Cyclometallation on platinum(II) complexes; the role of the solvent and added base donor capability on the reaction mechanisms

Article abstract of DOI:10.1039/b306243k

The reaction of PhCH=NCH₂CH₂NMe₂ with [PtCl₂S₂] (S = SMe₂ or dmso) produces [PtCl₂(PhCH=NCH₂CH₂-NMe₂)] (1) in which the imine acts as a bidentate [N,N′] ligand. The cyclometallated [C,N,N′] platinum(II) complex [PtCl(C₆H₄CH=NCH₂CH₂NMe₂) ] (2) has been obtained by refluxing a methanol solution of complex 1 in the presence of base; both complexes 1 and 2 have been characterized crystallographically. The cyclometallation process has been studied kinetically at variable temperature and pressure as a function of the base added in methanol and ethanol solutions. In all cases, the actual C-H bond activation is preceded by the Z to E isomerization of the imine; detection of this species by proton NMR has also been achieved. The kinetic, thermal and baric activation parameters associated with two processes are indicative of a different reaction mechanism operating for the reactions when the "external" base added (NaCH₃CO₂) can, in fact, react with the starting material to form a new acetato complex which allows for the reactions to be intramolecular.

Full text of DOI:10.1039/b306243k

View Article Online / Journal Homepage / Table of Contents for this issue
Cyclometallation on platinum(II) complexes; the role of the solvent
and added base donor capability on the reaction mechanisms†
Margarita Crespo,a Mercè Font-Bardia,b Jaume Granell,a Manuel Martíneza and
Xavier Solansb
a Departament de Química Inorgànica, Universitat de Barcelona, Martí i Franquès, 1-11,
E-08028 Barcelona, Spain. E-mail: manel.martinez@qi.ub.es
b Departament de Cristalؒlografia, Mineralogia i Dipòsits Minerals, Universitat de Barcelona,
Martí i Franquès s/n, E-08028 Barcelona, Spain
Received 4th June 2003, Accepted 25th July 2003
First published as an Advance Article on the web 26th August 2003
The reaction of PhCH᎐NCH CH NMe with [PtCl S ] (S = SMe or dmso) produces [PtCl (PhCH᎐NCH CH -
᎐
᎐
2
2
2
2
2
2
2
2
2
NMe2)] (1) in which the imine acts as a bidentate [N,NЈ] ligand. The cyclometallated [C,N,NЈ] platinum() complex
[PtCl(C H CH᎐NCH CH NMe )] (2) has been obtained by refluxing a methanol solution of complex 1 in the
᎐
6
4
2
2
2
presence of base; both complexes 1 and 2 have been characterized crystallographically. The cyclometallation process
has been studied kinetically at variable temperature and pressure as a function of the base added in methanol and
ethanol solutions. In all cases, the actual C–H bond activation is preceded by the Z to E isomerization of the
imine; detection of this species by proton NMR has also been achieved. The kinetic, thermal and baric activation
parameters associated with the two processes are indicative of a different reaction mechanism operating for the
reactions when the “external” base added (NaCH3CO2) can, in fact, react with the starting material to form a new
acetato complex which allows for the reactions to be intramolecular.
allow the investigation at variable temperatures and pressures
Introduction
of the reaction mechanism operating for the intramolecular
The reaction mechanisms leading to electrophilic cyclo-
C–H bond activation process. Given the important charge
metallation reactions on Pd() complexes have been extensively
separation involved in the metallation process, the study of the
mechanism of the C–H bond activation on these complexes has
studied.1–5 Comparison with similar mechanisms for cyclo-
metallation reactions on Rh() has also been reported,6–8 and
been monitored as a function of the polarity of the solvent
the conclusions drawn from the results obtained for both sys-
used.29 Furthermore, the study of the differences between the
tems have been found equivalent. On the other hand, the study
processes carried out in the presence of coordinating and non-
of the reaction mechanisms operating for metallations on Pt()
coordinating bases has also been pursued.8 In all cases the C–H
complexes is dominated by oxidative addition–reductive elim-
bond activation process is always preceded by a relatively fast
ination processes.9–11 In this respect, we have been involved in
isomerisation reaction of the imine ligand from an unreactive Z
the full mechanistic study of such reactions.12–14
conformation to a better oriented E conformation.30 The pres-
Recent reports on cycloplatination reactions proved that
ence of acetate ions in the reaction medium has been shown to
cis-[PtCl2(dmso)2] is a reasonable metalating agent leading to
higher yields and cleaner reactions than the more classical
lead to the formation of a much more reactive acetato complex
that enables a complete intramolecular reaction process.
K2[PtCl4] starting material. Successful reactions have been
reported for the synthesis of mono- and bis-cycloplatinated
complexes containing aryl or ferrocenyl [C,N] ligands,15–21 while
in some cases deoxygenation of dimethyl sulfoxide affords
platinum() dimethyl sulfide complexes as final products.22,23
Cycloplatination of aryl oximes using cis-[PtCl2(dmso)2] as well
as [PtCl2(SMe2)2] has been recently reported.24
Results
Compounds
Ligand PhCH᎐NCH CH NMe (L) was prepared as previously
᎐
2
2
2
reported28 and characterized by 1H NMR in CDCl3. According
to NMR data only one isomer was present in solution and the
expected E conformation was confirmed by a NOESY experi-
ment in which cross-peak signals between imine and methylene
protons (δ 8.32 and 3.75 ppm respectively) were observed.
