Mechanism of CO displacement from an unusually labile rhenium complex: An experimental and theoretical investigation

Article abstract of DOI:10.1021/ic302182b

The displacement of a CO ligand from an unusually labile rhenium carbonyl complex containing a bidentate carboxyaldehyde pyrrolyl ligand by PPh ₃ and pyridine has been investigated. The reaction is found to proceed by an associative, preequilibrium mechanism. Theoretical calculations support the experimental data and provide a complete energetic profile for the reaction. While the Re-CO bond is found to be intrinsically weak in these complexes, it is postulated that the unusual lability of this species is due to the presence of a weak aldehyde Re-O link that can easily dissociate to open a coordination site on the metal center and accommodate an incoming ligand prior to CO loss. The resulting intermediate complex has been identified by IR spectroscopy. The presence of the hemilabile pyrrolyl ligand provides a lower-energy reaction channel for the release of CO and may be of relevance in the design of CO-releasing molecules.

Full text of DOI:10.1021/ic302182b

Article
pubs.acs.org/IC
Mechanism of CO Displacement from an Unusually Labile Rhenium
Complex: An Experimental and Theoretical Investigation
Sohail Muhammad,† Veeranna Yempally,† Muhammad Anas,† Salvador Moncho,† Samuel J. Kyran,‡
Edward N. Brothers,† Donald J. Darensbourg,*,‡ and Ashfaq A. Bengali*,†
†Department of Chemistry, Texas A&M University at Qatar, Doha, Qatar
‡Department of Chemistry, Texas A&M University, College Station, Texas 77843, United States
S
* Supporting Information
ABSTRACT: The displacement of a CO ligand from an
unusually labile rhenium carbonyl complex containing a
bidentate carboxyaldehyde pyrrolyl ligand by PPh3 and
pyridine has been investigated. The reaction is found to
proceed by an associative, preequilibrium mechanism.
Theoretical calculations support the experimental data and
provide a complete energetic profile for the reaction. While the
Re−CO bond is found to be intrinsically weak in these complexes, it is postulated that the unusual lability of this species is due to
the presence of a weak aldehyde Re−O link that can easily dissociate to open a coordination site on the metal center and
accommodate an incoming ligand prior to CO loss. The resulting intermediate complex has been identified by IR spectroscopy.
The presence of the hemilabile pyrrolyl ligand provides a lower-energy reaction channel for the release of CO and may be of
relevance in the design of CO-releasing molecules.
Surprisingly, CO substitution by the weakly coordinating
tetrahydrofuran solvent also occurred readily, even at room
temperature. As noted by the authors, the ease of CO loss in
these complexes may lead to their application as CORMs.
Given this promising application, it becomes important to
investigate both the mechanism and energetics of the CO loss
pathway. The CO lability of 1 is in marked contrast to some
other rhenium carbonyl complexes such as the thermally robust
CpRe(CO)3 molecule. An important characteristic of 1 that
might assist in facile CO replacement is the presence of the
presumed weak Re−O interaction. This bond could potentially
be disrupted by an incoming ligand, thereby opening up a
coordination site on the rhenium metal center and providing a
low-energy pathway for CO loss. This process would be
somewhat analogous to an associative “ring-slip” mechanism10
in the substitution of CO from molecules such as (η5-
C10H9)Mn(CO)3, whereby an open coordination site is
generated upon an η5 → η3 hapticity shift of the hydro-
naphthalene ring system, facilitating incoming ligand binding
prior to CO loss.11
INTRODUCTION
■
While transition-metal carbonyl complexes are often used as
stoichiometric or catalytic reagents for a variety of chemical
transformations,1 there is increasing interest in their utility as
reagents in systems of biological relevance.2 For example, new
radiopharmaceuticals based on the Re(CO)3 and Tc(CO)3
fragments that bind to estrogen receptors have been
synthesized.3 The strong IR absorption of the CO ligands
also provides a useful spectroscopic tag that has led to the
development of carbonylmetalloimmunoassay procedures.4
Some metal carbonyl complexes also show promise as
anticancer drugs.5 Recently, CO has been identified as having
an important pharmacological function and, like NO, is thought
to possess vasodilatory properties.6 Several other beneficial
physiological effects of CO have been well documented.2a,6a,b,d
As a result, there have been some early investigations into the
use of metal carbonyl complexes as sources of solid CO and
therefore as CO-releasing molecules (CORMs).5b,6c For
example, among other complexes, [Ru(CO)3Cl2]2 and Ru-
(CO)3Cl(glycinate) have been shown to readily transfer CO to
mygolobin (Mb) to form Mb-CO under physiological
conditions.7 Other CORMs demonstrate a therapeutic value
in reducing sepsis-induced lethality by inhibiting bacterial
growth and respiration.8
We report in this paper an experimental and theoretical study
aimed at understanding the mechanism of CO displacement
(Scheme 1) from 1a and 1b by PPh3 and pyridine to form
compounds 2a and 2b and compounds 3a and 3b, respectively.
The results provide details about the energetics of this process
and lead to identification of an important intermediate along
the reaction pathway. Detailed calculations using density
Recently, Bideau and co-workers reported the synthesis of a
series of bidentate pyrrole-based rhenium tetracarbonyl
complexes that have potential utility as radiopharmaceutical
reagents because of the incorporation of 186Re or 188Re as a
radionuclide (1).9 Our interest in these complexes was
motivated by the reported unusual lability of the cis-CO
ligands in the presence of donor molecules such as PPh3.
