Transition-state effects of ionic liquids in substitution reactions of PtII complexes

Article abstract of DOI:10.1002/anie.200501329

When normal is a surprise: Studies of a ligand-substitution reaction of a Pt^II complex (see picture; apa: 2,6-bis(aminomethyl) pyridine) in water, methanol, and the ionic liquid 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)amide reveal that the ionic liquid behaves normally , that is, as any other solvent. In the ionic liquid, changes in the polarity of the transition state only play a minor role. (Chemical Equation Presented)

Full text of DOI:10.1002/anie.200501329

Angewandte
Chemie
Ionic Liquids
introduction of s-donor and p-acceptor spectator chelates.[7,8]
Rate and activation parameters clearly supported the asso-
ciative nature of the underlying substitution mechanism.
In recent years, ionic liquids have become a popular
research area that is expected to grow as a result of their
potential industrial application.[9] Not only are such liquids
interesting in terms of applications in separation techniques
but also they may be used to improve the solubility of weakly
soluble gases and enhance the catalytic activity of metal
complexes.[10] The first industrial process based on the
application of ionic liquids has been launched.[11]
DOI: 10.1002/anie.200501329
Transition-State Effects of Ionic Liquids in
Substitution Reactions of PtII Complexes**
Christian F. Weber, Ralph Puchta,
Nico J. R. van Eikema Hommes, Peter Wasserscheid,
and Rudi van Eldik*
Mechanistic studies on chemical reactions in ionic liquids
have only been performed to a limited degree. Skrzypczak
and Neta determined the rate constants for the reaction of
1,2-dimethylimidazole with benzyl bromide (Menschutkin
reaction) in a series of ionic liquids and organic solvents. The
rate constants in the ionic liquids are comparable with those
for polar aprotic solvents, but much larger than those for
weakly polar organic solvents and alcohols.[12] Chiappe and
Pieraccini determined rate constants and activation parame-
Ligand-substitution reactions of square-planar complexes of
d8 metal ions such as PtII and PdII play a fundamental role in
many chemical processes, for instance, in the treatment of
tumors and in homogeneous catalysis.[1,2] Detailed mechanis-
tic studies have demonstrated that such complexes usually
undergo associative ligand-substitution reactions as a result of
the 16-valence-electron structure of the square-planar com-
plex.[3] The transition state has either a semi-five-coordinate
character in the case of an associative interchange (Ia)
mechanism, or forms a short-lived, five-coordinate intermedi-
ate in the case of a limiting associative (A) mechanism.
Exceptions are found in cases in which two neighboring
spectator ligands form metal–carbon bonds with the metal
center and induce a dissociative mechanism.[4]
In terms of the biological or catalytic application of such
metal complexes, the fundamental understanding of the
ligand-substitution mechanisms enables a systematic tuning
of the lability of the metal center, that is, the efficiency of the
chemical process. A distinction between the three mecha-
nisms can usually be made on the basis of activation
parameters derived from the temperature and pressure
dependence of the reaction, that is, DHꢀ, DSꢀ, and DVꢀ.[3,5,6]
In recent work performed in our laboratories, the lability of
PtII complexes could be systematically tuned through the
ters (DHꢀ and DSꢀ) for the reactions of Br3 and ICl2 ions
with some alkenes and alkynes in a series of ionic liquids and
in 1,2-dichloroethane.[13] McLean and co-workers[14] studied
the effect of ionic liquids on bimolecular rate constants for the
displacement of solvent from [(C6H6)Cr(CO)2(solv)]. Inhib-
ition of the catalytic activity of a RuII catalyst in ionic liquids
was recently reported by Daguenet and Dyson for a reaction
system in which chloride dissociation is a crucial step.[15] More
recently, work in our laboratory demonstrated that this
inhibition could be related to the selected anionic component
of the employed ionic liquids, since other anions in our hands
clearly lead to a faster chloride dissociation step from a PtII
complex in the catalytic process studied.[16] Following the
submission of this work, Shaughnessy and co-workers
reported kinetic data for oxidative-addition and ligand-