When the ligand was treated with [PtCl2(SMe2)2], obtained as
a mixture of cis and trans isomers, in refluxing methanol,
Following our interest in cycloplatination of potentially
terdentate ligands [C,N,NЈ]25–28 we now report the results
obtained for the reactions of bis-sulfoxide and -sulfide dichloro-
platinum() complexes with PhCH᎐NCH CH NMe . An
additional interest in this system arises from the fact that the
presence of two nitrogen atoms would facilitate the isolation
᎐
2
2
2
of a coordination compound [PtCl (PhCH᎐NCH CH NMe )]
complex cis-[PtCl (PhCH᎐NCH CH NMe )] (1), depicted in
᎐
2
2
2
2
᎐
2
2
2
2
analogous to N-coordinated compounds trans-[PtCl2-
L(dmso)]15–21 which have been proposed as intermediates from
starting materials to cyclometallated complexes. This would
Chart 1, was formed as a white solid. For comparison, cis-
[PtCl2(dmso)2] was also tested as platinum substrate and analo-
gous results were obtained. In spite of the fact that sulfides are
better ligands for platinum() than sulfoxides,22,23 and that the
dimethyl sulfide complex consists of a mixture of isomers,5 the
replacement of the sulfide ligands by both nitrogen atoms of
ligand L is easily achieved.
† Electronic supplementary information (ESI) available: Listings of
selected bond lengths (Å) and angles (Њ) for complex 2 with estimated
standard deviations, crystallographic and refinement data for complex 2
and listing of observed rate constants for the cyclometallation reaction
of complex 1. Plot of the molecular structure of complex 2 and that of
the observed rate constants corresponding to the first reaction step
Our next goal was to test whether it is possible to obtain the
cyclometallated complex [PtCl(C H CH᎐NCH CH NMe )] (2)
᎐
6
4
2
2
2
from complex 1. The process involves the activation of a C–H
bond and the release of HCl via an electrophilic substitution on
the aromatic carbon. The most widely used conditions for
1
determined when proton sponge or neat solvent were used. H NMR
spectra of the mixture of Z and E isomers of complex 1. See http://
www.rsc.org/suppdata/dt/b3/b306243k/
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 3
D a l t o n T r a n s . , 2 0 0 3 , 3 7 6 3 – 3 7 6 9
3763

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for complex 1 with
estimated standard deviations
Pt–N(1)
Pt–N(2)
Pt–Cl(2)
Pt–Cl(1)
N(1)–C(7)
N(1)–C(8)
N(2)–C(10)
N(2)–C(11)
N(2)–C(9)
2.014(7)
2.084(8)
2.288(4)
2.305(4)
1.196(10)
1.559(12)
1.246(17)
1.423(14)
1.482(11)
C(1)–C(2)
C(1)–C(6)
C(2)–C(3)
C(3)–C(4)
C(4)–C(5)
C(5)–C(6)
C(6)–C(7)
C(8)–C(9)
1.281(13)
1.421(13)
1.396(15)
1.352(14)
1.478(13)
1.343(13)
1.550(12)
1.529(14)
N(1)–Pt–N(2)
N(1)–Pt–Cl(2)
N(2)–Pt–Cl(2)
87.0(3)
90.1(3)
176.3(2)
N(1)–Pt–Cl(1)
N(2)–Pt–Cl(1)
Cl(2)–Pt–Cl(1)
179.4(2)
93.0(2)
90.01(15)
(7.51 ppm) and methylene (4.06 ppm) protons was observed in
2D-NOESY NMR experiments. Therefore, since an E-form is
required for the metallacycle formation in complex 2, isomeris-
ation of the ligand should take place in order to achieve the
cyclometallation process.
3
The J(H–Pt) value for the methylamine protons decrease
from 1 (3J(H–Pt) = 34 Hz) to 2 (3J(H–Pt) = 13 Hz) as a result of
the higher trans influence of the phenyl vs. the chloro ligands.
Conversely, an increase in the coupling constant of the imine
proton to platinum on going from 1 (3J(H–Pt) = 62 Hz) to 2
(3J(H–Pt) = 143 Hz) is observed; this effect may be attributed
both to the higher electron density at the metal in the cyclo-
metallated complex, and to the formation of the structure
corresponding to an endo metallacycle. The 195Pt NMR spec-
trum of complex 1 shows one single resonance at (δ Ϫ2265
ppm), which is consistent with the presence of two nitrogen and
two chloro atoms in the coordination sphere of platinum().35
A comparison with the value for complex 2 (δ Ϫ3462 ppm)
indicates a shift towards higher fields upon metallation, in
agreement with an increase in ligand covalency from Cl (in 1) to
Caryl (in 2).
Chart 1 NMR data for the reported complexes; δ values in ppm,
3J(H–Pt), in parenthesis, in Hz.
converting complexes analogous to 1 into cycloplatinated
derivatives are refluxing for several hours in either toluene or in
a donor solvent such as methanol or ethanol, in some cases in
the presence of an “external” base such as NaCH3CO2.15–21,31,32
Several experimental conditions were tested with the aim of
producing cyclometallation of the coordinated imine. Complex
1 is stable in refluxing methanol; however, treatment for 12 h in
the presence of an equimolar amount of sodium acetate gave
complex 2 which was isolated in 58% yield. Conversion of
complex 1 to complex 2 was also achieved under analogous
conditions when external bases such as 1,8-bis(dimethyl-
amino)naphtalene, commonly used as a proton sponge, or tri-
ethylamine were used instead of sodium acetate, integration
of the NMR signals (Chart 1) of the final mixtures indicate
40–60% conversion yields. Reaction in refluxing toluene, even in
the presence of sodium acetate, produces the recovery of the
unchanged starting complex 1 after 18 h; the extreme insolu-
bility of the complex in this solvent could be responsible for
this fact.
