Thiolato-ligand substitution reaction with halide ions of five-coordinate trigonal-bipyramidal palladium(ii) complexes with tris(2-(diphenylphosphino)ethyl)phosphine. Electronic and steric effects on the reaction mechanism

Article abstract of DOI:10.1246/bcsj.75.91

Mononuclear and dinuclear thiolato complexes with five-coordinate trigonal-bipyramidal palladium(II), [Pd(pt or tt)(pp₃)](BF₄) and [Pd₂(pdt)(pp₃)₂](BF₄)₂ (pp₃ = tris(2

Full text of DOI:10.1246/bcsj.75.91

Bull. Chem.
Soc. Jpn.
75
1
© 2002 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 75, 91–97 (2002)
91
The Chemical
Society of Ja-
pan
The Chemical
Society of Ja-
pan
AI
1
15
2002
91
Thiolato-Ligand Substitution Reaction with Halide Ions of
Five-Coordinate Trigonal-Bipyramidal Palladium(Ⅱ) Complexes with
Tris(2-(diphenylphosphino)ethyl)phosphine. Electronic and Steric
Effects on the Reaction Mechanism
Substitution of Five-Coordinate Pd(Ⅱ) Complexes
S. Aizawa et al.
Sen-ichi Aizawa,* Takashi Iida,† Yasushi Sone,†† Tatsuya Kawamoto,††† Shigenobu Funahashi,†
Shinkichi Yamada,†† and Motoshi Nakamura††
97
BCSJA8
0009-2673
01236
7
Faculty of Engineering, Toyama University, 3190 Gofuku, Toyama 930-8555
23
2001
10
†Laboratory of Analytical Chemistry, Graduate School of Science, Nagoya University, Chikusa, Nagoya 464-8602
††Faculty of Engineering, Shizuoka University, Johoku, Hamamatsu 432-8561
9
2001
†††Department of Chemistry, Graduate School of Science, Osaka University, 1-16 Machikaneyama, Toyonaka,
Osaka 560-0043
2.5
2.8
(Received July 23, 2001)
Mononuclear and dinuclear thiolato complexes with five-coordinate trigonal-bipyramidal palladium(Ⅱ), [Pd(pt or
tt)(pp3)](BF4) and [Pd2(pdt)(pp3)2](BF4)2 (pp3 = tris(2-(diphenylphosphino)ethyl)phosphine, pt = 1-propanethiolate, tt
= α-toluenethiolate, pdt = 1,3-propanedithiolate) have been synthesized. The solid-state structure of the dinuclear pdt
complex was confirmed by an X-ray crystal structure analysis and the structures of the pt, tt, and pdt complexes in solu-
tion were characterized by 31P NMR spectroscopy. It is indicated from the 31P NMR chemical shifts of the bound pp3
ligand that the axially coordinated thiolato ligands are strong σ and π donors compared with the halo ligands in the axial
position of the corresponding trigonal-bipyramidal palladium(Ⅱ) complexes, [PdX(pp3)]+ (X− = Cl−, Br−, I−). The ki-
netic parameters for the one-step and successive two-step thiolato-ligand substitution reactions with halide ions in chlo-
roform were obtained for the mononuclear and dinuclear complexes, respectively. These results have revealed that the
reaction mechanism of the present thiolato-ligand substitutions is much more dissociative than that of the corresponding
halo-ligand substitutions using trimethyl phosphite as the entering ligand, and that the reaction mechanism of the two-
step substitution of the dinuclear thiolato complex is more dissociative than that of the mononuclear complexes. It has
been concluded that the reaction mechanism of the five-coordinate trigonal-bipyramidal palladium(Ⅱ) complexes with
the 18-electron ground state is quite sensitive to the electronic properties of the entering and leaving ligands and the ster-
ic environment of the reaction site compared with that of four-coordinate square-planar palladium(Ⅱ) complexes with the
16-electron ground state, which is generally associatively activated.
The reaction mechanism of four-coordinate square-planar
palladium(Ⅱ) and platinum(Ⅱ) complexes has been extensively
studied and an associative mechanism via the trigonal-bipyra-
midal 18-electron transition state has been generally pro-
posed.1,2 On the other hand, five-coordinate trigonal-bipyrami-
dal d8 metal complexes which have the 18-electron ground
state may bring about great electronic repulsion in the associa-
tive activation state, while dissociative bond breaking may re-
quire considerably great enthalpic activation for the palladi-
um(Ⅱ) and platinum(Ⅱ) complexes. Therefore, a thorough in-
vestigation is needed in order to make the above vague points
clear. Furthermore, such mechanistic studies will provide fur-
ther insight into the kinetic properties of the palladium(Ⅱ) and
platinum(Ⅱ) complexes.
