Mechanistic studies on monodentate-ligand substitution of five-coordinate trigonal-bipyramidal platinum (II) complexes with tris[2-(diphenylphosphino) ethyl]phosphine

Article abstract of DOI:10.1016/j.ica.2004.11.052

Reaction of the five-coordinate trigonal-bipyramidal platinum(II) complex, [Pt(pt)(pp₃)](BF₄) (pt = 1-propanethiolate, pp₃ = tris[2-(diphenylphosphino)ethyl]phosphine), with I^- in chloroform gave the five-coordinate square-pyramidal complex with a dissociated terminal phosphino group and an apically coordinated iodide ion in equilibrium. The thermodynamic parameters for the equilibrium between the trigonal-bipyramidal and square-pyramidal geometries, [Pt(pt)(pp₃)]⁺ + I ^- ? [PtI(pt) (pp₃)], and the kinetic parameters for the chemical exchange were obtained as follows: Kex298=7.5×10-1mol-1kg, ΔH⁰ = - 10 ± 2.4 kJ mol^-1, ΔS ⁰ = - 36 ± 10 J K^-1 mol^-1, kex298=1.3×104s-1, ΔH^? = 34 ± 4.7 kJ mol^-1, ΔS^? = - 50 ± 21 J K ^-1 mol^-1. The square-planar trinuclear platinum(II) complex was formed by bridging reaction of one of the terminal phosphino groups of trigonal-bipyramidal [PtCl(pp₃)]Cl with trans-[PtCl ₂(NCC₆H₅)₂] in chloroform. From these facts, ligand substitution reactions of [PtX(pp₃)]⁺ (X = monodentate anion) are expected to proceed via an intermediate with a dissociated phosphino group. The rate constants for the chloro-ligand substitution reactions of [PtCl(pp₃)]⁺ with Br^- and I^- in chloroform approached the respective limiting values as concentrations of the entering halide ions are increased. These kinetic results confirmed the preassociation mechanism in which the square pyramidal intermediate with a dissociated phosphino group and an apically coordinated halide ion is present in the rapid pre-equilibrium.

Full text of DOI:10.1016/j.ica.2004.11.052

Inorganica Chimica Acta 358 (2005) 2319–2326
www.elsevier.com/locate/ica
Mechanistic studies on monodentate–ligand substitution of
five-coordinate trigonal-bipyramidal platinum (II) complexes
with tris[2-(diphenylphosphino)ethyl]phosphine
a,
a
b
Sen-ichi Aizawa *, Tadashi Kobayashi , Tatsuya Kawamoto
a
Faculty of Engineering, Toyama University, 3190 Gofuku, Toyama 930-8555, Japan
Department of Chemistry, Graduate School of Science, Osaka University, 1-16 Machikaneyama, Toyonaka, Osaka 560-0043, Japan
b
Received 30 October 2004; accepted 24 November 2004
Abstract
Reaction of the five-coordinate trigonal-bipyramidal platinum(II) complex, [Pt(pt)(pp3)](BF4) (pt = 1-propanethiolate,
pp3 = tris[2-(diphenylphosphino)ethyl]phosphine), with Iꢀ in chloroform gave the five-coordinate square-pyramidal complex with
a dissociated terminal phosphino group and an apically coordinated iodide ion in equilibrium. The thermodynamic parameters for
the equilibrium between the trigonal-bipyramidal and square-pyramidal geometries, [Pt(pt)(pp3)]+ + Iꢀ ¢ [PtI(pt) (pp3)], and the
kinetic parameters for the chemical exchange were obtained as follows: K298 ¼ 7:5 ꢁ 10ꢀ1 molꢀ1 kg, DH0 = ꢀ 10 2.4 kJ molꢀ1
,
ex
DS0 = ꢀ 36 10 J Kꢀ1 molꢀ1, k298 ¼ 1:3 ꢁ 104 sꢀ1, DHꢀ = 34 4.7 kJ molꢀ1, DSꢀ = ꢀ 50 21 J Kꢀ1 molꢀ1. The square-planar tri-
ex
nuclear platinum(II) complex was formed by bridging reaction of one of the terminal phosphino groups of trigonal-bipyramidal
[PtCl(pp3)]Cl with trans-[PtCl2(NCC6H5)2] in chloroform. From these facts, ligand substitution reactions of [PtX(pp3)]+ (X = monod-
entate anion) are expected to proceed via an intermediate with a dissociated phosphino group. The rate constants for the chloro-ligand
substitution reactions of [PtCl(pp3)]+ with Brꢀ and Iꢀ in chloroform approached the respective limiting values as concentrations of the
entering halide ions are increased. These kinetic results confirmed the preassociation mechanism in which the square pyramidal inter-
mediate with a dissociated phosphino group and an apically coordinated halide ion is present in the rapid pre-equilibrium.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Trigonal-bipyramidal platinum(II) complex; Trigonal-bipyramidal–square-pyramidal equilibrium; Preassociation mechanism
1. Introduction
cently, we have carried out a few kinetic studies on li-