substitution reactions in ionic liquids.[17,18]
À
À
A fundamental question, however, does remain: Do ionic
liquids really behave as “normal” solvents, or are there
unknown aspects that could affect the substitution behavior of
square-planar PtII complexes? We report herein a detailed
kinetic study of the substitution reactions of the [PtII(apa)Cl]+
complex 1 (apa = 2,6-bis(aminomethyl)pyridine; Scheme 1)
with thiourea and iodide as nucleophiles, in water, methanol,
and the ionic liquid 1-butyl-3-methylimidazolium bis(trifluor-
[*] Dr. C. F. Weber, Prof. Dr. R. van Eldik
Institut für Anorganische Chemie
Universität Erlangen-Nürnberg
Egerlandstrasse 1, 91058 Erlangen (Germany)
Fax: ( +49)9131-8527387
E-mail: vaneldik@chemie.uni-erlangen.de
Dr. R. Puchta, Dr. N. J. R. van Eikema Hommes
Computer Chemistry Center
Universität Erlangen-Nürnberg
Nägelsbachstrasse 25, 91052 Erlangen (Germany)
Prof. Dr. P. Wasserscheid
Lehrstuhl für Chemische Reaktionstechnik
Universität Erlangen-Nürnberg
Egerlandstrasse 3, 91058 Erlangen (Germany)
[**] The authors gratefully acknowledge financial support from the
Deutsche Forschungsgemeinschaft, Fonds der Chemischen Indus-
trie, Max-Buchner-Forschungsstiftung and the European Commis-
sion within the framework of the RTN contract no. MRTN-CT-2003-
503864. Prof. Tim Clark, Computer Chemistry Center, is kindly
acknowledged for his support of this work. We thank the Regionales
Rechenzentrum Erlangen (RRZE) for a generous allocation of
computer time.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Scheme 1. Chemical formula and DFT structure of complex 1;
distances in pm.
Angew. Chem. Int. Ed. 2005, 44, 6033 –6038
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6033

Communications
omethylsulfonyl)amide ([BMIM]BTA) as solvents. This ionic
The activation parameters were determined from the
effect of temperature and pressure on k2. Spectral changes
observed for the reaction in [BMIM]BTA at 100 MPa (see
Supporting Information) are in excellent agreement with
those observed at ambient pressure. The thermal activation
parameters, DHꢀ and DSꢀ, were obtained from Eyring plots
(see example in the Supporting Information), whereas DVꢀ
was calculated from the gradient (ÀDVꢀ/RT) of the plot of
lnk versus pressure (see Supporting Information). Table 1
summarizes the values of the dielectric constants (e), ENT, k2,
DHꢀ, DSꢀ, and DVꢀ for the studied reactions in the three
selected solvents. It should be noted that the reactions in the
ionic liquid showed the same behavior as that found for other
more conventional solvents as a function of the experimental
variables.
liquid has a melting point of À48C[19] and a polarity close to
that of ethanol.[9d] We studied the reaction as a function of the
concentration of the entering ligand, the temperature, and the
pressure to determine all the rate and activation parameters
(DHꢀ, DSꢀ, and DVꢀ). This is, as far as we know, the first time
that a systematic pressure-dependence study was undertaken
for a chemical reaction in an ionic liquid. The reported rate
and activation parameters show a rather “normal” behavior
for the employed ionic liquid and demonstrate some unique
properties of the ionic liquid as medium for the investigated
substitution reactions.
The reaction of 1 with TU, a strong nucleophile, is
considerably faster than the reaction with iodide in the same
solvent. The k2 values show that the substitution reactions
depend significantly on the polarity of the solvent, namely,
k2(H2O) @ k2(MeOH) ꢀ k2(ionic liquid). A likely reason is an
increase in the dipole moment during the course of the
The solvent dependence of the ligand-substitution reac-
tions of 1 with thiourea (TU) and iodide (Scheme 2) were
À
reaction due to partial lengthening of the Pt Cl bond in the
five-coordinate trigonal-bipyramidal transition state. In gen-
eral, a decrease in solvent polarity results in a decrease in the
rate of a reaction that involves an increase in dipole moment
in going from the reactant to the transition state.[22] The trend
in the values of k2 clearly indicates that the ionic liquid
behaves like methanol and that no drastic acceleration or
deceleration of the substitution of chloride, as reported for
other PtII complexes in this and closely related ionic liquids,
can be observed.[15,16]
Scheme 2. The investigated substitution reaction for thiourea (Nu0)
and iodide ion (NuÀ).
