Hard/soft selectivity in ligand substitution reactions of β-diketonate platinum(II) complexes

Article abstract of DOI:10.1039/b909209a

The reactivity of platinum(II) complexes of the type [PtCl(O,O′-acac) (L)] (1) and [Pt(O,O′-acac)(γ-acac)(L)] (2) (L = DMSO, a; DMS, b), with a range of hard and soft nucleophiles such as dimethylsulfide (DMS, b), triphenylphosphine, (PPh₃, c),

Full text of DOI:10.1039/b909209a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Hard/soft selectivity in ligand substitution reactions of b-diketonate
platinum(II) complexes†
Sandra A. De Pascali,a Paride Papadia,a Serena Capoccia,a Luciano Marchio`,b Maurizio Lanfranchi,b
Antonella Ciccaresea and Francesco P. Fanizzi*a,c
Received 11th May 2009, Accepted 7th July 2009
First published as an Advance Article on the web 31st July 2009
DOI: 10.1039/b909209a
The reactivity of platinum(II) complexes of the type [PtCl(O,O¢-acac)(L)] (1) and [Pt(O,O¢-acac)-
(g-acac)(L)] (2) (L = DMSO, a; DMS, b), with a range of hard and soft nucleophiles such as
2
dimethylsulfide (DMS, b), triphenylphosphine, (PPh3, c), ethylene (h -C2H4, d), carbon monoxide (CO,
e), pyridine (py, f), and guanosine (Guo, g) has been investigated. Interestingly, the complexes 1a and 1b
undergo selective substitution of the chloro or sulfur ligand depending on the hard/soft character of
the incoming nucleophile. The soft incoming ligand replaces the softer one and the hard ligand replaces
the harder one, giving [PtCl(O,O¢-acac)(L)] complexes (1b, 1c, 1d and 1e in the reaction of 1a with L =
2
DMS, PPh3, h -C2H4, CO, respectively), and [Pt(O,O¢-acac)(DMSO)(L¢)] (3f, 3g) and [Pt(O,O¢-acac)-
(DMS)(L¢)] (4f, 4g) species in the reaction of 1a and 1b with L¢ = py and guo, respectively. In the cases
of 2a and 2b complexes, where the s-bonded acac (g-acac) replaces the chloro ligand, only in the
presence of an incoming soft nucleophile substituting the soft sulfur ligand the reaction occurs.
Equilibrium constants for the substitution reactions were measured by 1H NMR spectroscopy. Variable
temperature 1H NMR spectroscopy studies, performed for the reaction of 1a and 2a complexes with
DMS, revealed that the selective substitution of DMSO with DMS takes place in both cases, according
to a second-order kinetic law. The calculated values of DH‡ and DS‡ are consistent with an associative
mechanism. NMR spectroscopic characterization (1H, 13C, 195Pt, 31P) for the complexes and crystal
structures of isolated complexes ([PtCl(O,O¢-acac)(L)] (1) and [Pt(O,O¢-acac)(g-acac)(L)] (2), L =
DMSO, 1a and 2a; L = DMS, 1b and 2b; L = PPh3, 1c and 2c) are herein reported and discussed.
Introduction
Recently, we have reported the synthesis of a novel family of
Pt(II) b-diketonate complexes, which exhibit cytotoxic activity
higher than cisplatin on several cancer cell lines.1 Interestingly, the
activity appears to be related to the reaction with non-genomic
biological targets.2–4 These complexes contain a single chelated
Scheme 1 Chemical structure of O,O¢-chelated acac complexes: 1a and
(O,O¢-acac) (1), or one chelated and one s-bonded (g-acac)5,6
2a, L = DMSO; 1b and 2b, L = DMS.
(2) acetylacetonate. The coordination sphere is completed by one
chloride (Cl-) and one sulfur ligand (dimethylsulfoxide, DMSO,
1a, or dimethylsulfide, DMS, 1b), or simply by the soft sulfur
ligand (DMSO, 2a or DMS, 2b) in type 1 and 2 complexes,
respectively (Scheme 1). The [PtCl(O,O¢-acac)(DMS)] (1b) and
[Pt(O,O¢-acac)(g-acac)(DMS)] (2b) complexes were easily pre-
pared by reaction of [PtCl(O,O¢-acac)(DMSO)] (1a) and [Pt(O,O¢-
acac)(g-acac)(DMSO)] (2a) with excess DMS. Interestingly, in
the case of [PtCl(O,O¢-acac)(DMSO)] (1a), this substitution is
selective for the softer DMSO among the two possible leaving
ligands (DMSO and Cl-).
Selectivity in ligand substitution reactions is an important topic,
widely studied in coordination and organometallic chemistry. The
trans7,8 and cis9,10 effects have often been used in order to explain
selectivity in substitution reactions involving square-planar Pt(II)
complexes.
aDipartimento di Scienze e Tecnologie Biologiche ed Ambientali, Universita`
del Salento, prov.le Lecce/Monteroni, I-73100, Lecce, Italy. E-mail:
fp.fanizzi@unisalento.it; Fax: +39 832 298 626; Tel: +39 832 298 867
However, in most cases the studied systems were characterized
by the presence of amines as non-leaving ligands. Selective
substitution reactions of Pt(II) complexes with soft ligands
have been previously reported for systems having four mon-
odentate ligands coordinated to the metal in a square planar
arrangement.11–14 Moreover, selective substitution of the p bonded
Hacac (enolic form) with soft (olefins, CO, PPh3) and hard
bDipartimento di Chimica Generale ed Inorganica, Chimica Analitica,
Chimica Fisica, Parco Area delle Scienze 17A, 43100, Parma, Italy
cConsortium C.A.R.S.O., Cancer Research Center, I-70100, Valenzano-Bari,
Italy
† Electronic supplementary information (ESI) available: Selected bond
lengths and angles for 1a–1c, selected bond lengths and angles for 2a–
2c, plot of pseudo-first-order rate constants as a function of [DMS]
for reaction 1a→1b. CCDC reference numbers 713088–713093. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/b909209a
2
(py) ligands, was also observed for [PtCl(O,O¢-acac)(h -Hacac)],
obtained from [PtCl(O,O¢-acac)(g-acac)]- by protonation of the s
7786 | Dalton Trans., 2009, 7786–7795
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
bonded g-acac.15 On the other hand the reaction of [PtCl(en)(L)]Cl
(en = N,N¢-chelate ethylenediamine, L = DMSO, DMS) with
a range of charged and neutral, soft or hard nucleophiles,16,17
regularly results in the substitution of the chloro ligand. In this
context, our O,O¢-acac chelate complexes represent, to the best
of our knowledge, the first Pt(II) complexes with a symmetric
O,O¢-donor carrier ligand showing selective reactivity towards
hard or soft nucleophiles. In this work, the peculiar reactivity
of the novel acac complexes has been studied from the kinetic
and thermodynamic viewpoint and extended with a range of
hard and soft nucleophiles such as dimethylsulfide (DMS),
1c-2c complexes reported in the experimental section were de-
veloped, taking advantage of the selective reactivity showed by
1a-2a compounds towards soft nucleophiles. Indeed, [PtCl(O,O¢-
acac)(DMSO)] (1a) and [Pt(O,O¢-acac)(g-acac)(DMSO)] (2a)
complexes give selective substitution of DMSO in the presence
of incoming soft ligand such as DMS or PPh3. For the synthesis
of 1b and 2b complexes, an excess of DMS was added to 1a and
2a, in order to complete the substitution reaction. However, in the
case of 1c and 2c, a stoichiometric amount of PPh3 is required,
in order to avoid the further chloride substitution leading to the
coordination of a second phosphine ligand. Interestingly, [Pt(O,O¢-
acac)(g-acac)(PPh3)] (2c) was previously obtained only by reaction
of [Pt(O,O¢-acac)2] with PPh3.18,19 The substitution reaction of
the soft sulfur ligand was also used to obtain the ethylene (1d-
2d) and CO (1e-2e) species by reacting 1a-2a or 1b-2b in the
presence of excess gaseous ligand. Complexes (1b-2b and 1c-2c)
2
triphenylphosphine (PPh3), ethylene (h -C2H4), carbon monoxide
(CO), pyridine (py), and guanosine (Guo). NMR characterization
(1H, 13C, 195Pt, and 31P) and crystal structures are reported and
discussed for all isolated complexes ([PtCl(O,O¢-acac)(L)] (1) and
[Pt(O,O¢-acac)(g-acac)(L)] (2), L = DMSO, 1a and 2a; L = DMS,
1b and 2b; L = PPh3, 1c and 2c). Where possible, equilibrium and
kinetic constants for the substitution reactions (monitored by 1H
NMR spectroscopy) have been measured. Moreover, activation
parameters (DH‡, DS‡ and DG‡) for the DMSO substitution by
DMS for 1a and 2a complexes have been obtained by variable
temperature 1H NMR spectroscopy.