gand substitution reactions of some trigonal-
bipyramidal palladium(II) complexes with tris[2-(diph-
enylphosphino)ethyl]phosphine (pp3) as a bound ligand
[2,3], because the pp3 ligand tends to form the stable tri-
gonal-bipyramidal complexes with d8 metal ions [4,5]. In
the case of the substitution reactions of [PdX(pp3)]X
(X = Cl, Br, I) with trimethyl phosphite in chloroform,
the observed rate constants were proportional to the
concentration of entering trimethyl phosphite and the
associative mechanism was expected from the obtained
activation parameters [2]. In contrast to the reaction of
the haro complexes with neutral phosphite, the rate con-
stants for the reaction of the thiolato complexes
Reaction mechanisms of four-coordinate square-pla-
nar palladium(II) and platinum(II) complexes have been
well investigated so far, and the associative mechanism
via a trigonal-bipyramidal 18-electron transition state
has been generally proposed [1]. On the other hand, ki-
netic properties of five-coordinate trigonal-bipyramidal
palladium(II) or platinum(II) complexes with an 18-elec-
tron ground state have not been sufficiently investigated
and the general reaction mechanism is still unclear. Re-
*
Corresponding author. Tel./fax: +81 76 445 6980.
E-mail address: saizawa@eng.toyama-u.ac.jp (S. Aizawa).
0020-1693/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2004.11.052

2320
S. Aizawa et al. / Inorganica Chimica Acta 358 (2005) 2319–2326
1
[Pd(SR)(pp3)](BF4) (SR = 1-propanethiolate, a-toluen-
ethiolate) with halide ions in chloroform were indepen-
dent of concentration of the entering halide ions,
which was tentatively attributed to the ion-pair forma-
tion of the positively charged thiolato complexes with
halide anions in the weakly polar solvent [3]. Further-
more, the reactions showed rather dissociative character
the degree of which depends largely on the entering ha-
lide ions. On the assumption that the attack of nega-
tively charged halide ions onto the palladium(II)
center with the electron-donating thiolato ligand brings
about greater electronic repulsion than that of relatively
electron-accepting phosphite onto the haro complexes,
the variation in the reaction mechanism can be ex-
plained by difference in the electronic repulsion in the
transition state. However, the argument may not be nec-
essarily convincing considering a possibility that an
undetectable intermediate is present in rapid pre-equilib-
rium. Accordingly, we have pursued further investiga-
tion by using the corresponding platinum(II)
complexes because the intermediates or unstable species
might be detectable due to the greater activation energy
for each elementary reaction.
ter, JP–Pt = 2506 Hz). UV–vis (in CHCl3): kmax/nm
(log(e/molꢀ1 kg cmꢀ1)) 417 (3.81), 369 (3.91). Complex
1ÆH2O was recrystallized from a mixture of acetonitrile
and chloroform to remove the crystallizing water. Single
crystals of 1ÆCH3CN were obtained1 and it was con-
1
firmed by the H NMR measurements that the crystal-
lizing acetonitrile is volatile in vacuo.
2.2.2. [PtBr(pp3)]Br (2)
To a solution of 1ÆH2O (0.11 g, 0.12 mmol) in chloro-
form (10 cm3) was added excess n-Bu4NBr (0.24 g,
0.74 mmol). The solution was stirred for 1 h at room
temperature and then concentrated gradually to a small
volume. The resultant yellow crystals were collected by
filtration and air-dried. Yield: 88%. Anal. Calc. for
C42H42P4Br2PtÆ2H2O: C, 47.52; H, 4.37; N, 0.00.
Found: C, 47.32; H, 4.46; N, 0.00%. 31P{1H} NMR
(CHCl3): r 23.6 (s, 3P, terminal, JP–Pt = 2575 Hz),
124.7 (s, 1P, center, JP–Pt = 2491 Hz). UV–vis (in
CHCl3): kmax/nm (log(e/molꢀ1 kg cmꢀ1)) 427 (3.86),
380 (3.91).
1
1
2.2.3. [PtI(pp3)]I (3)
In this work, we have disclosed the pre-equilibrium
and the intermediate for the substitution reactions of
the trigonal-bipyramidal platinum(II) complexes with
the pp3 ligand by investigating the reactions of [Pt(pt)-
(pp3)](BF4) (pt = 1-propanethiolate) and [PtCl(pp3)]Cl
with halide ions in completely dry conditions. These re-
sults can also provide us further insight into the kinetic
properties of the trigonal-bipyramidal palladium (II)
complexes.
