Mechanistic Studies on Halo-Ligand Substitution of Five-Coordinate Trigonal-Bipyramidal Palladium(II) Complexes of Tris(2-(diphenylphosphino)ethyl)phosphine with Trimethyl Phosphite in Chloroform at Various Temperatures and Pressures

Article abstract of DOI:10.1021/ic9601596

Five-coordinate trigonal-bipyramidal palladium(II) complexes [Pd(pp₃)X]X (pp₃ = tris(2-(diphenylphosphino)-ethyl)phosphine, X^-- = Cl^-, Br^-, I^-) have been synthesized, and their structures i

Full text of DOI:10.1021/ic9601596

Inorg. Chem. 1996, 35, 5163-5167
5163
Mechanistic Studies on Halo-Ligand Substitution of Five-Coordinate Trigonal-Bipyramidal
Palladium(II) Complexes of Tris(2-(diphenylphosphino)ethyl)phosphine with Trimethyl
Phosphite in Chloroform at Various Temperatures and Pressures
Sen-ichi Aizawa, Takashi Iida, and Shigenobu Funahashi*
Laboratory of Analytical Chemistry, Faculty of Science, Nagoya University, Chikusa,
Nagoya 464-01, Japan
ReceiVed February 9, 1996X
Five-coordinate trigonal-bipyramidal palladium(II) complexes [Pd(pp3)X]X (pp3 ) tris(2-(diphenylphosphino)-
ethyl)phosphine, X- ) Cl-, Br-, I-) have been synthesized, and their structures in the solid state and in solution
have been confirmed by X-ray crystal structure analysis and 31P NMR spectroscopy, respectively. The 31P NMR
chemical shifts of the axial and equatorial phosphorus atoms indicate that the σ- and π-donating abilities of the
axial monodentate ligands are in the order Cl- > Br- > I- . P(OCH3)3. The thermodynamic parameters for the
equilibria between the halo complexes, [Pd(pp3)Cl]+ + X- a [Pd(pp3)X]+ + Cl- (X- ) Br-, I-), have been
determined as follows: K298 ) 1.70, ∆H° ) -5.66 ( 0.07 kJ mol-1, and ∆S° ) -14.6 ( 0.2 J K-1 mol-1 for
Br-; K298 ) 16.8, ∆H° ) -16 ( 1 kJ mol-1, and ∆S° ) -30 ( 3 J K-1 mol-1 for I-. It is revealed that the
relative stability of the halo complexes, I- > Br- > Cl- for the axial ligands, is determined by the difference in
enthalpy. The second-order rate constants at 25 °C and activation parameters for the halo-ligand substitution
with trimethyl phosphite, [Pd(pp3)X]+ + P(OCH3)3 f [Pd(pp3)(P(OCH3)3)]2+ + X- (X- ) Cl-, Br-, I-), have
been obtained as follows: k298 ) 1.19 × 10-1 mol-1 kg s-1, ∆Hq ) 19 ( 1 kJ mol-1, ∆Sq ) -198 ( 3 J K-1
mol-1, and ∆Vq ) -25.5 ( 0.5 cm3 mol-1 at 302.5 K for X- ) Cl-; k298 ) 6.20 × 10-2 mol-1 kg s-1, ∆Hq )
23 ( 3 kJ mol-1, ∆Sq ) -191 ( 3 J K-1 mol-1, and ∆Vq ) -24.5 ( 0.7 cm3 mol-1 at 302.5 K for X- ) Br-;
k298 ) 1.09 × 10-2 mol-1 kg s-1, ∆Hq ) 57 ( 4 kJ mol-1, ∆Sq ) -89 ( 15 J K-1 mol-1, and ∆Vq ) -22.6
( 0.8 cm3 mol-1 at 302.9 K for X- ) I-. The associative mechanism is proposed on the basis of the kinetic
behavior and the activation parameters. The kinetic properties are discussed in terms of the electronic properties
of the axial and entering ligands.
Introduction
substitution of the axial ligand in the trigonal-bipyramidal
complexes. Though such studies can give us further insight
into the kinetic properties of palladium(II) and platinum(II)
complexes, so far there have been only a few mechanistic studies
of trigonal-bipyramidal complexes of these metals.3
In the present work, we have selected the tripodal tetradentate
ligand, tris(2-(diphenylphosphino)ethyl)phosphine (pp3), as a
bound ligand because the pp3 ligand tends to form the trigonal-
bipyramidal structure for the d8 metal ions4-8 and have
synthesized the halo complexes [Pd(pp3)X]X (X- ) Cl-, Br-,
I-). We report the structural characterization of the halo
complexes in the crystal form and in solution and the equilibria
and kinetics for the axial monodentate-ligand substitution as
the simple reaction for the mechanistic study of the trigonal-
bipyramidal palladium(II) complexes. The present reuslts make
it possible to compare the effects of the leaving halo ligands
on the reaction mechanism.