1
were isolated and characterized by H, 13C, 195Pt, and 31P NMR
spectroscopy and X-ray diffraction crystal structure. The ethylene
(1d-2d) and CO (1e-2e) complexes were characterized in solution
by 1H NMR. Relevant NMR data for all complexes are reported
in Table 1 (1H), Table 2 (13C), and Table 3 (195Pt). 1H NMR spectra
of 1c, 1d, 1e, 2c and 13C NMR spectra of 2c were found consistent
with those already reported.15,18,19 The 1H NMR spectra in CDCl3
of 1b-e and 2b-e complexes show the characteristic signal pattern
of an asymmetric O,O¢-acac chelate already observed for 1a-2a.
This consists of two singlets, assigned to the two methyl groups,
and one more deshielded singlet for gCH, in agreement with the
aromatic character of the O,O¢-acac chelate metalla-cycle. NOE
cross-peaks with DMS or PPh3 hydrogens in the 2D 1H NOESY
spectra of 1b-2b and 1c-2c allow to assign the lower frequency
Results and discussion
Synthesis and NMR spectroscopy characterization
The synthesis and the spectroscopic characterization of 1a and
2a complexes were previously reported,2 their crystal structures
will be here discussed. The synthetic procedures of 1b-2b and
Table 1 1H NMR chemical shifts of acac complexes with soft/hard liganda
2
Solvent
H/Cg
Me/acac
L = DMSO, DMS, PPh3, h -C2H4, py, guo
1a
2a
CDCl3
CD3OD
CDCl3
5.56 (O-bonded)
5.56 (O-bonded)
5.53 (O-bonded)
4.79 [120] (g-bonded)
5.69 (O-bonded)
4.82 [120] (g-bonded)
5.48 (O-bonded)
5.62 (O-bonded)
5.47 (O-bonded)
4.88 [124] (g-bonded)
5.62 (O-bonded)
4.90 [120] (g-bonded)
5.39 (O-bonded)
5.42 (O-bonded)
3.94 dd (5.8)ꢀ [110] (g-bonded)
5.55 (O-bonded)
5.53 (O-bonded)
3.88[114] (g-bonded)
5.68 (O-bonded)
5.63 (O-bonded)
4.59 [132] (g-bonded)
5.48 (O-bonded)
5.92 (O-bonded)
5.99 (O-bonded)
5.95 (O-bonded)
5.50 (O-bonded)
5.78 (O-bonded)
2.06; 2.02# (O-bonded)
2.05; 2.02# (O-bonded)
2.00; 1.95# (O-bonded)
2.29 (g-bonded)
3.44 [40]* (DMSO)
3.45 [20] (DMSO)
3.31 [19] (DMSO)
CD3OD
2.01; 2.00# (O-bonded)
2.26 (g-bonded)
3.35 [20] (DMSO)
1b
2b
CDCl3
CD3OD
CDCl3
1.97; 1.88# (O-bonded)
1.92; 1.90# (O-bonded)
1.95; 1.89# (O-bonded)
2.20 (g-bonded)
2.34 [48] (DMS)
2.31 [47] (DMS)
2.29 [51] (DMS)
CD3OD
1.96; 1.95# (O-bonded)
2.19 (g-bonded)
2.31 [50] (DMS)
1c
2c
CDCl3
CDCl3
2.02; 1.47# (O-bonded)
1.98; 1.52# (O-bonded)
2.13 (g-bonded)
7.73m o, 7.39 m m, 7.45 m p, (PPh3) d31P = 0.40, JPt-P = 4195
7.69m o, 7.40 m m, 7.45 m p (PPh3) d31P = 4.51, JPt-P = 4492
2
1d
2d
CDCl3
CDCl3
2.16; 1.95# O-bonded
2.10; 1.96# O-bonded
2.22 g-bonded
4.47[64] (h -C2H4)
2
4.02[65] (h -C2H4)
1e
2e
CDCl3
CDCl3
2.12; 2.10# (O-bonded)
2.08; 2.06# (O-bonded)
2.27 (g-bonded)
3f
CDCl3
CD3OD
D2O
CD3OD
CDCl3
CD3OD
1.94; 1.87 O-bonded
2.14; 2.06 O-bonded
2.12; 2.00 O-bonded
2.14; 2.05 O-bonded
1.93; 1.86 O-bonded
2.06; 1.97 O-bonded
3.46 [20] (DMSO); 8.75 [40] m o, 7.43 m m, 7.87 m p (py)
3.48 [20] (DMSO); 8.84 [40] m o, 7.62 m m, 8.13 m p (py)
3.52–3.51; 3.40–3.39 (DMSO); 8.62 H8, 5.99 H1¢ (guo)
3.51–3.50; 3.40–3.49 (DMSO); 8.72 H8, 5.95 H1¢ (guo)
2.43 [42] (DMS); 8.83 [34]m o, 7.36 m m, 780 m p (py)
2.41; 2.39 (DMS); 8.61 H8, 5.93 H1¢ (guo)
3g
4f
4g
a # indicates the acac methyl group cis to the L ligand. JH-P (ꢀ) and JH-Pt (*) are reported in parentheses (Hz) and in square brackets [Hz], respectively,
where measurement was possible.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 7786–7795 | 7787
©

View Article Online
Table 2 13C NMR chemical shifts (d) of isolated complexes
account for the ethylene group. The 13C NMR spectra, acquired
for isolated complexes, confirmed the structures assigned on the
C/(O,O¢-acac)
C/(s-acac)
C/L ligand
1
basis of H NMR spectroscopic data. One bond and long range
1
2D H-13C HETCOR experiments allowed correct assignments
1a
2a
26.3 (Me), 102.3 (gCH)
44.3 (DMSO)
42.95 (DMSO)
for all the 13C resonances. In particular, O,O¢-chelated and Cg-
bonded acac groups show very different chemical shifts for the
Cg atom. The pseudo-aromatic Cg carbons of the acac chelates
resonate downfield with respect to the corresponding carbon in
the s-bonded acac (see Table 2). Carbonyls of the acac chelate
involved in the pseudo-aromatic metallacycle resonate at lower
frequencies with respect to the s-bonded acac ones. The long range
2D 1H-13C HETCOR spectra, together with the above mentioned
NOESY data, were used to correlate the carbonyls and the methyl
groups belonging to the same half of the asymmetric chelate acac.