The iodo complex (3) was prepared by the haro-li-
gand substitution reaction similar to that for the bromo
complex (2) using 0.22 g (0.24 mmol) of 1ÆH2O and ex-
cess n-Bu4NI (0.37, 1.00 mmol) instead of n-Bu4NBr.
Yield: 84%. Anal. Calc. for C42H42P4I2PtÆH2O: C,
44.34; H, 3.90; N, 0.00. Found: C, 44.50; H, 3.73; N,
0.00%. 31P{1H} NMR (CHCl3): r 18.5 (s, 3P, terminal,
1
1JP–Pt = 2572 Hz), 129.4 (s, 1P, center, JP–Pt
=
2395 Hz). UV–vis (in CHCl3): kmax/nm (log(e/
molꢀ1 kg cmꢀ1)) 441 (3.89), 409 (3.92).
2. Experimental
2.2.4. [Pt(pt)(pp3)](BF4) (4)
To a solution of 1ÆH2O (0.51 g, 0.54 mmol) in chloro-
form (50 cm3) was added AgBF4 (0.23 g, 1.18 mmol).
After removal of precipitated AgCl by filtration, to the
colorless filtrate was added a solution of 1-propanethiol
(Hpt, 0.045 g, 0.54 mmol) in chloroform (5 cm3), and
then a 0.5 M NaOH aqueous solution (10 cm3), water
(5 cm3) and ethanol (30 cm3). The mixture was stirred
for 5 min at room temperature and NaOH was extracted
with several 150 cm3 portions of water. Finally, to the
separated chloroform layer was added ethanol and
water gradually. The resultant yellow needle crystals
were collected by filtration and air-dried. Yield: 60%.
Anal. Calc. for C45H49BF4P4SPtÆH2O: C, 51.68; H,
4.92; N, 0.00. Found: C, 51.69; H, 4.70; N, 0.00%.
31P{1H} NMR (CHCl3): r 20.2 (s, 3P, terminal,
2.1. Reagents
Chloroform (Wako, 1 pure) and deuterated chloro-
form (CDCl3, Aldrich) were dried over activated 4A
Molecular Sieves. Tris[2-(diphenylphosphino)ethyl]
phosphine (pp3, Aldrich), potassium tetrachloroplati-
nate(II) (Wako), tetra-n-butylammonium bromide and
iodide (n-Bu4NBr and n-Bu4NI), 1-propanethiol (Hpt,
Wako)
and
dichlorobis(benzonitrile)platinum(II)
(Strem) were used for preparation without further
purification.
2.2. Preparation of complexes
2.2.1. [PtCl(pp3)]Cl (1)
Complex 1ÆH2O was prepared by a similar procedure
described in the literature [5]. Yield 98%. Anal. Calc. for
C42H42P4Cl2PtÆH2O: C, 51.38; H, 4.31; N, 0.00. Found:
C, 51.37; H, 4.53; N, 0.00%. 31P{1H} NMR(CHCl3): r
26.8 (s, 3P, terminal, 1JP–Pt = 2588 Hz), 120.1 (s, 1P, cen-
1
The X-ray diffraction measurements were carried out using the
crystal sealed in a Lindemenn glass-capillary tube with the mother
liquor. Because the crystal structure of 1Æ6H2O was reported recently
in [5], our crystallographic data for 1ÆCH3CN were merely deposited
as a supplementary material.

S. Aizawa et al. / Inorganica Chimica Acta 358 (2005) 2319–2326
2321
1JP–Pt = 2650 Hz), 126.3 (s, 1P, center, 1JP–Pt
= 1940 Hz). UV–vis (in CHCl3): kmax/nm (log(e/
molꢀ1 kg cmꢀ1)) 449 (3.95), 360 (3.52 sh).
temperatures of the NMR sample solutions were deter-
mined by direct measurements of the solvent tempera-
ture in an NMR tube set in the probe using a Technol
Seven D617 digital thermister and were controlled with-
in 0.1 K.
2.2.5. [Pt3Cl4(pp3)2]Cl2 (5)
The Pt(II) trinuclear complex was prepared by
bridging reaction of the five-coordinate trigonal-bipy-
ramidal Pt(II) complex 1ÆH2O (0.32 g, 0.34 mmol)
with trans-[PtCl2(NCC6H5)2] (0.080 g, 0.17 mmol) as
a metal complex to be bridged in chloroform. The
reaction solution was allowed to stand at room tem-
perature for 2 days and to this was added diethylether
to give pale yellow crystals. Yield: 79%. Anal. Calc.