The kinetics for ligand substitution reactions of metal
complexes in solutions have been extensively investigated1
because it is essential to clarify the reaction mechanism for metal
complexes in order to understand their properties in solutions
and in biological systems. So far, the kinetic properties of
square-planar palladium(II) and platinum(II) complexes have
been well established and the associative mechanism via a
trigonal-bipyramidal transition state has been generally pro-
posed.1 In this mechanism, since the vacant nonbonding pz
orbital perpendicular to the square plane is able to accommodate
an electron pair, an entering ligand may attack the vacancy.2
2
On the other hand, because the dz orbital is occupied, the
entering ligand shifts to the equatorial plane to reduce the
electronic repulsion and, consequently, pushes the leaving ligand
away from the equatorial plane to form the trigonal plane in
the transition state.2
In contrast with the square-planar complexes, the trigonal-
bipyramidal complexes of d8 metal ions have no vacant orbital.
Therefore, the lone pair of the entering ligand exclusively brings
about repulsive interaction with the occupied d orbitals in the
associative mechanism. On the other hand, if the dissociative
mechanism is promoted for the trigonal-bipyramidal complexes
with the leaving ligand in the axial position, the bond-breaking
energy is considerably large because of strong σ bonding with
Experimental Section
Reagents. Chloroform (Nakarai, Sp. Gr.), trimethyl phosphite
(Wako, Sp. Gr.), deuterated chloroform (CDCl3, Aldrich), and deuter-
ated dichloromethane (CD2Cl2, Aldrich) were dried over activated 4A
(3) (a) Meakin, P.; Jesson, J. P. J. Am. Chem. Soc. 1974, 96, 5751. (b)
Pearson, R. G.; Muir, M. M.; Venanzi, L. M. J. Chem. Soc. 1965,
5521.
(4) King, R. B.; Kapoor, R. N.; Saran, M. S.; Kapoor, P. N. Inorg. Chem.
1971, 10, 1851.
2
dz orbital. Therefore, it is significant to investigate the
(5) Ghilardi, C. A.; Midollini, S.; Sacconi, L. Inorg. Chem. 1975, 14,
1970.
(6) Orlandini, A.; Sacconi, L. Inorg. Chem. 1976, 15, 78.
(7) Sacconi, L.; Dapporto, P.; Stoppioni, P.; Innocenti, P.; Benelli, C.
Inorg. Chem. 1977, 16, 1669.
(8) Hohman, W. H.; Kountz, D. J.; Week, D. W. Inorg. Chem. 1986, 25,
616.
X Abstract published in AdVance ACS Abstracts, August 1, 1996.
(1) (a) Wilkins, R. G. Kinetics and Mechanism of Reaction of Transition
Metal Complexes, 2nd ed.; VCH: Weinheim, Germany, 1991; Chapter
4. (b) Cross, R. J. In Mechanisms of Inorganic and Organometallic
Reactions; Twigg, M. V., Ed.; Plenum: New York, 1988; Chapter 5.
(2) Lin, Z.; Hall, M. B. Inorg. Chem. 1991, 30, 646.
S0020-1669(96)00159-0 CCC: $12.00 © 1996 American Chemical Society

5164 Inorganic Chemistry, Vol. 35, No. 18, 1996
Aizawa et al.
Table 1. Crystallographic Data for [Pd(pp3)X]X‚CH3CN
Molecular Sieves and then distilled in a vacuum line. Tris(2-
(diphenylphosphino)ethyl)phosphine (pp3, Strem), bis(2-(diphenylphos-
phino)ethyl)phenylphosphine (p3, Strem), and tetrakis(acetonitrile)-
palladium(II) tetrafluoroborate (Aldrich) were used for the preparation
of the palladium(II) complexes without further purification.