Due to the substitution of the Cl- with the strong s-donor g-
acac ligand, the 195Pt spectra of the isolated species 2a-c show a
more deshielded signal (ca. 600–800 ppm Dd) with respect to 1a-c
complexes (Table 3). As already observed for other Pt phosphino-
complexes,24 the 1c-2c phosphino species exhibit the greater 195Pt
shielding with respect to the sulfur complexes 1a-2a and 1b-2b.
Furthermore, for both type 1 and 2 complexes the shielding of
195Pt chemical shift follows the series PPh3 > DMSO > DMS
according to the soft ligand substitution (DMSO in a, DMS in b,
and PPh3 in c).
#
=
185.1 , 185.9 (C O)
27.50, 27.3# (Me),
30.9 (Me) 42.0 (gCH)
=
102.2 (gCH)
208.5 (C O)
#
=
185.8 , 184.9 (C O)
1b
2b
1c
2c
26.5, 26.1# (Me),
22.1 (DMS)
101.7 (gCH)
#
=
184.9, 182.9 (C O)
27.1, 27.5# (Me),
30.9 (Me), 40.5 (gCH) 22.1 (DMS)
=
101.6 (gCH)
207.2 (C O)
#
=
183.7, 184.7 (C O)
27.3, 25.8# (Me),
134.6 (o),
128.0 (m),
130.9 (p)(PPh3)
101.4 (gCH)
#
=
185.7, 183.0 (C O)
27.8, 26.9# (Me),
30.9 (Me), 41.1 (gCH) 134.4 (o),
=
100.9 (gCH)
209.1 (C O)
127.9 (m),
130.7 (p)(PPh3)
#
=
183.9, 184.1 (C O)
a # indicates the acac methyl group cis to the L ligand.
Table 3 195Pt NMR chemical shifts of isolated acac complexes
d
1a
2a
1b
2b
1c
2c
-2399
-3198
-2096
Crystal structure
-2905
-2888 [4195]a
-3532 [4492]a
Molecular drawings and crystal packing of 1a, 1b, and 1c are
depicted in Fig. 1–3, whereas selected bond lengths and angles
are reported in Table S1.† In all complexes the Pt atom shows a
square-planar coordination achieved by means of a O,O¢-chelate
acetylacetonate ligand, a chlorine atom, and a S-coordinated
DMSO (1a), a S-coordinated DMS (1b), or a PPh3 molecule
(1c). Two independent molecules stacked through the Pt-acac
hexa-atomic ring are present in 1a. The Pt atoms of the two
complexes exchange two long contacts with the oxygen atoms
trans to the chlorine atom of the other complex (Pt–O range
a JP-Pt are reported in square brackets [Hz].
signal of the two methyl groups to the methyl in cis to the soft
ligand (DMS/PPh3).
Due to the PPh3 phenyl groups’ anisotropic effect, the O,O¢-
acac chelate methyl cis to PPh3 is significantly more shielded
with respect to the analogous methyl in the other complexes.
Therefore, the two singlets of the O,O¢-acac methyl groups in
the phosphine complexes 1c and 2c (2.02 and 1.47, 1.98 and
1.52 ppm for 1c and 2c, respectively) exhibit the greater Dd
(~0.5 ppm) with respect to the analogous complexes. The lower
frequency resonance of O,O¢-acac methyl group was assigned to
˚
3.610(7)–3.783(7) A). This generate 1D chains that run parallel to
the b crystallographic axis (Fig. 1). On the contrary, in the crystal
packing of 1b, adjacent molecules are arranged perpendicularly to
each other (Fig. 2). In the case of 1c, the asymmetric unit contains
two independent complexes (Fig. 3) and the main difference
between them is represented by the arrangement of the phenyl
residues of the PPh3 ligand. In fact, the C(32) atom of one phenyl
ring, belonging to unit-1, points towards the Cl(2) atom of the
other complex, whereas the corresponding C(36) atom of unit-2
points towards the O(21) oxygen atom. In 1a, 1b and 1c complexes,
the Pt-O bond distance trans to the chlorine atom is slightly shorter
than the one trans to the sulfur or phosphorous atom, suggesting a
greater trans influence of the DMSO, DMS, or PPh3 moieties with
respect to Cl-.25 The bond lengths within the acetylacetonate unit
are in agreement with the p delocalization of the six-membered
ring. In 1a the two independent molecules show practically the
same bond distances and angles, and they are comparable with
those of 1b, apart from the Pt-S bond distance that reflects the
different electronic configuration of the S atoms in DMSO and
DMS (Table S1†).26
1
the methyl in cis to the L ligand, by comparing the H NMR
data of 2d and 2e species with those already reported for other
[PtCl(O,O¢-acac)(L)] complexes (L = olefin, tertiary phosphine,
carbon monoxide).15,20–23 The 1H NMR spectra of 2b-e complexes
exhibit for the Cg-H acac the characteristic deshielded signal with
195Pt coupling for the methine and only one singlet relative to the
methyl groups, integrating for one and six protons, respectively.
Only for 2c the methine resonance is a doublet, due to the 31P
3
2
coupling (d = 3.94, JH-P = 5.8 Hz and a JH-Pt = 110 Hz). The
spectra of 1b and 2b show only one singlet for two methyl groups
of the DMS ligand with the characteristic 195Pt coupling (d = 2.29
with a 3JH-Pt of 51 Hz and d = 2.31 with a 3JH-Pt of 50 Hz for 1b and
2b, respectively). The 1c-2c complexes exhibit at high frequencies
the characteristic triphenylphosphine signal pattern and in the 31
P
NMR spectra they show only one signal at 0.40 and 4.51 ppm
1
with a JP-Pt of 4195 and 4492 Hz for 1c and 2c, respectively.
