Kinetics and mechanism of the monomerization of a Re(V) dithiolato dimer with monodentate ligands. Electronic and steric effects

Article abstract of DOI:10.1021/ic991431p

The sulfur-bridged dimeric dithiolato rhenium(V) chelate [CH₃(O)Re(η²,μ-o-SCH₂C₆H₄S)]₂ (D), derived from 2-mercaptothiophenol, was monomerized to give [CH₃(O)Re(η²-o-SCH₂C₆H₄S)]L (M-L) in benzene upon reaction with various neutral and anionic monodentate ligands (L) such as pyridine and its substituted derivatives, triarylphosphines, dimethyl sulfoxide, 4-picoline-N-oxide, and halide ions. The kinetic observations can readily be interpreted for all ligands by a unified mechanism in which the initial fast formation of a 1:1 (DL) and 1:2 (DL₂) adduct is followed by the slow monomerization of each species so formed. The use of different ligands gave insight into different steps of the same multistep mechanism. The kinetics of ligand exchange between free L and the monomeric complexes was also studied; an associative pathway has been proposed to interpret the results. The crystal structures of two new monomeric ML complexes (with L = 4-acetylpyridine and 1,3-diethylthiourea) are reported.

Full text of DOI:10.1021/ic991431p

Inorg. Chem. 2000, 39, 1311-1319
1311
Kinetics and Mechanism of the Monomerization of a Re(V) Dithiolato Dimer with
Monodentate Ligands. Electronic and Steric Effects
Ga´bor Lente, Ilia A. Guzei, and James H. Espenson*
Ames Laboratory and Department of Chemistry, Iowa State University of Science and Technology,
Ames, Iowa 50011
ReceiVed December 15, 1999
The sulfur-bridged dimeric dithiolato rhenium(V) chelate [CH3(O)Re(η2,µ-o-SCH2C6H4S)]2 (D), derived from
2-mercaptothiophenol, was monomerized to give [CH3(O)Re(η2-o-SCH2C6H4S)]L (M-L) in benzene upon reaction
with various neutral and anionic monodentate ligands (L) such as pyridine and its substituted derivatives,
triarylphosphines, dimethyl sulfoxide, 4-picoline-N-oxide, and halide ions. The kinetic observations can readily
be interpreted for all ligands by a unified mechanism in which the initial fast formation of a 1:1 (DL) and 1:2
(DL2) adduct is followed by the slow monomerization of each species so formed. The use of different ligands
gave insight into different steps of the same multistep mechanism. The kinetics of ligand exchange between free
L and the monomeric complexes was also studied; an associative pathway has been proposed to interpret the
results. The crystal structures of two new monomeric ML complexes (with L ) 4-acetylpyridine and
1,3-diethylthiourea) are reported.
Introduction
certain mechanistic aspects of MTO-catalyzed atom-transfer
reactions. D can be monomerized readily with various ligands,
giving Re(V) complexes having a monodentate ligand in
addition to the chelating dithiolato ligand.
Methyltrioxorhenium(VII) (CH3ReO3, abbreviated as MTO)
is a versatile catalyst. It is air- and water-stable, and soluble in
a large variety of solvents in which it can be used in a number
of useful transformations.1-4 In one group of reactions MTO is
used to catalyze oxidations by hydrogen peroxide. In another,
it catalyzes various atom-transfer reactions from a donor to an
acceptor. In this second group of reactions, lower oxidation-
state rhenium intermediates, most notably those of Re(V), have
been postulated from kinetics evidence and sometimes de-
tected.5,6 Recent work on the MTO-catalyzed stereospecific
desulfurization of thiiranes7 focused our attention on thiolato
rhenium(V) complexes in which the rhenium-carbon bond
remains fully intact. Reaction of MTO with the chelating ligand
l,2-mercaptothiophenol in toluene gave a sulfur-bridged dithi-
olato methyloxorhenium(V) dimer (designated here as D):8
(2)
In this paper we report detailed studies of the kinetics and
mechanism of the monomerization reaction with a series of
ligands and of exchange reactions between the free ligands and
the ML complexes. This includes the results of kinetics and
equilibrium experiments, X-ray crystallography, and matrix rank
analysis for the determination of the number of species present.
Experimental Section
Materials. MTO was synthesized from sodium perrhenate and
tetramethyltin (Strem) according to the procedure published earlier.9
2-Mercaptomethylthiophenol10,11 and the dimer D8 were synthesized
as described earlier. Other reagents, including substituted pyridines
(Aldrich) and triphenylphosphines (Strem), were purchased from
commercial sources and used without further purification except for
4-pyridinecarboxaldehyde that was distilled in a vacuum and handled
under argon. Benzene (Fisher spectraanalyzed) was used as a solvent
throughout this study. Unless otherwise stated, the rate constants
reported in this paper were determined at 25.0 °C.
Our principal objective has been to characterize the reactivity
of the dimer D because such studies may further enlighten
(1) Roma˜o, C. C.; Ku¨hn, F. E.; Herrmann, W. A. Chem. ReV. 1997, 97,
3197-3246.
(2) Conry, R. R.; Mayer, J. M. Inorg. Chem. 1990, 29, 4862-4867.
(3) Espenson, J. H. Chem. Commun 1999, 479-488.
(4) Espenson, J. H.; Abu-Omar, M. M. AdV. Chem. Ser. 1997, 253, 99-
134.
(5) Abu-Omar, M. M.; Appelmann, E. H.; Espenson, J. H. Inorg. Chem.
1996, 35, 7751-7757.
(6) Abu-Omar, M. M.; Espenson, J. H. Inorg. Chem. 1996, 34, 6239-
6240.
(7) Jacob, J.; Espenson, J. H. Chem. Commun. 1999, 1003-1004.
(8) Jacob, J.; Guzei, I. A.; Espenson, J. H. Inorg. Chem. 1999, 38, 1040-
1041.
(9) Herrmann, W. A.; Kratzer, R. M.; Fischer, R. W. Angew. Chem., Int.
Ed. Engl. 1997, 36, 2652-2654.
(10) Klingsberg, E.; Screiber, A. M. J. Am. Chem. Soc. 1962, 84, 2941-
2944.
(11) Hortmann, A. G.; Aron, A. J.; Bhattacharya, A. K. J. Org. Chem.
1978, 43, 3374-3378.
10.1021/ic991431p CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/01/2000

1312 Inorganic Chemistry, Vol. 39, No. 6, 2000
Lente et al.