Received: October 7, 2012
Published: November 15, 2012
© 2012 American Chemical Society
13041
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049

Inorganic Chemistry
Article
(s, 1 C, CO), 188.5 (s, 1 C, CO). Anal. Calcd for C10H3F3N1O5Re1: C,
26.1; H, 0.7; N, 3.0. Found: C, 26.2; H, 0.5; N, 3.0.
Synthesis of {2-(CF3CO)C4H3N}Re(CO)3(PPh3) (2b). A solution of
1b (0.101 g, 0.219 mmol) in 20 mL of hexane was treated with PPh3
(0.060 g, 0.230 mmol) and refluxed for 1 h under an atmosphere of
argon. Upon cooling to room temperature, formation of a yellow
precipitate was observed. The solution was reduced to a third of the
volume, further cooled in a freezer overnight, and filtered to give a
yellow-orange product (0.103 g, 68%). IR data in CH2Cl2 (νCO): 1582
(m), 1905 (s), 1936 (s), 2032 (s). NMR data in CDCl3: 1H δ 6.34 (d,
J = 4.4 Hz, 1 H), 7.01 (m, 1 H), 7.20 (m, 6 H, Ho), 7.30 (m, 1 H), 7.33
(t, J = 8.3 Hz, 6 H, Hm), 7.41 (t, J = 7.3 Hz, 3 H, Hp); 31P δ 20.16;
13C{1H} δ 117.9 (q, 1JCF = 280 Hz, 1 C, CF3), 120.6, 126.1, 128.6 (d,
Scheme 1
J
CP = 10 Hz, 6 C, C6H5), 129.3 (d, JCP = 44 Hz, 3 C, C6H5), 130.5 (d,
JCP = 2 Hz, 3 C, C6H5), 133.5 (d, JCP = 11 Hz, 6 C, C6H5), 139.0,
150.0, 170.5 (q, 2JCF = 37 Hz, 1 C, CO), 187.6 (d, JCP = 69 Hz, 1 C,
CO), 195.7 (d, JCP = 67 Hz, 1 C, CO), 195.8 (d, JCP = 67 Hz, 1 C,
CO). Anal. Calcd for C27H18F3N1O4P1Re1: C, 46.7; H, 2.6; N, 2.0.
Found: C, 46.6; H, 2.5; N, 2.0.
Synthesis of {2-(CHO)C4H3N}Re(CO)3(pyridine) (3a). A solution of
1a (0.390 g, 0.99 mmol) in 50 mL of hexane was treated with pyridine
(0.1 mL, 1.24 mmol) and refluxed for 2 h under an atmosphere of
argon. Upon cooling to room temperature, formation of a yellow
precipitate was observed. The solution was further cooled in a freezer
overnight and filtered to give a yellow product (0.410 g, 93%). IR data
in CH2Cl2 (νCO): 1570 (m), 1906 (s), 1927 (s), 2030 (s). NMR data
functional theory (DFT) provide additional support for the
experimental data.
1
in CDCl3: H δ 6.42 (dd, J = 4.1 and 1.4 Hz, 1 H), 7.08 (dd, J = 4.1
EXPERIMENTAL AND THEORETICAL DETAILS
and 0.9 Hz, 1 H), 7.28 (m, 2 H, pyridine), 7.64 (d, J = 1.4 Hz, 1 H),
7.77 (tt, J = 7.7 and 1.7 Hz, 1 H, pyridine), 8.38 (m, 2 H, pyridine),
8.73 (d, J = 1.4 Hz, 1 H, CHO); 13C{1H} δ 118.1, 124.4, 125.4 (s, 2 C,
pyridine), 138.3 (s, 1 C, pyridine), 144.3, 145.0, 152.0 (s, 2 C,
pyridine), 181.2 (s, 1 C, CHO), 193.1 (s, 1 C, CO), 197.3 (s, 1 C,
CO), 197.6 (s, 1 C, CO). Anal. Calcd for C13H9N2O4Re: C, 35.2; H,
2.1; N, 6.3. Found: C, 35.2; H, 2.0; N, 6.4.
X-ray Crystal Structure Analyses. Single crystals of 2b and 3a were
obtained by the slow evaporation of a hexane solution. A Leica MZ7.5
stereomicroscope was used to identify suitable crystals of the same
habit. Each crystal was coated in paratone, affixed to a Nylon loop, and
placed under streaming nitrogen (110 K) in a SMART Apex CCD
diffractometer (see details in the CIF files). The space groups were
determined on the basis of systematic absences and intensity statistics.
The structures were solved by direct methods and refined by full-
matrix least squares on F2. Anisotropic displacement parameters were
determined for all non-hydrogen atoms. Hydrogen atoms were placed
at idealized positions and refined with fixed isotropic displacement
parameters. The following is a list of programs used: data collection
and cell refinement, APEX2;12 data reductions, SAINTPLUS, version
6.63;13 absorption correction, SADABS;14 structural solutions,
SHELXS-97;15 structural refinement, SHELXL-97;16 graphics and
publication materials, SHELXTL.17
c. DFT Calculations. All calculations were performed in the
development version of the Gaussian suite of programs18 using DFT.