studied by dissolving the isolated complex 1-Cl in the
selected solvents followed by rapid mixing with a
stock solution of the nucleophile. Typical examples of
the observed spectral changes and kinetic traces are
shown in Figure 1. The overall spectral changes were
found to be rather similar for the studied solvents,
with the exception of [BMIM]BTA due to the
absorption of this liquid in the wavelength range l <
350 nm (see Figure 1c). All kinetic traces (Figur-
e 1b,d) showed excellent fits to a single-exponential
function, in line with pseudo-first-order behavior. The
calculated pseudo-first-order rate constants, kobs, were
plotted against the concentration of the entering
nucleophiles, TU and iodide, and are linear with a
zero intercept under all conditions (see typical
examples given in the Supporting Information). The
rate constant kobs can be expressed as a function of the
concentration of the entering ligand, namely, kobs
=
k2[Nu], in which k2 is the second-order rate constant.
We also studied the effect of the water content of
the ionic liquid on the second-order rate constant for
the reaction of 1 with TU in [BMIM]BTA. The
following k2 values were obtained at 258C (water
content in ppm in brackets): 0.260 (23), 0.245 (100),
and 0.259 mÀ1 sÀ1 (200). It follows that the water
content has no significant effect on the rate constant
in the investigated concentration range.
Figure 1. UV/Vis spectra (a,c) and typical kinetic traces at 320 nm (b,d) recorded
for the reaction of 1 with iodide in methanol (a,b) and [BMIM]BTA (c,d). The
experimental kinetic traces can be perfectly fitted with a single-exponential
function. Experimental conditions: a) [PtII]=9.5”10À5 m, LiCl (0.002m),
[IÀ]=0.015m, ionic strength=0.2m (LiClO4), pH 2 (0.01m trifluoromethanesul-
fonic acid), T=258C, optical path length=10 mm. b) [PtII]=9.5”10À5 m, LiCl
(0.002m), [IÀ]=0.0090m, ionic strength=0.2m (LiClO4), pH 2 (0.01m trifluoro-
methanesulfonic acid), T=258C. c,d) [PtII]=3.3”10À4 m, [IÀ]=0.0392m,
T=258C, optical path length=5 mm.
6034 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6033 –6038

Angewandte
Chemie
Table 1: Summary of second-order rate constants and activation parameters[a] for the displacement of
chloride by iodide and thiourea in 1 in different solvents.
mechanism and represents the
intrinsic volume collapse associated
with the formation of a five-coor-
dinate transition state.[5,6]
Solvent
e(298 K), ENT [b]
103 k2 at 258C
[mÀ1 sÀ1
DHꢀ
DSꢀ
DVꢀ
]
[kJmolÀ1
]
[JKÀ1 molÀ1
]
[cm3 molÀ1
]
To gain further insight into the
details of the ligand-substitution
mechanism, in particular, the ques-
tion of associative interchange (Ia)
versus associative (A), we calcu-
lated the pathways for the reactions
IÀ
water
78.30, 1.000
32.66, 0.762
–, 0.642
78.30, 1.000
32.66, 0.762
–, 0.642
253Æ2
15.3Æ0.2
32.1Æ0.2
1620Æ10
385Æ2
58Æ1
69Æ1
66Æ1
49Æ1
63Æ1
49Æ1
À62Æ3
À51Æ4
À53Æ5
À77Æ4
À42Æ4
À92Æ2
À6.9Æ0.3
À10.0Æ0.2
À14.1Æ0.4
À10.4Æ0.5
À6.6Æ0.1
À13.9Æ0.2
IÀ
methanol
[BMIM]BTA
water
methanol
[BMIM]BTA
IÀ
TU
TU
TU
277Æ3
[a] Activation parameters were calculated from the temperature and pressure dependence of the second-
order rate constant in the usual way. No corrections were made for changes in solvent density[20] or
viscosity[21] as a function of temperature and pressure. [b] Dielectric constant (e) of the solvent and its
polarity (ENT ).