Instrumentation. Shimadzu scanning spectrophotometers and a
Shimadzu Multispec 1501 diode-array spectrophotometer were used
to record UV-vis spectra and kinetic curves. A 1 cm cell was used
except for the study with tetrabutylammonium iodide where its low
solubility required much lower concentrations and a 10 cm cell. An
Applied Photophysics DX-17 MV sequential stopped-flow apparatus
was used in single mixing mode for fast reactions. The optical path
length in the stopped-flow instrument was 1 cm. In every kinetics
experiment (stopped-flow and conventional UV-vis) the measured
initial absorbances were compared with corresponding blank curves,
in which one of the reagents was absent, to detect possible reactions
occurring in the mixing time. Single-wavelength kinetic studies for the
monomerization reactions were usually run at 610 nm, where all the
ML complexes have considerable absorption and D absorbs very little.
Kinetics data for the ligand exchange reactions were evaluated at ∼400
nm to obtain the maximum absorbance change.
Chart 1
dissolved in 2 mL of benzene was added 100 µL of ethyl isothiocyanate,
which is known to react with moisture giving 1,3-diethylthiourea. The
solution was stored at room temperature; after 1 month crystals formed.
X-ray crystal data: monoclinic, C2/c, a ) 26.5431(13) Å, b ) 9.4650-
(5) Å, c ) 15.6416(7) Å, â ) 117.385(1)°, V ) 3489.3(3) Å3, Z ) 8,
T ) 173(2) K, Dcalc ) 1.918 Mg/m3, R(F) ) 1.93% for 3555
independently observed (I g 2σ(I)) reflections (4° e 2θ e 53°).
Crystallography. The crystal evaluation and data collection were
performed using a Bruker CCD-1000 diffractometer with Mo KR (λ
) 0.710 73 Å) radiation. The diffractometer-to-crystal distance was
5.08 cm. All atoms other than hydrogen were refined with anisotropic
displacement coefficients. All hydrogen atoms were included in the
structure factor calculation at idealized positions and were allowed to
ride on the neighboring atoms with relative isotropic displacement
coefficients. The software and sources of the scattering factors are
contained in the SHELXTL (version 5.1) program library.16 Absorption
corrections were carried out by programs SADABS.17
1H NMR spectra were recorded with Varian VXR 300 MHz and
1
Bruker DRX 400 MHz spectrometers. The H chemical shifts were
measured relative to the residual 1H resonance of the deuterated solvent
C6D6 (δ ) 7.16 ppm). Because all the reactions were accompanied by
substantial UV-vis spectral changes, the NMR technique was used
only to monitor reactions qualitatively and to identify the products.
Computation. Stopped-flow kinetics curves were fitted with the
nonlinear least-squares routine in the software package provided by
the manufacturer. Each rate constant was determined as the average of
at least five values from replicate kinetic runs and was reproducible
within 5%. The software PSEQUAD12 was used to obtain equilibrium
constants from multiwavelength absorbance data. These calculations
are based on the net absorbance change from beginning to end and are
independent of the kinetic measurements. Typically, absorbance read-
ings were recorded at 25-35 different wavelengths in the 320-700
nm range; this set of values was used in a global fit to extract the
equilibrium constant for each pyridine derivative. The concentration
ranges were [Re]T ) 0.2-2 mmol L-1 and [L]T ) 2-800 mmol L-1
and were designed for each individual ligand specifically to optimize
the precision of equilibrium measurements.
Results
Preliminary Considerations. The monomerization of D is
believed to occur via two intermediates, DL and DL2, the
suggested structures of which are given in Chart 1. The
following scheme was proposed earlier to interpret monomer-
ization reactions with pyridine and triphenylphosphine:18,19
D + L yk1z DL
(3)
(4)
Rate constants from spectrophotometric experiments were obtained
by nonlinear least-squares fitting. Linearized forms were used only to
visualize data. The method of flooding was used whenever applicable,
and the functional dependence of the pseudo-first-order rate constants
on the concentration of the reagent in excess was used to derive the
rate law. With a few ligands, multiwavelength kinetic data were fitted
simultaneously in a global fit to determine pseudo-first-order rate
constants. That is, from data at n wavelengths, one obtains from this
method a single kobs, n amplitudes (Abs0 - Abs∞), and n end points
(Abs∞).13
k-1
DL + L yk2z DL2
k-2
Calculations for matrix rank analysis14 were carried out by the
software MRA.15 This method was used to determine the minimum
number of absorbing species from UV-vis spectral data based on the
singular values of the data matrix. An example of this process is given
in the Supporting Information.
Isolation and Crystallographic Parameters of M-L, L )
4-Acetylpyridine, C15H16NO2ReS2. Preparation. To 3 mg of D
dissolved in 2 mL of benzene was added 30 µL of 4-acetylpyridine.
The color of the solution turned a bright greenish-yellow in about 15
min. The product was isolated in crystalline form with slow diffusion
of hexane into the solution. X-ray crystal data: monoclinic, P21/c, a
) 16.8291(9) Å, b ) 7.4348(4) Å, c ) 12.9117(7) Å, â ) 98.710(1)°,
V ) 1596.89(15) Å3, Z ) 4, T ) 173(2) K, Dcalc ) 2.049 Mg/m3, R(F)
) 2.34% for 3761 independently observed (I g 2σ(I)) reflections (4°
e 2θ e 57°).
(5)
k4
DL2
98 2ML
(6)
This scheme has been able to account for all of the
monomerization reactions studied. Let us assume that reactions
3 and 4 are rapidly established equilibria in all instances because
that has been shown to be the case directly for every case in
which DL or DL2 was formed in detectable amounts. With that
assumption, then one can derive the following rate law for
experiments done with a large excess of ligand:
k3K1[L] + k4K1K2[L]2
1 + K1[L] + K1K2[L]2
d[D]
dt
d[ML]
dt
1
2
-
)
)
[D] (7)
Isolation and Crystallographic Parameters of M-L, L ) 1,3-
Diethylthiourea, C13H21N2OReS3. Preparation. To 3 mg of D
where the parameters are those given in eq 3-6, with K1 )
(12) Ze´ka´ny, L.; Nagypa´l, I. In Computational Methods for the Determi-
nation of Formation Constants; Legett, D. J., Ed.; Plenum Press: New
York, 1985; pp 291-299.
(16) Sheldrick, G. Bru¨ker Analytical X-ray Systems; Madison, WI,.
(17) Blessing, R. H. Acta Crystallogr. 1995, A51, 33-38.
(18) Jacob, J.; Lente, G.; Guzei, I. A.; Espenson, J. H. Inorg. Chem. 1999,
38, 3762-3763.
(19) Lente, G.; Jacob, J.; Guzei, I. A.; Espenson, J. H. Inorg. React. Mech.,
in press.
(13) Scientist, version 2.0; Micromath Software, 1995.
(14) Peintler, G.; Nagypa´l, I.; Jancso´, A.; Epstein, I. R.; Kustin, K. J. Phys.
Chem. A 1997, 101, 8013-8020.