for C84H84P8Cl6Pt3 Æ2H2O: C, 46.38; H, 4.08; N,
3. Results and discussion
3.1. Rapid equilibrium
The 31P NMR spectra of 1–4 in chloroform indicate
the trigonal-bipyramidal structure with C3 symmetry
exhibiting two signals coupled with 195Pt which are as-
signed to the three equivalent terminal phosphorus
atoms of the pp3 ligand in the equatorial position and
the central phosphorus atom in the axial position. The
coupling between the Pt(II) ion and the axial phospho-
rus atom of the pt complex 4 (1JP–Pt = 1940 Hz) is defi-
nitely weak compared with those of the haro complexes
1–3 (1JP–Pt = 2506–2395 Hz) due to the strong trans
inference of the thiolato sulfur atom. The electronic
absorption spectra of 1–4 in chloroform are also consis-
tent with the five-coordinate trigonal-bipyramidal geom-
etry showing two d–d absorption bands in the visible or
near UV0 regions assigned to the transitions from
0.00. Found: C, 45.91; H, 4.04; N, 0.00%. 31P{1H}
1
11.4 (m, 2P, terminal, JP–Pt
NMR (CHCl3):
2
r
= 3670 Hz, JP–P = 15.5 Hz, JP–P = 48.0 Hz), 42.0 (s,
3
1
4P, terminal, JP–Pt = 2492 Hz), 88.2 (m, 2P, center,
1JP–Pt = 2925 Hz). 195Pt{1H} NMR (CHCl3): r 285.3
1
(t, 1Pt, center, JPt–P = 3675 Hz), ꢀ99.8 (dt, 2Pt, termi-
1
nal, JPt–P(center) = 2920 Hz, JPt–P(terminal) = 2494 Hz).
1
2.3. Measurements
Sample preparations for kinetics of 1 were carried out
under a nitrogen atmosphere using completely dried
chloroform, n-Bu4NX (Xꢀ = Brꢀ, Iꢀ) and 1 obtained
from 1ÆCH3CN in vacuo. The kinetic measurements
were performed at 298 K under pseudo-first-order con-
0
1A1 to E and E (see Section 2) [6]. The 31P NMR
spectrum of 4 in chloroform was gradually changed un-
der a luminescent lamp to give quantitatively the square-
planar complex with a dissociated phosphine oxide
00
1
1
ditions where the concentrations of Xꢀ (2.5 · 10ꢀ2
–
2.5 · 10ꢀ1 mol kgꢀ1) were in large excess over that of 1
(7.8 · 10ꢀ5 mol kgꢀ1). The temperature of the reaction
solutions was controlled within 0.1 K. The absorption
spectral changes were recorded on a Perkin–Elmer
Lambda 19 spectrophotometer.
group, [Pt(pt)(pp3O)]+ (Scheme 1), exhibiting 31P
3
NMR signals at 32.3 ppm (phosphine oxide, JP–P
=
37 Hz), 44.6 ppm (terminal, JP–Pt = 2648 Hz) and
1
1
91.7 ppm (center, JP–Pt = 2122 Hz) (Fig. S1, Supple-
mentary material), while the oxidation reaction hardly
proceeded in the dark or in the deoxygenated solution.
The substitution reaction of 4 with Iꢀ proceeded in an
unsealed chloroform solution to give the iodo complex
3 and the square-planar iodo complex with a dissociated
1
195Pt, 31P and H NMR spectra were recorded on a
JEOL JNM-A400 FT-NMR spectrometer operating at
85.48, 160.70 and 399.65 MHz, respectively. In order
to determine the chemical shifts of 195Pt and 31P NMR
signals, a 3 mm o.d. NMR tube containing the sample
solution was coaxially mounted in a 5 mm o.d. NMR
tube containing deuterated water as a lock solvent and
K2[Pt(CN)4] and phosphoric acid as references, respec-
tively. NMR samples for equilibrium and kinetic exper-
iments for 4 were prepared by vacuum distillation of
dried CDCl3 into NMR tubes containing the completely
dried sample that were then flame-sealed after degassing.
The equilibrium constants were evaluated from initial
concentrations of 4 and n-Bu4NI and relative intensities
of the non-decoupled 31P NMR signals for the equili-
brated species obtained by taking a sufficiently long
acquisition time in the temperature range from 206.5
to 236.7 K. The chemical exchange rate constants were
given by increase in transverse relaxation rates obtained
from the spectral simulation of the line-broadening data
in the temperature range from 219.5 to 234.9 K. The
P
+
+
I
P
P
P
P
P
P
I-
P
H2O
+ pt-
P
Pt
Pt
P
P
Pt
P
pt
I
pt
I
3
4
h
ν
+
+
O
O
P
P
+ pt-
P
P
P
P
Pt
Pt
P
P
pt
[Pt(pt)(pp3O)]+
I
[PtI(pp3O)]+
Scheme 1. Substitution and oxidation reactions of 4 in chloroform.