X-
Cl-
Br-
I-
formula
fw
a/Å
b/Å
c/Å
R/deg
â/deg
γ/deg
V/Å3
Z
PdCl2P4NC44H45 PdBr2P4NC44H45 PdI2P4NC44H45
889.10
978.00
10.470(3)
10.482(5)
20.37(1)
103.53(2)
96.53(4)
104.71(4)
2089(2)
2
1072.00
10.417(4)
10.520(3)
20.158(6)
96.69(2)
103.98(1)
104.86(2)
2058(1)
2
10.853(4)
39.69(2)
10.430(2)
Preparation of Complexes. [Pd(pp3)I]I. To a solution of [Pd(CH3-
CN)4](BF4)2 in acetonitrile was added a solution of an equimolar amount
of pp3 in dichloromethane, followed by concentrating the solution and
keeping it in a refrigerator to give a pale yellow powder of [Pd(pp3)]-
(BF4)2.9 To a suspension of [Pd(pp3)](BF4)2 (1.42 g, 1.49 mmol) in a
dichloromethane/nitromethane (5:1 v/v) mixture (60 cm3) was added
n-Bu4NI (1.35 g, 3.66 mmol), whereupon the solution immediately
turned from orange to deep red. The solution was concentrated to ca.
30 cm3 and then kept in a refrigerator for 1 day. The resultant deep
red crystals were collected by filtration and recrystallized from a
dichloromethane/nitromethane (3:1 v/v) mixture. Yield: 90%. Single
crystals were obtained by slow evaporation of a dichloromethane/
nitromethane/acetonitrile solution and air-dried. Anal. Calcd for
[Pd(pp3)I]I: C, 48.93; H, 4.11; N, 0.00. Found: C, 46.75; H, 3.99; N,
0.36.
106.14(2)
4316(3)
4
space group P1 (No. 1)
P1 (No. 1)
295
0.710 73
1.58
25.15
0.089
P21 (No. 4)
295
0.710 73
1.65
20.16
0.070
0.089
T/K
λ/Å
295
0.710 73
F
calc/g cm-3 1.45
µ/cm-1
7.60
0.064
0.075
Ra
b
Rw
0.101
2
a R ) ∑||Fo| - |Fc||/∑|Fo|. b Rw ) [∑w(|Fo| - |Fc|)2/∑w|Fo| ]1/2
.
[Pd(pp3)Br]Br. To a solution of K2[PdBr4] (1.72 g, 3.41 mmol) in
distilled water (10 cm3) were added acetonitrile (80 cm3) and a solution
of pp3 (2.52 g, 3.76 mmol) in chloroform (10 cm3). The resultant red
solution was separated into two phases by addition of water and
chloroform. The chloroform phase was concentrated to obtain red
crystals. Yield: 75%. This complex was also prepared by a procedure
similar to that for [Pd(pp3)I]I using n-Bu4NBr instead of n-Bu4NI.
Yield: 75%. Single crystals were obtained by slow evaporation of an
acetonitrile/water solution and air-dried. Anal. Calcd for [Pd(pp3)-
Br]Br: C, 53.84; H, 4.52; N, 0.00. Found: C, 52.67; H, 4.71; N, 0.25.
[Pd(pp3)Cl]Cl. The chloro complex was prepared by procedures
similar to those for the bromo complex by using K2[PdCl4] instead of
K2[PdBr4] and for the iodo complex by using n-Bu4NCl instead of
n-Bu4NI. Yields: 89 and 80%, respectively. Single crystals were
obtained from an acetonitrile/water solution and air-dried. Anal. Calcd
for [Pd(pp3)Cl]Cl: C, 59.49; H, 4.99; N, 0.00. Found: C, 56.84; H,
5.01; N, 0.15.
crystallographic data are summarized in Table 1. The crystallographic
details, positional and thermal parameters for all atoms containing
hydrogen atoms observed, and bond distances and angles for non-
hydrogen atoms are listed in Tables S1-S4 (Supporting Information).
Measurements. The kinetic measurements for the substitution
reactions of the chloro and bromo complexes with trimethyl phosphite
in chloroform were carried out with a Shimadzu UV-265FW spectro-
photometer. The measurements for the kinetics of the iodo complex
and equilibria between halo complexes in deuterated chloroform were
performed by 31P NMR spectroscopy on a JEOL JMN-GX270 FT-
NMR spectrometer operating at 109.3 MHz with sufficient acquisition
time of at least 0.4 s. The substitution reactions at various pressures
were followed spectrophotometrically by using a high-pressure static
vessel equipped with a le Noble type quartz cell.13 The temperature
of the reaction solution was controlled within (0.1 °C. Though about
15 min was required for the temperature and/or pressure equilibration,
the reactions were sufficiently slow to obtain accurate rate constants.
