found:
E-2Cl. The structure was determined using 4478 reflections.
2
2
The function minimized was Rw[(|Fo| − |Fc| )2] where w =
=
C, 35.51; H, 4.66; N, 7.07%. IR (KBr): 1590 (CH N); 1105,
1045 (amine) cm−1. MS (FAB) m/z 331 ([M − 2Cl]+). 1H NMR
(500 MHz) d 9.14 (d, 8.5 Hz, 2H), 8.11 (s, 1H), 7.59 (d, 8.5 Hz,
1H), 4.90 (m, 1H), 3.77 (q, 5.8 Hz, 1H), 3.0 (m, 1H), 2.77 (s,
3H), 2.57 (s, 3H), 2.5 (m, 1H), 2.4 (ddd, 13 Hz, 7.5 Hz, 3 Hz,
1H), 1.96 (m, 1H). 13C NMR (100.6 MHz) d 169.9 (CH), 141.2
(C), 133.2 (CH), 129.8 (CH), 129.3 (C), 62.0 (CH2), 61.1 (CH2),
53.6 (CH3), 51.7 (CH3), 26.2 (CH2) ppm.
2
2
[r2(I) + (0.304P)2]−1 and P = (|Fo| + 2|Fc| )/3; f , f ꢀꢀ and f ꢀꢀ were
taken from International Tables of X ray Crystallography.35 Four
hydrogen atoms were located from a difference synthesis and
refined with an overall isotropic temperature factor. 13 hydrogen
atoms were computed and refined using a riding model, with an
isotropic temperature factor equal to 1.2 times the equivalent
temperature factor of the atom which are linked.
Cyclometallated complexes, 1AcO-CM, 1Cl-CM, 2AcO-CM
and 2Cl-CM
CCDC reference numbers 238987 (E-1AcO), 238988 (E-
2AcO), 238989 (E-2Cl) and 238990 (Z-1AcO).
See http://www.rsc.org/suppdata/dt/b4/b415613g/ for cry-
stallographic data in CIF or other electronic format.
These were prepared from the corresponding coordination
complexes by the same procedure further described. 5 × 10−5 mol
of the appropriate coordination complex (E-1AcO, 21.5 ×
10−3 g; E-1Cl, 19.4 × 10−3 g; Z-1AcO, 21.5 × 10−3 g; Z-1Cl,
19.4 × 10−3 g; E-2AcO, 22.0 × 10−3 g; E-2Cl, 20.0 × 10−3 g)
were dissolved in 10 cm3 of acetic acid and the solution heated
Kinetics
The reactions at atmospheric pressure were followed by UV-
Vis spectroscopy in the full 750–300 nm range on a HP8542A
or Cary 50 instruments equipped with a multicell transport,
thermostated ( 0.1 EC) with a circulation bath. Observed rate
constants were derived from the absorbance vs. time traces at
the wavelengths where a maximum increase and/or decrease
of absorbance was observed. No dependence of the values on
the selected wavelengths was detected, as expected for reactions
where a good retention of isosbestic points is observed (Fig. 3).
For runs at elevated pressure, a previously described pressurizing
system and high-pressure cell were used.13,22 In these cases
the absorbance vs. time traces were recorded on a Cary 50
◦
at 60 C for 5–6 h. The solvent was then reduced to ca. 1 cm3
under vacuum. The corresponding cyclometallated compound
was precipitated by addition of diethyl ether. The products were
obtained in quantitative yields. Analytical data described in
ref. 14.
Crystallography
Orange crystals of E-1AcO, Z-1AcO, E-2AcO and E-2Cl
were grown by diffusing diethyl ether into a dichloromethane
solution. They were mounted on a MAR345 with image
1 3 0
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2

View Article Online
instrument at a fixed wavelength chosen from the atmospheric
pressure experiments; alternatively a J&M TIDAS instrument
was also used in the full wavelength range.
The general kinetic technique is that previously described;5,13,31
solutions for the kinetic runs were prepared by dissolving the
calculated amounts of the palladium compounds in acetic acid.
In all cases no dependence on the concentration of palladium
was detected and it was kept in the (2–5) × 10−4 M margin,
given the low solubility of the complexes in acetic acid. In
some cases concentrated solutions of the palladium complexes in
CH2Cl2 were prepared, and 1–2% was added to prethermostated
acetic acid. Rate constants were derived from exponential least-
square fitting by the standard routines. Tables S1 and S2
(ESI†) collect all the obtained kobs values for all the complexes
studied as a function of the metallating starting complex, added
acid, temperature and pressure. Least-square errors for the rate
constants were always in the range of 10–15% of the calculated
value. All post-run fittings to rate laws were done by standard
fitting programs commercially available.
Acknowledgements
We would like to thank the Ministerio de Ciencia y Tecnolog´ıa
(BQU2001-3358, CTQ2004-00954/BQU and BQU2003-00906)
and the Generalitat de Catalunya for financial support.
References
1 A. D. Ryabov, Chem. Rev., 1990, 90, 403–424.
2 L. M. Rendina and R. J. Puddephatt, Chem. Rev., 1997, 97, 1735–
1754.
3 A. J. Canty and G. van Koten, Acc. Chem. Res., 1995, 28, 406–413.
4 G. R. Newkome, W. E. Puckett, W. K. Gupta and G. E. Kiefer, Chem.
Rev., 1986, 86, 451–489.