Geometries were optimized using the ωB97XD19 functional, which
includes different fractions of exact exchange in the long and short
ranges, as well as a dispersion correction. All atoms were described
with the def2-TZVPP basis set,20 which describes the core electrons of
the heavy atom (rhenium) using an effective core potential. The
computed geometries were confirmed to be ground-state structures or
transition states according to their number of imaginary frequencies.
The enthalpies or free energies were computed at 298.15 K and 1 atm
and are expressed in kcal/mol. Figures of computed geometries
included in this work were rendered using CYLview.21
■
a. Kinetic Experiments. All kinetic experiments were performed
with a Bruker Vertex 80 Fourier transform infrared (FTIR)
spectrometer. Spectra were obtained at 4 cm−1 resolution. A
temperature-controlled 0.75-mm IR cell with CaF2 windows was
used to obtain all spectra. A typical kinetic run was performed as
follows. An appropriate amount of PPh3 or pyridine was added to a 3
mM heptane solution of 1a or 1b. A background spectrum was taken
immediately, followed by the acquisition of sample spectra at defined
time intervals over a 303−333 K temperature range. Observed rate
constants (kobs) were obtained from first-order fits to the temporal
profile of either reactant decay or product growth. The stated errors in
the rate constants and activation parameters were obtained from a
least-squares fit to the data, as reported by the data analysis program
Kaleidagraph.
b. Synthesis. The tetracarbonylrhenium complex with the 2-
formylpyrrolyl ligand {2-(CHO)-C4H3N}Re(CO)4 (1a) was prepared
as per the literature.9 Solvents were either anhydrous grade (Aldrich)
or purified by an MBraun Manual Solvent Purification System packed
with an Alcoa F200 activated alumina desiccant. NMR spectra were
recorded on a Varian INOVA 500 (operating at 499.42, 202.17, and
125.59 MHz for 1H, 31P, and 13C, respectively) or a Bruker Advance II
400 spectrometer. 1H and 13C NMR spectra were referenced to
residual solvent resonances, while 31P NMR spectra were referenced to
an external H3PO4 in D2O at 0.0 ppm. IR spectra were obtained on a
Bruker Tensor 27 or Vertex 80 FTIR spectrometer. Elemental analyses
for 1b, 2b, and 3a were determined by Atlantic Microlab (Norcross,
GA).
Synthesis of {2-(CF3CO)-C4H3N}Re(CO)4 (1b). Re(CO)5Br (0.770 g,
1.90 mmol), 2-(trifluoroacetyl)pyrrole (0.312 g, 1.91 mmol), and
sodium tert-butoxide (0.206 g, 2.14 mmol) in toluene (25 mL) were
heated to reflux for 2 h under an argon atmosphere. The resulting
vermillion solution was filtered to remove NaBr and the solvent
concentrated to give a viscous oil. This was purified by flash
chromatography on silica (6:4 pentane/ether) to obtain a dark-orange
oil (the small presence of Re2(CO)10 can be removed by a second flash
chromatography with 9:1 pentane/CH2Cl2), which crystallizes as a
dark-red solid after the complete removal of residual solvent on a
rotovap (0.260 g, 30%). IR data in heptane (νCO): 1583 (m), 1958 (s),
2001 (s), 2014 (s), 2115 (m). NMR data in CDCl3: 1H δ 6.61 (dd, J =
4.6 and 1.3 Hz, 1 H), 7.46 (m, 1 H), 7.70 (m, 1 H); 13C{1H} δ 117.9
(q, 1JCF = 280 Hz, 1 C, CF3), 121.5, 128.0 (q, 4JCF = 3 Hz, 1 C), 139.6,
151.9, 172.7 (q, 2JCF = 38 Hz, 1 C, CO), 183.6 (s, 2 C, CO), 188.3
RESULTS AND DISCUSSION
■
a. 1a + PPh3. The thermal reaction of 1a with PPh3 results
in the spectral changes shown in Figure 1. The CO stretching
bands of 1a at 2112, 2010, 1993, and 1949 cm−1 and the
aldehyde absorbance at 1571 cm−1 exhibit a first-order
exponential decrease in intensity and are consistent with
13042
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049

Inorganic Chemistry
Article
Figure 2. Difference FTIR spectra obtained at the start of the reaction
between a 3 mM solution of 1a and 0.17 M PPh3 at 303 K in heptane.
The peaks marked with IM are due to the formation of an
intermediate species that grows in rapidly and then decays over the
course of the reaction at the same rate as the growth of the product
complex 2a. The inset shows spectral changes in the 1700−1500 cm−1
region due to the CO stretching of the aldehyde.
Figure 1. Difference FTIR spectra observed as a function of time upon
reaction of a 3 mM solution of 1a with 0.17 M PPh3 in heptane at 303
K. The positive peaks are due to the growth of the product complex,
2a, while the negative peaks are due to the disappearance of the
reactant, 1a.
literature values;9 product peaks due to 2a at 2032, 1939, 1912,
and 1580 cm−1 increase at the same rate. The locations of the
relevant IR bands are presented in Table 1. The observation of
three CO stretching absorptions in the product complex is
consistent with the previously determined crystal structure of
{2-(CH3CHO)C4H3N}Re(CO)3PPh3.9 While the CO ligands
have local C3v symmetry, a reduction in the symmetry due to
the pyrrolyl ligand results in the splitting of the E band, yielding
absorptions at 1939 and 1912 cm−1 (see Table 1). The facial
geometry of 2a indicates loss of a CO ligand that is cis to both
coordinating atoms of the bidentate pyrrolyl ligand and, as
discussed previously,9 is consistent with the mutual trans
influence of the CO ligands.