Attempts to achieve one-pot synthesis of cyclometallated
complex 2 without prior isolation of complex 1 gave poorer
results. The reaction of cis-[PtCl2(dmso)2] and equimolecular
amounts of ligand L and sodium acetate in refluxing methanol
produced after 18 h a mixture of complex 1 and complex 2 in
low yields (13 and 20%, respectively), together with small
amounts of benzaldehyde and unreacted platinum substrate.
Even lower yields of complexes 1 and 2 were obtained when
[PtCl2(SMe2)2] was used as starting material in agreement with
the fact that SMe2 is a worse leaving ligand than dmso for Pt()
centres.24
Suitable crystals of complexes 1 and 2 were obtained from
methanol concentrated solutions on standing. The crystal
structures are composed of discrete molecules separated by van
der Waals distances. The structures are shown in Figs. 1 and S1
(ESI†) and selected molecular dimensions are listed in Tables 1
and S1 (ESI†). For complex 1, the platinum atom has the
expected square planar coordination with the bidentate (N,NЈ)
ligand and two mutually cis chloro ligands. The Pt–Cl bonds
are similar in length to those found in related complexes.15,20
Complex 1 was characterized by elemental analysis, NMR
and mass spectrometry. The synthesis of complex 2 following
an alternative route has been reported elsewhere25 along with
characterisation of the cyclometallated complex. Crystal struc-
ture determinations were carried out for complex 1 and for
previously reported complex 2. For complex 1, evidence of co-
ordination of both nitrogen atoms to platinum is obtained from
the fact that both methylamine and imine protons are coupled
to platinum (3J(Pt–H) = 34 and 62 Hz, respectively). The down-
field shift of methylamine protons (δ 3.15 ppm) confirms the
coordination of this group to platinum. On the other hand, the
chemical shift of the imine resonance (δ 9.50 ppm) indicates,
according to previous studies for platinum or palladium com-
plexes with coordinated imines,31–34 a Z-conformation around
the C᎐N bond with the imine proton close to the platinum
᎐
nucleus. Accordingly, a cross-peak signal between aromatic
Fig. 1 Molecular structure of complex 1.
D a l t o n T r a n s . , 2 0 0 3 , 3 7 6 3 – 3 7 6 9
3764

View Article Online
The Pt–NMe2 distance (2.084(8) Å) is longer than the Pt–NCH
(2,014(7) Å), which is consistent with the relatively weak co-
ordinating ability of tertiary amines for platinum. Bond angles
at platinum are close to the ideal 90Њ, the smallest (87.0(3)Њ)
corresponding to the chelate (N,NЈ). In agreement with the
above NMR data, the conformation of the imine ligand is
Z, the torsion angle C(6)–C(7)–N(1)–C(8) being Ϫ9.1(14)Њ.
Description of the structure for complex 2, which is analogous
to those described for similar complexes,28 is included in the
ESI†(Fig. S1).
order to observe the second process indicated. Even for this
second process, no meaningful dependence of the observed rate
constants on the concentration of sodium acetate was detected
(5–20 fold margin). Although these were not the preparative
conditions indicated above, proton NMR monitoring of the
reaction under kinetic conditions indicated that the conversion
of complex 1 to its cyclometallated derivative is quantitative
without the formation of side products. As shown in Fig. 3, the
value determined for the first observed reaction rate constant in
methanol and ethanol solutions was found the same within
error, while for the second process dramatic differences were
obtained on changing the solvent. Table 2 collects all the values
of the relevant kinetic and thermal and baric activation param-
eters derived for these processes. As stated above, in all cases 1H
NMR spectra of the final reaction mixture indicated that the
complete process measured corresponds to the activation of the
C–H bond of the coordinated imine stated above. The oxid-
ative-addition pathway can be discarded due to the fact that
platinum hydride resonances were not observed in the proton
NMR spectra.
Mechanistic studies
When repetitive UV-Vis spectra of sodium acetate-containing
methanol or ethanol solutions of the starting complex were
recorded, the spectral changes indicated in Fig. 2 were obtained.
As indicated in the Figure, these changes can be separated into
two well-defined processes with absorbance vs. time traces
showing a good first-order exponential behaviour on the plat-
inum complex concentration. At low excess (1–2 fold) concen-
trations of sodium acetate, the magnitude of the first spectral
changes appear to be dependent on the amount of sodium
acetate added; nevertheless, the values found for the derived
first-order rate constants are found independent on these con-
centrations. Practically no further changes could be observed in
these cases and, consequently, further studies were carried out
under pseudo-first-order conditions (20-fold minimum excess
given the two possible substitution positions at the Pt centre), in
Fig. 3 Eyring plots for the rate constants corresponding to the two
processes occurring on cyclometallation of complex 1 assisted by
acetate in different solvents. Solid symbols correspond to methanol
solution, empty symbols to ethanol solution; squares to the first step
and circles to the second.
In view of the above data the cyclometallation reaction of
complex 1 was monitored in the presence of other bases such as
NEt3, both in methanol and ethanol solutions. The same type
of spectral changes than those observed for the process carried
out in the presence of sodium acetate were obtained. Further-
more, the same dependences with the solvent and concentration
of the base used were also made evident, Table 2 collects the
relevant kinetic and activation parameters, and Fig. 4 indicates
the pressure dependence of the two steps observed. When
methanol or ethanol solutions of the starting material, with no
sodium acetate or any base added, were monitored by UV-Vis
Fig. 2 Spectral changes (a first step, b second step) in the UV-Vis
spectra of a solution of the starting complex, 1, in a solution of sodium
acetate (50-fold excess) in ethanol; 324 K. Symbols correspond to the
experimental data, lines to the exponential fitting.