coordinated halide ion is replaced by monodentate phosphite
to form [Pd(pp3)(P(OCH3)3)]2+ maintaining the trigonal-bipy-
ramidal geometry (Eq. 1).3
(1)
From the activation parameters, it was indicated that the reac-
tion proceeds by a quite associative activation mode compared
with the reaction mechanism of the square-planar palladi-
um(Ⅱ) complexes generally reported.4 This result was rather
surprising, because it would be possible that the 18-electron
ground state of the five-coordinate trigonal-bipyramidal palla-
dium(Ⅱ) complexes provides a much greater electronic repul-
sion with the entering ligand than the 16-electron ground state
For a clear-cut investigation, we previously performed ki-
netic studies on the substitution reaction of [PdX(pp3)]+ (pp3
= tris(2-(diphenylphosphino)ethyl)phosphine; X− = Cl−, Br−,
I−) with trimethyl phosphite in chloroform, where the axially

92
Bull. Chem. Soc. Jpn., 75, No. 1 (2002)
Substitution of Five-Coordinate Pd(ꢀ) Complexes
Found: C, 58.10; H, 4.95; N, 0.32%. Calcd for [Pd(tt)(pp3)]-
(BF4)•H2O: C, 58.55; H, 5.11; N, 0.00%. 31P NMR (in CHCl3): δ
(relative to D3PO4 in external D2O) 30.65 (d, equatorial), 135.57
(q, axial); JP–P = 10 Hz.
of the four-coordinate square-planar ones, and promotes disso-
ciation of the leaving ligand in the activation state. In our pre-
vious paper,3 we proposed that the interaction between the oc-
cupied dxz and dyz orbitals on the palladium(Ⅱ) center and the
entering electron-accepting trimethyl phosphite is of great ad-
vantage to the associative mechanism in that particular case.
In addition, we also suggested that the large void volume
around the substitution site surrounded by the six phenyl
groups can accommodate the large volume of the entering
ligand in the activation state, and shows the associative charac-
ter in the activation parameters.
In order to obtain general views on the reaction mechanism
of the five-coordinate trigonal-bipyramidal palladium(Ⅱ) com-
plexes, it is necessary to investigate the reaction system with a
strong electronic repulsion in the activation state. In addition,
steric effects of the surroundings around the reaction site
should also be elucidated. From these circumstances, we have
selected thiolates, such as 1-propanethiolate (pt) and α-tolu-
enethiolate (tt), as the leaving ligands, where the thiolate sulfur
atoms are expected to be relatively strong σ and π donors to in-
crease the electron density of the palladium(Ⅱ) center; we used
halide ions as the entering ligands, which can cause an elec-
tronic repulsive interaction in the activation state. To investi-
gate the steric effects, we undertook the corresponding substi-
tution reaction by using the dinuclear trigonal-bipyramidal pal-
ladium(Ⅱ) complex with a bridging dithiolato ligand, 1,3-pro-
panedithiolate (pdt), where the substitution site is apparently
blocked by the surrounding twelve phenyl groups of the bound
pp3 ligands.
[Pd2(pdt)(pp3)2](BF4)2. To a suspension of [Pd(pp3)](BF4)2
(1.036 g, 1.09 mmol) in acetonitrile (80 cm3) was added 1,3-pro-
panedithiol (H2pdt, 0.0504 g, 0.466 mmol), and then 0.25 M aque-
ous NaOH (5 cm3). To the resultant red solution were added
dichloromethane (40 cm3) and water (100 cm3), and the dichlo-
romethane phase was evaporated to dryness. The resultant red
complex was recrystallized from dichloromethane by adding ace-
tonitrile and a small amount of ethanol and water, and air-dried.
Yield: 0.849 g (85%). Anal. Found: C, 55.61; H, 4.93; N, 0.00%.
Calcd for [Pd(pdt)(pp3)2](BF4)2: C, 56.97; H, 4.95; N, 0.00%. 31P
NMR (in CHCl3): δ (relative to D3PO4 in external D2O) 30.33 (d,
equatorial), 134.50 (q, axial); JP-P = 10 Hz.
X-ray Structure Analysis. As the red crystals of the pdt
complex lost the crystallizing solvent molecules when exposed to
the atmosphere, one of them was sealed in a Lindemann glass-
capillary tube with the mother liquor.6 X-ray diffraction measure-
ments were performed on a Mac Science MXC3 diffractometer
with Mo Kα radiation. The unit-cell parameters and orientation
matrix were determined by a least-squares refinement of 17 inde-
pendent reflections collected in the range 18.0° < 2θ < 26.2°.