(15) Peintler, G. version 3.04; Attila Jo´zsef University: Szeged, 1997.

Monomerization of Re(V) Dithiolato Dimer
Inorganic Chemistry, Vol. 39, No. 6, 2000 1313
Table 1. Kinetic and Equilibrium Data for Monomerization with Pyridine Derivatives at 25.0 °C in Benzene
σ
KD/L mol-1
kb/L2 mol-2 s-1
ka/L mol-1 s-1
4-Me2NC5H4N
4-methoxypyridine
4-tert-butylpyridine
4-picoline
3-picoline
4-phenylpyridine
pyridine
3-phenylpyridine
3-chloropyridine
4-pyridinecarboxaldehyde
4-acetylpyridine
3-cyanopyridine
4-cyanopyridine
pyrazine
-0.83
-0.27
-0.20
-0.17
-0.07
-0.01
0.00
0.06
0.37
0.45
0.50
0.56
0.66
b
>5 × 104 c
(6.41 ( 0.03) × 104
(3.76 ( 0.05) × 102
(9.7 ( 0.3) × 101
(1.33 ( 0.02) × 102
(6.19 ( 0.06) × 101
(7.0 ( 0.06) × 101
(1.31 ( 0.03) × 101
(7.3 ( 0.5) × 100
(7.7 ( 0.1) × 10-2
(3.3 ( 0.4) × 10-1
(1.5 ( 0.1) × 100
(2.9 ( 0.2) × 10-3
(2.7 ( 0.2) × 10-2
(2.5 ( 0.8) × 10-2
(1.9 ( 0.3) × 10-3
a
<6 × 101 a
(4.0 ( 0.1) × 103
(4.07 ( 0.09) × 103
(1.00 ( 0.05) × 103
(1.07 ( 0.02) × 103
(1.15 ( 0.03) × 103
(1.74 ( 0.04) × 102
(8.9 ( 0.2) × 101
(2.5 ( 0.2) × 10-1
(8.9 ( 0.8) × 100
(1.2 ( 0.2) × 101
∼3 × 10-3
<5 × 100 a
<2 × 100 a
<3 × 100 a
<1 ×100 a
<1 × 100 a
<3 × 10-1 a
(1.1 ( 0.3) × 10-1
(4.0 ( 1.0) × 10-3
(2.7 ( 0.2) × 10-2
(4.9 ( 0.5) × 10-2
(4.3 ( 0.8) × 10-4
(2.8 ( 0.7) × 10-3
(3.3 ( 0.5) × 10-3
(1.5 ( 0.1) × 10-3
(9.0 ( 0.1) × 10-3
(1.9 ( 0.2) × 10-1
(1.3 ( 0.1) × 10-1
(2.2 ( 0.2) × 10-2
∼1 × 10-2
4-methylthiazole
2-picoline
b
b
a Too small to be determined reliably. b Not defined. c Practically complete reaction.
k1/k-1 and K2 ) k2/k-2. Over any particular range of ligand
concentration, this equation gives rise to a number of limiting
rate laws depending on the particular values of equilibrium and
rate constants that are dominant. These forms are of considerable
use in resolving certain steps in the reaction scheme; the concept
of these limiting rate laws will be elaborated in the following
sections.
L ) Pyridine. The reaction between D and pyridine produces
the indicated monomeric product, the crystal structure of which
was reported earlier.19 In solution, however, this reaction does
not proceed to completion. With both NMR and UV-vis
detection, an increase in pyridine concentration caused more
monomer to form. A series of experiments was carried out with
various pyridine derivatives containing substituents in the 3 or
4 position. Both types of data were consistent with reaction 2
being an equilibrium characterized by the equilibrium constant
KD.
to pseudo-first-order kinetics. In these cases the rate law requires
the inclusion of a term for the reversion reaction. Then
d[D]
dt
) -kobs+[D] + kobs-[M-L]2
(10)
in which kobs+ and kobs- are the respective pseudo-nth-order rate
constant for the reaction in either direction. The resulting
differential equation was solved exactly with a method similar
to that derived by King.20 The solution was converted to
absorbance readings:
eqt
(1 + q)[Abs0 - Abs∞] e-k
Abst ) Abs∞ +
(11)
eqt
1 + q e-k
where Abs0 is the initial absorbance and Abs∞ the final
absorbance. The rate constant keq represents the value for the
approach to equilibrium and was determined along with q from
least-squares fitting to eq 11. The other two constants can be
expressed in terms of keq as
[M-L]2
[D][L]2
KD )
(8)
keq ) kobs+ + 4kobs-[M-L]∞,
4kobs-[M-L]∞
The equilibrium constants so determined are listed in Table 1.
With the exception of L ) 4-(dimethylamino)pyridine,
considered later, the following expression proved applicable to
each of the pyridine ligands. It expresses the rate of the forward
reaction in terms of concentrations:
q )
(12)
2kobs+ + 4kobs-[M-L]∞
Thus, a straightforward combination of the determined
parameters gave kobs+
:
Vf ) {ka[L] + kb[L]2}[D]
(9)
1 - q
k
obs+ ) k
(13)
eq1 + q
This rate law is a limiting form of eq 7 applicable when K1[L]
+ K1K2[L]2 , 1, the conditions under which the concentrations
of the complexes DL and DL2 remain much lower than that of
D throughout the reaction. Comparison of eqs 7 and 9 yields ka
) k3K1 and kb ) k4K1K2. These composite constants cannot be
further resolved for the pyridine derivatives without an inde-
pendent evaluation of K1 and K2.
For the more reactive pyridine derivatives (X ) 3-Me, 4-Ph,
4-MeO, 4-t-Bu, 4-Me, and H) ka was too small to be determined
reliably; thus, the overall third-order term dominated in the rate
law.
The measured kinetic curves were fitted to eq 11. The quantity
kobs+ holds the same significance that kobs does for complete
reactions (see Supporting Information for further details). This
treatment was preferred over the alternative where one fits only
the last portion of the kinetic curves with an exponential function
to determine an equilibration rate constant because the method
outlined afforded direct information on the forward rate
constants derived from data over the full time course of the
experiment.
The results for all derivatives are summarized in Table 1.
The values of the determined equilibrium constants KD span a
remarkably broad range, >6 orders of magnitude. Reliable
determinations throughout this broad range were possible
because of the second-order dependence on ligand concentration
and the excellent solubility of the pyridines in benzene. A linear
The least reactive pyridine derivatives (X ) 3-CN, 4-CN,
3-Cl, 4-CHO) did not allow the use of a ligand in high enough
concentration to shift the equilibrium of reaction 2 to completion.
Also, in these cases the absorbance-time data did not adhere
(20) King, E. L. Int. J. Chem. Kinet. 1982, 14, 1285-1286.