Fig. 4 Pressure dependence of the rate constants of the two processes
involved in the cyclometallation reaction of complex 1 in ethanol
solution in the presence of NEt3. Square symbols correspond to the first
step (336 K) and circles to the second (343 K).
D a l t o n T r a n s . , 2 0 0 3 , 3 7 6 3 – 3 7 6 9
3765

View Article Online
Table 2 Values of the kinetic and activation parameters for the two processes involved in the acetate- and triethylamine-assisted cyclometallation
reaction of complex 1, as a function of the solvent used.
Step
First
Solvent
Base
105 × 323k/sϪ1
∆H≠/kJ molϪ1
∆S≠/J KϪ1 molϪ1
∆V ≠/cm3 molϪ1
MeOH or EtOH
NaCH3CO2
NEt3
12.1
8.9
63
2
3
Ϫ128
6
Ϫ14 2 (332 K, MeOH)
Ϫ11 2 (342 K, EtOH)
19 1 (336 K, EtOH)
109
13 11
Second
MeOH
EtOH
NaCH3CO2
NEt3
NaCH3CO2
NEt3
2.2
0.33
0.9
76
64
83
94
5
7
9
5
Ϫ101 16
Ϫ156 22
Ϫ86 26
Ϫ59 14
Ϫ10 1 (332 K)
Ϫ5.1 1.9 (338 K)
Ϫ13 3 (342 K)
13 1 (343 K)
0.54
spectroscopy vs. time, only very small spectral changes corre-
sponding to the first reaction step could be detected. The
changes were attributed to the possible small amounts of
alkoxy bases present in the solvents used, but the values derived
from the absorbance vs. time traces agree very well with the
values determined in pseudo-first-order (10-fold excess) condi-
tions of NEt3. In the same respect, when the base used was
proton sponge, despite the dramatic interferences in the UV-Vis
spectrum, the values determined for the observed rate constant
corresponding to the first reaction step agree very well with the
values determined when triethylamine or neat solvent were used
(Fig. S2, ESI†). Unfortunately the spectral changes leading to
the final cyclometallated product, 2, could not be monitored
due to interferences from the electronic spectrum of the proton
sponge. Attempts to follow the cyclometallation process in
other solvents with different characteristics (toluene, chloro-
form and acetone) were unsuccessful given the extremely low
solubility of complex 1 in them (see above).
It is clear from the data obtained that the complete cyclo-
metallation process takes place very slowly with an initial,
relatively faster, formation of an intermediate species. The form-
ation of such species takes place independly of the solvent
used, but it strongly depends on the coordinating capabilities of
the base added. The process switches from one having a high
ordered/compressed transition state, for the reactions carried
out in the presence of sodium acetate, to one having an
expanded transition state where only a small increase of the
ordering is apparent from the values of ∆S≠. The second step
leading to the final cyclometallated complex 2, also shows
dramatic differences in the activation parameters, especially in
the values determined for ∆V ≠, depending of the coordinating
capabilities of the proton abstractor used. In this case, though,
even the relatively small differences in the solvent used produce
changes in the energetics leading to the transition state. It seems
clear that the processes taking place with NaCH3CO2 or NEt3
present in the reaction medium depart from a different starting
material. Given the UV-Vis spectral changes registered during
the cyclometallation reaction studied and indicated in Fig. 2, 1H
NMR monitoring of the reaction under similar conditions was
carried out. Effectively the NMR spectra of ethanol reaction
mixtures of complex 1 with a 10-fold excess of NEt3 showed
a new series of peaks that further evolve to form the final
cyclometallated complex 2. The 2D-NOESY NMR spectrum
showed a cross-peak signal, confirmed by a 1D-NOESY
experiment, between the imine and the methylene resonances of
a minor compound which, on these bases, was identified as the
E isomer (Fig. S3, †).36 As a result, the full reaction of the
cyclometallation reaction of imine L can be interpreted as indi-
cated in Scheme 1. The reaction times, nevertheless, are found
to be extremely dependent on the solvent and coordination
capability of the added base, as shown by the data in Table 2.
Probably at high sodium acetate concentrations and elevated
temperatures, as used in this study, complex 1 is transformed
very rapidly into a mixture of acetato complexes, as expected
from the known substitution processes on square-planar Pt()
complexes (see Scheme 1).37,38 In this respect, 1H NMR spectra
of freshly prepared solutions of complex 1 with a five-fold
excess of sodium acetate showed three resonances corre-
sponding to the methynic proton that can be assigned to differ-
ent acetato complexes. A singlet at 9.45 ppm can be assigned to
a monoacetato complex with the imine in a Z form, while two
other singlets at 8.70 and 8.40 ppm were assigned, respectively,
to mono- and bis-acetato complexes having the imine already in
a E conformation. Accordingly in the acetate region the reson-
ances at 2.06 and 2.07 ppm were assigned to the two mono-
acetato complexes (Z and E imine forms) while the two signals
of equal intensity at 2.04 and 2.05 ppm were assigned to the
bis-acetato complex.
Given the extreme differences observed in the mechanistic
data, the discussion of the results of the reactions followed with
Scheme 1
D a l t o n T r a n s . , 2 0 0 3 , 3 7 6 3 – 3 7 6 9
3766

View Article Online
added acetate or triethylamine has to be carried out separately.
The above mentioned immediate (according to the cyclo-
metallation timescale) formation of the acetato derivatives of 1
under the kinetic conditions used, supports this differentiation.