Data collection was performed with the θ/2θ scan mode with three
standard reflections measured after every 100 scans, which
showed no significant decay. An empirical absorption (ψ scan)
correction was applied. The solution and refinement were carried
out using the CRYSTAN-GM (version 6.3.3) software package.7
The structure was solved by the direct method using the SIR 92
program.8 Because of a limited data set collected from the thin
plate of the crystal, only the palladium, sulfur, and phosphorus at-
oms were reasonably refined anisotropically. All atoms of the
counter anion of the complex cation, BF4−, and crystallizing ace-
tonitrile and ethanol were constrained to the appropriate positions.
The hydrogen atoms were included in the calculated positions.
Experimental
Reagents. Chloroform (Wako, Sp. Gr.) and deuterated chlo-
roform (Aldrich, 99.8 atom% D) were dried over activated 4A
Molecular Sieves, and then distilled in a vacuum line. Chloroform
(Wako, ∞ pure) and deuterated chloroform (Wako, 100%) were
dried over activated 4A Molecular Sieves. Tetrabuthylammonium
halides (Wako), Bu4NX (X− = Cl−, Br−, and I−), were dried un-
der a vacuum. Tris(2-(diphenylphosphino)ethyl)phosphine (pp3,
Aldrich and Strem), tetrakis(acetonitrile)palladium(Ⅱ) tetrafluo-
roborate (Aldrich), 1-propanethiol (Hpt, Wako), α-toluenethiol
(Htt, Aldrich), and 1,3-propanedithiol (H2pdt, Wako) were used to
prepare of palladium(Ⅱ) complexes without further purification.
Preparation of Complexes. [Pd(pt)(pp3)](BF4). To a sus-
−
The occupancy factor of each atom of BF4 was fixed to 0.5, ex-
cept for B1, B2, F3, F4, and F10. The atomic scattering factors
were taken from Ref. 9.
The crystallographic data are summarized in Table 1. The crys-
tallographic details, positional and thermal parameters for all at-
Table 1. Crystallographic Data for [Pd2(pdt)(pp3)2](BF4)2•
CH3CN•C2H5OH
5
pension of [Pd(pp3)](BF4)2 (0.205 g, 0.216 mmol) in acetonitrile
Formula
C91H99B2F8NOP8Pd2S2
(20 cm3) was added dropwise 1-propanethiol (Hpt, 0.10 g, 1.31
mmol), and then 0.1 M aqueous NaOH (1.2 cm3). To a resultant
red solution was added a small amount of ethanol and then water
to give red crystals. The crystals were collected by filtration and
washed with ethanol and water. Red needle crystals were ob-
tained by recrystallization from a dichloromethane-ethanol mix-
ture, and air-dried. Yield: 0.128 g (63%). Anal. Found: C, 56.88;
H, 5.29; N, 0.00%. Calcd for [Pd(pt)(pp3)](BF4): C, 57.56; H,
5.26; N, 0.00%. 31P NMR (in CHCl3): δ (relative to D3PO4 in ex-
ternal D2O) = 30.34 (d, equatorial), 134.51 (q, axial); JP-P = 10
Hz. UV–vis (in CH2Cl2):λmax/nm (log(ε/mol−1 kg cm−1)) 551
(3.31), 409 (3.82), 315 (4.43).
Fw
1921.16
monoclinic
P21/n
35.26(1)
18.45(2)
13.817(5)
91.47(3)
8989(9)
4
Cryst syst
Space group
a/Å
b/Å
c/Å
β/deg
V/Å3
Z
ρ
calcd/g cm−3
1.420
0.10×0.25×0.50
6.358
Cryst dimens/mm
µ/cm−1
[Pd(tt)(pp3)](BF4). Red needle crystals of the tt complex
were obtained by a procedure similar to that of the pt complex by
using 0.353 g (0.371 mmol) of [Pd(pp3)](BF4)2 and 0.070 g (0.56
mmol) of α-toluenethiol (Htt). Yield: 0.291 g (81%). Anal.
b)
R,a) Rw
0.097, 0.109
a) R = Σ||Fo| − |Fc||/Σ|Fo|. b) Rw = [Σw(|Fo| − |Fc|)2/
2
2
Σw|Fo| ]1/2. w = (σ(Fo)2 + 0.001Fo )−1
.

S. Aizawa et al.
Bull. Chem. Soc. Jpn., 75, No. 1 (2002)
93
oms, and bond distances and angles for non-hydrogen atoms are
provided as Supporting Data.