1314 Inorganic Chemistry, Vol. 39, No. 6, 2000
Lente et al.
Figure 3. Hammett plot for the equilibrium between D and M with
ring-substituted pyridines. Equilibrium constants were determined by
UV-vis spectrophotometry in benzene at 25.0 °C. The three substit-
uents that lie off the correlation in this plot are the same three that
deviate from the lines in Figure 2.
Figure 1. Linear free energy plot for monomerization of D with ring-
substituted pyridines. Shown are values of ka and kb, eq 9, against KD
for eqs 2 and 8, all on logarithmic scales. Rate and equilibrium constants
were determined in benzene at 25.0 °C: (9) ka; (b) kb; (0) pyrazine
ka; (O) pyrazine kb; (") 2-methylthiazole ka; (]) 2-methylthiazole kb.
Slopes of straight lines are 0.55 ( 0.03 (ka) and 0.83 ( 0.04. (kb).
Figure 2. Hammett plot (log ka and log kb, eq 9, as squares and circles,
respectively) against the substituent constant σ for the kinetics of
monomerization of D with ring-substituted pyridines. Rate constants
were determined in benzene at 25.0 °C.
Figure 4. UV-vis spectra for two monomeric ML complexes, with
L ) pyridine and 4-acetylpyridine in benzene at 25.0 °C.
monomers. The UV-vis spectra of the pyridine and 4-acetylpy-
free energy plot for the monomerization reaction 2 is shown in
Figure 1, which presents log k vs log KD. All the points for 3
and 4-substituted pyridine derivatives are on the lines, including
those that deviated in the Hammett treatment, next described.
A second approach was in terms of Hammett plots for the
equilibrium and rate constants, shown in Figures 2 and 3. Of
the 12 data, 9 define straight lines. Three points, those for
4-acetyl, 4-formyl, and 4-cyano derivatives, deviate significantly
from the straight line fitted to the remaining derivatives. The
slopes of the Hammett plots, neglecting these three for the
present, give the reaction constants F ) -4.80 for ka, -6.09
for kb, and -7.51 for KD with correlation coefficients of 0.9997,
0.984, and 0.987. These values of F imply a large electronic
effect on the reaction. The deviations in the Hammett plot are
not likely to be a consequence of a mechanism change, however,
since the plots of log k vs log KD are linear with these three
data included.
ridine monomers are given in Figure 4 as an example. For the
4-acetylpyridine monomer, values of λmax/nm (ꢀ/L mol-1 cm-1
)
are 444 (8620) and 586 (569); for the pyridine monomer, they
are 368 (7540) and 608 (256). The monomeric adduct with
4-acetylpyridine was crystallized and its structure determined
by X-ray crystallography to learn whether the different color
and the deviation on the Hammett plot might not be a
consequence of the product having a different composition or
structure. The solved structure is shown in Figure 5. Comparison
with the crystal structure of the monomeric adduct with
pyridine19 shows that this is not so. The two structures are
analogous; the Re-N distances are very similar in the two
structures, 215.3 and 214.9 pm. The rhenium atom exhibits a
distorted trigonal bipyramidal geometry with atoms S(1), O(1),
and C(1) in the equatorial plane and atoms S(2) and N in the
apical positions. Atoms Re, S(1), O(1), and C(1) are coplanar
within 10 pm. The angle S(2)-Re-N spans 152.68(8)°,
differing significantly from linearity. As a result of the trans
location of atoms S(2) and N the Re-S(2) distance, 229.95(9)
pm, is longer than the Re-S(1) bond length, 228.15(9) pm, and
The monomeric complexes ML with L ) 4-acetylpyridine,
4-pyridinecarboxaldehyde, and 4-cyanopyridine (the deviating
points on the Hammett plot) were bright greenish-yellow,
significantly different from the usual green color of pyridine

Monomerization of Re(V) Dithiolato Dimer
Inorganic Chemistry, Vol. 39, No. 6, 2000 1315
2-picoline (2-methylpyridine) was not enough to monomerize
D completely. Only ka contributed significantly to the rate; its
value is given in Table 1. Consistent with that, treatment of D
with excess 2,6-lutidine (2,6-dimethylpyridine), 2,6-di-tert-butyl-
4-methylpyridine, or 2-phenylpyridine gave no evidence for
monomerization.
4-(Dimethylamino)pyridine, DAP, an extremely nucleophilic
pyridine derivative, gave the usual green color with D in a
reaction that was complete even at the lowest concentrations.
N,N-Dimethylaniline, on the other hand, did not react with D,
leading us to conclude that the nitrogen in the pyridine ring of
DAP is the one coordinated to rhenium forming M-DAP, as
with the other pyridine derivatives. Monomerization of D with
DAP could be followed only by the stopped-flow technique.
An initial and very rapid (τ < 1 ms) reaction was noted within
the mixing time. The accompanying initial absorbance jump
provided evidence for the significant formation of D-DAP in
a fast equilibrium step. The corresponding spectra are shown
in Figure 6. The analysis of the data according to eq 3 afforded
the value K1 ) 126 ( 3 L mol-1 based on a global fit. The
data collected during the subsequent kinetics stage of the
reaction followed first-order kinetics. The values of kobs for
monomerization varied with [DAP] according to the following
rate law:
Figure 5. Top: perspective view of the monomeric 4-acetylpyridine
complex ML with thermal ellipsoids of 50% probability level. Selected
bond lengths (pm) and angles (deg) are the following: Re-O(1), 168.6;
Re-C(1), 210.9; Re-N, 215.3; Re-S(1), 228.2; Re-S(2), 230.0;
O(1)-Re-N, 99.90; C(1)-Re-N, 81.57; S(1)-Re-N, 82.20; S(2)-
Re-N, 152.68. Bottom: perspective view of the monomeric 1,3-
diethylthiourea complex ML with thermal ellipsoids of 50% probability
level. Selected bond lengths (pm) and angles (deg) are the following:
Re-O, 169.8; Re-C(1), 214.5; Re-S(1), 229.7; Re-S(2), 232.5; Re-
S(3), 237.24; O-Re-S(3), 105.50; C(1)-Re-S(3), 78.96; S(1)-Re-
S(3), 86.58; S(2)-Re-S(3), 148.71.
kb[L]2
d[ML]
dt
) 2
[D]
(14)
1 + K1[L]
The rate constant kb was determined with K1 fixed at its value
known from the equilibrium measurements. Least-squares fitting
gave kb ) (6.41 ( 0.03) × 104 L2 mol-2 s-1
.
Pyrazine and 4-methylthiazole were also used in monomer-
ization studies. They gave the same rate law as pyridine
derivatives. The rate constants are listed in Table 1. Both ligands
are included in Figure 1, illustrating the linear free energy
relationship. The value of kb for 4-methylthiazole is the only
datum that deviates significantly from the line defined by the
pyridine derivatives. This can be explained by the fact that
4-methylthiazole is substituted on the ring carbon next to the
nitrogen. Thus, steric effects may also influence its reaction,
especially the pathway involving DL2, in which there are two
ligands in the transition state.