1
In the H NMR spectra of 1d and 2d complexes the singlets at
The molecular drawings of 2a, 2b and 2c are reported in
Fig. 4–6 and selected bond lengths and angles are reported in
4.47 (2JH-Pt of 64 Hz) and 4.02 ppm (2JH-Pt of 65 Hz), respectively,
7788 | Dalton Trans., 2009, 7786–7795
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Fig. 3 Molecular structure of 1c. Thermal ellipsoids are drawn at the
30% probability level. The two complex molecules that represent the
asymmetric unit are reported.
Fig. 1 Molecular structure of 1a showing the two independent complexes.
Thermal ellipsoids are drawn at the 30% probability level (above); crystal
packing of 1a projected along the b crystallographic axis. Hydrogen atoms
were removed for clarity (below).
Fig. 4 Molecular structure of 2a. Thermal ellipsoids are drawn at the
30% probability level.
Table S2.† The molecular structure of 2a was recently reported
showing an asymmetric unit comprised of three molecular units
(space group P-1).2 In the present case, the asymmetric unit
is represented by one complex molecule (space group P21/n)
and the geometric parameters are essentially equivalent with
those previously reported (Fig. 4). In these complexes the square
planar coordination of the metal is defined by one O,O¢ chelate
acetylacetonate ligand, a Cg-bonded acetylacetonate, and a
S-coordinated DMSO (2a), a DMS (2b), or a PPh3 molecule
(2c). The bidentate acetylacetonate moiety presents bond lengths
that are in agreement with the p delocalization as found for
1a-c, whereas the Cg bonded one exhibits a tetrahedral geometry
at the Cg carbon atom, and C-O distances consistent with a
Fig. 2 Molecular structure of 1b. Thermal ellipsoids are drawn at the
30% probability level (above); crystal packing of 1b projected along the a
crystallographic axis (below).
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 7786–7795 | 7789
©

View Article Online
Fig. 5 Molecular structure of 2b. Thermal ellipsoids are drawn at the
30% probability level.
Fig. 7 Portion of the crystal packing of 2c, with indicated the pseudo–
center of symmetry. Symmetry codes: ¢ = x; -y; 1/2 + z, ¢¢ = x; 1 - y;
z - 1/2.
DMS) account for the different trans influence of the soft ligands
measured by crystallographic data in other model systems.27
Thermodynamic and kinetic studies of the substitution reactions
Thermodynamic and kinetic studies of ligand substitution re-
actions with hard/soft nucleophiles for 1a-2a and 1b-2b were
performed by 1H NMR spectroscopy, as described in the ex-
perimental section. Reaction of [PtCl(O,O¢-acac)(DMSO)] (1a)
complex with DMS to give [PtCl(O,O¢-acac)(DMS)] (1b) is fast
in CDCl3, reaching almost immediately the equilibrium, while
reaction of [Pt(O,O¢-acac)(g-acac)(DMSO)] (2a) to give [Pt(O,O¢-
acac)(g-acac)(DMS)] (2b) in the same solvent needs approximately
3 days to reach the steady-state (Fig. 8). On the other hand,
Fig. 6 Molecular structure of 2c. Thermal ellipsoids are drawn at the 30%
probability level. One of the two independent molecules is reported.
double bond character (Table S2†). The main difference between
2a and 2b is represented by the orientation of the Cg bonded
acetylacetonate ligand with respect to the S-donor one. In fact,
in 2a, the Cg-H is directed towards the DMSO molecule and
is bisecting the O(5)-S-C(11) angle, whereas in 2b it is oriented
towards the O(3) atom of the chelate acetylacetonate ligand.
As far as the Pt-O distances are concerned, 2a exhibits the
longer distance for the bond in trans to the DMSO molecule,
whereas in 2b and 2c, the longer Pt-O distances are those in
trans to the Cg carbon. This is a consequence of the greater
trans influence of the Cg-bonded acetylacetonate group with
respect to the chloride ion that is present in 1a-c. Interestingly,
the asymmetric unit of 2c is comprised of two complex molecules
that are strictly related by a pseudo-center of symmetry, and on a
first stage, this cast some doubts on the space group assignment
(Fig. 7). Nevertheless, by inspecting the crystal packing it became
evident that this symmetry is only local, since it does not apply
to the rest of the lattice but only to the two molecular units.
As an example, the C(251) atom of unit-1 points towards the
center of the penta-atomic ring containing Pt(2)¢¢, whilst the
pseudo-centrosymmetric C(212) atom points towards O(11)¢.
Interestingly, the Pt-O distances trans to the phosphorus or sulfur
in type 1 (PPh3 > DMSO ≥ DMS) and 2 (PPh3 > DMSO >
1
comparison of the equilibrium constants evaluated by H NMR
spectroscopy at 25◦C, for the two reactions 1a→1b (K1a→1b
=
6.5 ¥ 101
1.3 ¥ 101) and 2a→2b (K2a→2b = 17.0 ¥ 101
3.4 ¥ 101) indicates that the substitution of the sulfur ligand is
thermodynamically favoured in the case of 2a with respect to 1a.
Therefore, although the two ligands cis to DMSO (Cl- and g-
acac in the case of 1a and 2a, respectively) formally share the
same charge, the strongly s-donating ligand g-acac favours the
DMSO substitution reaction in the presence of the incoming DMS
ligand (K2a→2b greater than K1a→1b). In order to rule out a solvent
effect (aprotic vs. protic)28 on the selectivity of the substitution, we
followed the reactions of 1a and 2a with DMS in CD3OD by 1H
NMR spectroscopy. The same substitution reactions (1a→1b and
2a→2b) and hard/soft selectivity (leading to the exclusive DMSO
over Cl- substitution for 1a→1b) were observed also in this case.
Moreover, the reactions 1a→1b and 2a→2b were performed on a
preparative scale to give, in the presence of excess DMS, the final
7790 | Dalton Trans., 2009, 7786–7795
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
is slowly replaced by the counterion, affording [PtCl2(en)], as
normally occurs in the absence of the excess sulfur ligand.29,30 The
kinetic constants and the activation parameters for the reactions
1a→1b and 2a→2b were also calculated by variable temperature
1H NMR spectroscopy (in the range 288.15–308.15 K). Under
the conditions reported in the experimental section, the reactions
were completed in 20–60 min, and a satisfactory pseudo-first-order
behaviour was observed for both reactions. The plot of pseudo-
first-order rate constants, kobs, against the concentration of DMS
indicates a first order in the incoming ligand for both 1a→1b and
2a→2b reactions without any significant intercept. Thus, both the
reactions follow the same rate law
kobs = k2 [DMS]
The second-order rate constants k2 were obtained at different
temperatures (Table 6) and were used for calculating the corre-
sponding activation parameters, DG‡, DH‡ and DS‡, reported in
Table 7. Interestingly, the obtained data show for both reactions,
within the experimental error, the same value of DS‡ (-124
6
and -124 12 for 1a→1b and 2a→2b, respectively) and only a
slight difference in the enthalpy contents (47 2 and 50 3 for
1a→1b and 2a→2b, respectively). These data imply an analogous
associative mechanism for DMS promoted DMSO substitution
reaction in 1a and 2a complexes, with a similar geometrical
arrangement in the transition state.