of
1
with chloride (chloride
exchange), iodide, and thiourea by
using B3LYP hybrid density func-
tional theory[24] and the LANL2D-
Z ECP basis set augmented with
polarization functions.[25,26] We have employed this theoretical
The activation parameters summarized in Table 1 show
level in previous work.[16,27]
some interesting trends. Throughout the series of reactions
studied, the negative activation entropy and volume data
clearly support the operation of a compact transition state in
terms of an associative mechanism.[3] For the reaction
between complex 1 and TU, significantly more negative
activation entropies and volumes are found for water and
[BMIM]BTA as the solvent irrespective of their significant
difference in polarity. This finding suggests that reorganiza-
tion in the transition state is not controlled by solvent
electrostriction. The less negative activation entropies and
volumes for the reaction in methanol once again illustrate
that solvent electrostriction does not contribute significantly
to the transition state, otherwise the opposite effect would
occur. In this case, TU is a neutral entering nucleophile and
the mentioned activation parameters mainly correspond to
intrinsic entropy and volume changes.
The first step in the reaction is the formation of a
precursor complex between the platinum complex 1 and the
incoming ligand. As 1 bears a positive charge, interaction
energies in the gas phase are dominated by electrostatics and
exaggerated interaction energies are calculated, particularly
for the complexes with the anionic nucleophiles chloride and
iodide. Nevertheless, the structures of the precursor com-
plexes for all incoming ligands indicate rather weak bonding.
The main contact consists of a hydrogen bond to one of the
amine hydrogen atoms (202, 249, and 221 pm for ClÀ, IÀ, and
S, respectively). On the other hand, the platinum–halide
distances, 386 (chloride) and 437 pm (iodide), and the
platinum–sulfur separation (for the reaction with thiourea),
458 pm, are very long. The other bonds around platinum
differ only to a minor extent from those in 1.
In the case of iodide as the entering nucleophile, the
values of the activation entropy are almost the same for the
different solvents, whereas the activation volumes are sig-
nificantly more negative in the case of methanol and
[BMIM]BTA as solvents. This result suggests that charge
neutralization in the five-coordinate transition state leads to a
decrease in electrostriction—that is, a less negative activation
volume—in the more polar solvents (water and methanol).
This trend, however, does not show up in the values of the
activation entropy, which is probably related to the usually
larger error limits of this parameter. Nevertheless, the trend in
the activation volumes for the different solvents is quite
convincing.
The activation volumes for the reactions with TU and
iodide in [BMIM]BTA are indeed very similar, which suggests
that volume changes in this ionic liquid do not respond to
changes in the overall charge of the transition state. We
suggest that this lack of response is due to strong hydrogen
bonding and possibly also van der Waals forces in ionic
liquids, which result in specific solvent structures that consist
of well aligned cation–anion aggregates. The formation of
such aggregates in ionic liquids has recently been investigated
by using microcalorimetry.[23] The stability of such aggregates
will show a resistance towards solvent structural changes
(electrostriction/solvation) as a result of charge neutralization
or charge creation. The value of about À14 cm3 molÀ1 for DVꢀ
is close to that expected for an associative ligand-substitution
Hydrogen bonds to an amine hydrogen atom are well
known in the solid state,[28] but may be unrealistic in the gas
phase. Therefore, we also studied the reactions for the
dimethylamino
derivative,
[Pt{2,6-bis(dimethylamino-
methyl)pyridine}Cl]+ ([PtII(mapa)Cl]+, 2). As expected, the
platinum–ligand bonds are shorter in the absence of the
À
À
À
hydrogen bonds, but only modestly: Pt Cl 346, Pt I 409, Pt S
378 pm.
All reactions proceed through a single transition-state
structure. Despite extensive searches along the reaction
coordinate, no stationary points other than the precursor
and product complexes could be located. Thus, the reactions
clearly proceed by an interchange-type mechanism. The
calculated activation barriers range from 64 to 89 kJmolÀ1
and show that the nucleophile has a remarkably small
influence (Table 2). Due to the presence of the hydrogen
bond in the precursor complex, the barriers for the reactions
of 1 are somewhat higher than for the reactions of 2. The
influence of N-methylation on the coordination sphere
around the platinum center appears to be negligible.