Data in Table 2 clearly indicates that both steps of the studied
process take place with a highly ordered and compressed transi-
tion state. The fact that for the first step, Z to E imine isomeris-
ation (Scheme 1), the kinetic and activation parameters are
found independent of the solvent used, indicate that the transi-
tion state has no influence from the reaction medium; insolubil-
ity of the starting complex prevented the study of the reaction
in solvents with more severe differences. From Scheme 1 it is
clear that the presence of an acetato ligand at the coordination
sphere of the platinum centre, and in a cis position with respect
to the iminic nitrogen, should favour an intramolecular stabilis-
ation of the polar ؒ ؒ ؒ CH(ϩ)–N(–)–CH2 ؒ ؒ ؒ form of the imine
needed for the isomerisation. That is, the dangling oxygen of
the monodentate acetato ligand is likely to stabilize the positive
charge on the iminic carbon, allowing an easy rotation of the
C–N bond to the desired E conformation. This process is
bound to be fairly independent of the solvent, as observed, and
should imply a highly ordered (∆S≠ = Ϫ128 J KϪ1 molϪ1) and
compressed (∆V ≠ = Ϫ11/Ϫ14 cm3 molϪ1) transition state. As for
the second reaction step, the activation parameters collected in
Table 2, indicate that, although the activation volume results
fairly independent on the solvent used, the values for ∆H≠ and
∆S≠ are discriminating with respect to the polarity of the sol-
vent. In all cases the activation parameters correspond to a
highly ordered and compressed transition state, with the polar-
ity of the solvent favouring the less enthalpy demanding and
more ordered one. Given the fact that this second step corre-
sponds to the actual C–H bond activation reaction, the proton
transfer from the aromatic carbon to the acetate implies
an important charge separation that should be favoured in
methanol, as observed. Even so, the differences in ethanol and
methanol solution are not found to be very high (especially with
respect to the activation volumes), which seems to indicate that
the solvent participation in the transition state is small. Again
the excellent positioning of the dangling oxygen of the plat-
inum-coordinated acetate favours an intramolecular reaction
via a tetracentered transition state involving the acetato ligand,
the metal centre, the cyclometallating carbon and the hydrogen,
as a whole, with charge separation being at a minimum. It this
respect the known cyclometallation reactions of equivalent
Pd() complexes3–5 have values for the thermal and baric activ-
ation parameters very similar to the ones in this study, whilst
the tetracentered transition state involved is equivalent. In
those cases, the cyclometallation process is strongly accelerated
by the presence of protons in the reaction medium, due to the
formation of coordinated acetic acid that facilitates the exiting
of the leaving ligand.4 In this respect, all attempts to run the
cyclometallation reaction from 1 to 2 in the presence of acetic
acid resulted in the full recovery of the starting material; even
no Z to E isomerisation of the ligand was detected. The
protonation of the dangling oxygen of the acetato ligand at the
change from negative to positive activation volume has to be
related to the liberation of solvent molecules from the solvated
NEt3 and the polar form of the imine occurring on going to the
ion-paired transition state. As for the actual C–H bond activ-
ation process, the reaction is extremely solvent-dependent; the
activation volumes having different sign and enthalpies of acti-
vation having a 1.5 fold difference (see Table 2). A very import-
ant charge separation should occur in the transition state, where
a proton has to be transferred from an aromatic carbon to an
external NEt3 molecule, allowing for an anionic ClϪ to leave the
Pt() centre. Consequently, the reaction in solvents with higher
polarity should be less enthalpy demanding but having a more
solvent ordered and compressed transition state, as effectively
indicated by the values of Table 2.
The absolute values for the activation volumes are, neverthe-
less, confusing; although the trend observed is that indicated
above, the values do not agree with the entropy data. These sort
of opposite trends between volumes and entropies of activation
have been repetitively associated with the formation of highly
solvent-structured transition states in aqueous solution, where
not only electrostriction is operating.12,39–42 The solvents used in
this study, nevertheless, are not prone to these type of inter-
actions, so some other explanation should be made available.
ϩ
Probably some ion pairing between NHEt3 and ClϪ is pro-
ducing a less solvent-ordered and compressed transition state
than expected.29 If so, the ion pairing should be much more
important in the less polar ethanol solutions43 producing a less
negative value for the entropy of activation and an important
positive contribution to the volume of activation based on the
intrinsic bond breaking process involved to go to the transition
state.
Conclusions
It is shown in this study that the preparation of cyclometallated
platinum() complexes containing terdentate [C,N,NЈ] ligands
via electrophilic substitution is feasible using [PtCl2(SMe2)2] or
cis-[PtCl2(dmso)2] as platinum substrates. The results here
described indicate the dramatic influence of the nature of the
base added in the cyclometallation processes. In particular the
reaction mechanism is proved to have an intramolecular char-
acter when the added base has coordination capabilities, such as
sodium acetate, vs. the intermolecular characteristics of the
process for reactions carried out in the presence of an authentic
“external” base such as NEt3 or proton sponge. Differences in
solvent implication in the process are more difficult to quantify
given the small differences between the solvents used and the
insolubility of the starting material in other solvents, which
prevented the monitoring of the reactions.
Experimental
Instruments
Mass and NMR spectra were performed by the Serveis
Científico-Tècnics de la Universitat de Barcelona. Micro-
analyses were performed by the Institut de Química Bio-
orgànica de Barcelona (Consejo Superior de Investigaciones
Científicas)
platinum centre, that should destabilize the polar ؒ ؒ ؒ CH(ϩ)
–
N(–)–CH2 ؒ ؒ ؒ form of the imine, can easily explain this fact.
If the E form of the imine is not present the further C–H
activation cannot take place.