Fig. 8 1H NMR spectra in CDCl3 (400 MHz) for the reaction of 2a with
DMS. t = 0, 2a and free DMS; t = 24 h, 2a reacts with DMS (#) giving 2b
and free DMSO (*); t = 3 d, steady-state of the reaction.
In order to confirm the reactivity and hard/soft selectivity in
the ligand substitutions for these O,O¢-chelated-acac platinum
systems, both the DMSO (1a and 2a) and DMS (1b and 2b)
complexes were checked by NMR in the reactions with other
products (1b and 2b). Interestingly, these specific reactivity and
selectivity have never been observed for the cationic platinum com-
plexes [PtCl(en)(DMSO)]Cl. Indeed, when [PtCl(DMSO)(en)]Cl
(the simplest analogue of 1a, having a chelating diamine ligand in
place of a O,O¢-chelated acetylacetonate) was reacted with DMS
in CD3OD, no DMSO substitution by DMS was detected. Even in
the presence of excess DMS, the DMSO in [PtCl(DMSO)(en)]Cl
2
soft nucleophiles such as triphenylphosphine (PPh3), ethylene (h -
C2H4) and carbon monoxide (CO) in CDCl3. In the presence of sto-
ichiometric amounts of PPh3, both 1a and 2a, as well as 1b and 2b,
gave in all cases almost immediately the substitution of the sulfur
ligand and the quantitative formation of [PtCl(O,O¢-acac)(PPh3)]
Table 4 Crystal data and structure refinement for 1a-1ca
1a
1b
1c
Empirical formula
Formula weight
T (K)
C7H13ClO3PtS
407.77
293(2)
0.71073
Monoclinic
P21/c
17.718(11)
7.245(8)
18.793(13)
90
111.57(2)
90
C7H13ClO2PtS
391.77
293(2)
0.71073
Monoclinic
P21/c
C23H22ClO2PPt
591.92
293(2)
0.71073
Triclinic
P-1
9.948(1)
12.738(2)
18.411(2)
101.68(1)
100.25(1)
106.83(2)
2116.0(5)
4, 1.858
˚
l (A)
Crystal system
Space group
˚
a (A)
6.268(2)
˚
b (A)
11.781(6)
15.140(16)
90
99.41(2)
90
1103(1)
4, 2.359
13.115
˚
c (A)
a (deg)
b (deg)
g (deg)
3
˚
V (A )
2243(3)
8, 2.415
12.907
Z, r(calc) (Mg m-3)
m, mm-1
6.849
q range (deg)
Refl. Coll./uniq.
data/restraints/param.
GOF on F2
R1
3.01 to 27.01
5007/4877 [Rint = 0.0711]
4877/0/235
1.018
0.0476
0.1284
3.23 to 30.00
3314/3213 [Rint = 0.0438]
3213/0/113
0.982
0.0321
0.0707
1.17 to 26.00
23015/8301 [Rint = 0.0566]
8301/0/508
0.937
0.0281
0.0449
wR2
-3
˚
largest diff. peak/hole (e A )
2.653 and -2.886
0.973 and -1.308
1.109 and -0.951
ꢀ
ꢀ
ꢀ
ꢀ
1
2
a R1 = ꢀFo| - |Fc|/ |Fo|, wR2 = [ [w(Fo - Fc2)2]/ [w(Fo2)2]] , w = 1/[s (Fo2) + (aP)2 + bP], where P = [max(Fo2,0) + 2Fc2]/3.
2
2
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 7786–7795 | 7791
©

View Article Online
Table 5 Crystal data and structure refinement for 2a-2ca
2a
2b
2c
Empirical formula
Formula weight
T (K)
C12H20O5PtS
471.43
293(2)
0.71073
Monoclinic
P21/n
C12H20O4PtS
455.43
293(2)
0.71073
Monoclinic
C2/c
C29H30Cl3O4PPt
774.94
293(2)
0.71073
Monoclinic
P2/c
17.067(6)
12.310(4)
27.577(9)
90
92.55(1)
90
˚
l (A)
Crystal system
Space group
˚
a (A)
7.910(4)
8.766(1)
˚
b (A)
13.000(5)
15.711(8)
90
101.63(2)
90
1582(1)
4, 1.979
9.010
14.185(2)
25.006(3)
90
93.561(2)
90
3103.4(7)
8, 1950
9.180
˚
c (A)
a (deg)
b (deg)
g (deg)
3
˚
V (A )
5788(3)
8, 1.779
5.214
Z, r(calc) (Mg m-3)
m, mm-1
q range (deg)
Refl. Coll./uniq.
data/restraints/param.
GOF on F2
R1
3.06 to 24.99
2878/2779 [Rint = 0.0765]
2779/0/172
1.004
0.0409
0.0615
1.63 to 26.39
4838/3256 [Rint = 0.0000]
3256/0/170
1.008
0.0493
0.1398
1.19 to 27.00
64056/12637 [Rint = 0.0859]
12637/0/621
1.002
0.0384
0.0526
wR2
-3
˚
largest diff. peak/hole (e A )
1.733 and -1.463
1.331 and -1.407
1.255 and -1.259
ꢀ
ꢀ
ꢀ
ꢀ
1
2
a R1 = ꢀFo| - |Fcꢀ/ |Fo|, wR2 = [ [w(Fo - Fc2)2]/ [w(Fo2)2]] , w = 1/[s (Fo2) + (aP)2 + bP], where P = [max(Fo2,0) + 2Fc2]/3.
2
2
Table 6 Kinetic constants vs. T for the substitution reaction of 1a and 2a with DMS
T (K)
k (1a)
Adj R2
k (2a)
Adj R2
308.15
303.15
298.15
293.15
288.15
2.7 ¥ 10-2 1.4 ¥ 10-2
2.0 ¥ 10-2 1.1 ¥ 10-2
1.4 ¥ 10-2 0.8 ¥ 10-2
1.0 ¥ 10-2 0.5 ¥ 10-2
0.7 ¥ 10-2 0.3 ¥ 10-2
0.9918
0.9920
0.9907
0.9973
0.9997
6.5 ¥ 10-3 2.9 ¥ 10-3
4.4 ¥ 10-3 1.7 ¥ 10-3
3.6 ¥ 10-3 1.3 ¥ 10-3
2.4 ¥ 10-3 0.8 ¥ 10-3
1.5 ¥ 10-3 0.5 ¥ 10-3
0.9771
0.9833
0.9951
0.9992
0.9989
Table 7 Activation parameters for the substitution reactions 1a→1b and
2a→2b
k (measured at
298.15 K)
DH‡
DS‡
DG‡
(298 K)
(kJ mol-1) (J K-1 mol-1)
1a→1b 1.4 ¥ 10-2 0.8 ¥ 10-2
2a→2b 3.6 ¥ 10-3 1.3 ¥ 10-3
47
50
2
3
-124
6
84
87
3
7
-124 12
(1c) and [Pt(O,O¢-acac)(g-acac)(PPh3)] (2c), respectively. It should
be noted that also in the case of 1a and 1b, the PPh3 substitution
is selective for the softer ligand (DMSO/DMS) among the two
possible leaving ligands (DMSO/DMS and Cl-) (Fig. 9). The
thermodynamic of the substitution reactions (1a/1b→1c and
2a/2b→2c) strongly favours the phosphino complexes 1c and 2c,
being the equilibrium constants K1a/1b→1c and K2a/2b→2c >> 103,
estimated from NMR data. The strongly favoured, both kinetically
and thermodynamically, substitution of the sulfur ligand (DMSO
or DMS) with PPh3 allowed us to set up an easy synthetic
procedure, in order to isolate in good yield 1c and 2c complexes.