All sample solutions were prepared under a nitrogen atmosphere. Tetra-
n-butylammonium bromide and iodide were used for the equilibrium
measurements to adjust the concentration of the halide. Rates were
measured under pseudo-first-order conditions where the concentrations
of trimethyl phosphite were in large excess over the concentrations of
the halo complexes. The leaving-ligand concentration dependence of
the rates was checked by use of a large excess of tetra-n-butylammo-
nium bromide for the bromo complex. The initial concentrations of
the reactants for each kinetic and equilibrium measurements are given
in Table S5 (Supporting Information).
[Pd(pp3)(P(OCH3)3](BF4)2. The trimethyl phosphite complex was
prepared by the procedure described previously.9
[Pd(p3)X]X (X ) Cl-, Br-, I-). The square-planar complexes were
prepared by procedures similar to those for the corresponding trigonal-
bipyramidal complexes by using p3 instead of pp3. Anal. Calcd for
[Pd(p3)Cl]Cl: C, 57.27; H, 4.67; N, 0.00. Found: C, 57.20; H, 4.70;
N, 0.00. Calcd for [Pd(p3)Br]Br: C, 50.00; H, 4.15; N, 0.00. Found:
C, 51.33; H, 4.20; N, 0.01. Calcd for [Pd(p3]I]I: C, 45.64; H, 3.72;
N, 0.00. Found: C, 44.54; H, 3.73; N, 0.04.
X-ray Structure Analysis. Each single crystal of chloro, bromo,
and iodo complexes suitable for diffraction measurements was sealed
in a 0.7 mm o.d. capillary tube without air-drying because acetonitrile
molecules in the crystal are volatile to collapse the single crystal.10
X-ray diffraction measurements were performed on a MAC Science
Rapid X-Ray Diffraction Image Processor (DIP 320N) with graphite-
monochromated Mo KR radiation (λ ) 0.710 73 Å) at ambient
temperature. Reflections were collected by using 30 continuous
Weissenberg photographs with a φ range of 6° (total φ range 0-180°).
The intensity data were corrected for the standard Lorentz and
polarization effects. An empirical absorption correction was not
applied. The structures were solved by direct methods and refined by
a full-matrix least-squares technique using Crystan-G (version 3).11 All
non-hydrogen atoms except for the carbon atoms for the iodo complexes
were refined with anisotropic thermal parameters, and the hydrogen
atoms were placed in the observed positions with fixed thermal
parameters. The atomic scattering factors were taken from ref 12. The
Results and Discussion
Characterization. Each crystal of the halo complexes
consists of two nonequivalent formula units, [Pd(pp3)X]X‚CH3-
CN (X- ) Cl-, Br-, I-). The perspective views of the chloro,
bromo, and iodo complexes are displayed in Figure 1a-c,
respectively. The selected bond distances and angles are
summarized in Table 2. The complex cations have a trigonal-
bipyramidal geometry with the halo ligand and the central
phosphorus atom of pp3 in axial positions and the three terminal
phosphorus atoms of pp3 in equatorial positions. The halide
counterions are located on sides opposite to the halo ligands. A
significant difference in the structures of the Pd(pp3) moieties
is not observed among the halo complexes, though the Pd-X
distances vary with the halide ion sizes. The average axial bond
distance (Pd-Pax ) 2.22-2.23 Å) is notably shorter than the
average equatorial bond distance (Pd-Peq ) 2.42-2.43 Å) for
each complex. This is attributed to the strong σ interaction of
(9) Aizawa, S.; Funahashi, S. Anal. Sci. 1996, 12, 27.
(10) It was confirmed by X-ray analyses of the sealed crystals that one
acetonitrile molecule was contained in each formula unit of the halo
complexes while elemental analyses of the air-dried samples did not
show the corresponding nitrogen content with reproducible values.
The loss of the acetonitrile molecules from the crystals hindered our
density measurements.
(12) International Tables for X-ray Crystallography; Kynoch: Birmingham,
England, 1974; Vol. IV.
(13) le Noble, W. J.; Schlott, R. ReV. Sci. Instrum. 1976, 47, 770.
(11) Programs of a structure determination package from MAC Science,
Yokohama, Japan, 1992.

Ligand Substitution Reactions of Pd(II) Complexes
Inorganic Chemistry, Vol. 35, No. 18, 1996 5165
Figure 1. ORTEP diagrams for [Pd(pp3)X]X‚CH3CN (X- ) Cl- (a), Br- (b), I- (c)). Acetonitrile molecules are omitted for clarity.