As shown in Figure 2, within the first few minutes of the
reaction, additional peaks are observed to first increase rapidly
and then decrease in intensity during the course of the reaction
at 303 K. The complex associated with these absorptions (IM)
exhibits the temporal profile shown in Figure 3. The reactant
species 1a also has a biexponential time profile with fast and
slow components that match, within a factor of 2, the fast
growth and slower decay of IM. These observations suggest the
presence of an intermediate species along the 1a → 2a reaction
pathway. As discussed later, DFT calculations also confirm the
viability of an intermediate complex along the CO displacement
channel.
Four CO stretching bands are observed for IM at 2103, 2019,
1990, and 1957 cm−1, confirming that generation of this
intermediate does not require the displacement of a CO ligand
from 1a. Furthermore, the peak at 1651 cm−1 associated with
this complex is in good agreement with the CO stretching
band of the uncoordinated acetylpyrrolyl ligand observed at
1650 cm−1. Put together, these data suggest that the incoming
PPh3 ligand initially displaces the weakly coordinated aldehyde
group of the pyrrolyl ligand to generate an intermediate with
the following structure:
This intermediate complex then proceeds to form the final
product 2a by loss of a CO molecule that must be cis to both
the phosphine and N-coordinated pyrrolyl ligands, followed by
recoordination of the aldehyde oxygen to the rhenium center.
The overall reaction then appears to follow a 1a → IM → 2a
pathway.
To determine whether IM is formed reversibly from 1a, a
heptane solution of 1a and 0.4 M PPh3 was allowed to react at
Table 1. Listing of All Relevant IR Absorptions, Experimental and Calculated, for the Complexes Studied
ν (cm−1
)
a
complex
experimental (heptanes)
calculated (scaled)
1a
1b
2112, 2010, 1993, 1949, 1571
2115, 2014, 2001, 1958, 1583
2103, 2019, 1990, 1957, 1651
2106, 2015, 1982, 1945, 1650
2032, 1939, 1912, 1580
2110, 2015, 1996, 1955, 1561
2101, 2012, 1985, 1946, 1655
2023, 1934, 1918, 1559
IM (PPh3)
IM (pyridine)
2a
2b
3a
2035, 1946, 1917, 1582
2028, 1926, 1908, 1569
a
Calculated values were scaled by a factor of 0.947 (see the text).
13043
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049

Inorganic Chemistry
Article
Figure 4. Spectral changes observed when a heptane solution initially
containing 1a, IM, and 2a is left standing at 298 K. The spectra are
consistent with the complete conversion of IM to the reactant 1a while
the concentration of 2a is left unchanged. Because of the overlap
between the CO bands of the various species, conversion is best
observed at 2103 cm−1 (IM) and 2112 cm−1 (1a). The intensities of
the product bands at 1939 and 1912 cm−1 due to 2a are unchanged.
Figure 3. Temporal profile of the intermediate complex, IM,
monitored at 2103 cm−1 obtained upon reaction of a heptane solution
of 1a with 0.17 M PPh3 at 303 K. The time dependence of this species
demonstrates a fast rise followed by a slower decay. The solid black
line represents a biexponential fit to the data.
issue is discussed later in light of the DFT calculations. During
the course of these studies, 2b was isolated and a crystal
structure obtained (Figure 6). The structure of this species is
very similar to that of the previously characterized {2-
(CH3CHO)C4H3N}Re(CO)3PPh3 complex.
b. 1a + Pyridine. A complete kinetic analysis over a wide
range of incoming ligand concentrations could not be
performed with PPh3 due to solubility limitations. Thus,
pyridine was chosen as the displacing ligand for these
experiments. As in the case of PPh3, the reaction of 1a with
pyridine results in complete conversion to the pyridine complex
3a (Figure 7), which was isolated and a crystal structure
obtained (Figure 8). The CO ligands are facially coordinated as
in 2b, and this geometry is consistent with the cis displacement
of CO from 1a. At early reaction times, an intermediate
complex is observed with the same spectral features as those
observed with PPh3, although it is significantly less stable and is
observable for only a few minutes at 303 K. In an effort to
obtain better spectroscopic evidence for this intermediate, the
reaction was conducted at 286 K with [pyridine] = 0.4 M, and
as shown in Figure 9, this species absorbing at 2106, 2015,
1982, 1945, and 1650 cm−1 is clearly observed for several
seconds. As with the PPh3 ligand, a 6-fold rate enhancement is
observed when pyridine is used to displace CO from the CF3
analogue (1b).
room temperature for a few minutes and the solution quenched
at 263 K. Evaporation of the solvent yielded an amorphous
yellow powder together with large white PPh3 crystals. After
hand separation of the PPh3 crystals, the yellow powder was
redissolved in heptane and spectra were acquired at 298 K over
a 30-min time period. As shown in Figure 4, the initial spectrum
clearly shows the presence of all three complexes, 1a, IM, and
2a. Under these conditions, IM converts completely to the
reactant 1a and there is no change in the concentration of 2a.