The same reactivity and selectivity were observed also when 1a-
2a and 1b-2b were reacted with a p-electron system nucleophile,
Fig. 9 1H NMR spectra in CDCl3 for the reaction of 1b with triph-
enylphosphine giving 1c.
2
of 2a and 2b to give [Pt(O,O¢-acac)(g-acac)(h -C2H4)] (2d). This
behaviour suggests the idea that, in these systems, the driving force
ruling the reactivity is the hard/soft nature of the incoming ligand.
In all the cases of sulfur ligand displacement by ethylene,
the reactions are less favoured both kinetically (ca. two weeks
are required to reach the equilibrium) and thermodynamically
2
such as ethylene (h -C2H4) in CDCl3. Again, the sulfur ligand
displacement (DMSO or DMS) is selectively preferred over the
chloride substitution in the case of 1a and 1b, to afford [PtCl(O,O¢-
2
acac)(h -C2H4)] (1d) (Fig. 10), and takes place also in the case
7792 | Dalton Trans., 2009, 7786–7795
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
carrier ligand, two leaving ligands with different hard/soft char-
acter (Cl- and DMSO or DMS) are present, hard/soft selectivity
is observed for the first substitutions. Indeed, in type 1 complexes,
in the first instance, substitution of the harder leaving group by
an incoming hard ligand and of the softer group by an incoming
soft ligand takes place. This seems to be the case of complexes
1a and 1b in the reactions with another sulfur ligand, phosphine,
ethylene, CO, py and Guo. On the other hand, when only a soft
ligand is available for substitution (being the s bonded g-acac
essentially unavailable for substitution in the reaction conditions)
as in the cases of [Pt(O,O¢-acac)(g-acac)(L)] (2) (L = DMSO, a;
DMS, b), the substitution reaction takes place only in the presence
of an incoming soft ligand, otherwise no reaction occurs at all
(Scheme 2).
Fig. 10 1H NMR spectra in CDCl3 for the reaction of 1a with ethylene
giving 1d.
with respect to the analogous reactions, where the displacing
nucleophile is another sulfur ligand or a phosphine (K1a→1d
=
9.0 ¥ 10-1
1.8 ¥ 10-1, K2a→2d = 4.9 ¥ 10-1
9.8 ¥ 10-2,
K1b→1d = 4.5 ¥ 10-2
9.0 ¥ 10-3 and K2b→2d = 3.0 ¥ 10-3
5.4 ¥ 10-4). The reaction of 1a-2a and 1b-2b with CO gave
again, in all cases, substitution on the sulfur ligand. Selective
displacement of DMSO or DMS over the chloro ligand from
1a and 1b, respectively, gave [PtCl(O,O¢-acac)(CO)] (1e), while
[Pt(O,O¢-acac)(g-acac)(CO)] (2e) was obtained from 2a and 2b.
However, in these cases, due to both the presence of CO excess
and the slowness in reaching the steady-state for all four reactions,
the carbonyl complexes (1e, 2e) could be only spectroscopically
characterized, since a further reaction takes place. Therefore,
the equilibrium constants for 1a/1b→1e 2a/2b→2e reactions
could not be measured, nor the pure complexes 1e and 2e could
be isolated. In order to confirm this unforeseen reactivity and
substitution selectivity for this class of compounds, we also studied
1
Scheme 2 Selective reaction of O,O¢-acac complexes with soft nucle-
ophiles and nitrogen ligands.
by H NMR spectroscopy the reaction of 1a-1b and 2a-2b, with
hard nitrogen donor ligands such as pyridine (py) and guanosine
(Guo, able to undergo easily N7 coordination). Interestingly,
while 2a and 2b show negligible reactivity with both py and
Guo, 1a and 1b slowly undergo chloride substitution as the first
reaction step giving [Pt(O,O¢-acac)(DMSO)(py)]+ (3f), [Pt(O,O¢-
acac)(DMSO)(Guo)]+ (3g) and [Pt(O,O¢-acac)(DMS)(py)]+ (4f)
[Pt(O,O¢-acac)(DMS)(Guo)]+ (4g), respectively. The Cl- substi-
tution reactions by py or Guo appear both kinetically and
thermodynamically less favoured with respect to the analogous
substitutions involving the soft ligands. Moreover, at least in
the case of py, even when used stoichiometrically as incoming
nucleophile, NMR spectra in CDCl3 account for further reactions
involving 3f and 4f. These include substitution of the sulfur ligand
by coordination of a second py molecule or reentering of the
chloride in the Pt coordination sphere.
Kinetic and thermodynamic analyses for selective DMSO
substitution by DMS indicate that both 1a→1b and 2a→2b
reactions take place with a second order kinetic law and an
associative mechanism with a similar geometrical configuration in
the transition state. In principle, the strongly favoured substitution
reactions with soft ligands, in both 1 and 2 systems can be used to
set up straightforward synthetic procedures giving new O,O¢-acac
chelate Pt(II) complexes.
Experimental
Physical measurements
Elemental analyses were performed using a Carlo-Erba elemental
1
1
analyser, model 1106. H NMR, 195Pt NMR, 31P NMR, H–1H
COSY, H–1H NOESY, H–195Pt HETCOR, H–13C HETCOR,
and H–13C HETCOR long range two-dimensional experiments
1
1
1
Conclusions
1
All data herein presented confirm the hypothesis that, in the
6-membered O,O¢-chelate ring systems, [PtCl(O,O¢-acac)(L)] (1)
(L = DMSO, a; DMS, b), where, together with the O,O¢-acac
were recorded on a Bruker Avance DPX 400 MHz and on a Bruker
Avance DRX 500 MHz (CARSO). CDCl3, CD3OD, and D2O
were used as solvents, and chemical shift were referenced to TMS
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 7786–7795 | 7793
©

View Article Online
for CDCl3 and CD3OD by the residual protic solvent peaks as
internal references (d = 7.24 ppm for CDCl3 and d = 3.30 ppm for
CD3OD) and to TSP (2,2¢,3,3¢-d(4)-3-(trimethyl-silyl)-propionic
acid sodium salt) for D2O (d = 0 ppm) as internal reference. 195Pt
chemical shifts were referenced to Na2[PtCl6] (d(Pt) = 0 ppm) in
D2O as an external reference.31
Kinetics curve fitting and activation parameters calculation
The kinetics of reactions 1a→1b and 2a→2b were studied in
CDCl3 using a Bruker Avance DRX 500 MHz spectrometer
equipped with an indirect broad band triple resonance probe and
a variable temperature unit BVT300 module on the temperature
range from 288.15 to 308.15 K. The 1H NMR spectra were
acquired using a 60◦ pulse of 6 ms and a D1 delay of 5 s and 8 scans.