FAB-mass spectra were carried out in a VG-Quattro
For the full cyclometallation process carried out in the pres-
ence of a non-coordinating base, such as NEt3, the differences
observed should be related to the impossibility of an intra-
molecular-assisted process, both for the first and second
reaction steps. Effectively, the thermal and baric activation
parameters for the Z to E isomerisation, collected in Table 2,
indicate a much more enthalpy demanding process, with an
important expansion and little change in organization, to go to
the transition state. We can speculate that the NEt3 base should
be responsible for the stabilisation of the positive charge
density of the ؒ ؒ ؒ CH(ϩ)–N(–)–CH2 ؒ ؒ ؒ form of the imine. The
1
spectrometer with 3-nitrobenzyl alcohol matrix. H and 195Pt
NMR spectra were recorded by using Varian Gemini 200 (1H,
200 MHz) and Bruker 250 (195Pt, 54 MHz) spectrometers, and
referenced to SiMe4 (1H) and H2PtCl6 in D2O (195Pt). 1D- and
2D-NOESY experiments were carried out in a Bruker 600
spectrometer; δ values are given in ppm and J values in Hz.
Compounds
Compounds cis-[PtCl2(dmso)2],44 [PtCl2(SMe2)2]45 and PhCH-
NCH2CH2NMe2 (L)28 were prepared as reported. L was
D a l t o n T r a n s . , 2 0 0 3 , 3 7 6 3 – 3 7 6 9
3767

View Article Online
1
Table 3 Crystallographic and refinement data for complex 1
characterized by H NMR (200 MHz, CDCl3): δ 2.31 [s, 6H,
Ha]; 2.64 [t, J(H–H) = 7, 2H, Hb]; 3.75 [t, J(H–H) = 7, 2H, Hc];
7.41 [m, 3H, aromatics]; 7.72 [m, 2H, aromatics]; 8.32 [s, 1H,
Hd].
Formula
Mw
C11H16Cl2N2Pt
442.25
293(2)
T/K
Complex [PtCl2(Me2NCH2CH2NCHPh)] (1) was obtained
from 0.100 g (2.56 × 10Ϫ4 mol) of [PtCl2(SMe2)2] and 0.045 g
(2.56 × 10Ϫ4 mol) of ligand PhCHNCH2CH2NMe2 in 25 cm3 of
methanol. The mixture was refluxed with continous stirring
during 4 h. The solvent was evaporated in vacuo and the remain-
ing yellow residue was recrystallized in CH2Cl2–MeOH to yield
white crystals. Yield 91 mg (80%). Anal. Found: C, 29.4; H, 3.7;
N, 6.3. Calc. for C11H16Cl2N2Pt : C, 29.87; H, 3.65; N, 6.33%. 1H
NMR (200 MHz, CDCl3): δ 2.68 [t, J(H–H) = 6, 2H, Hb] 3.15
[s, J(Pt–H) = 34, 6H, Ha]; 4.06 [t, J(H–H) = 6, 2H, Hc]; 7.51
[m, 5H, aromatics]; 9.50 [s, J(H–Pt) = 62, 1H, Hd]. 195Pt NMR
(54 MHz, CDCl3): δ Ϫ2265 [s, br]. FAB-MS, m/z: 441 [M], 406
[M Ϫ Cl], 371 [M Ϫ 2Cl].
Complex [PtCl(Me2NCH2CH2NCHC6H4)] (2) was obtained
using the following procedure: The mixture formed by 0.050 g
(1.13 × 10Ϫ4 mol) of complex 1, the equimolar amount (0.09 g)
of sodium acetate and 25 cm3 of methanol was heated under
reflux during 12 h. The yellow solution was evaporated to ca. 10
cm3, a small amount of complex 1 was filtered off, and the
solution was allowed to crystallize at room temperature. Within
several hours, red crystals of complex 2 were collected. Yield 25
mg (58%). 1H NMR (200 MHz, CDCl3): δ 2.88 [s, J(Pt–H) = 13,
6H, Ha] 3.07 [t, J(H–H) = 6, 2H, Hb]; 3.86 [t, J(H–H) = 6,
J(H–Pt) =34, 2H, Hc]; 6.99 [t, J(H–H) = 7, 1H]; 7.17 [d, J(H–H)
=8, 1H]; 7.21 [t, J(H–H) = 7, 1H]; 7.69 [d, J(H–H) = 8, J(H–Pt)
= 42]; 8.21 [s, J(H–Pt) = 143, 1H, Hd].
λ/Å
0.71069
Monoclinic
P21/a
7.919(15)
15.528(15)
11.099(9)
97.72(10)
1352(3); 4
2.172
10.746
832
3892/3892 (0.0686)
3892 / 0 / 145
0.812
Crystal system
Space group
a/Å
b/Å
c/Å
β/Њ
V/Å3; Z
Dc/Mg mϪ3
µ/mmϪ1
F(000)
Rflns. coll./unique (Rint
Data/restraints/param.
GOF on F 2
R1 (I > 2σ(I ))
wR2 (all data)
)
0.0356
0.0985
0.588 and Ϫ0.624
Largest peak and hole/e ÅϪ3
parameters were determined from automatic centering of 25
reflections (12 < θ < 21Њ) and refined by least-squares methods.
Intensities were collected with graphite monochromated
Mo-Kα radiation, using the ω–2θ scan-technique. For 1, 3892
reflections were measured in the range 2.27 < θ < 30.05Њ; 1865
of which were assumed as observed applying the condition
I > 2σ(I ). For 2, 7009 reflections were measured in the range
2.31 < θ < 29.96Њ; 3467 of which were non-equivalent by sym-
metry (Rint = 0.069) and 1590 reflections were assumed as
observed applying the condition I > 2σ(I ). Three reflections
were measured every 2 h as orientation and intensity controls,
significant decay was not observed. Lorentz polarization but
not absorption corrections were made. Further details are given
in Tables 3 and S3 (ESI†).