The data are therefore consistent with the reversible formation
of IM from 1a, with the equilibrium condition favoring the
reactant complex. Furthermore, because conversion of IM to 2a
is not observed under these conditions, the overall CO
displacement mechanism is consistent with a preequilibrium
process with the IM → 1a reaction barrier lower than that for
IM → 2a. The overall data therefore suggest the reaction
mechanism shown in Scheme 2.
The reaction of 1b with L = PPh3 was also studied to
ascertain the effect of the electron-withdrawing CF3 group
upon the reaction rate and stability of the intermediate. As
shown in Figure 5, the rate of CO displacement from 1b is
almost 6 times faster than that from 1a. The available data do
not allow for the origin of this rate enhancement to be
determined with certainty. It is possible that the faster rate is
due to a shift in the equilibrium toward IM as a result of weaker
Re−O binding in 1b or because of differences in the stability of
the transition state connecting IM to the product complex. This
As discussed earlier, the kinetic data are consistent with a
preequilibrium between 1a and IM followed by slower
conversion to the product complex (Scheme 2). Application
of the preequilibrium assumption yields the dependence of kobs
on [pyridine] shown in eq 1.
⎛
⎞
Keqk2[pyridine]
k1
k−1
kobs
=
K
=
⎜
⎝
⎟
⎠
eq
1 + K [pyridine]
eq
(1)
A plot of kobs versus [pyridine] is therefore expected to
demonstrate saturation behavior as the [pyridine] becomes
13044
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049

Inorganic Chemistry
Article
Scheme 2
Figure 5. Plot of Abs versus time for the reaction of 1a and 1b with
PPh3 monitored at 2010 and 2001 cm−1, respectively. A heptane
solution of 1a or 1b was reacted with 0.17 M PPh3 at 303 K.
Figure 6. Crystal structure of 2b. Selected bond lengths (Å) and
angles (deg): Re1−C1 = 1.905(4), Re1−C2 = 1.918(4), Re1−C3 =
1.950(4), Re1−N1 = 2.152(3), Re1−O4 = 2.185(2), Re1−P1 =
2.4989(13); C1−Re1−C2 = 89.91(14), C1−Re1−C3 = 88.76(15),
C2−Re1−C3 = 88.88(15), C1−Re1−N1 = 97.52(13), C2−Re1−N1
= 172.32(12), C3−Re1−N1= 93.18(13), C1−Re1−O4 = 172.27(11),
C2−Re1−O4 = 97.64(12), C3−Re1−O4 = 92.99(13), N1−Re1−O4
= 74.88(10), C1−Re1−P1 = 92.57(11), C2−Re1−P1 = 86.24(10),
C3−Re1−P1 = 174.94(11), N1−Re1−P1 = 91.49(8), O4−Re1−P1 =
86.33(8).
large enough to force the equilibrium toward the intermediate,
at which point kobs is limited by the rate constant for conversion
of IM to the product (k2). However, as shown in Figure 10a,
the observed dependence of kobs on [pyridine] is linear,
implying that Keq ≪ 1, so that even at the highest [pyridine]
(4.8 M) employed Keq[pyridine] < 1. As mentioned earlier, the
observation that the intermediate converts completely to 1a
indicates that Keq is, in fact, less than 1. Under these conditions,
eq 1 reduces to
To provide support for the experimental findings, detailed
DFT calculations were performed. The energetic profile of the
substitution reaction, together with the calculated spectroscopic
properties of all of the species involved, was found to be in
close agreement with the experimental values.
kobs = Keqk2[pyridine]
(2)
and a plot of kobs versus [pyridine] should be linear, as
observed, with a slope of Keqk2. The relevant mechanistic
parameters are listed in Table 2. Unfortunately, the kinetic data
do not allow for an independent measurement of Keq or k2 but
rather their product. Thus, the activation enthalpy from the
Eyring analysis, ΔHtotal, shown in Figure 10b, is expected to
yield the sum of ΔHeq and ΔH2⧧ rather than individual values.
As discussed below, the experimental value of 15.4 0.4 kcal/
mol is in good agreement with a theoretical estimate of 13.4
c. Theoretical Modeling. All calculations were performed
using pyridine as the incoming ligand for comparison with the
experimental results. To rule out the possibility of a dissociative
mechanism of CO displacement in these systems, initial
calculations focused on the determination of the CO binding
enthalpy in 1a. The calculated Re−CO bond dissociation
energy (BDE) of 30.2 kcal/mol is considerably higher than the
experimental activation enthalpy of 15.4 kcal/mol, supporting
the experimental conclusion that the reaction does not occur by
a dissociative but rather a lower-energy associative pathway. It
should be noted, however, that the Re−CO bond is relatively
weak, especially for a third-row metal. For example, in
molecules such as Fe(CO)5 and M(CO)6 (M = Cr, Mo, W),
the metal−carbonyl BDEs range from 37 to 46 kcal/mol,22
while in the thermally robust CpMn(CO)3 complex, the BDE is
⧧
kcal/mol for ΔHeq + ΔH2 . Similarly, ΔStotal = ΔSeq + ΔS2⧧ and
is determined to have a value of −22.4 1.2 eu. While ΔSeq is
expected to be negative and ΔS2⧧ positive, the magnitude of the
latter value depends on the structure of the transition state and
may not be positive enough to counter the negative entropy
change for the equilibrium step. Thus, the overall negative
entropy is not unreasonable given the proposed mechanism.