Attempts were made to study the kinetics of the reverse
reaction by adding a strong acid that would protonate free ligand,
causing the reversal of eq 2. Experiments were carried out with
4-picoline in which excess trifluoromethanesulfonic acid was
added to a solution of the monomeric 4-picoline complex. The
green color of the monomer diminished, and the yellow color
of the dimer reappeared in some hours. The system was not
quite ideal for kinetic studies owing to the limited miscibility
of benzene and trifluoromethanesulfonic acid. However, mean-
ingful but limited kinetics data could be obtained by UV-vis
spectrophotometry. The data gave a reasonable fit to an
integrated second-order rate equation, and the second-order rate
constant was estimated to be ∼0.15 L mol-1 s-1. This gives a
kinetic estimate of the equilibrium constant for monomerization
of KD ≈ 0.89 × 103 L mol-1, in good agreement with the value
of 1.0 × 103 L mol-1 found in equilibrium studies.
Figure 6. UV-vis spectra recorded 1.0 ms after mixing D with
4-(dimethylamino)pyridine. Each spectrum was constructed using the
first reliable absorbance readings from 29 independent stopped-flow
kinetic traces at a 10 nm wavelength interval. Initial concentrations in
mmol L-1 are the following: [D] ) 0.408 for all spectra; [4-Me2N-
Py] ) 0 (a), 2.51 (b), 6.89 (c), 13.8 (d), 27.6 (e) at 25.0 °C in benzene.
the difference is statistically significant. However, both distances
are well within the typical range for rhenium-sulfur single
bonds. The rhenium-oxygen and rhenium-carbon bond lengths
are in good agreement with corresponding parameters in similar
complexes.
The 2-substituted pyridine derivatives were found to be less
reactive and led to reactions in which the equilibrium in eq 2
lay more toward D, most probably because of steric hindrance.
That is, the interaction with the rhenium atom of the fairly
crowded D is disfavored. Addition of a 100-fold excess of
Triarylphosphines. With these ligands the monomerization
reactions proceeded to completion even at the smallest concen-
trations applicable. The rate law given in eq 9 was confirmed
for every derivative. The rate constants are summarized in Table
2 and are plotted against the Hammett σ value in Figure 7. The
F value is -1.26 for ka and -1.89 for kb, with respective
correlation coefficients of 0.934 and 0.992.

1316 Inorganic Chemistry, Vol. 39, No. 6, 2000
Lente et al.
Table 2. Kinetic Data for Monomerization with Triarylphosphines
and Other Ligands at 25.0 °C in Benzene
3σ
kb/L2 mol-2 s-1
ka/L mol-1 s-1
(4-XC6H4)3P, X )
CH3O
-0.81 (4.35 ( 0.03) × 101 (1.29 ( 0.04) × 10-1
-0.51 (1.35 ( 0.05) × 100 (7.0 ( 0.2) × 10-2
0.00 (5.1 ( 0.6) × 10-2 (1.57 ( 0.05) × 10-2
0.18 (1.1 ( 0.2) × 10-2 (4.1 ( 0.2) × 10-3
0.69 (1.3 ( 0.3) × 10-2 (2.2 ( 0.2) × 10-3
1.62 (4.0 ( 1.2) × 10-4 (1.2 ( 0.2) × 10-4
CH3
H
F
Cl
CF3
(Me2N)2CS
8.6 ( 0.7
(2.4 ( 0.1) × 10-1
Tris(p-methoxyphenyl)phosphine gave rise to an intermediate
that was detected by spectrophotometry and NMR. It is proposed
to be an isomer of ML, designated ML*. The concentration of
this intermediate remains low throughout the process. Its
characteristics could be studied much better from exchange
reactions and will be dealt with in that section. It is postulated
that ML* is the immediate product of monomerization in every
reaction between D and triphenylphosphine derivatives, but it
undergoes isomerization to give ML in a process that is fast
relative to the monomerization itself.
Figure 7. Hammett plot for the kinetics of monomerization reaction
between D and triarylphosphines, (XC6H4)3P. Rate constants were
determined in benzene at 25.0 °C.
Ligand-Catalyzed Monomerization. As explained in the
preceding sections, triarylphosphines are thermodynamically
favored over pyridine complexes, the phosphines being stronger
Lewis bases, whereas pyridines are favored kinetically. It
seemed possible, therefore, that pyridine might catalyze the
formation of M-PPh3. Experiments were therefore done starting
with D, Ar3P, and (added initially or mid-course) Py. Figure 8
depicts the results from such an experiment. Without pyridine,
M-PPh3 forms slowly; at the point where pyridine was added,
its buildup rate accelerated sharply.
Matrix rank analysis of time-resolved UV-vis spectra (some
of which are given in Figure 21 of Supporting Information)
indicated the presence of only two absorbing species throughout
the time course; they were the dimer and the monomeric
phosphine complex; the monomeric pyridine complex was too
fleeting. The positions of the isosbestic points of the D + PPh3
reaction did not change after the addition of pyridine. This
conclusion was further affirmed by NMR experiments in which
only the signals corresponding to the D and M-PPh3 were
detected. The rate law confirmed for this reaction exhibits a
mixed term that is first-order with respect to each of tri-
phenylphosphine and pyridine:
Figure 8. Sample kinetic curve to demonstrate the catalytic effect of
pyridine in the monomerization of D with triphenylphosphine. A defined
amount of pyridine was added at one time during the experiment. Initial
concentrations are the following: [D] ) 0.508 mmol L-1; [PPh3] )
14.4 mmol L-1; [Py] ) 9.9 mmol L-1 after injection; 25.0 °C, benzene.
The second step was confirmed directly, as presented in a
subsequent section. It is indeed more rapid than the mixed-
ligand process, consistent with our failure to detect M-Py
during these experiments.
Other Neutral Ligands. No reaction of the dimer was
detected with diethyl ether, tetrahydrofuran, propylene oxide,
1,1,3,3-tetramethylurea, acetone, triphenylphosphine oxide,
triphenylphosphine sulfide, triphenylarsine, ethyl thiocyanate,
thiophene derivatives, and acetonitrile. Reaction of n-butyl
isocyanide gave a mixture of at least five products, none of
which was identified.
d[M-PPh3]
) 2kmixed[D][Py][PPh3]
(15)
dt
This rate law implies the intervention of a DLL′ complex with
two different ligands. Other terms that apply to the separate
reactions of PPh3 and Py separately proved to be negligible with
respect to the mixed term, in particular,
k
mixed[Py][PPh3] .
kbPy[Py]2 > kaPPh3[PPh3] + kbPPh3[PPh3]2 (16)
The crystal structure of ML with L ) 1,1,3,3-tetramethyl-
thiourea (Me4tu) was reported earlier.19 The rate law governing
the reaction between D and Me4tu is given by eq 9, with ka )
(2.4 ( 0.1) × 10-1 L mol-1 s-1 and kb ) 8.6 ( 0.7 L2 mol-2
s-1. The crystal structure of the monomer with 1,3-diethyl-
thiourea was also determined. An ORTEP view of the structure
is shown in Figure 5. The monomeric complex also possesses
a distorted trigonal pyramidal geometry about the rhenium atom.