2322
S. Aizawa et al. / Inorganica Chimica Acta 358 (2005) 2319–2326
sponds to the free terminal phosphorus atom. Consider-
ing that the relative strength of these signals increases
with an increase of the Iꢀ concentration, the above find-
ing is attributable to formation of the square-pyramidal
Pt(II) complex with a dissociated terminal phosphino
group and an apically coordinated Iꢀ (I in Scheme 1).
Coordination of the pt ligand in the square-plane is con-
firmed by the chemical shifts and the JP–Pt values for the
coordinated terminal and central phosphorus atoms of I
that are in good agreement with those of [Pt(pt)(pp3O)]+
but entirely different from those of [PtI(pp3O)]+. Be-
cause the two isomeric structures, one with the iodo li-
gand and the pendant diphenylphosphinoethyl group
on the same side of the Pt(II) square plane and the other
on the opposite side, are possible for I, we can assume
that the two relatively broad resonance lines observed
in the region of the free diphenylphosphino group at
the lower temperature are due to the two different steric
surroundings.
By addition of a small amount of water to the equili-
brated solution, the substitution reaction of the pt ligand
with Iꢀ proceeded to give the five-coordinate trigonal-
bipyramidal iodo complex 3 (Scheme 1). This fact indi-
cates that re-coordination of the free terminal phosphino
group accompanies release of the pt ligand bound to
Pt(II) into the moist chloroform because of the greater
solvation energy for ptꢀ compared with dry chloroform.
Formation of [PtI(pp3O)]+ in the unsealed solution in the
dark indicates that the dissociated terminal phosphino
group in I is partially oxidized in the course of the pt
ligand substitution reaction with Iꢀ (Scheme 1). The
equilibrium constant at 298 K and the enthalpy and
entropy changes for the formation of I in CDCl3 were
obtained by using the change in ratio of the 31P NMR
intensity with temperature at which the 31P NMR signals
Fig. 1. 31P NMR spectra of CDCl3 solution of 4 (2.2 · 10ꢀ2 mol kgꢀ1
)
containing a large excess of NBu4I (4.1 · 10ꢀ1 mol kgꢀ1) at 298.3 (a),
250.0 (b), 236.9 (c) and 206.7 (d) K. 4s and Is denote the signals for the
corresponding complexes.
phosphine oxide group, [PtI(pp3O)]+ (Scheme 1), exhib-
iting 31P NMR signals at 32.2 ppm (phosphine oxide,
for 4 and I are well resolved (see Section 2) as follows:
¼ 7:5 ꢁ 10ꢀ1 molꢀ1 kg, DH0 = ꢀ 10 2.4 kJ molꢀ1
1
3JP–P = 37 Hz), 42.2 ppm (terminal, JP–Pt = 2434 Hz),
298
ex
K
,
1
and 95.0 ppm (center, JP–Pt = 2800 Hz) (Fig. S2, Sup-
plementary material). On the other hand, the sealed
sample solution containing 4 and a large excess of n-
Bu4NI in completely deoxygenated and dried CDCl3
showed extremely broad 31P NMR signals at room tem-
perature, which were sharpened as the temperature was
lowered (Fig. 1). Three sets of signals other than the sig-
nals of 4 were clearly observed at the lower temperature
and the spectrum was changed reversibly with tempera-
ture. By comparison with the 31P NMR spectra of
square-planar [Pt(pt)(pp3O)](BF4) (see above) and the
free pp3 ligand2, the two signals at 44.0 and 89.6 ppm
coupled with 195Pt (JP–Pt = 2644 and 2116 Hz, respec-
tively) are assigned to the two terminal and central phos-
phorus atoms on the Pt(II) square-plane, respectively,
and a pair of signals at ꢀ9.1 and ꢀ11.4 ppm corre-
Fig. 2. 31P NMR spectra of 5 in CHCl3. Pc, Pt and Pb are assigned to
the signals for the central, terminal and bridging phosphorus atoms,
respectively. Asterisks and ꢁrefꢂ denote satellite peaks due to 195Pt and
the signal for D3PO4 in the outer D2O, respectively.
31P{1H} NMR (CHCl3): r ꢀ13.0 (d, 3P, terminal), ꢀ17.5 (q, 1P,
2
3
center); JP–P = 26 Hz.

S. Aizawa et al. / Inorganica Chimica Acta 358 (2005) 2319–2326
2323
Fig. 3. Observed (upper) and simulated (lower) 31P NMR spectra for the bridging phosphorus atoms of 5 (a) and the intermediate (b). The simulated
2
spectra are depicted by using JP–P=15 Hz and JP–P = 48 Hz for 5 and JP–P = 490 Hz and JP–P = 48 Hz for the intermediate.