13045
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049

Inorganic Chemistry
Article
Figure 9. Difference FTIR spectra obtained immediately upon
reaction of 1a with 0.4 M pyridine at 286 K. The absorptions due
to IM persist for less than 90 s under these conditions. The inset
shows the aldehyde CO stretch for IM, and its position at 1650
cm−1 is consistent with an uncoordinated oxygen atom.
Figure 7. Difference FTIR spectra obtained at several time intervals
upon reaction of 1a with 1.6 M pyridine at 333 K. The inset shows the
decay and growth of 1a and 3a, respectively.
intermediate complex yields estimates that are within 5 cm−1 of
the observed values for IM (Table 1). The close agreement
between the calculated and experimental vibrational values for
this complex provides further confirmation for the correctness
of its structural assignment. The calculated structures of 1a, IM,
and 3a are shown in Figure 11. Interestingly, the Re−pyridine
bond lengths are similar in both IM (2.255 Å) and 3a (2.257
Å), indicating that this bond is already fully formed in the
intermediate complex. The average Re−CO bond length for
the CO fragments trans to the pyrrolyl ligand in 1a is 1.935 Å,
while it is calculated to be 2.015 Å for the cis-CO ligands. These
bond lengths are consistent with the experimental finding of
preferential cis-CO displacement and are indicative of a weaker
Re−CO interaction in this case.
A reaction coordinate diagram (Figure 12) was obtained by
calculating the lowest-energy path connecting the reactant,
intermediate, and product complexes for both the 1a → 3a (R
= H) and 1b → 3b (R = CF3) reactions. A listing of all of the
relevant thermodynamic parameters is presented in Table 3.
R = H. The product is found to be 2.2 kcal/mol more stable
(ΔG°) than the reactant species. Relative to 1a, the
intermediate complex is calculated to be enthalpically favored
by 7.3 kcal/mol but entropically disfavored, resulting in ΔGeq =
+4.7 kcal/mol. As discussed earlier, the linearity of the kobs
versus [pyridine] plot suggests that Keq ≪1. The modeling
results predict Keq ≈ 4 × 10−4 at 303 K and are therefore
consistent with the kinetic data. The intermediate is found to
lie in a well with the barrier for conversion to 1a ≈ 4 kcal/mol
lower than its reaction to 3a, thereby justifying the validity of
the preequilibrium assumption in the kinetic analysis. Another
important agreement between theory and experiment comes
from a comparison of the enthalpic parameters associated with
Figure 8. Molecular structure of 3a. Selected bond lengths (Å) and
angles (deg): Re1−C3 = 1.915(5), Re1−C2 = 1.916(5), Re1−C1 =
1.920(5), Re1−N1 = 2.140(4), Re1−O4 = 2.186(3), Re1−N2 =
2.227(4); C3−Re1−C2 = 87.3(2), C3−Re1−C1 = 89.4(2), C2−Re1−
C1 = 91.3(2), C3−Re1−N1 = 92.85(19), C2−Re1−N1 = 98.00(19),
C1−Re1−N1 = 170.56(17), C3−Re1−O4 = 95.14(19), C2−Re1−O4
= 173.71(16), C1−Re1−O4 = 94.55(17), N1−Re1−O4 = 76.12(14),
C3−Re1−N2 = 177.25(17).
calculated to be >50 kcal/mol.23 The calculations also indicate a
higher free energy (ΔG⧧ = 28.1 kcal/mol) for the direct
displacement of CO by pyridine through an interchange
pathway compared to the associative mechanism considered
here.
The intermediate complex, with the structure suggested by
the experimental data, was found to be a local minimum on the
potential energy surface. Because the identities of 1a and 3a are
known unambiguously, a comparison of the experimental
vibrational values of these species with the calculated values
results in a scaling factor of 0.9473. Application of this scaling
factor to the calculated CO vibrational frequencies of the
the substitution reaction. The experimental estimate of ΔHeq +
ΔH2⧧ = 15.4 kcal/mol is in good agreement with the predicted
value of 13.4 kcal/mol. The structure of TS1 shows a significant
lengthening of the Re−O bond, and the relatively low value of
⧧
ΔH1 (9.1 kcal/mol) is consistent with the expected weak
nature of this interaction (Figure 13). For the departing CO,
the Re−CO bond length in TS1 is similar to that in the reactant
13046
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049

Inorganic Chemistry
Article
Figure 12. Reaction coordinate diagram for the reaction of 1a with
pyridine to form 3a (R = H). The ΔG°298 values were obtained from
the values listed in Table 3.