Atoms S(1), O, and C(1) define the equatorial plane with the
rhenium atom being within 2 pm from this plane. The apical
The value kmixed ) 37.1 ( 0.6 L2 mol-2 s-1 was determined.
M-Py was not detected because it reacted immediately with
triphenylphosphine to give M-PPh3 (see a later section about
this reaction studied directly). We suggest, therefore, that these
steps occur:
D + L + L′ ) {DLL′} f M-PPh3 + M-Py
M-Py + PPh3 f M-PPh3 + Py

Monomerization of Re(V) Dithiolato Dimer
Inorganic Chemistry, Vol. 39, No. 6, 2000 1317
positions are occupied by atoms S(2) and S(3) with the S(2)-
Re-S(3) angle spanning 148.71(3)°. As in the previous case,
this angle differs considerably from 180°, making the coordina-
tion environment of the rhenium atom intermediate between
trigonal bipyramidal and square pyramidal. The rhenium-sulfur,
rhenium-oxygen, and rhenium-carbon bond distances are
within the ranges for corresponding bond lengths. As expected,
the trans bonds Re-S(2), 232.50(7) pm, and Re-S(3), 237.24-
(8) pm, are noticeably longer than the equatorial Re-S(1) bond,
229.73(7) pm. All three rhenium-sulfur bonds are statistically
different.
Dimethyl sulfoxide (DMSO) and 4-picoline-N-oxide were
also confirmed to monomerize D. However, further reactions
ensue following monomerization that involve the oxidation of
Re(V) to Re(VII), thus giving MTO as a final product. These
redox reactions are not relevant to the present topic and
constitute a topic in their own right, which will be reported
separately.
Chart 2
Table 3. Rate Constants for the Monomerization with Halides at
25.0 °C in Benzene
halide ion
k3/s-1
Cl-
Br-
I-
(3.87 ( 0.04) × 10-4
(1.10 ( 0.07) × 10-3
(1.7 ( 0.1) × 10-3
repulsion between negatively charged DL- and L- in the
nonpolar solvent benzene. The value K1 > 5 × 104 L mol-1
was estimated as a lower limit. The experimental rate law for
ligand excess is
d[M-L]
The crystal structure for the DL intermediate with L )
DMSO was published earlier.18 The formation of DL occurred
too rapidly to be measured even by stopped-flow techniques,
and so k1 > 3 × 105 L mol-1 s-1 was concluded. Spectropho-
) 2kc[D]
(18)
dt
Setting K1[L] . 1 + K1K2[L]2 and k3 . k4K2[L] in eq 7 gives
this limiting rate law with kc ) k3. This equation is mathemati-
cally correct even without a large excess of the ligand. The rate
constants determined for halide ions are given in Table 3.
Activation parameters for k3 with chloride are ∆Hq ) 79.9 (
tometric and NMR determinations gave K1 ) 120 ( 6 L mol-1
.
Evaluation of K1 at different temperatures gave ∆H1° ) -35.6
( 0.8 kJ mol-1 and ∆S1° ) -80 ( 3 J K-1 mol-1 18
.
The experimentally confirmed rate law for the kinetic phase
of the reaction is
0.4 kJ mol-1 and ∆Sq ) -43 ( 1 J K-1 mol-1 19
.
With thiocyanate ion stoichiometric formation of DL- was
also confirmed. However, the initial absorbance readings and
kd[L]
d[M-L]
dt
2-
the kinetics indicated that DL2 was equilibrated with DL-
) 2
[D]
(17)
and L- during the time of mixing (Figures 23 and 24 in
Supporting Information). The experimental rate law with L in
large excess this time is
1 + Kd[L]
Because K1 is known independently for this system from 10.0
to 60.0 °C, it can be shown that very little D remained
uncomplexed even at the lowest DMSO concentration used in
the kinetic measurements displayed in Figure 10. Thus, the
inequality 1 , K1[L] + K1K2[L]2 applies. Setting k3 , k4K2-
[L] in eq 7 converts it into another limiting form that agrees
with eq 17. According to that, kd ) k4K2 and Kd ) K2.
Consequently, K2 ) 0.60 ( 0.04 L mol-1 and k4 ) (9.5 ( 0.5)
× 10-4 s-1 were determined at 50.0 °C in benzene. A higher
temperature was used in this case because the reaction was
inconveniently slow at lower temperatures.
4-Picoline-N-oxide gave results very similar to those of
DMSO, the only difference being the time scale of the reaction.
The initial absorbance readings from stopped-flow experiments,
prior to the kinetic phase, were used to determine the stability
constant of DL, K1 ) (4.1 ( 0.3) × 103 L mol-1. From the
kinetics, and eq 17, K2 ) (1.7 ( 0.1) × 102 L mol-1 and k4 )
(5.2 ( 0.3) × 10-2 s-1 were determined.
kc + kd[L]
1 + Kd[L]
d[ML]
dt
) 2
[D]
(19)
This is another limiting form of eq 7 under conditions where 1
, K1[L] + K1K2[L]2, kc ) k3, kd ) k4K2, and Kd ) K2. Similar
to the halide ions, K1 > 5 × 104 L mol-1 can be claimed, the
remaining constants being K2 ) (1.3 ( 0.3) × 102 L mol-1, k3
) (3.2 ( 0.5) × 10-1 s-1, and k4 ) 1.3 ( 0.3 s-1
.
Detailed experiments were also carried out in the chloride
ion system with D in excess. Matrix rank analysis of time-
resolved spectra recorded in several different kinetic experiments
gave the number of absorbing species as 4, in contrast to the
expected value of 3 (D, D-Cl-, and M-Cl-). Indeed, a new
set of 1H NMR signals was seen when D was in stoichiometric
excess over chloride. The new species is proposed to be a chloro
bridged dimer {M-Cl-M}-, an isomer of D-Cl-, without
sulfur bridges so that its spectroscopic properties are more
similar to those of the monomeric complexes. The proposed
structure is shown in Chart 2; the equivalency of the benzylic
protons requires a species more symmetric than, for example,
one with one S and Cl bridging groups.