Table 2. Selected Bond Distances (Å) and Angles (deg) for
[Pd(pp3)X]X‚CH3CN
X- ) Cl-
X- ) Br-
X- ) I-
Pd1-P1 2.436(4) Pd1-P4
2.214(4) I1-Pd1
2.690(2)
2.500(8)
2.373(9)
2.418(6)
2.214(6)
2.684(3)
2.440(8)
2.420(6)
2.359(9)
2.241(5)
Pd1-P2
Pd1-P3
Pd1-P4
Pd1-Cl1
Pd2-P5
Pd2-P6
Pd2-P7
Pd2-P8
Pd2-Cl2
2.449(4) Pd1-P1
2.388(4) Pd1-P2
2.237(3) Pd1-P3
2.414(4) Pd1-Br1
2.446(4) Pd2-P8
2.419(4) Pd2-P5
2.456(4) Pd2-P6
2.214(3) Pd2-P7
2.422(3) Pd2-Br2
2.389(6) Pd1-P1
2.460(6) Pd1-P2
2.462(5) Pd1-P3
2.516(2) Pd1-P4
2.222(4) I2-Pd2
2.386(6) Pd2-P5
2.411(5) Pd2-P6
2.412(6) Pd2-P7
2.519(2) Pd2-P8
P4-Pd1-P1 83.3(1) P4-Pd1-P1 82.7(2) P4-Pd1-P1 84.9(2)
P4-Pd1-P2 84.1(1) P4-Pd1-P2 83.1(2) P4-Pd1-P2 81.5(3)
P4-Pd1-P3 83.6(1) P4-Pd1-P3 84.0(2) P4-Pd1-P3 85.8(2)
P4-Pd1-Cl1 178.7(2) P4-Pd1-Br1 178.0(2) P4-Pd1-I1 176.4(3)
P1-Pd1-P2 113.3(1) P1-Pd1-P2 121.9(2) P3-Pd1-P1 113.4(3)
P1-Pd1-P3 121.4(1) P1-Pd1-P3 121.7(2) P3-Pd1-P2 123.6(3)
P1-Pd1-Cl1 98.0(1) P1-Pd1-Br1 95.7(1) P3-Pd1-I1 95.7(2)
P2-Pd1-P3 121.6(1) P2-Pd1-P3 112.2(2) P2-Pd1-P1 119.8(2)
P2-Pd1-Cl1 95.3(1) P2-Pd1-Br1 98.7(1) P2-Pd1-I1 95.0(2)
P3-Pd1-Cl1 95.7(2) P3-Pd1-Br1 96.0(1) P1-Pd1-I1 97.4(1)
P8-Pd2-P5 83.2(1) P8-Pd2-P5 84.0(2) P8-Pd2-P5 84.4(2)
P8-Pd2-P6 83.0(1) P8-Pd2-P6 84.0(2) P8-Pd2-P6 82.3(2)
P8-Pd2-P7 83.7(1) P8-Pd2-P7 83.9(2) P8-Pd2-P7 84.8(2)
P8-Pd2-Cl2 178.2(2) P8-Pd2-Br2 178.2(2) P8-Pd2-I2 177.5(2)
P5-Pd2-P6 121.0(1) P5-Pd2-P6 121.8(2) P7-Pd2-P5 120.6(2)
P5-Pd2-P7 113.5(1) P5-Pd2-P7 121.2(2) P7-Pd2-P6 124.0(3)
P5-Pd2-Cl2 98.5(1) P5-Pd2-Br2 94.8(1) P7-Pd2-I2 92.9(2)
P6-Pd2-P7 121.5(1) P6-Pd2-P7 113.8(2) P6-Pd2-P5 112.0(3)
P6-Pd2-Cl2 96.1(1) P6-Pd2-Br2 95.5(1) P6-Pd2-I2 98.2(1)
P7-Pd2-Cl2 95.5(1) P7-Pd2-Br2 97.8(1) P5-Pd2-I2 97.6(1)
Figure 2. 31P NMR spectra of the chloro (a) and iodo (b) complexes
in the equilibrium in deuterated chloroform.