Table 2. Selected Bond Distances (Å) and Angles (deg) for
[Pd2(pdt)(pp3)2]2+
Measurements. The kinetic measurements for the substitu-
tion reactions of the pt, tt, and pdt complexes with halide ions in
chloroform were carried out under the pseudo-first-order condi-
tions where the concentrations of Bu4NX (X− = Cl−, Br−, I−)
were in large excess over those of the thiolato complexes (Tables
S5–S9, Supporting Data). The sample preparation and kinetic
measurements were performed under a nitrogen atmosphere. The
absorption spectral changes were recorded on Shimadzu UV-260
FW or JASCO V-570 spectrophotometers. The temperature of the
reaction solution was controlled within 0.1 K. The substitution
reactions at various pressures were followed by using a high-pres-
sure static vessel equipped with a le-Noble-type quartz cell.10 The
substitution rates of the pt and tt complexes were sufficiently slow
to obtain accurate rate constants, though about 20 min was taken
up for temperature and pressure equilibration in the high pressure
Pd(1)–S(1)
2.392(16) Pd(2)–S(2)
2.398(15)
Pd(1)–P(1)
Pd(1)–P(2)
Pd(1)–P(3)
Pd(1)–P(4)
2.256(16) Pd(2)–P(5)
2.430(15) Pd(2)–P(6)
2.384(16) Pd(2)–P(7)
2.474(16) Pd(2)–P(8)
2.240(17)
2.342(17)
2.359(17)
2.637(18)
S(1)–Pd(1)–P(1) 176.5(6)
S(1)–Pd(1)–P(2) 97.9(6)
S(1)–Pd(1)–P(3) 93.8(6)
S(1)–Pd(1)–P(4) 96.8(6)
P(1)–Pd(1)–P(2) 85.0(6)
P(1)–Pd(1)–P(3) 83.0(6)
P(1)–Pd(1)–P(4) 83.6(6)
P(2)–Pd(1)–P(3) 123.4(6)
P(2)–Pd(1)–P(4) 116.6(6)
P(3)–Pd(1)–P(4) 116.7(6)
S(2)–Pd(2)–P(5) 166.5(7)
S(2)–Pd(2)–P(6) 90.4(6)
S(2)–Pd(2)–P(7) 94.5(6)
S(2)–Pd(2)–P(8) 110.0(6)
P(5)–Pd(2)–P(6) 83.3(6)
P(5)–Pd(2)–P(7) 83.5(7)
P(5)–Pd(2)–P(8) 83.2(6)
P(6)–Pd(2)–P(7) 142.9(6)
P(6)–Pd(2)–P(8) 104.8(6)
P(7)–Pd(2)–P(8) 107.9(6)
1
vessel. The 31P and H NMR spectra were recorded on a JEOL
JNM-A400 FT-NMR spectrometer operating at 160.7 and 399.65
MHz, respectively. In order to determine the chemical shift of the
31P NMR, a 3 mm o.d. NMR tube containing the sample solution
was coaxially mounted in a 5 mm o.d. NMR tube containing deu-
terated water and phosphoric acid as a lock solvent and a refer-
ence, respectively.
Pax–C–C–Peq chelates to form small chelate bite angles (Pax–
Pd–Peq = 83.0–85.0°). Furthermore, the average axial and
equatorial bond distances (Pd–Pax = 2.25 Å and Pd–Peq = 2.44
Å, respectively) are in a similar range to those of the halo com-
plexes, indicating a strong σ interaction of the axial Pd–P
bond. The average axial Pd–S bond distance (2.40 Å) is com-
parable to the axial Pd–Cl bond distance (2.42 Å).3 The no-
ticeable characteristic of the pdt complex is the fact that the ax-
ial thiolato coordination sites are surrounded by twelve phenyl
groups, in which the shortest C–C distance between the phenyl
groups in the adjacent Pd(pp3) moieties is ca. 3.4 Å.
Results and Discussion
Characterization. A perspective view of the pdt complex
cation is displayed in Fig. 1 along with the atomic labeling for
the palladium and the coordinated atoms. The selected bond
distances and angles are summarized in Table 2. The complex
cation has a dinuclear structure bridged by the pdt ligand.
Each palladium(Ⅱ) center has a trigonal-bipyramidal geometry
with the thiolato sulfur atom of pdt and the central phosphorus
atom of pp3 (Pax) in the axial position and the three terminal
phosphorus atoms of pp3 (Peq) in the equatorial positions. The
structure of the Pd(pp3) moiety is quite comparable to those of
the halo complexes, [PdX(pp3)]+ (X− = Cl−, Br−, I−), as pre-
viously reported.3 For example, the palladium(Ⅱ) ions in the
pdt and halo complexes are located out of the equatorial plane
defined by the three equatorial phosphorus atoms by 0.25–0.28
Å toward the axially coordinated thiolato sulfur atom or halide
ions, since the chelate strain is operating in the five-membered
As shown in Fig. 2, the 31P NMR spectra of the pdt complex
in chloroform exhibit two signals, indicating that each Pd(pp3)
moiety of the dinuclear structure has C3 symmetry with three
equivalent terminal phosphorus in the equatorial position (Peq)
and the central phosphorus in the axial position (Pax); the two
Pd(pp3) moieties are also equivalent to each other in solution.