Anionic Ligands. Tetrabutylammonium salts are usually
reasonably soluble in benzene, and thus we carry out mono-
merization studies with a series of anionic ligands. No mono-
merization was detected with weakly coordinating, nonbasic
anions such as trifluoromethanesulfonate, hexafluorophosphate,
and tetrafluoroborate. Perrhenate ion (ReO4-) and tetrathio-
perrhenate ion (ReS4-) did not react with the dimer either. As
reported for chloride earlier,19 halide ions gave green products
upon reaction with the yellow dimer solution. First-order kinetic
curves with rate constants independent of the halide ion
concentration were recorded when the halide ion was in excess
over the dimer. Investigation of the spectra obtained immediately
after mixing revealed stoichiometric formation of DL- (see
Figure 11 in Supporting Information). The species with two
halides, DL22-, was not detected owing to the strong electrostatic
Ligand Exchange with Monomeric Complexes. Separate,
sharp NMR signals were seen for the free and complexed ligand
in any monomerization reaction. This observation shows that
the ligand exchange between ML and free L is slow on the
NMR time scale. To study a reaction very close to true
exchange (an identity), a pseudo-exchange was created with the
monomeric complex with pyridine (L) and excess 4-methylpy-
ridine (L′):
ML + L′ ) ML′ + L
(20)

1318 Inorganic Chemistry, Vol. 39, No. 6, 2000
Lente et al.
Figure 10. Sample kinetic curve in the exchange reaction between
Figure 9. Temperature dependence of the rate constants in the
exchange reactions ML + L′ with L ) pyridine and L′ ) 4-meth-
ylpyridine; L ) triphenylphosphine and L′ ) tri(p-tolyl)phosphine.
ML (1.15 mmol L-1) and L′ in solutions containing 17.9 mmol L-1
L
(pyridine) and 25.7 mmol L-1 L′ (triphenylphosphine). Conditions are
in benzene at 25.0 °C. The rising portion of the curve represents the
pyridine-catalyzed formation of M-PPh3*, and the slower, falling part
represents its conversion to the stable M-PPh3 in a reaction that is
also pyridine-catalyzed.
Small differences in the spectra of the monomeric adducts made
it possible to study the kinetics of the reaction with the stopped-
flow technique. The following rate law was confirmed:
pyridine to give an M-PPh3*, an isomer of M-PPh3. This
isomer has a UV-vis spectrum and 1H and 31P NMR resonances
different from that determined for M-PPh3 in independent
studies (see Table 5 in Supporting Information for details).
Isomerization of M-PPh3* follows in which M-PPh3 is
produced. This isomerization is catalyzed by pyridine. The
following scheme was confirmed to interpret the kinetic data:
d[ML′]
) k+[ML][L′] - k-[ML′][L]
(21)
dt
The values k+ ) 95 ( 2 L mol-1 s-1 and k- ) 19 ( 6 L mol-1
s-1 were determined by least-squares fitting. The equilibrium
constant for reaction 20, Kex ) [ML′][L]/([ML][L′]) could be
estimated in three independent ways. First, Kex could be
calculated from the data given in Table 1 as the square root of
the ratio of the KD values for ML′ and ML, giving Kex ) 2.4
( 0.1. Second, direct fitting to the equilibrium absorbance
readings in reaction 20 resulted in Kex ) 2.0 ( 0.3. Finally,
from kinetics, Kex ) k+/k- ) 5.0 ( 2.5. The high standard
deviation of the kinetic value reflects the high uncertainty in
k-, although the agreement among these three values is
acceptable. A preferable value of k- can be obtained from k+/
M-Py + PPh3 { k } M-PPh3* + Py
(22)
5
k-5
k6
M-Py + PPh3
9
8 M-PPh3 + Py
(23)
(24)
k7
M-PPh3* + PPh3
98 M-PPh3 + PPh3
Kex, which gives 40 ( 4 L mol-1 s-1
.
The values k5 ) 88.0 ( 0.8 L mol-1 s-1 and k-5 ) 16.1 ( 0.8
L mol-1 s-1 were determined from the kinetic analysis of the
first step, giving the equilibrium constant for reaction 22 because
K5 (or K22) ) k5/k-5 ) 5.5 ( 0.3. This equilibrium constant
was used in the evaluation of the second step, and the remaining
rate constants were resolved as k6 ) 6.27 ( 0.09 L mol-1 s-1
and k7 ) 0.2 ( 0.1 L mol-1 s-1 (see Supporting Information
for details). It can be seen that reaction 24 plays but a marginal
role in the isomerization of M-PPh3*. Adding the unassisted
first-order isomerization reaction M-PPh3* f M-PPh3 to the
scheme given in eq 22-24 did not improve the fit at all, whereas
substituting this first-order step for reaction 24 resulted in a
somewhat inferior fit.
It was also possible to estimate the overall equilibrium
constant for this exchange. This equilibrium constant is defined
as K6 ) [M-PPh3][Py]/([M-Py][PPh3]). The value obtained
from NMR based on integration of signals is K6 ) 930 ( 30,
whereas UV-vis gave K6 ) 860 ( 20. Evaluating the two sets
of results together gave K6 ) 900 ( 50.
The temperature-dependent values of k+ are shown in Figure
9, plotted according to the Eyring equation. The activation
parameters were determined as ∆Hq ) 36 ( 2 kJ mol-1 and
∆Sq ) -85 ( 7 J K-1 mol-1 (8-43 °C). The large negative
value for the activation entropy suggests an associative Ia
pathway for this process.
The same experiments were done with triphenylphosphine
(L) and tri(p-tolyl)phosphine (L′) as well. This exchange was
much slower. The rate law given in eq 20 was confirmed, and
k+ ) 0.053 ( 0.003 L mol-1 s-1 and k- ≈ 0.02 L mol-1 s-1
were determined. Their ratio gives Kex ≈ 2.5, a value that is
consistent with the equilibrium absorbance readings. The
temperature dependence of k+ is shown in an Eyring equation
format in Figure 9. ∆Hq ) 41 ( 2 kJ mol-1 and ∆Sq ) -132
( 6 J K-1 mol-1 (17-50 °C). The activation entropy is also
negative in this case; its value is even larger than for the
analogous pyridine reaction. Pyridine had a considerable
catalytic effect on phosphine exchange (Figure 17 in Supporting
Information). The mechanism of this catalytic reaction was
found to be surprisingly complex and is not reported here.
Exchange with L ) pyridine and L′ ) triphenylphosphine
was also studied. Biphasic kinetic curves were recorded; a
typical experiment on dual time bases is shown in Figure 10.