3
2
3
DS0 = ꢀ 36 10 J Kꢀ1 molꢀ1 (Fig. S3, Supplementary
material). These thermodynamic parameters reveal that
I is enthalpically a little stable but entropically unstable
compared with 4 due to coordination of free Iꢀ. The
observed chemical exchange rate constant at 298 K and
the activation enthalpy and entropy were estimated to
195Pt NMR spectrum showed two signals: one is
coupled with the two bridging phosphorus atoms
(1JPt–P = 3675 Hz) and the other is coupled with the
central phosphorus atom (1JPt–P = 2920 Hz) and the
two terminal phosphorus atoms (1JPt–P = 2494 Hz)
(Fig. 4). These 31P and 195Pt NMR spectra confirmed
be 1.3 · 104 sꢀ1
,
34 4.7 kJ molꢀ1 and ꢀ50 21
the
structure.
Change in the 31P NMR spectrum for the reaction
square-planar Pt(II)–Pt(II)–Pt(II) trinuclear
J Kꢀ1 molꢀ1, respectively, by using the 31P NMR line-
broadening data of 4 under the same conditions de-
scribed in Fig. 1 (Fig. S4, Supplementary material). This
exchange rate is extremely faster than the usual substitu-
tion rates for square-planar Pt(II) complexes [7].
solution of trans-[PtCl2(NCC6H5)2] with 1 showed ini-
tial formation of the trinuclear complex different from
5 (Fig. 5), which was finally converted to 5 quantita-
tively in two days. A notable difference in the 31P
NMR spectra is the signal pattern for the central
and bridging terminal phosphorus atoms, that is, the
intermediate complex gives a relatively stronger cou-
pling between the two bridging phosphorus atoms
through the central Pt(II) (2JP–P = 490 Hz) compared
with that of 5 (Fig. 3(b)). Such a difference in the cou-
pling constants can be attributed to a difference in
geometry around the central Pt(II) because similar val-
3.2. Bridging reaction
The complex 5 obtained by the reaction of trans-
[PtCl2(NCC6H5)2] with 2 equiv of 1 in chloroform
exhibited three 31P NMR signals coupled with
195Pt at 88.2 ppm (1JP–Pt = 2925 Hz), 42.0 ppm
(1JP–Pt = 2492 Hz) and 11.4 ppm (1JP–Pt = 3675 Hz),
which can be assigned to the central and two terminal
phosphorous atoms of the pp3 ligand on a square-pla-
nar Pt(II) ion and the other terminal phosphorus
atom bridging to another Pt(II) ion, respectively
(Fig. 2). The NMR spectral simulation3 revealed a
weak coupling (2JP–P = 15 Hz) between the two bridg-
ing phosphorus atoms (Fig. 3(a)) which are chemically
equivalent4 and bound to the same Pt(II) ion. The
2
ues of JP–P (13 and 400 Hz) have been observed for
the bridging phosphorus atoms in the cis and trans iso-
mers of the Pd(II)–Pt(II)–Pd(II) heterotrinuclear com-
plexes, respectively [8]. Consequently, the 31P NMR
spectral change indicates that the thermodynamically
stabler cis isomer is finally formed by the isomerization
on the Pt(II) center (Scheme 2). It should be empha-
sized here that predissociation of one of the terminal
phosphino groups on 1 followed by attack onto the
Pt(II) center of trans-[PtCl2(NCC6H5)2] is required
for the bridging reaction, considering the associative
substitution mechanism of square-planar Pt(II)
complexes [1].
3
gNMR version 5.0, Adept Scientific plc, Herts, UK, 2003.
2
4
The virtual JP–P coupling can be observed in the AA0XX0 spin
system, where A and A0 and X and X0 correspond to the two bridging
and two central phosphorus atoms, respectively.

2324
S. Aizawa et al. / Inorganica Chimica Acta 358 (2005) 2319–2326
1
Fig. 4. 195Pt NMR spectrum of 5 in CHCl3 showing the JPt–P values.
3.3. Substitution reaction
were determined by nonlinear least-squares analyses for
the exponential time courses of the absorbances at
340 nm for Xꢀ = Brꢀ and 360 nm for Xꢀ = Iꢀ.