The transition-state structures are compact, with rela-
tively short bonds to the ligands that are being exchanged
(Figure 2). This compactness is best seen in the reaction of 1
À
À
with Cl : the Pt Cl bonds of the transition state are 259 pm,
À
only 26 pm longer than the Pt Cl bond in 1 itself and 23 pm
longer than the Pt Cl bond in the precursor complex. The
situation in the other transition-state structures is very similar.
À
Angew. Chem. Int. Ed. 2005, 44, 6033 –6038
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6035

Communications
Table 2: Computational results for the reactions 1+X and 2+X (X=ClÀ, IÀ, TU).
whatsoever were observed. For the
first time it has been possible to
determine the activation volume for
a ligand-substitution reaction on a
metal complex in an ionic liquid.
An analysis of the thermal and
pressure activation parameters indi-
cates that the ionic liquid exhibits
rather unique properties in terms of
electrostriction/solvation of transi-
tion states that involve different
degrees of polarity as reflected by
similar activation parameters in
conventional solvents. The appar-
ent insensitivity of the ionic liquid
Relative energies [kJmolÀ1 [a]
]
Distances [pm] (Pt Cl/Pt X)
À
À
Reaction
Ecpx
Ereact
Eact
precursor
transition state
product
1+ClÀ
1+IÀ
1+TU
451
364
86
0
3
13
89
77
72
236/386
235/437
235/458
259/259
259/287
259/268
386/236
385/266
298/238
2+ClÀ
2+IÀ
2+TU
415
334
59
0
21
13
66
75
64
237/346
237/409
237/378
260/260
262/291
265/268
346/237
345/269
318/239
[a] Ecpx: complexation energy released on the formation of the precursor complex; Ereact: reaction energy
for the formation of the product complex; Eact: activation energy for the formation of the transition state
relative to the precursor complex.
towards changes in polarity on going to the
transition state can be ascribed to its ten-
dency to form well-structured cation–anion
aggregates. As a result of the apparent
absence of solvent reorganization due to
changes in electrostriction in this medium,
the observed activation volume can be
directly correlated with intrinsic volume
changes that result from changes in bond
lengths and bond angles on going to the
transition state. This factor can, in principle,
considerably simplify the accurate assign-
ment of intimate mechanisms in cases in
which changes in the polarity of the tran-
sition state cause large solvent effects in
terms of electrostriction/solvation in con-
ventional solvents and complicate the mech-
anistic assignment on the basis of activation
parameters, especially activation entropy
and volume data. This unique property of
ionic liquids may come in very handy to
resolve mechanistic discrepancies in con-
ventional solvents.
Figure 2. Calculated transition-state structures for the reaction of 1 with ClÀ, IÀ, and
thiourea. Distances are given in pm.
Hence, the reactions are best characterized as proceeding by
an associative interchange, Ia, mechanism. In addition, in the
reactions with thiourea, a hydrogen bond between chloride
and one of the thiourea hydrogen atoms is preserved
throughout the reaction.
Experimental Section
All chemicals used were of analytical reagent grade and of the highest
purity commercially available. K2[PtCl4]was obtained from Strem
Chemical Co.; [BMIM]Cl and lithium bis(trifluoromethylsulfonyl)-
amide were obtained from Solvent Innovation. Ultrapure water was
used for the spectroscopic and kinetic measurements. The solvato-
chromic Reichardtꢀs dye, (2,6-diphenyl-4-(2,4,6-triphenylpyridinium-
1-yl)phenolate hydrate, was obtained from Aldrich.