Table 3. Calculated Thermodynamic Parameters (kcal/mol)
for the Relevant Species in the CO Displacement Reaction at
298.15 K
R = H
1a
TS1
IM
4.7
−7.3
R = CF3
IM
2.3
−8.1
TS2
3a
ΔG°
ΔH°
0
0
20.2
9.1
24.2
13.4
−2.2
−4.1
1b
TS1
TS2
3b
ΔG°
ΔH°
0
0
17.3
7.7
20.8
12.1
−4.9
−4.7
complex, indicating little disruption of this bond as the pyridine
ligand approaches the metal center. The final step in the
mechanism, k2, has the largest activation enthalpy (ΔH2⧧ = 20.7
kcal/mol), which is consistent with the disruption of the Re−
CO interaction that is largely broken in TS2.24 Nonetheless,
Figure 10. (a) Plot of kobs versus [pyridine] at several temperatures for
the reaction of 1a with pyridine to form 3a. (b) Eyring plot.
Table 2. Kinetic Parameters for the Reaction of 1a with
Pyridine at Several Temperatures
because of the chelate effect, the slowest step in the mechanism
⧧
is calculated to be the breaking of the Re−O bond (ΔG1
=
temperature (K)
Keqk2 (s−1) × 1000
20.2 kcal/mol) rather than dissociation of the more strongly
303
313
323
333
0.49 0.03
1.25 0.03
2.71 0.08
5.56 0.24
⧧
bound CO ligand (ΔG2 = 19.5 kcal/mol).
R = CF3. The energetic profile for the reaction of 1b with
pyridine is similar to that for 1a. The expected weaker Re−O
interaction in 1b due to the inductive effect of the CF3 group is
Figure 11. Calculated structures for 1a, IM, and 3a.
13047
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049

Inorganic Chemistry
Article
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: djdarens@mail.chem.tamu.edu (D.J.D.), ashfaq.
bengali@qatar.tamu.edu (A.A.B.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This publication was made possible by funding from the Qatar
National Research Fund (member of Qatar Foundation). The
experimental work was supported by NPRP Grant 09-157-1-
024 and the theoretical studies by NPRP Grant 09-143-1-022.
Figure 13. Calculated structures of the transition states in the reaction
of 1a with pyridine.
REFERENCES
■
(1) (a) Herrmann, W. A.; Cornils, B. Applied Homogeneous Catalysis
with Organometallic Compounds; Wiley-VCH: Berlin, 2002. (b) Mas-
ters, C. Homogeneous Transition-Metal Catalysis; Chapman Hall:
London, 1981. (c) Parshall, G. W.; Ittel, S. D. Homogeneous
CatalysisThe Application and Chemistry of Catalysis by Soluble
Transition-Metal Complexes; John Wiley and Sons: New York, 1980.
(d) Cornils, B.; Herrmann, W. A. J. Catal. 2003, 216, 23.
(2) (a) Johnson, T. R.; Mann, B. E.; Clark, J. E.; Foresti, R.; Green,
C. J.; Motterlini, R. Angew. Chem., Int. Ed. 2003, 43, 3722. (b) Jaouen,
G. Bioorganometallics. Biomolecules, Labelling, Medicine; Wiley-VCH:
Weinheim, Germany, 2005.
⧧
manifested in both a smaller activation barrier to TS1 (ΔH1
=
7.7 vs 9.1 kcal/mol) and increased stability of the resulting
intermediate (ΔHeq = −8.1 vs −7.3 kcal/mol) relative to the
hydrogen analogue. As mentioned earlier, the product Keqk2,
and hence the reaction rate, is almost 6 times larger for the
displacement of CO from 1b compared to 1a. The calculations
confirm this reactivity difference and indicate that, in the case of
the CF3 system, both a more stable intermediate (larger Keq)
and a lower barrier for its conversion to product (larger k2)
contribute to increased CO lability.
̀
(3) (a) Ferber, B.; Tos, S.; Vessieres, A.; Welter, R.; Jaouen, G.
Organometallics 2006, 25, 5730. (b) Le Bideau, F.; Salmain, M.; Top,
S.; Jaouen, G. Chem.Eur. J. 2001, 7, 2289.
̀
(4) (a) Jaouen, G.; Vessieres, A.; Butler, I. S. Acc. Chem. Res. 1993,
CONCLUSIONS
26, 361. (b) Jaouen, G.; Top, S.; Vessier
Organomet. Chem. 2000, 600, 23. (c) Salmain, M.; Vessier
̀
es, A.; Alberto, R. J.
es, A.;
■
̀
The thermal displacement of a CO molecule from a rhenium
tetracarbonyl complex bearing a hemilabile pyrrolyl ligand by
PPh3 and pyridine has been investigated. As reported
previously, the CO ligand is found to be unusually labile in
this species, with the reaction complete within a few hours at
room temperature. During early times in the reaction, an
intermediate complex that is generated rapidly and then
converts to the product tricarbonyl species is observed. This
i n t e r m e d i a t e i s i d e n t i fi e d a s t h e ( η 1 - N -
pyrrolecarboxyaldehyde)Re(CO)4L (L = PPh3 or pyridine)
complex, which forms upon displacement of the aldehyde Re−
O bond by the incoming ligand. This species then converts to
the final complex by loss of a CO ligand. Theoretical modeling
of the reaction using DFT confirms the presence of the
intermediate complex along the reaction pathway, and the
calculated energetic parameters agree well with the exper-
imental values. While the Re−CO BDE is calculated to be 30.2
kcal/mol, the reaction proceeds with a 15.4 kcal/mol enthalpic
barrier. It is postulated that the unusual lability of this complex
is due to the presence of the weak aldehyde Re−O link that can
easily dissociate to open a coordination site on the metal center
and accommodate an incoming ligand prior to CO loss. This
associative pathway provides a lower-energy reaction channel
for the release of CO.