Table 3. 31P NMR Chemical Shifts for the Halo and Trimethyl
Phosphite Complexesa
chemical shift/ppmb
complex
Pax
Peq
[Pd(pp3)Cl]+
136.50
142.57
147.35
149.18c
31.62
31.76
31.98
43.48
[Pd(pp3)Br]+
[Pd(pp3)I]+
[Pd(pp3)(P(OCH3)3)]2+
a In deuterated dichloromethane. b Relative to 85% D3PO4 in D2O.
c Center of the doublet.
angles, Pax-Pd-Peq, are reduced to 83.5-84.0°. Since the
distortion of the trigonal-bipyramidal structure does not depend
on the size of the halo ligands, the chelate strain of the five-
membered P-C-C-P chelate is expected.
the axial bond in the trigonal-bipyramidal geometry. A similar
trend in difference in the axial and equatorial bond distances is
observed for the analogous d8 metal complexes [Co(pp3)-
(P(OCH3)3)]+ and [Ni(pp3)(P(OCH3)3)]2+ 8
.
The difference in
The 31P NMR spectra of the halo complexes in deuterated
chloroform and dichloromethane exhibit two peaks, the weaker
peak in the lower field and the more intense peak in the upper
field (Figure 2). The NMR spectra indicate that the halo
complexes have trigonal-bipyramidal geometry at least with C3
symmetry where the weaker and more intense peaks correspond
to those for the central phosphorus atom in the axial position
(Pax) and the three equivalent terminal phosphorus atoms in the
equatorial position (Peq), respectively. The values of the 31P
NMR chemical shift in dichloromethane are listed in Table 3.
The considerable lower field shift for Pax compared with the
chemical shift for Peq is observed for each halo complex though
the bond distances (Co-Pax ) 2.125 Å and Co-Peq ) 2.211
Å; Ni-Pax ) 2.181 Å and Ni-Peq ) 2.275-2.325 Å) is
obviously smaller than that for the Pd(II) complexes, though
the axial ligand, P(OCH3)3, trans to Pax is a weaker σ donor
compared to the halo ligands (vide infra). This finding indicates
that the σ interaction of the phosphorus atom with the d8 ions
in the second-row transition metal series is relatively larger than
that in the first-row transition metal series. For the present
complexes, each Pd(II) ion is displaced by 0.25-0.28 Å from
the equatorial plane defined by the three equatorial phosphorus
atoms toward the halo ligand, and consequently, the chelate bite

5166 Inorganic Chemistry, Vol. 35, No. 18, 1996
Aizawa et al.
Table 4. Absorption Spectral Data for the Halo and Trimethyl
Table 6. Activation Parameters for the Substitution Reactions of
[Pd(pp3)X]+ (X- ) Cl-, Br-, I-) with Trimethyl Phosphite in
Chloroform
Phosphite Complexesa
transition energy/103 cm-1 (log(ꢀ/M-1 cm-1))
k298/mol-1
∆Hq/kJ
∆Sq/JK-1
∆Vq/cm3
complex
1A1′ f 1E′
1A1′ f 1E′′
X-
kg s-1
mol-1
mol-1
mol-1
[Pd(pp3)Cl]+
20.75 (3.74 sh)
20.37 (3.80 sh)
20.77 (3.80)
23.47 (3.85 sh)
23.69 (3.90 sh)
28.03 (4.03)
Cl-
Br-
I-
11.9 × 10-2
6.20 × 10-2
1.09 × 10-2
19 ( 1
23 ( 1
57 ( 4
-198 ( 3
-191 ( 3
-89 ( 15
-25.5 ( 0.5a
-24.5 ( 0.7a
-22.6 ( 0.8b
[Pd(pp3)Br]+
[Pd(pp3)I]+
[Pd(pp3)(P(OCH3)3)]2+
24.10 (4.09 sh)
26.94 (4.32)
a At 302.5 K. b At 302.9 K.
a In dichloromethane.
In process of the substitution reaction of halo complexes with
trimethyl phosphite in deuterated chloroform (reaction 3), the
Table 5. Thermodynamic Parameters for the Equilibrium between
[Pd(pp3)Cl]+ and [Pd(pp3)X]+ (X- ) Br-, I-)
X-
Br-
I-
K298
∆H°/kJ mol-1 ∆S°/J K-1 mol-1 ∆G°/kJ mol-1 a
[Pd(pp3)X]+ + P(OCH3)3 f [Pd(pp3)P(OCH3)2]2+ + X-
1.70
16.8
-5.66 ( 0.07
-16 ( 1
-14.6 ( 0.2
-30 ( 3
-1.3 ( 0.1
-7.0 ( 2.1
(X- ) Cl-, Br-, I-)
(3)
a At 25 °C.