1
Starting materials
The progress of reactions was monitored by integrating the H
NMR signals from the O,O¢-acac methine of the Pt complexes. The
pseudo-first-order conditions were ensured by using at least 10-
folds excess of DMS with respect to the Pt starting complex. The
reactions were started by mixing solutions of Pt complex (0.0085 g;
~0.02 M) and DMS in CDCl3 to give the total solution a volume of
1 mL. The pseudo-first-order constants kobs were calculated from
the slope of linear plots ln[M] (where [M] is the concentration of
starting complex) versus time (s). After preliminary fittings of the
data with both the exact rate law integrated formula and the first
order approximation for the studied reaction, we determined that,
within the range of concentrations explored, no significant error
occurred within the pseudo-first-order approximation. Therefore,
for each temperature, concentration were fitted to a simple single
exponential decay function, of the type [C] = C0 exp(-kobst),
where kobs = k2[DMS]. In order to exclude a first order process,
we performed multiple experiments at selected temperatures
using different DMS concentration. By plotting the obtained
kobs against concentrations, we obtained a straight line with no
significant 0th order term (see supplementary material Figure S3†).
In order to calculate the activation parameters, lnk was fitted
against 1/T, obtaining a straight line with equation ln(k/T) =
DH‡/R(1/T) + DS‡/R + ln (kB/h). DG‡ werecalculated at 298.15 K
according to the equation DG‡ = DH‡ - TDS‡. All data were
fitted with the computer program OriginPro 8 by OriginLab
Corporation on a PC running Windows.
Acetylacetone (2,4-pentanedione), pyridine, guanosine, DMSO,
DMS, PPh3, carbon monoxide, ethylene and deuterated solvents
(CDCl3, CD3OD, and D2O) were used without further purifica-
tion. [PtCl(en)(DMSO)]Cl,32 [PtCl(O,O¢-acac)(DMSO)]2 (1a) and
[Pt(O,O¢-acac)(g-acac)(DMSO)]2 (2a) were prepared according to
previously reported procedures.
Syntheses of [PtCl(O,O¢-acac)(DMS)] (1b) and [Pt(O,O¢-
acac)(c-acac)(DMS)] (2b). To a chloroform (3 mL) solution of
1a or 2a (0.1 g, 0.24 mmol for 1a and 0.1 g, 0.21 mmol for 2a) a
large DMS excess (0.224 g, 3.6 mmol for 1a and 0.263 g, 4.24 mmol
for 2a) was added. The reaction mixture was left under stirring at
room temperature, overnight. The resulting yellow solution was
added of pentane (10 mL) and kept at 5 ◦C for one day up to
the formation of a yellow needles crystals for 1b and pale yellow
crystals for 2b. Finally, the crystals were isolated, washed with
pentane, and dried under vacuum. (Yield 0.075 g, 0.191 mmol,
80% for 1b. Anal. Calcd for C7H13ClO2PtS (391.773): C 21.46; H
3.34. Found: C 21.27; H 3.20; yield 0.078 g, 0.171 mmol, 82%
for 2b. Anal. Calcd for C7H13ClO2PtS (455.428): C 31.65; H 4.43.
Found: C 31.72; H 4.56).
Syntheses of [PtCl(O,O¢-acac)(PPh3)] (1c) and [Pt(O,O¢-acac)(c-
acac)(PPh3)] (2c). To a chloroform (30 mL) solution of 1a or
2a (0.1 g, 0.246 mmol for 1a and 0.1 g, 0.212 mmol for 2a) a
stoichiometric amount of PPh3 previously dissolved in 20 mL of
CHCl3 was added. The mixture was left under stirring at room
temperature for ca. 6 h. The resulting solution was concentrated
(5 mL) and added of pentane (20 mL) to allow the precipitation of
a yellow powder for 1c and a pale-yellow solid for 2c. The products
(1c and 2c) were then collected, washed and dried under vacuum.
By slow evaporation from CHCl3 solution, yellow crystals of 1c
were obtained. Yellow crystals of 2c were formed by addition of
pentane to a CHCl3 solution. (Yield 0.120 g, 0.202 mmol, 82%
for 1c. Anal. Calcd for C23H22ClO2PPt (591.924): C 46.67 H 3.75,
found C 46.10 H 3.77; yield 0.106 g, 0.162 mmol, 76% for 2c. Anal.
Calcd for C28H29O4PPt (655.579): C 51.30; H 4.46, found C 50.93;
H 4.45).
X-Ray crystallography
A summary of data collections and structure refinements is
reported in Table 4 and Table 5. Single crystal data were collected
on a Philips PW 1100 diffractometer (1a, 2a, 1b), on Bruker Smart
1000 area detector diffractometer (1c, 2c), and on Bruker Smart
APEXII area detector diffractometer (2b). All data collection
˚
were performed with the Mo Ka radiation (l = 0.71073 A).
For 1a, 2a and 1b the cell constants were obtained from a least-
square refinement of the setting angles of 24 randomly distributed
and carefully centered reflections, whereas for 1c and 2c the
cell constants were obtained from a least-square refinement of
10944, and 14330 reflections, respectively. Empirical absorption
correction was applied using the program NEWABS92333 (1a, 2a,
1b), and with the program SADABS (1c, 2c).34 The crystals of 2b
were identified as non-merohedral twins using RLATT.35 Two unit
cells could be determined where the twin law corresponds to [-1 0
0, 0 -1 0, 0.35 0 1]. Integration of the data of 2b was performed with
SAINT-Plus36 using both orientation matrices. The absorption
correction for 2b was applied using the program TWINABS,37
and each component was scaled separately. The R(int) (point
group 2/m) for the first component refined to 0.0651, and for the
second component it refined to 0.0648. Equivalent reflections were
merged if they derived from the same twin component. The ratio
Reactions in NMR tube. 0.010 g of complex (1a, 2a, 1b,
and 2b) were dissolved in deuterated solvent (1 mL) and the
resulting solution placed in a NMR tube. The reaction with
added soft (DMS, DMSO, PPh3, ethylene, carbon monoxide) and
hard (pyridine, guanosine) ligands in a stoichiometric amount was
time-monitored recording 1H NMR spectra. The thermodynamic
measurements were performed using a Bruker Avance DPX
1
400 MHz at 298.15 K. The H NMR spectra were conducted
using a 60◦ pulse of 4.8 ms and a D1 delay of 5 s and 16 scans. The
progress of reactions was monitored by integrating the 1H NMR
signals from the O,O¢-acac methine of the Pt complexes.