The reactions of 1 with IÀ and TU are both calculated to
be mildly endothermic (+ 3 and + 13 kJmolÀ1, respectively)
in the gas phase. This is due to the lack of stabilizing solvation
of the leaving chloride ion. For the reaction of 1 with iodide,
inclusion of the influence of the bulk solvent through the
isodensity polarizable continuum model (IPCM) gives a
reaction energy of À7 kJmolÀ1.[29]
The complex 1-Cl·H2O, was synthesized following a reported
procedure.[7] The ligand, 2,6-bis(aminomethyl)pyridine, was prepared
according to a method used to synthesize 6-(aminomethyl)-2,2’-
bipyridine[30] by using 2,6-bis(bromomethyl)pyridine[31] as the starting
material. The corresponding platinum complex was prepared as
described previously and was obtained in the same quality (see
Supporting Information).[7]
The ionic liquid [BMIM]BTAwas synthesized by using a reported
procedure.[32,33] A solution of [BMIM]Cl (705 g, 4.04 mol) in H2O
(200 mL) was added to a solution of Li(bis(trifluoromethylsul-
fonyl)amide) (1.054 kg, 3.67 mol) in H2O (200 mL; 1.1:1 molar
ratio). The reaction media was mixed whereupon two phases
formed: the bottom, slightly yellow phase was [BMIM]BTA; the
In conclusion, the studied substitution reaction was
selected as a test system to investigate the behavior of ionic
liquids as reaction medium for ligand-substitution reactions in
comparison to conventional solvents. The observed spectral
changes, kinetic traces, rate, and activation parameters clearly
demonstrate the “normal” behavior and innocence of the
selected ionic liquid in ligand-substitution reactions of PtII
complexes. The kinetic data can be treated according to the
transition-state theory as for any other solvent. No deviations
6036 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6033 –6038

Angewandte
Chemie
top phase was aqueous LiCl. The reaction mixture was stirred
overnight at room temperature and the top phase was decanted off.
CH2Cl2 (200 mL) was then added and the mixture was washed three
times with H2O (500 mL). The addition of CH2Cl2 led to an improved
separation of the phases. After removal of CH2Cl2 and addition of
activated carbon, the mixture was stirred for over 48 h at room
temperature. Filtration of the mixture through a normal frit followed
by Millipore filtration (pore size 0.22 mm) afforded a colorless liquid,
which was washed with water once more, dried at 50–608C under
vacuum for 48 h and filtered through a Millipore filter (pore size
0.22 mm). The product, [BMIM]BTA, was a clear colorless liquid; its
NMR data are in close agreement with those reported for an acetone
solution (see Supporting Information).[33] The water content of
[BMIM]BTA was determined by a Karl–Fischer titration by using a
Metrohm 756 KF coulometer and found to be 23 ppm.
[3]M. L. Tobe, J. Burgess,
Inorganic Reaction Mechanisms,
Longman, Essex, 1999, p. 46.
[4]R. Romeo, A. Grassi, L. M. Scolora, Inorg. Chem. 1992, 31, 4383.
[5]R. van Eldik, C. Dücker-Benfer, F. Thaler, Adv. Inorg. Chem.
2000, 49, 1.
[6]R. van Eldik, C. D. Hubbard in High Pressure Chemistry:
Synthetic, Mechanistic, and Supercritical Applications (Eds.: R.
van Eldik, F.-G. Klärner), Wiley-VCH, Weinheim, 2002,
chap. 1.
[7]A. Hofmann, D. Jaganyi, O. Q. Munro, G. Liehr, R. van Eldik,
Inorg. Chem. 2003, 42, 1688.
[8]A. Hofmann, L. Dahlenburg, R. van Eldik, Inorg. Chem. 2003,
42, 6528.
[9]a) T. Welton, Coord. Chem. Rev. 2004, 248, 21; T. Welton, Coord.
Chem. Rev. 2004, 248, 2459; b) J. Dupont, R. F. deSouza, P. A. Z.
Suarez, Chem. Rev. 2002, 102, 3667; c) M. Freemantle, Chem.
Eng. News 2004, 82(45), 44; d) Ionic Liquids in Synthesis (Eds.:
P. Wasserscheid, T. Welton), Wiley-VCH, Weinheim, 2003.
[10]E. T. Silveira, A. P. Umpierre, L. M. Rossi, G. Machado, J.
Morias, G. V. Soares, I. J. R. Baumvol, S. R. Teixeira, P. F. P.
Fichtner, J. Dupont, Chem. Eur. J. 2004, 10, 3734.
[11]M. Volland, V. Seitz, M. Maase, M. Flores, R. Papp, K.
Massonne, V. Stegmann, K. Halbritter, R. Noe, M. Bartsch, W.