5 G. Gonza´lez, P. Lahuerta, M. Mart´ınez, E. Peris and M. J. Sanau´,
Chem. Soc., Dalton Trans., 1994, 545–550.
6 J. Vicente, I. Saura-Llamas, M. G. Palin, P. G. Jones and M. C.
Ram´ırez de Arellano, Organometallics, 1997, 16, 826–833.
7 B. J. O’Keefe and P. J. Steel, Organometallics, 2003, 22, 1281–1292.
8 C. Lo´pez, A. Caubet, S. Pe´rez, X. Solans and M. Font-Bard´ıa, Chem.
Commun., 2004, 540–541.
9 J. Albert, M. Cadena, A. Gonza´lez, J. Granell, X. Solans and M.
Font-Bard´ıa, Chem. Commun., 2003, 528–529.
10 J. M. Valk, R. van Belzen, J. Boersma, A. L. Speck and G. J. van
Koten, Chem. Soc., Dalton Trans., 1994, 2293–2302.
11 M. L. Tobe and J. Burgess, Inorganic Reaction Mechanisms, Addison
Wesley Longman, Harlow, 1999.
12 R. G. Wilkins, Kinetics and Mechanisms of Reactions of Transition
Metal Complexes, VCH, Weinheim, 1991.
13 M. Go´mez, J. Granell and M. J. Mart´ınez, Chem. Soc., Dalton Trans.,
1998, 37–43.
14 M. Go´mez, J. Granell and M. Mart´ınez, Inorg. Chem. Commun.,
2002, 5, 67–70.
15 A. Ferna´ndez, P. Uria, J. J. Ferna´ndez, M. Lopez Torres, A. Sua´rez,
B. Va´quez Garc´ıa, M. T. Pereira and J. M. J. Vila, Organomet. Chem.,
2001, 620, 8–19.
16 J. Mart´ınez, M. T. Pereira, I. Buceta, G. Alberdi, A. Amoedo, J. J.
Ferna´ndez, M. Lo´pez-Torres and J. M. Vila, Organometallics, 2003,
22, 5581–5584.
17 S. Stoccoro, B. Soro, A. Minghetti, A. Zucca and M. A. J. Cinellu,
Organomet. Chem., 2003, 679, 1–9.
18 L. Xu, Q. Shi, X. Li, X. Jia, X. Huang, R. Wang, Z. Zhou, Z. Lin
and A. S. Chan, Chem. Commun., 2003, 1666–1667.
19 A. Zucca, M. A. Cinellu, M. V. Pinna, S. Stoccoro, A. Minghetti, M.
MAnasero and M. Sanso´n, Organometallics, 2000, 19, 4295–4304.
20 M. Crespo, M. Font-Bard´ıa and X. Solans, Organometallics, 2004,
23, 1708–1713.
21 M. Go´mez, J. Granell and M. Mart´ınez, Organometallics, 1997, 16,
2539–2546.
22 M. Crespo, M. Font-Bard´ıa, J. Granell, M. Mart´ınez and X. Solans,
Dalton Trans., 2003, 3763–3769.
23 M. Crespo, M. Mart´ınez and E. J. Pablo, Chem. Soc., Dalton Trans.,
1997, 1231–1235.
24 J. G. Dorsey, W. T. Cooper, B. A. Siles, J. P. Foley and H. G. Barth,
Anal. Chem., 1998, 70, 591–644.
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2
1 3 1

View Article Online
25 Y. Q. Feng, M. Shibukawa and K. Oguma, Anal. Sci., 1997, 13,
217–223.
26 J. M. Vila, M. Gayoso, M. T. Pereira, T. Lo´pez, J. J. Ferna´ndez, A.
Ferna´ndez and J. M. J. Ortigueira, Organomet. Chem., 1997, 532,
171–180.
30 P. V. Bernhardt, F. Bozoglia´n, B. P. Macpherson, M. Mart´ınez, G.
Gonza´lez and B. Sienra, Eur. J. Inorg. Chem., 2003, 2512–2518.
31 P. V. Bernhardt, C. Gallego, M. Mart´ınez and T. Parella, Inorg. Chem.,
2002, 41, 1747–1754.
32 G. K. Anderson and M. Lin, Inorg. Synth., 1990, 28, 60–63.
33 G. M. Sheldrick, SHELXS. A computer program for determination of
crystal structure, 1997, Germany, University of Go¨ttingen.
34 G. M. Sheldrick, SHELX97. A computer program for determination
of crystal structure, 1997, Germany, University of Go¨ttingen.
35 International tables of X ray crystallography, Kynoch Press, 1974, vol.
IV, pp. 99–100 and 149.
27 M. Go´mez, J. Granell and M. Mart´ınez, Eur. J. Inorg. Chem., 2000,
217–224.
28 F. Estevan, G. Gonza´lez, P. Lahuerta, M. Mart´ınez, E. Peris and R. J.
van Eldik, Chem. Soc., Dalton Trans., 1996, 1045–1050.
29 M. Mart´ınez, M. Pitarque and R. van Eldik, Inorg. Chim. Acta, 1997,
256, 51–59.
1 3 2
D a l t o n T r a n s . , 2 0 0 5 , 1 2 3 – 1 3 2