Kinetics
The reactions were followed by UV-Vis spectroscopy in the
500–300 nm range where none of the solvents absorb. Runs at
room pressure were recorded on an HP8452A or Cary-50
instruments equipped with a thermostated multicell transport.
For the kinetic runs at elevated pressure, a previously described
pressurizing system46 connected to the Cary-50 spectrophoto-
meter was used. Observed rate constants were derived from
absorbance vs. time traces at the wavelengths where a maximum
increase and/or decrease of absorbance was observed. In some
cases the SPECFIT47 program was used in order to evaluate
two consecutive reaction rate constants. In all cases the differ-
ences between the two observed rate constants has been found
large enough (Fig. 2) to evaluate them separately. Table S2
(ESI†) collects all the obtained kobs values for all the reactions
studied as a function of the starting complex, added base, sol-
vent, temperature and pressure. No dependence of the observed
rate constant values on the selected wavelengths was detected,
as expected for reactions where a good retention of isosbestic
points is observed. The general kinetic technique is that pre-
viously described,13,48 in all cases no dependence of the values
obtained for the observed rate constant on the concentration of
starting material or added base was observed; nevertheless
pseudo-first-order conditions were maintained (at least 20-fold)
and the platinum concentration was maintained at (2–3) × 10Ϫ4
M; values in Table S1 (ESI†) correspond to the average value of
the data obtained under the specified conditions, in all cases
errors were within 10% of the stated value. Given the very long
periods during which the reactions were monitored (2–3 half
lives in all cases), special care was taken to avoid solvent evap-
oration, plunger-stoppered or pill-box cells were used. Even for
runs at temperatures close to the boiling point of the solvents
used, these were carried out within the variable-pressure setup,
in order to avoid lack of homogeneity in the solutions.
Structure solution and refinement. The structures were solved
by direct methods using SHELXS computer program49 and
refined by the full-matrix least-squares method, with the
SHELX97 computer program49 using 3892 (1) or 3467 (2)
reflections (very negative intensities were not assumed). The
function minimised was Σw||Fo|2 Ϫ |Fc|2, where w = [σ2(I ) ϩ
(0.0375P)2]Ϫ1 (1) and w = [σ2(I ) ϩ (0.0084P)2]Ϫ1 (2) and P = (|Fo|2
ϩ 2|Fc|2)/3. f, f Ј and f Љ were taken from International Tables of
X-Ray Crystallography.50 All hydrogen atoms were computed
and refined with using a riding model with an isotropic temper-
ature factor equal to 1.2 times the equivalent temperature factor
of the atom to which they are linked. Further details are given
in Tables 3 and S3 (ESI†).
CCDC reference numbers 212048 and 212049.
See http://www.rsc.org/suppdata/dt/b3/b306243k/ for crystal-
lographic data in CIF or other electronic format.
Acknowledgements
We acknowledge financial support for the Ministerio de Ciencia
y Tecnología (project numbers BQU2001-3205 and BQU2000-
0652) and the CIRIT from the Generalitat de Catalunya
(project number SGR2001-00054).
References
1 A. D. Ryabov, I. K. Sakodinskaya and A. K. Yatsimirsky, J. Chem.
Soc., Dalton Trans., 1985, 2629.
2 A. D. Ryabov, Chem. Rev., 1990, 90, 403.
3 M. Gómez, J. Granell and M. Martínez, Organometallics, 1997, 16,
2539.
X-Ray structure analysis
4 M. Gómez, J. Granell and M. Martínez, J. Chem. Soc., Dalton
Trans., 1998, 37.
5 M. Gómez, J. Granell and M. Martínez, Eur. J. Inorg. Chem., 2000,
217.
Data collection. Prismatic crystals were selected and mounted
on an Enraf-Nonius CAD4 four-circle diffractometer. Unit cell
D a l t o n T r a n s . , 2 0 0 3 , 3 7 6 3 – 3 7 6 9
3768

View Article Online
6 G. Gonzalez, M. Martinez, F. Estevan, Garcia, P. Lahuerta and
E. Peris, New J. Chem., 1996, 20, 83.
7 G. Gonzalez, P. Lahuerta, M. Martinez, E. Peris and M. Sanau,
J. Chem. Soc., Dalton Trans., 1994, 545.
8 F. Estevan, G. González, P. Lahuerta, M. Martínez, E. Peris and
R. van Eldik, J. Chem. Soc., Dalton Trans., 1996, 1045.
9 L. M. Rendina and R. J. Puddephatt, Chem. Rev., 1997, 97, 1735.
10 R. J. Puddephatt, Coord. Chem. Rev., 2001, 219–221, 157.
11 R. A. Gossage, A. D. Ryabov, A. L. Spek, D. J. Stufkens, J. A. M.
van Beek, R. van Eldik and G. van Koten, J. Am. Chem. Soc., 1999,
121, 2488.
28 C. M. Anderson, M. Crespo, M. C. Jennings, A. J. Lough and
R. J. Puddephatt, Organometallics, 1991, 10, 2672.
29 F. K. Fleischmann and H. Kelm, Tetrahedron Lett., 1973, 14, 3773.
30 Z/E conformation has been assigned not taking into account the
platinum coordination to the iminic nitrogen.
31 X. Riera, C. López, A. Caubet, V. Moreno, X. Solans and M. Font-
Bardía, Eur. J. Inorg. Chem., 2001, 2135.