Siegel, M. Becker, O. Huttenloch (BASF AG), WO 062251,200,
2003.
[12]A. Skrzypczak, P. Neta, Int. J. Chem. Kinet. 2004, 36, 4.
[13]C. Chiappe, D. J. Pieraccini, Org. Chem. 2004, 69, 6059.
[14]K. Swiderski, A. McLean, C. M. Gordon, D. H. Vaughan, Chem.
Commun. 2004, 590.
[15]C. Daguenet, P. J. Dyson, Organometallics 2004, 23, 6080.
[16]P. Illner, A. Zahl, R. Puchta, N. J. R. van Eikema Hommes, P.
Wasserscheid, R. van Eldik, J. Organomet. Chem. 2005, 690,
3567.
Instrumentation and measurements: [BMIM]BTA was dried and
degassed before use. Its polarity was determined by using the
solvatochromic Reichardtꢀs dye (see the Supporting Information).
The dimensionless normalized ENT values of the solvent were
calculated according to reference [34].
Carlo Erba Elemental Analysers 1106 and 1108, and Bruker
Avance DPX 300 and DRX 400 NMR spectrometers were used for
chemical analysis and compound characterization, respectively. The
UV/Vis spectra for the study of slow reactions were recorded on a
Varian Cary 1G spectrophotometer equipped with a cell holder and
thermostat. For the kinetic measurements on fast reactions an
Applied Photophysics SX 18MV stopped-flow instrument was used.
The wavelengths used are listed in the Supporting Information.
Kinetic experiments at elevated pressure (1 to 130 MPa) were
performed in a laboratory-made high-pressure stopped-flow instru-
ment[35] or a Shimadzu UV-2101-PC spectrophotometer equipped
with a high-pressure cell designed for the study of slow reactions.[36]
The temperature of the instruments was controlled within an accuracy
of Æ 0.1 K. TU and IÀ (as LiI) were used as entering nucleophiles.
Their high nucleophilicity prevents the back reaction. The starting
complex was dissolved in water and methanol that contained LiCl
(0.002m) to avoid its solvolysis. Trifluoromethanesulfonic acid (0.01m)
was added to avoid formation of hydroxo or alcoholate ions, which
could compete in the chloride displacement. In the case of iodide as
nucleophile and methanol and water as solvent, the ionic strength was
kept at 0.2m (LiClO4). The reactions were studied under pseudo-first-
order conditions by using at least a tenfold excess of the nucleophiles.
All listed rate constants represent an average value of at least five
kinetic runs for each experimental condition.
Calculations: Structures were fully optimized at B3LYP/
LANL2DZp[24–26] and characterized by calculation of vibrational
frequencies. Relative energies include corrections for zero-point
vibrational energy differences. Solvent effects were probed with
IPCM single-point energy calculations by using the default param-
eters, that is, water as solvent.[29] The Gaussian program was used.[37]
Supporting Information available: Yields, elemental analyses,
and 1H NMR data for synthesized compounds; plots showing the
concentration, temperature, and pressure dependence of the
observed rate constants for the reactions between complex 1 and
thiourea or iodide.
[17]S. J. PꢀPool, M. A. Klingshirn, R. D. Rogers, K. H. Shaughnessy,
J. Organomet. Chem. 2005, 690, 3522.
[18]M. D. Sliger, S. J. PꢀPool, R. K. Traylor, J. McNeill III, S. H.
Young, N. W. Hoffman, M. A. Klingshirn, R. D. Rogers, K. H.
Shaughnessy, J. Organomet. Chem. 2005, in press.
[19]D. J. Adams, P. J. Dyson, S. J. Tavener, Chemistry in Alternative
Reaction Media, Wiley, Chichester, 2004, p. 81.
[20]It has been pointed out that corrections made to second-order
rate constants to compensate for changes in solvent density are
incorrect as all kinetic parameters are defined in terms of
concentration units under ambient conditions, that is, the
conditions under which all solutions were prepared, see: S. D.
Hamann, W. J. le Noble, J. Chem. Educ. 1984, 61, 658.