Because the substitution reaction of the pt complex 4
with halide ions did not proceed in completely dry con-
ditions, the reactions of the chloro complex 1 with halide
ions were examined. The 31P NMR spectra of solutions
containing vacuum-dried 1 and a large excess of n-
Bu4NX (Xꢀ = Brꢀ and Iꢀ) in dried chloroform changed
to the spectra of 2 and 3, respectively. The absorption
spectra of the reaction solutions changed with isosbestic
points at 448 and 312 nm for Xꢀ = Brꢀ and 438 and
340 nm for Xꢀ = Iꢀ and finally agreed with those of 2
and 3, respectively. The observed pseudo-first-order rate
As shown in Fig. 6, the observed rate constants at
298 K approach the respective limiting values as concen-
trations of the entering halide ions are increased. Be-
cause formation of an intermediate with a dissociated
terminal phosphino group is probable considering the
rapid equilibrium and the oxidation reaction of 4 and
the bridging reaction of 1 (vide supra), two reaction
mechanisms can be expected for such a saturation
behavior: one is the limiting dissociative mechanism in
which the unstable four-coordinate square-planar inter-
mediate is generated by the predissociation of one of the
constants of the haro ligand substitution reactions (kobs
)
Fig. 5. 31P NMR spectrum for the solution containing 1 (3.0 · 10ꢀ2 mol kgꢀ1) and trans-[PtCl2(NCC6H5)2] (1.5 · 10ꢀ2 mol kgꢀ1) in chloroform at
45 min after preparation. 5s denote the signals for the corresponding complex. Asterisks and ꢁrefꢂ denote satellite peaks due to 195Pt and the signal for
D3PO4 in the outer D2O, respectively.

S. Aizawa et al. / Inorganica Chimica Acta 358 (2005) 2319–2326
2325
+
P
Cl
PhCN
P
NCPh
Pt
+
2
Pt
P
Cl
P
Cl-
Cl
1
2+
P
P
Cl
Pt
Cl
P
P
P
Pt
P
P
Cl
Pt
2Cl-
Cl
P
intermediate
2+
Cl
Cl
P
P
Cl
Pt
P
Pt
P
P
P
Pt
P
Cl
2Cl-
P
5
Fig. 6. Dependence of the observed rate constants for the chloro-
ligand substitution reactions of 1 in chloroform at 298 K on the
concentrations of the entering Brꢀ (d) and Iꢀ (j). The solid lines were
depicted by using the equation for Mechanism II and the correspond-
ing K and k values in Table 1.
Scheme 2. Formation reaction of 5.
terminal phosphino groups (Mechanism I in Scheme 3)
and the other is the preassociation mechanism in which
the square-pyramidal intermediate similar to I detected
for 4 is present in rapid pre-equilibrium (Mechanism II
in Scheme 3). The limiting value of kobs obtained by a
least-squares analysis for Xꢀ = Iꢀ is about several times
that for Xꢀ = Brꢀ. Because the limiting values of kobs in
Mechanism I are equal to the predissociation rate con-
stant k1 that is independent of the entering halide ions,
Mechanism II via the preassociation intermediate with
a dissociated terminal phosphino group and an apically
coordinated halide ion is reasonably expected. The val-
ues of the pre-equilibrium constant (K) and the succes-
sive dissociation rate constant (k) for each entering
halide ion were obtained by a least-squares analysis
according to the equation: kobs = Kk[Xꢀ]/(K[Xꢀ] + 1)
(Table 1). Assuming that the equilibrium constants for
the trigonal-bipyramidal–square-pyramidal geometrical
change of the haro complexes 1–3 are comparable to
the pre-equilibrium constants obtained kinetically here,
concentrations of the square-pyramidal species for 1–3
are quite low under the concentration conditions for
the NMR and absorption spectral measurements in
the present work. The latter dissociation step probably
proceeds by a reaction mechanism similar to that for
general square-planar Pt(II) and Pd(II) complexes in
Mechanism I
+
+
+
P
P
P
X-
P
k1
k2
P
P
=
+ Cl-
Pt
P
Pt
P
P
k3
Pd
P
P
Cl
P
Cl
X
k1k3[X-]
1
2 or 3
kobs
k2 + k3[X-]
Mechanism II
+
+
X
P
P
P
P
X-
K
P
k
P
+ Cl-
Pt
P
Pt
P
P
Pd
P
P
Cl
P
Cl
X
1
2 or 3
Kk[X-]
kobs
=
K[X-] + 1
X = Br- (2), I- (3)
Scheme 3. Possible mechanisms for the substitution reactions of 1.

2326
S. Aizawa et al. / Inorganica Chimica Acta 358 (2005) 2319–2326
Table 1
Pre-equilibrium constants (K) and dissociation rate constants (k) for
5. Acknowledgments
the substitution reactions of
chloroform at 298 K.
1
with Bu4NX (Xꢀ = Brꢀ, Iꢀ) in
This work was supported by a grant from Kurita
Water and Environment Foundation and Grant-in-Aid
for Scientific Research (No. 15550104) from the Minis-
try of Education, Culture, Sports, Science and Technol-
ogy of Japan.