Because the 31P NMR spectral pattern of the pt and tt complex-
es in chloroform agrees with that of the pdt complex, it is obvi-
ous that the pt and tt complexes have almost the same trigonal-
bipyramidal structure as that of the Pd(thiolato)(pp3) moiety in
the dinuclear pdt complex.
Fig. 2. 31P NMR spectra of the pdt complex in chloroform.
Asterisk denotes the signal for D3PO4 in the external D2O.
Fig. 1. Structure of the pdt complex, [Pd2(pdt)(pp3)2]2+, in
the crystal.

94
Bull. Chem. Soc. Jpn., 75, No. 1 (2002)
Substitution of Five-Coordinate Pd(ꢀ) Complexes
Table 3. 31P NMR Chemical Shifts for Halo and Thiolato
Complexes
Chemical shift/ppma)
Complex
Pax
Peq
[PdCl(pp3)]+ b), c)
[PdBr(pp3)]+ b), c)
[PdI(pp3)]+ b), c)
[Pd2(pdt)(pp3)2]2+ d)
[Pd(pt)(pp3)]+ d)
[Pd(tt)(pp3)]+ d)
136.50
142.57
147.35
134.50
134.51
135.57
31.62
31.76
31.98
30.33
30.34
30.65
a) Relative to D3PO4 in external D2O. b) Reference 3.
c) In dichloromethane. d) In chloroform.
In the case of the free pp3 ligand, the 31P NMR signal of the
central phosphorus atom shows a greater upfield shift than that
of the terminal phosphorus atoms in the diphenylphosphino
groups.11 However, the downfield shift of Pax is much larger
than that of Peq. This is explained by the greater deshielding
(diamagnetic contribution) and configurational ununiformity
of the valence electrons (paramagnetic contribution) on the
phosphorus atoms caused by the donation of Pax stronger than
that of Peq, which is observed in the Pd–P bond distances.12,13
As previously described,3 the σ donation of the axial ligand X
in [PdX(pp3)]+ competes with that of Pax, mainly through the
dz2 and pz orbitals of the palladium(Ⅱ) ion; also the π donation
of the axial ligand increases the electron density in the dxz and
Fig. 3. Pressure dependences of ln k for the reaction of the pt
complex with Cl− (a), Br− (b), and I− (c) and the tt com-
plex with I− (d).
methyl phosphite, the observed rate constants were first order
with respect to the entering-ligand concentration.3 Because the
limiting dissociative mechanism is not consistent with the fact
that the present kinetic parameters depend on the entering ha-
lide ions (vide infra), it is probable that the ion pairs of the pos-
itively charged thiolato complexes with halide ions existing in
large excess are quantitatively formed in chloroform. Each
temperature dependence of the first-order rate constant was fit-
ted to the Eyring equation to give the values of the activation
enthalpy (∆H‡) and activation entropy (∆S‡) (Table S6 and Fig.
S2, Supporting Data). The values of the activation volume
(∆V‡) were given by fitting the pressure dependence of k to
(∂ln k/∂P)T = −∆V‡/RT, as shown in Fig. 3 (Table S7, Sup-
porting Data). The activation parameters obtained for the pt
and tt complexes are summarized in Table 4. If an undetect-
able square-planar intermediate was formed by pre-equilibri-
um accompanied by the dissociation of one of the terminal
diphenylphosphino groups, the observed rate would be much
slower than that for the square-planar complex, which has a
structure corresponding to that of the square-planar intermedi-
ate, because the pre-equilibrium constant should be much
smaller than unity.3 Considering the fact that the present rate
constants are comparable to or larger than those of the corre-
sponding square-planar complex, [Pd(pt)(p3)]+ (p3 = bis(2-
(diphenylphosphino)ethyl)phenylphosphine),14 we can reason-
dyz orbitals of the palladium(Ⅱ) ion, which have π interactions
with the empty d orbitals of Peq. Accordingly, the strength of
the σ and π donation of the axial ligand X can be separately es-
timated by the upfield shift of Pax and Peq, respectively. The
chemical shifts listed in Table 3 show that the σ and π dona-
tions of the present thiolato ligands are stronger than those of
the halo ligands.
Kinetics of Mononuclear Complexes.