31P NMR peaks for the halo complexes and free trimethyl
phosphite decreased as the peaks for the trimethyl phosphite
complex9 increased and the other peaks were not observed. The
rates of the substitution reaction for the iodo complex were
measured by the increase in peak height for the equatorial
phosphorus atoms of the trimethyl phosphite complex under
pseudo-first-order conditions. Since the substitution reactions
for the chloro and bromo complexes are a little too fast to be
followed by NMR measurements, the reactions were followed
spectrophotometrically in chloroform solution. The isosbestic
points in their spectral changes were observed at 324 and 403
nm for the bromo complex and at 321 and 401 nm for the chloro
the equatorial terminal phosphorus atoms have two electron-
withdrawing phenyl groups. This fact is consistent with the
stronger σ donation from the axial phosphorus atom to the
palladium(II) center as observed in the crystal structures.
Furthermore, the electronic absorption spectra of [Pd(pp3)X]X
(X- ) Cl-, Br-, I-) and [Pd(pp3)(P(OCH3)3)](BF4)2 in dichlo-
romethane are compatible with those for the low-spin trigonal-
bipyramidal complexes of d8 metal ions.14 The two absorption
bands in the visible region for each complex are assigned to
1
1
1
the transitions from A1′ to E′ and E′′ (Table 4).
As shown in Table 3, the 31P NMR chemical shift is sensitive
to the axially coordinated halo ligands. The order of the upfield
shifts of Pax and Peq for [Pd(pp3)X]+ (X- ) Cl-, Br-, I-) and
[Pd(pp3)(P(OCH3)3)]2+ is Cl- > Br- > I- > P(OCH3)3 for the
axial ligands. The σ donation of the axial ligand weakens the
σ Pd-Pax bond because of the competition for σ donation to
complex. The observed pseudo-first-order rate constants, kobs
,
were determined by a least-squares analysis for the time course
of the absorbance. The deviations of the kobs values obtained
at different wavelengths, 480, 440, and 350 nm, are less than
(2%. Since the observed rate constants are proportional to the
concentration of trimethyl phosphite as shown in Table S7 and
Figure S2 (Supporting Information), reaction 3 is first order with
respect to the entering-ligand concentration. The temperature
dependence of the second-order rate constant, k, was fitted to
the Eyring equation to give the values of activation enthalpy
(∆Hq) and activation entropy (∆Sq) (Figure S3, Supporting
Information). Since the plots of ln k vs pressure for the
substitution reactions were linear within the experimental errors
(Figure S4, Supporting Information), the values of activation
volume (∆Vq) were determined from the slopes. All the
activation parameters obtained are summarized in Table 6.
As apparent from the ∆G° values at 25 °C in Table 5, the
relative stability of the halo complexes is in the order I- > Br-
> Cl- for the axial halo ligands, and the order is determined
by the difference in enthalpy. The contribution of entropy to
the stability is in the reverse order because the larger halo ligand
surrounded by the phenyl groups brings about more steric
crowding. On the basis of the 31P NMR chemical shifts, the
enthalpic stabilization is attributable to the weakness of the π
donation of the halo ligands rather than the strength of the σ
donation. Such π interaction reduces the electronic repulsion
between the occupied dxz and dyz orbitals and halo ligand and
consequently strengthens the Pd-halo ligand bonding, while
the σ donation of the halo ligands competitively weakens the σ
bonding in the trans position.
2
the dz orbital of the Pd(II) ion and, consequently, causes the
shielding of the axial phosphorus atom. On the other hand,
the π donation of the axial ligand increases the electron density
in the dxz and dyz orbitals of the Pd(II) ion which have direct π
interaction with the empty d orbitals of the equatorial phosphorus
atoms. It should be mentioned that the influence of π donation
of the axial ligand toward the axial phosphorus through the dxz
and dyz orbitals is minor compared with that of the σ donation.
Accordingly, the 31P NMR chemical shifts for the present
complexes demonstrate that the strengths of σ and π donations
are both in the order Cl- > Br- > I- > P(OCH3)3.
Equilibrium and Kinetics. It was confirmed by the 31P
NMR spectra that the equilibria, (1) and (2), between halo
K1
\
[Pd(pp3)Cl]+ + Br- {
} [Pd(pp3)Br]+ + Cl-
(1)
(2)
K2
[Pd(pp3)Cl]+ + I- { } [Pd(pp3)I]+ + Cl-
\
complexes in deuterated chloroform were reached without any
byproduct. The equilibrium constants, K1 and K2, were evalu-
ated by the initial concentrations of the halo complexes and
tetra-n-butylammonium halide and the relative intensity of the
31P NMR signals for the axial phosphorus atoms of each halo
complex in an equilibrium state which are well resolved (Figure
2). The temperature dependence of the equilibrium constants
for each reaction is shown in Table S6 and Figure S1 (supporting
information). The thermodynamic parameters obtained are listed
in Table 5.