The two steps were separated enough to be handled indepen-
dently, however. In the first step triphenylphosphine replaces
Discussion
The steps for dimer and ligand reactions represented by k1
and k2 are assumed to be very rapid for all the ligands. This
was explicitly proved for those ligands with which DL attains
a detectable concentration (for DMSO,18 4-pyridine-N-oxide,

Monomerization of Re(V) Dithiolato Dimer
Inorganic Chemistry, Vol. 39, No. 6, 2000 1319
4-(dimethylamino)pyridine, and halide and thiocyanate ions).
In each of these cases, the reaction occurred too rapidly for the
stopped-flow method. We thus infer that no significant
structural rearrangement occurs around the rhenium center
during that step of the reaction. The confirmed fact that the
reaction D + L ) DL is extremely fast renders it implausible
to assign the reaction between DL + L as a rate-controlling
step in monomerization, which would be equivalent to omitting
the equilibrium formation of DL2. Kinetic results with L )
DMSO, picoline-N-oxide, and thiocyanate also require DL2 as
an intermediate; only small UV-vis spectral changes for the
conversion of DL to DL2 were seen for these ligands.
tetrabutylammonium halides, so higher ligand concentrations
2-
could be used favoring the formation of DL2
.
1
The structure of the M-PPh3* species remains unclear. H
NMR and 31P NMR show without doubt that it has the same
composition as M-PPh3. However, any investigation other than
NMR or UV-vis spectroscopy has not proved feasible because
the lifetime of the species is about 2-3 min at room temperature.
A geometric isomer must be considered, since the two S atoms
of the dithiolate are inequivalent; we plan to study this further.
K6 ) 900 was determined from exchange studies. One can
but assume that monomerization reactions of triarylphosphines
are also equilibrium reactions, with a KD (PAr3). One can
calculate a value for triphenylphosphine from the relation
The rate constants for monomerization with pyridine deriva-
tives containing electron-withdrawing substituents in the para
position (4-cyano, 4-acetyl, 4-formyl) show significant deviation
from other pyridine derivatives on Hammett plots. The basicity
of pyridine derivatives in nitromethane21 showed the same kind
of deviation when plotted against the usual Hammett σ values
(Figure 18 in Supporting Information). Thus, the usual Hammett
σ values do not seem to be applicable for pyridine derivatives
containing electron-withdrawing substituents in the para position.
Kinetics results on re-dimerization following the addition of
triflic acid suggest that two M-Py molecules associate to give
a dimer before liberating pyridine rather than dissociation of
M-Py into Py and M. This is in agreement with the microscopic
reverse of monomerization, showing that the acid has not opened
a new pathway. It can also be concluded that the possible
reaction M-Py + H+ f M + HPy+ has no significance. Most
probably such a step would be thermodynamically unfavorable
in that the rate of pyridine exchange suggests that it should be
reasonably fast if the reaction were thermodynamically allowed.
Similar re-dimerization experiments with less reactive pyridine
derivatives may have given some information on M itself.
Unfortunately, these experiments were not feasible owing to
the limited solubility of triflic acid in benzene.
Py
KPDPh ) KD K6
(25)
3
giving 1.6 × 105 L mol-1. This value implies an almost
complete reaction under the accessible conditions and could not
be determined directly because of the experimental difficulties
connected with slow equilibration, especially at the very low
phosphine concentrations needed to obtain a balanced system.
On the basis of the data presented in this paper, it can be
estimated that several weeks might be needed to reach a
detectable true equilibrium in the reaction of D with PPh3.
On the basis of the value of K6, it is also possible to calculate
the equilibrium constant for the isomerization of M-PPh3* to
M-PPh3; Kiso ) [M-PPh3]/[M-PPh3*] ) K6/K5 ) 164. This
value is consistent with the fact that only M-PPh3 is detected
in equilibrium.
The oxo-bridged Re(V) dimer {Cp*Re(O)}2(µ-O)2, reported
by Gable,23 provides an informative comparison. The reaction
of that dimer with styrene oxide was studied by quantitative
kinetics. It was found to proceed via the corresponding
monomeric form, and a monomer-dimer equilibrium was
established. No signs of such an equilibrium between D and its
monomer M were seen even at the highest temperatures and
concentrations applicable, giving a lower limit for the dimer-
ization equilibrium constant K ) [D]/[M]2 > 107 L mol-1. It is
assumed that M reacts very rapidly with ligands because it could
not be detected (i.e., its concentration was lower than 10-6 mol
L-1) in our studies. Coordination of a ligand to one of the metal
centers of {Cp*Re(O)}2(µ-O)2 seems to be less likely because
the electron count of its Re is 18, in contrast with the electron
count of 14 for Re in D. This suggests why the reactions of
{Cp*Re(O)}2(µ-O)2 are predominantly dissociative (D), whereas
they are associative (A) for D.
The large negative activation entropies determined for ligand-
exchange reactions of monomeric complexes indicate that M
is not a likely intermediate in these reactions either. The
transition state most probably contains a six-coordinated Re-
(V) center. This fact is seen as further evidence to prove that
the dissociation of ML into L and M is unimportant.
Efforts to crystallize anionic derivatives were unsuccessful.
Chloride is known to bridge between rhenium centers in earlier
examples,22 giving some support to a postulated M-Cl-M-
structure. The NMR and UV-vis spectroscopic properties of
the proposed M-Cl-M- are much closer to those of M-Cl-
Acknowledgment. This research was supported by the
National Science Foundation. G.L. also thanks the U.S.-
Hungary Science and Technology Program for financial support.
Some of this research was carried out in the facilities of the
Ames Laboratory. Three anonymous reviewers are also ac-
knowledged for their useful suggestions.
than to D-Cl-. Unlike the halides, the kinetic role of DL2
2-
was confirmed with thiocyanate. Its formation constant could
also be determined. This reflects the well established fact that
thiocyanate usually forms much more stable complexes than
2-
halides. So formation of DL2 is possible with thiocyanate
despite the electronic repulsion that represses this process with
the halides. It should also be noted that tetrabutylammonium
thiocyanate is considerably more soluble in benzene than
Supporting Information Available: Graphs and tables containing
additional kinetic and spectroscopic data referred to in the text, detailed
description of the kinetic equations used to obtain rate constants with
less reactive pyridine derivatives, and X-ray crystallographic tables.
X-ray crystallographic data files for the monomeric complexes ML
with L ) 4-acetylpyridine and diethylthiourea in CIF format. This
material is available free of charge via the Internet at http://pubs.acs.org.
(21) Korolev, B. A.; Osmolovskaya, L. A.; Dyumaev, K. M. Zh. Obshch.
Khim. 1979, 49, 898-904.
(22) Dunbar, K. R.; Powell, D.; Walton, R. A. Inorg. Chem. 1985, 24,
2842-2846.
(23) Gable, K. P.; Juliette, J. J. J.; Gartman, M. A. Organometallics 1995,
14, 3138-3140.
IC991431P