The 31P NMR
spectra of solutions containing the pt or tt complexes and a
large excess of Bu4NX (X = Cl−, Br−, I−) in chloroform grad-
ually changed to those of the corresponding halo complexes.
The absorption spectra of the diluted solutions (2.46×10−5
–
1.98×10−4 mol kg−1 of the pt and tt complexes and 7.13×10−3
–5.32×10−1 mol kg−1 of Bu4NX in chloroform) changed with
isosbestic points, and finally exhibited the spectra of the corre-
sponding halo complexes.3 The observed pseudo-first-order
rate constants of the thiolato-ligand substitution reactions with
halide ions (k) were determined by a least-squares analysis for
the exponential time course of the absorbance at 480 nm.
The values of k were independent of the halide-ion concen-
tration (Table S5 and Fig. S1, Supporting Data). In the case of
substitution reactions of the halo complexes with neutral tri-
Table 4. Activation Parameters for Substitution Reactions of [Pd(thiolato)(pp3)]+ with Halide Ions
Thiolato ligand Halide ion
k
298/10−4 s−1 ∆H‡/kJ mol−1
∆S‡/J mol−1 K−1
∆V‡/cm3 mol−1
pt
pt
pt
tt
Cl−
Br−
I−
3.55
2.76
2.14
6.12
70
82
89
86
2
2
1
2
−77
−37
−16
−18
5
7
2
8
6.1 0.3a)
9.0 1b)
11.0 0.8c)
14.7 1.3d)
I−
a) At 302.2 K. b) At 302.7 K. c) At 303.7 K. d) At 298.2 K.

S. Aizawa et al.
Bull. Chem. Soc. Jpn., 75, No. 1 (2002)
95
ably expect one-step substitution, as previously proposed (Eq.
2).3
(2)
As we reported previously, the halo-ligand substitution reac-
tions of the same type of the trigonal-bipyramidal palladi-
um(Ⅱ) complexes, [PdX(pp3)]+ (X = Cl−, Br−, I−), proceed by
a surprisingly associative activation mode, showing large nega-
tive values of ∆S‡ (−89–−198 J K−1 mol−1) and ∆V‡ (−22.6–
−25.5 cm3 mol−1), probably due to the electron-accepting
ability of the entering phosphite.3 As shown in Table 4, the
present substitution mechanism is obviously much more disso-
ciative, showing large ∆H‡, less negative ∆S‡, and considerably
large positive ∆V‡ values compared with the previous case.
Considering that the axially coordinated thiolato sulfur atom is
a fairly strong σ and π donor to the orbitals on the palladi-
um(Ⅱ) center (vide supra), the change in the reaction mecha-
nism is mainly attributed to an electronic repulsion between
the shielded metal center and the entering anion. This is sup-
ported by the fact that the steric difference in the axial thiolato
ligands, pt and tt, does not vary the reaction mechanism, ex-
cept for a slight deviation of the ∆V‡ value. It may be conclud-
ed that the reaction mechanism of the five-coordinate trigonal-
bipyramidal palladium(Ⅱ) complexes with the 18-electron sys-
tem is quite variable with the electronic repulsion in the transi-
tion state, while four-coordinate square-planar ones with emp-
ty pz orbital are generally activated in the associative mode to
form an 18-electron transition state. On the other hand, en-
hancement of the dissociative character from Cl− to I− is as-
cribed to an increase in the size of the entering halide ion, be-
cause a greater steric hindrance promotes the dissociation of
the leaving ligand (vide infra).
Fig. 4. Absorption spectral change of chloroform solution
containing [Pd2(pdt)(pp3)2]2+ and excess I− at 0, 520, 1040,
1560, 3161, 6239, 12477, and 24028 s after the tempera-
ture equilibration. The direction of the change is denoted
by arrows.
excess Bu4NCl and Bu4NBr instead of Bu4NI. On the basis of
the above absorption and NMR spectral behaviors, the pseudo-
first-order rate constants for the two-step substitution reactions
(Eq. 3) were obtained by using chloroform solutions contain-
ing the pdt complex (6.03–6.33×10−5 mol kg−1) and a large
excess of Bu4NX (X− = Cl−, Br−, I−) at various concentra-
tions (3.45–36.1×10−2 mol kg−1).