We should consider some reaction mechanisms only on the
basis of the good linearity of the entering-ligand concentration
dependence of the pseudo-first-order rate constants. It has been
confirmed that the pre-equilibrium with the dissociation of the
halo ligand (mechanism I in Scheme 1) does not exist in the
process of the substitution reaction from the fact that the reaction
is not retarded by an increase in the halide ion concentration
(14) Lever, A. B. P. Inorganic Electronic Spectroscopy, 2nd ed.; Elsevier:
Amsterdam, 1984; Chapter 6.

Ligand Substitution Reactions of Pd(II) Complexes
Inorganic Chemistry, Vol. 35, No. 18, 1996 5167
Scheme 1
acceptor and very weak σ donor, it is the most plausible that
occupied d orbitals such as dxz and dyz of the Pd(II) ion interact
with the empty d orbital of the phosphorus atom in trimethyl
phosphite in the transition state and that the electronic repulsion
between d orbitals of the Pd(II) ion and lone pair of the
phosphorus atom is weak. A comparison of the activation
parameters for the three halo complexes reveals that the reaction
mechanism becomes more associative in the order I- < Br- <
Cl- for the axial ligands. This order corresponds to that for
the π-donating ability of the halo ligands which increase the
electron density of the dxz and dyz orbitals and is advantageous
to the interaction with the electron-accepting ligand. Therefore,
the change in the activation parameters for the halo complexes
supports the reaction mechanism proposed above. In addition,
the distinguishably large negative ∆Vq values compared with
those of the substitution reactions of the square-planar Pd(II)
complexes15 are due to the large void volume around the
substitution site surrounded by the six phenyl groups which can
accommodate the large volume of the entering ligand in the
transition state.
Acknowledgment. This research was supported by Grants-
in-Aid for Scientific Research (Nos. 06640779, 07454199, and
07504003) from the Ministry of Education, Science, and Culture
of Japan. S.A. gratefully acknowledges receipt of a grant from
the Kurata Foundation.
Supporting Information Available: X-ray crystallographic data
(Table S1), atomic coordinates and isotropic thermal parameters (Table
S2), anisotropic thermal parameters (Table S3), bond distances and
angles (Table S4), initial compositions of the sample solutions for the
kinetic and equilibrium measurements (Table S5), temperature depend-
ences of the equilibrium constants (Table S6 and Figure S1), entering-
and leaving-ligand concentration dependences of the observed rate
constants (Table S7 and Figure S2), temperature dependences of the
second-order rate constants (Table S8 and Figure S3), and pressure
dependences of the second-order rate constants (Table S9 and Figure
S4) (37 pages). Ordering information is given on any current masthead
page.
(Table S7d). Judging from the much more negative values of
∆Sq and ∆Vq and relatively small values of ∆Hq for the present
ligand substitution of the trigonal-bipyramidal Pd(II) complexes
compared with those for the associative ligand substitution and
solvent exchange of the square-planar Pd(II) complexes so far
studied,15 it is not probable that the square-planar intermediate
is formed by the pre-equilibrium with the dissociation of one
of the terminal diphenylphosphino groups (mechanism II in
Scheme 1), because the first step in the dissociation should give
positive entropy and volume changes and enthalpic destabiliza-
tion due to bond cleavage.16 Consequently, we can reasonably
expect that the associative activation mode with the entering
trimethyl phosphite attack on the trigonal-bipyramidal Pd(II)
complex operates in the transition state (mechanism III in
Scheme 1). Considering that trimethyl phosphite is a strong π
IC9601596
(16) We have obtained preliminary data for halo-ligand substitution
reactions with trimethyl phosphite by using the corresponding square-
planar halo complexes [Pd(p3)X]+ (p3 ) bis(2-(diphenylphosphino)-
ethyl)phosphine; X- ) Cl-, Br-, I-). The second-order rate constant
at 25 °C (6.59 × 10-4 mol-1 kg s-1) for the chloro complex obtained
by the spectrophotometric method is much smaller than that for the
corresponding five-coordinate complex in the present work under the
same conditions (Table S7e). This fact is not consistent with
mechanism II, Scheme 1, because k1/k2 should be much smaller than
unity.
(15) Kotowski, M.; van Eldik, R. In Inorganic High Pressure Chemistry;
van Eldik, R., Ed.; Elsevier: Amsterdam, 1986; Chapter 4.