7794 | Dalton Trans., 2009, 7786–7795
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
of the twin components was refined to 0.87/0.13. The structures
of all compounds were solved by direct methods (SIR9738 and
SIR200439), and refined with full-matrix least-squares (SHELX-
97),40 using the WinGX software package.41 Non-hydrogen atoms
were refined anisotropically for all compounds, and the hydrogen
atoms were placed at their calculated positions. Graphical material
was prepared with the ORTEP3 for Windows42 and Mercury CSD
2.043 programs. Full tables of bond lengths and angles, atomic
positional parameters, anisotropic displacement parameters are
given in the supplementary material. Details of the crystal
structure investigations (excluding structure factors) are deposited
in the Cambridge Crystallographic Data Centre: CCDC 713088–
713093.†
17 M. Bonivento, L. Canovese, L. Cattalini, G. Marangoni, G. Michelon
and M. L. Tobe, Inorg. Chem., 1981, 20, 3728–3730.
18 T. Ito, T. Kiriyama and A. Yamamoto, Bull. Chem. Soc. Jpn., 1976, 49,
3250–3256.
19 T. Ito, T. Kiriyama, Y. Nakamura and A. Yamamoto, Bull. Chem. Soc.
Jpn., 1976, 49, 3257–3264.
20 S. Baba, T. Ogura and S. Kawaguchi, Bull. Chem. Soc. Jpn., 1974, 47(3),
665–668.
21 A. C. Jesse, D. J. Stufkens and K. Vrieze, Inorg. Chim. Acta, 1979, 32,
87–91.
22 J. Ashley-Smith, Z. Douek, B. F. G. Johnson and J. Lewis, J. Chem.
Soc. Dalton, 1974, 2, 128–133.
23 C. E. Holloway, G. Hulley, B. F. G. Johnson and J. Lewis, J. Chem. Soc.
A, 1970, 10, 1653–1658.
24 P. S. Pregosin, Coord. Chem. Rev., 1982, 44, 247–291.
25 S. A. Cotton, Chemistry of Precious Metals, Blackie Academic and
Professional, 1997, London, UK.
26 F. P. Fanizzi, N. Margiotta, M. Lanfranchi, A. Tiripicchio, G. Pacchioni
and G. Natile, Eur. J. Inorg. Chem., 2004, 8, 1705–1713.
27 F. P. Fanizzi, N. Margiotta, M. Lanfranchi, A. Tiripicchio, G. Pacchioni
and G. Natile, Eur. J. Inorg. Chem., 2004, 8, 1705–1713.
28 When the complex and ligand were both soluble, the reactions were
performed also in D2O, giving the same outcome.
29 F. P. Fanizzi, F. P. Intini, L. Maresca, G. Natile and G. Uccello-Barretta,
Inorg. Chem., 1990, 29(1), 29–33.
30 F. P. Fanizzi, L. Maresca, G. Natile, M. Lanfranchi, A. M. Manotti-
Lanfredi and A. Tiripicchio, Inorg. Chem., 1988, 27(14), 2422–31.
31 R. K. Harris, E. D. Becker, S. M. Cabral De Menezes, R. Goodfellow
and P. Granger, Pure Appl. Chem., 2001, 73, 1795–1818.
32 R. Romeo, D. Minniti, S. Lanza and M. L. Tobe, Inorg. Chim. Acta,
1977, 22, 87–91.
Notes and references
1 D. Wang and S. J. Lippard, Nat. Rev. Drug Disc., 2005, 4, 307–320.
2 S. A. De Pascali, P. Papadia, A. Ciccarese, C. Pacifico and F. P. Fanizzi,
Eur. J. Inorg. Chem., 2005, 4, 788–796.
3 A. Muscella, N. Calabriso, S. A. De Pascali, L. Urso, A. Ciccarese, F. P.
Fanizzi, D. Migoni and S. Marsigliante, Biochem. Pharmacol., 2007,
74(1), 28–40.
4 A. Muscella, N. Calabriso, F. P. Fanizzi, S. A. De Pascali, L. Urso, A.
Ciccarese, D. Migoni and S. Marsigliante, Br. J. Pharmacol., 2008, 153,
34–49.
5 D. Gibson, J. Lewis and C. Oldham, J. Chem. Soc. A, 1967, 1, 72–
77.
33 (a) N. Walker and D. Stuart, Acta Crystallogr. Sect. A, 1983, 39, 158;
(b) F. Ugozzoli, Comput. Chem., 1987, 11, 109.
34 Area-Detector Absorption Correction, Siemens Industrial Automation,
Inc., Madison, WI, (1996).
6 B. N. Figgis, J. Lewis, R. F. Long, R. Mason, R. S. Nyholm, P. J. Pauling
and G. B. Robertson, Nature, 1962, 195, 1278–1279.
7 F. Basolo, Coord. Chem. Rev., 1996, 154, 151–161, and references cited
therein.
35 RLATT, Bruker, AXS Inc., Madison, Wisconsin, USA.
36 SAINT-Plus (Version 7.51A), Bruker AXS Inc., Madison, Wisconsin,
USA.
8 F. P. Fanizzi, N. Margiotta, M. Lanfranchi, A. Tiripicchio, G. Pacchioni
and G. Natile, Eur. J. Inorg. Chem., 2004, 8, 1705–1713.
9 P. D. Braddock, R. Romeo and M. L. Tobe, Inorg. Chem., 1974, 13(5),
1170–1175.
37 G. M. Sheldrick, (2007). TWINABS (Version 2007/5), University of
Go¨ttingen, Germany.
10 L. Cattalini, F. Guidi and M. L. Tobe, J. Chem. Soc. Dalton Trans.,
38 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacov-
azzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and R. J. Spagna,
Appl. Crystallogr., 1999, 32, 115.
1993, 2, 233–236.
11 P. Nilsson, F. Plamper and O. F. Wendt, Organometallics, 2003, 22(25),
5235–5242.
39 M. C. Burla, R. Caliandro, M. Camalli, B. Carrozzini, G. L. Cascarano,
L. De Caro, C. Giacovazzo, G. Polidoria and R. Spagna, J. Appl.
Crystallogr., 2005, 38, 381–388.
40 G. M. Sheldrick, SHELX97. Programs for Crystal Structure Analysis
(1997) (release 97–2), University of Go¨ttingen, Germany.
41 L. J. Farrugia, Appl. Crystallogr., 1999, 32, 837.
42 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
43 C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock, G. P. Shields,
R. Taylor, M. Towler and J. van de Streek, J. Appl. Crystallogr., 2006,
39, 453–457.
12 O. F. Wendt, K. Oskarsson, J. G. Leipoldt and L. I. Elding, Inorg.
Chem., 1997, 36(20), 4514–4519.
13 C. Eaborn, K. Kundu and A. Pidcock, J. Chem. Soc. Dalton Trans.,
1981, 4, 933–938.
14 R. Romeo, A. Grassi and L. Monsu` Scolaro, Inorg. Chem., 1992, 31(21),
4383–4390.
15 G. Hulley, B. F. G. Johnson and J. Lewis, J. Chem. Soc. A, 1970, 10,
1732–1738.
16 M. Bonivento, L. Cattalini, G. Marangoni, G. Michelon, A. P. Schwab
and M. L. Tobe, Inorg. Chem., 1980, 19, 1743–1746.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 7786–7795 | 7795
©