Kinetics of Dinuclear Complex. The absorption spectral
change of the reaction solution containing the pdt complex
(6.0×10−5 mol kg−1) and a large excess of Bu4NI (2.4×10−2
mol kg−1) in chloroform is shown in Fig. 4. Though the final
spectrum agreed with that of the iodo complex,3 the isosbestic
point at ca. 375 nm was gradually shifted to ca. 340 nm. As
shown in Fig. 5, the absorbance at 365 nm first decreased and,
subsequently increased; each absorbance change for the first
several minutes and after the elapse of ca. 1 h can be fitted to
an exponential curve. From these facts, the reaction of the pdt
complex with I− can be regarded as successive substitution. In
(3)
The rate constant for the first-step reaction (k1) with I− was ob-
tained from the exponential time course for the first several
minutes at 350 nm, where the absorbance change for the sec-
ond step was quite small, and that for the second step (k2) was
obtained from the sufficiently large absorbance change at 478
nm after a sufficient elapse of time. The values of k1 and k2 for
the substitution reactions with Cl− and Br− were similarly ob-
tained by using the large absorbance change at 350 nm.
The rate constants for the first and second steps were inde-
pendent of the entering-halide-ion concentration in chloroform
(Table S8 and Fig. S3, Supporting Data), as observed for the
substitution reaction of the pt and tt complexes. The activation
parameters for the two steps were obtained from the tempera-
ture dependence of the k1 and k2 values, as listed in Table 5.
Comparing the activation parameters between the pt and pdt
complexes, it is obvious that the reaction mechanism of the di-
1
the initial stage of the reaction, the H NMR signals for the
methylene protons of the bridging pdt ligand (δ = 1.7 for the
terminal –CH2–S– and 1.4 for the central –CH2–) were over-
lapped by large signals of the excess Bu4NI (Fig. 6a). As the
reaction proceeded, the signals assigned to the methylene pro-
tons of the pendant pdt ligand (δ = 2.1–2.3 for the free –CH2–
S− and 1.8–1.9 for the bound –CH2–S–) increased (Fig. 6b),
followed by a subsequent decrease with increasing signals for
the free pdt ligand (δ = 3.1 for the terminal –CH2–S− and 2.3
for the central –CH2–) (Fig. 6e). This 1H NMR spectral
change, which is consistent with the successive two-step sub-
stitution, was also observed in the case of a solution containing

96
Bull. Chem. Soc. Jpn., 75, No. 1 (2002)
Substitution of Five-Coordinate Pd(ꢀ) Complexes
Table 5. Activation Parameters for Successive Substitution Reactions of [Pd2(pdt)(pp3)2]2+ with Halide Ions
First step
Second step
∆H‡/kJ mol−1
130 1.3
126 2.6
93.7 1.5
Halide ion
Cl−
k
298/10−4 s−1
16.9
∆H‡/kJ mol−1
∆S‡/J mol−1 K−1
24 18
k
298/10−4 s−1
5.69
∆S‡/J mol−1 K−1
96
97 16
115
5
129 4.4
Br−
29.5
5.90
32 15
79 18
2.90
4.14
110
9
I−
5
4.6 4.9
Fig. 5. Absorbance change of chloroform solution contain-
ing [Pd2(pdt)(pp3)2]2+ and excess I− at 365 nm.
nuclear pdt complex is more dissociative than that of the
mononuclear pt complex, showing large ∆H‡ and positive ∆S‡
values. In the initial state of the first step, the substitution site
is surrounded by the twelve phenyl groups, as shown in the
crystal structure, where the entering ligand is apparently
blocked. Consequently, the leaving ligand dissociation is re-
quired to reach the activation state. In fact, the dissociative
character is enhanced by an increase in the size of the entering
ion. In addition, the mutual steric repulsion between the phen-
yl groups in the two palladium(Ⅱ) terminals also promotes
dissociation of the palladium(Ⅱ)–sulfur bond. In the case of
the second step, an assumption that the negative charge of the
pendant thiolate group in the intermediate interferes with the
approach of the entering halide ion is consistent with the fact
that the dissociative character is enhanced compared with the
reaction of the positive mononuclear pt complex, and is pro-
moted by an increase in the charge density of the entering ha-
lide ion.
Fig. 6. 1H NMR spectral change of chloroform solution con-
taining [Pd2(pdt)(pp3)2]2+ (7.1×10−3 mol kg−1) and excess
I− (5.0×10−1 mol kg−1) at 5 (a), 20 (b), 45 (c), 120 (d),
and 1200 (e) min after preparation of the sample solution.
Asterisks and ss’s denote the signals for the excess Bu4NI
and their spinning side bands, respectively. The chemical
shifts for the terminal and central methylenes of the bridg-
ing pdt are shown by a1 and a2, respectively. The signals
for the free and bound terminal methylenes of the pendant
pdt are denoted by b1ꢀ and b1ꢀ respectively, and the signal
for the central methylene is overlapped by the signals of
the excess Bu4NI. The signals for the terminal and central
methylenes of the free pdt are denoted by c1 and c2, respec-
tively. The signals for ethanol and acetonitrile are due to
the solvent molecules contained in the crystals of the pdt
complex.