32 X. Riera, A. Caubet, C. López, V. Moreno, X. Solans and M. Font-
Bardía, Organometallics, 2000, 19, 1384.
33 J. Albert, R. M. Ceder, M. Gómez, J. Granell and J. Sales,
Organometallics, 1992, 11, 1536.
12 M. Crespo, M. Martínez and E. Pablo, J. Chem. Soc., Dalton Trans.,
1997, 1231.
13 M. Crespo, M. Martinez, J. Sales, X. Solans and M. Font-Bardía,
Organometallics, 1992, 11, 1288.
14 M. Crespo, M. Martinez and J. Sales, Organometallics, 1993, 12,
4297.
15 Y. J. Wu, L. Ding, H. X. Wang, Y. H. Liu, H. Z. Yuan and
X. A. Mao, J. Organomet. Chem., 1997, 535, 49.
34 J. Albert, M. Gómez, J. Granell, J. Sales and X. Solans,
Organometallics, 1990, 9, 1405.
35 P. S. Pregosin, Coord. Chem. Rev., 1982, 44, 247.
36 NMR data obtained for the E isomer in CDCl3 (ppm, Hz): δ 3.08
[s,3J(HPt) = 31, 6H]; 2.83 [t,3J(HH) = 6, 2H]; 4.25 [t,3J(HH) = 6, 2H];
8.49 [d,3J(HH) = 7, 1H]; 8.75 [s,3J(HPt) = 124.1, 1H].
37 M. L. Tobe and J. Burgess, Inorganic Reaction Mechanisms, Addison
Wesley, Edinburgh, 1999.
16 L. Ding, D. P. Zou and Y. J. Wu, Polyhedron, 1998, 17, 2511.
17 M. D. Meijer, A. W. Kleij, M. Lutz, A. L. Spek and G. van Koten,
J. Organomet. Chem., 2001, 640, 166.
18 A. W. Kleij, R. J. M. K. Gebbink, M. Lutz, A. L. Spek and G. van
Koten, J. Organomet. Chem., 2001, 621, 190.
19 A. D. Ryabov, I. M. Panyashkina, V. A. Polyakov and A. Fisher,
Organometallics, 2002, 21, 1633.
20 P. R. R. Ranatunge-Bandarage, B. H. Robinson and J. Simpson,
Organometallics, 1994, 13, 500.
21 P. R. R. Ranatunge-Bandarage, N. W. Duffy, S. M. Johnston,
B. H. Robinson and J. Simpson, Organometallics, 1994, 13, 511.
22 A. D. Ryabov, G. M. Kazankov, I. M. Panyashkina, O. V.
Grozovsky, O. G. Dyachenko, V. A. Polyakov and L. G. Kuz’mina,
J. Chem. Soc., Dalton Trans., 1997, 4385.
38 R. G. Wilkins, Kinetics and Mechanisms of Reactions of Transition
Metal Complexes, VCH, Weinheim, 1991.
39 M. Martínez, M. A. Pitarque and R. van Eldik, Inorg. Chim. Acta,
1997, 256, 51.
40 M. Martínez, M. A. Pitarque and R. van Eldik, J. Chem. Soc.,
Dalton Trans., 1994, 3159.
41 M. Martínez and M. A. Pitarque, J. Chem. Soc., Dalton Trans.,
1995, 4107.
42 P. V. Bernhardt, F. Bozoglian, B. P. Macpherson, M. Martínez,
G. González and B. Sienra, Eur. J. Inorg. Chem., 2003, 2512.
43 M. Albrecht, R. A. Gossage, U. Frey, A. W. Ehlers, E. J. Baerends,
A. E. Merbach and G. van Koten, Inorg. Chem., 2001, 40, 850.
44 J. M. Price, A. N. Williamson, R. F. Schramm and B. B. Wayland,
Inorg. Chem., 1972, 11, 1280.
23 L. Alexandrova, O. G. D’yachenko, G. M. Kazankov, V. A.
Polyakov, P. V. Samuleev, E. Sansores and A. D. Ryabov, J. Am.
Chem. Soc., 2000, 122, 5189.
24 A. D. Ryabov, S. Otto, P. V. Samuleev, V. A. Polyakov,
L. Alexandrova, G. M. Kazankov, S. Shova, M. Revenco,
J. Lipkowski and M. M. Johansson, Inorg. Chem., 2002, 41,
4286.
45 G. S. Hill, M. J. Irwin, C. J. Levy, L. M. Rendina and R. J.
Puddephatt, Inorg. Synth., 1998, 32, 149.
46 M. G. Basallote, J. Durán, M. J. Fernández-Trujillo, G. González,
M. A. Máñez and M. Martínez, J. Chem. Soc., Dalton Trans., 1999,
3379.
47 R. A. Binstead, A. D. Zuberbühler and B. Jung, SPECFIT version
3.0.32, Spectrum Software Associates, 2002.
25 M. Crespo, J. Granell, X. Solans and M. Font-Bardía,
Organometallics, 2002, 21, 5140.
26 O. López, M. Crespo, M. Font-Bardía and X. Solans,
Organometallics, 1997, 16, 1233.
27 C. M. Anderson, M. Crespo, G. Ferguson, A. J. Lough and
R. J. Puddephatt, Organometallics, 1992, 11, 1177.
48 P. V. Bernhardt, C. Gallego and M. Martínez, Organometallics,
2000, 19, 4862.
49 G. M. Sheldrick, SHELXS97: A computer program for crystal
structure determination, University of Göttingen, 1977.
50 International Tables of X-Ray Crystallography, Kynoch Press,
Birmingham, UK, 1974.
D a l t o n T r a n s . , 2 0 0 3 , 3 7 6 3 – 3 7 6 9
3769