Xꢀ
K298 (molꢀ1 kg)
k298 (sꢀ1
)
Brꢀ
Iꢀ
(1.53 0.05) · 101
(2.02 0.04) · 101
(5.48 0.10) · 10ꢀ3
(2.40 0.08) · 10ꢀ2
which the entering ligand Xꢀ shifts to the square-plane
and pushes away the leaving chloro ligand via the trigo-
nal-bipyramidal transition state. Accordingly, the
apparently greater rate constant k for Xꢀ = Iꢀ is attrib-
uted to the p-accepting ability of the iodo ligand, which
can stabilize the transition state with the entering Xꢀ
and leaving chloro ligands and the central phosphorus
atom in the trigonal plane. In the case of the present
pp3 complexes, the substitution reactions are completed
by the chelate-ring closure of the dissociated terminal
phosphino group.
In general, the equatorial phosphorus atoms are
weakly coordinated to d8 metal ions compared with
the axial one for the trigonal-bipyramidal pp3 complexes
[2,4e,5]. Furthermore, it should be mentioned here that
the average equatorial Pd–P distance for [PdCl(pp3)]Cl
References
[1] (a) R.G. Wilkins, Kinetics and mechanism of reaction of transition
metal complexes, second ed., VCH, Weinheim, 1991 (Chapter 4);
(b) R.J. Cross, M.V. Twigg, Mechanisms of Inorganic and
Organometallic Reactions, Plenum Press, New York, 1988
(Chapter 5);
(c) Z. Lin, M.B. Hall, Inorg. Chem. 30 (1991) 646.
[2] S. Aizawa, T. Iida, S. Funahashi, Inorg. Chem. 35 (1996) 5163.
[3] S. Aizawa, T. Iida, Y. Sone, T. Kawamoto, S. Funahashi, S.
Yamada, M. Nakamura, Bull. Chem. Soc. Jpn. 75 (2002) 91.
[4] (a) R.B. King, R.N. Kapoor, M.S. Saran, P.N. Kapoor, Inorg.
Chem. 10 (1971) 1851;
(b) C.A. Ghilardi, S. Midollini, L. Sacconi, Inorg. Chem. 14 (1975)
1790;
(c) A. Orlandini, L. Sacconi, Inorg. Chem. 15 (1976) 78;
(d) L. Sacconi, P. Dapporto, P. Stoppioni, P. Innocenti, C. Benelli,
Inorg. Chem. 16 (1977) 1669;
˚
(2.432 A) [2] is definitely longer than the corresponding
˚
(e) W.H. Hohman, D.J. Kountz, D.W. Meek, Inorg. Chem. 25
(1986) 616;
(f) P. Bruggeller, Inorg. Chem. Acta 129 (1987) 27;
¨
Pt–P distances for 1ÆCH3CN (2.375 A) and 1Æ6H2O
˚
(2.369 A) [5], while the axial Pt–P distances are compa-
˚
rable (2.221–2.237 A) [2,5]. These facts leave place for
(g) C. Bianchini, A. Meli, M. Peruzzini, F. Zanobini, Organomet-
allics 9 (1990) 241;
argument that the substitution reactions of the trigo-
nal-bipyramidal Pd(II) complexes with the pp3 ligand
also proceed via similar intermediates with a dissociated
terminal phosphino group.
(h) C. Bianchini, A. Meli, M. Peruzzini, F. Vizza, Organometallics
9 (1990) 1146;
(i) C. Bianchini, P. Innocenti, A. Meli, M. Peruzzini, F. Zanobini,
Organometallics 9 (1990) 2514;
(j) M.C. Vaira, M. Peruzzini, P. Stoppioni, Inorg. Chem. 30 (1991)
1001.
[5] D. Ferna´ndez, M.I.G. Seijo, T. Ke´gl, G. Peto˜cz, L. Kolla´r,
4. Supplementary material
´
M.E.G. Fernandez, Inorg. Chem. 41 (2002) 4435.
[6] A.B.P. Lever, Inorganic Electronic Spectroscopy, second ed.,
Elsevier, Amsterdam, 1984 (Chapter 6).
31P NMR spectra of [Pt(pt)(pp3O)](BF4) (Fig. S1)
and the reaction solution containing 4 and N–Bu4NI
(Fig. S2) in chloroform, detailed thermodynamic and ki-
netic data for the reactions of 4 with Iꢀ in chloroform
(Fig. S3 and S4, respectively) and crystallographic data
of 1ÆH2O in CIF format are available from the author
upon request.
[7] (a) U. Belluco, R. Ettorre, F. Basolo, R.G. Pearson, A. Turco,
Inorg. Chem. 5 (1996) 591;
(b) J.R. Paxson, D.S. Martin Jr., Inorg. Chem. 10 (1971) 1551;
(c) L. Helm, L.I. Elding, A.E. Merbach, Helv. Chim. Acta 67
(1984) 1453.
[8] S. Aizawa, T. kawamoto, K. Saito, Chem. Lett., to be
submitted.