Conclusion
We have newly prepared mononuclear and dinuclear five-
coordinate trigonal-bipyramidal thiolato palladium(Ⅱ) com-
plexes, [PdX(pp3)]+ (X = pt, tt) and [Pd2(pdt)(pp3)2]2+, for the
purpose of mechanistic studies. These complexes were char-
acterized by X-ray crystal structure analysis and 31P NMR
spectroscopy. The activation parameters of the thiolato-ligand
substitution reactions with halide ions in chloroform were ob-
tained for the one-step substitution of the mononuclear com-
plexes, and for the successive two-step substitution of the dinu-
clear complex. A comparison between the present results and
those for the halo-ligand substitution reaction with trimethyl
phosphite3 indicates that an increase in the electronic repulsion
and steric hindrance in the activation state changes the reaction
mechanism sensitively to the dissociative mode of the reaction.
This phenomenon is a clear difference between the five-coordi-
nate trigonal-bipyramidal palladium(Ⅱ) complexes with the
18-electron ground state and four-coordinate square-planar
ones with the 16-electron ground state, which is activated asso-

S. Aizawa et al.
Bull. Chem. Soc. Jpn., 75, No. 1 (2002)
97
lic Reactions,” ed by M. V. Twigg, Plenum, New York (1988),
Chap. 5.
ciatively. Therefore, the fine tuning of the reaction mechanism
will be possible by controlling the electronic and steric envi-
ronments in the case of the five-coordinate trigonal-bipyrami-
dal palladium(Ⅱ) complexes.
2
3
Z. Lin and M. B. Hall, Inorg. Chem., 30, 646 (1991).
S. Aizawa, T. Iida, and S. Funahashi, Inorg. Chem., 35,
5163 (1996).
4
M. Kotowski and R. van Eldik, “Inorganic High Pressure
This research was supported by a Grant-in-Aid for Scientific
Research (No. 12640583) from the Ministry of Education,
Culture, Sports, Science and Technology.
Chemistry,” ed by R. van Eldik, Elsevier, Amsterdam (1986),
Chap. 4.
5
6
S. Aizawa and S. Funahashi, Anal. Sci., 12, 27 (1996).
It was confirmed by 1H NMR measurements of the freshly
Supporting Data
prepared crystals dissolved in deuterated chloroform that acetoni-
trile and ethanol molecules were contained in the crystals. The
signals of the solvent molecules for the air-dried powdery sample
were apparently decreased.
X-ray crystallographic data (Table S1), atomic coordinates and
isotropic thermal parameters (Table S2), anisotropic thermal pa-
rameters (Table S3), bond distances and angles (Table S4), enter-
ing-ligand concentration (Table S5 and Fig. S1), temperature (Ta-
ble S6 and Fig. S2), and pressure (Table S7) dependences of the
substitution rate constants for the mononuclear pt and tt complex-
es, entering-ligand concentration (Table S8 and Fig. S3) and tem-
perature (Table S9 and Fig. S4) dependences of the successive
substitution rate constants for the dinuclear pdt complex (in total
20 pages) are deposited as Document No.75004 at the Office of
the Editor of Bull. Chem. Soc. Jpn. and are also available from the
author upon request. Crystallographic data have been deposited at
the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK and copies
can be obtained on request, free of charge, by quoting the publica-
tion citation and the deposition number 174601.
7
Programs for X-ray crystal structure determination, MAC
Science, Yokohama (1994).
8
A. Altomare, G. Cascarano, C. Giacovazzo, and A.
Guagliardi, J. Appl. Crystallogr., 26, 343 (1993).
9
“International Tables for X-ray Crystallography,” Kynoch,
Birmingham (1974), Vol. ꢁ.
10 W. J. le Noble and R. Schlott, Rev. Sci. Instrum., 47, 770
(1976).
11 31P NMR (in CHCl3): δ (relative to D3PO4 in D2O) −17.48
(q, central), −12.99 (d, terminal); JP-P = 27 Hz.
12 C. J. Jameson and H. S. Gutowsky, J. Chem. Phys., 40,
1714 (1964).
13 N. F. Ramsey, Phys. Rev., 77, 567 (1950).
14 The substitution rate constants at 298.1 K for the reaction
of [Pd(pt)(p3)]+ with Br− and Cl− are 1.68×10−4 s−1 and
7.22×10−4 s−1 in the same conditions as those in this work, re-
spectively (unpublished results).
References
1
R. G. Wilkins, “Kinetics and Mechanism of Reaction of
Transition Metal Complexes,” 2nd ed, VCH, Weinheim (1991),
Chap. 4; R. J. Cross, “Mechanisms of Inorganic and Organometal-