[21]No corrections were made to compensate for the temperature
and pressure dependence of the solvent viscosity as the inves-
tigated bimolecular reaction is slow and far away from a
diffusion-controlled chemical process for which viscosity
dependence is to be expected. It has been demonstrated
elsewhere that slow bimolecular reactions in general do not
exhibit meaningful viscosity dependence, see: C. F. Weber, R.
van Eldik, J. Phys. Chem. A 2002, 106, 6904.
[22]C. Reichardt,
Solvents and Solvent Effects in Organic
Received: April 16, 2005
Revised: June 19, 2005
Published online: August 22, 2005
Chemistry, Wiley-VCH, Weinheim, 2003, p. 163.
[23]C. S. Consorti, P. A. Z. Suarez, R. F. de Souza, R. A. Burrow,
D. H. Farrar, A. J. Lough, W. Loh, L. H. M. da Silva, J. Dupont,
J. Phys. Chem. B 2005, 109, 4341.
[24]a) P. J. Stevens, F. J. Devlin, C. F. Chablowski, M. J. Frisch, J.
Phys. Chem. 1994 98, 11623; b) A. D. Becke, J. Chem. Phys.
1993, 98, 5648; c) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988,
37, 785.
Keywords: density functional calculations · ionic liquids ·
.
kinetics · platinum · reaction mechanisms
[25]P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270, 284, 299.
[1]B. Lippert, Cisplatin. Chemistry and Biochemistry of a Leading
[26]S. Huzinaga,
Gaussian Basis Sets for Molecular Calculations,
Anticancer Drug, Wiley-VCH, Weinheim, 1999.
[2]U. Fekl, K. I. Goldberg, Adv. Inorg. Chem. 2003, 54, 260.
Elsevier, Amsterdam, 1984.
Angew. Chem. Int. Ed. 2005, 44, 6033 –6038
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6037

Communications
[27]The performance of the computational level employed in this
study is well documented, see for example: S. Klaus, H.
Neumann, H. Jiao, A. Jacobi von Wangelin, D. Gördes, D.
Strübing, S. Hübner, M. Hately, C. Weckbecker, K. Huthmacher,
T. Riermeier, M. Beller, J. Organomet. Chem. 2004, 689, 3685 –
3700; R. W. Saalfrank, C. Deutscher, H. Maid, A. M. Ako, S.
Sperner, T. Nakajima, W. Bauer, F. Hampel, B. A. Heß, N. J. R.
van Eikema Hommes, R. Puchta, F. W. Heinemann, Chem. Eur.
J. 2004, 10, 1899 – 1905.
[28]S. W. A. Bligh, A. Bashall, C. Garrud, M. McPartlin, N. Wardle,
K. White, S. Padhye, V. Barve, G. Kundu, Dalton Trans. 2003,
184.
[29]J. B. Foresman, T. A. Keith, K. B. Wiberg, J. Snoonian, M. J.
Frisch, J. Phys. Chem. 1996, 100, 16098 – 16104; the default e =
78.39 (H2O) was used throughout.
[30]R. Ziessel, J.-M. Lehn, Helv. Chim. Acta 1990, 73, 1149.
[31]W. Baker, K. M. Buggle, J. F. W. McOmie, D. A. M. Watkins, J.
Chem. Soc. 1958, 3594.
[32]J. G. Huddleston, A. E. Visser, W. M. Reichert, H. D. Willauer,
G. A. Broker, R. D. Rogers, Green Chem. 2001, 3, 156.
[33]P. Bonhꢁte, A.-P. Dias, N. Papageorgiou, K. Kalyanasundaram,
M. Grätzel, Inorg. Chem. 1996, 35, 1168.
[34]Reference [22,]p. 417.
[35]R. van Eldik, W. Gaede, S. Wieland, J. Kraft, M. Spitzer, D. A.
Palmer, Rev. Sci. Instrum. 1993, 64, 1355.
[36]M. Spitzer, F. Gärtig, R. van Eldik, Rev. Sci. Instrum. 1988, 59,
2092.
[37]Gaussian98 (RevisionA11.3), M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G.
Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C.
Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin,
M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R.
Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J.
Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma,
N. Rega, P. Salvador, J. J. Dannenberg, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V.
Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong,
J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle,
J. A. Pople, Gaussian, Inc., Pittsburgh, PA, 1998.
6038 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6033 –6038