Brossier, P.; Butler, I. S.; Jaouen, G. J. Immunol. Methods 1992, 148, 65.
(5) (a) Schmidt, K.; Jung, M.; Keilitz, R.; Schnurr, B.; Gust, R. Inorg.
Chim. Acta 2000, 306, 6. (b) Jaouen, G.; Top, S.; Vessier
P.; Leclercq, G.; Laios, I. Chem. Commun. 2001, 383.
̀
es, A.; Pigeon,
(6) (a) Romao, C. C.; Blattler, W. A.; Seixas, J. D. Chem. Soc. Rev.
2012, 41, 3571. (b) Motterlini, R.; Otterbein, L. E. Nat. Rev. Drug
Discov. 2010, 9, 728. (c) Alberto, R.; Motterlini, R. Dalton Trans. 2007,
1651. (d) Motterlini, R.; Mann, B. E.; Johnson, T. R.; Clark, J. E.;
Foresti, R.; Green, C. J. Curr. Pharm. Des. 2003, 9, 2525.
(7) (a) Motterlini, R.; Clark, J. E.; Foresti, R.; Sarathchandra, P.;
Mann, B. E.; Green, C. J. Circ. Res. 2002, 90, e17. (b) Clark, J. E.;
Naughton, P.; Shurey, S.; Green, C. J.; Johnson, T. R.; Mann, B. E.;
Foresti, R.; Motterlini, R. Circ. Res. 2003, 93, e2.
(8) (a) Smith, H.; Mann, B. E.; Motterlini, R.; Poole, R. K. IUBMB
Life 2011, 63, 363. (b) Nobre, L. S.; Al-Sahahrour, F.; Dopazo, J.;
Saraiva, L. M. Escherichia coli Microbiol. 2009, 155, 813. (c) Nobre, L.
S.; Seixas, J. D.; Romao, C. C.; Saraiva, L. M. Antimicrob. Agents
Chemother. 2007, 51, 4303. (d) Davidge, K. S.; Sanguimetti, G.; Yee,
C. H.; Cox, A. G.; McLeod, C. W.; Monk, C. E.; Mann, B. E.;
Motterlini, R.; Poole, R. K. J. Biol. Chem. 2009, 284, 4516.
(9) Mirebeau, J.-H.; Bideau, F. L.; Marrot, J.; Jaouen, G.
Organometallics 2008, 27, 2911.
(10) (a) Basolo, F. J. Organomet. Chem. 1990, 383, 579.
(b) O’Connor, J. M.; Casey, C. P. Chem. Rev. 1987, 87, 307.
(c) Jordan, R. B. Reaction Mechanisms of Inorganic and Organometallic
Systems; Oxford University Press: New York, 1998; p 138.
(d) Schuster-Woldan, H. G.; Basolo, F. J. Am. Chem. Soc. 1966, 88,
1657.
ASSOCIATED CONTENT
(11) Son, S. U.; Paik, S.-J.; Lee, I. S.; Lee, Y.-A.; Chung, Y. K.; Seok,
W. K.; Lee, H. N. Organometallics 1999, 18, 4114.
(12) APEX2, version 2009.7-0; Bruker AXS, Inc.: Madison, WI, 2007.
(13) SAINTPLUS: Program for Reduction of Area Detector Data 1034,
version 6.63; Bruker AXS Inc.: Madison, WI, 2007.
■
S
* Supporting Information
CIF file giving X-ray structural data for 2b and 3a and a
complete listing of all Cartesian coordinates for the modeled
complexes. This material is available free of charge via the
Internet at http://pubs.acs.org.
(14) Sheldrick, G. M. SADABS: Program for Absorption Correction of
1036 Area Detector Frames; Bruker AXS Inc.: Madison, WI, 2001.
13048
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049

Inorganic Chemistry
Article
(15) Sheldrick, G. M. SHELXS-97: Program for Crystal Structure 1038
̈
̈
̈
Solution; Universitat Gottingen: Gottingen, Germany, 1997.
(16) Sheldrick, G. M. SHELXL-97: Program for Crystal Structure 1040
̈
̈
̈
Refinement; Universitat Gottingen: Gottingen, Germany, 1997.
(17) Sheldrick, G. M. SHELXL, version 6.10; AXS Inc.: Madison, WI,
2000.
(18) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Parandekar, P. V.; Mayhall, N. J.;
Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; ,
Fox, D. J. Gaussian Development Version, revision H.13; Gaussian, Inc.:
Wallingford, CT, 2009.
(19) Chai, J.-D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10,
6615−6620.
(20) Weigend, F.; Ahlrichs, R. Phys . Chem. Chem. Phys. 2005, 7,
3297−3305.
́
(21) Legault, C. Y. CYLview, 1.0b; Universite de Sherbrooke:
Sherbrooke, Quebec, Canada, 2009; http://www.cylview.org.
(22) Lewis, K. E.; Golden, D. M.; Smith, G. P. J. Am. Chem. Soc.
1984, 106, 3905.
(23) (a) Swennenhuis, B. H. G.; Poland, R.; Fan, W. Y.;
Darensbourg, D. J.; Bengali, A. A. Organometallics 2010, 49, 7597.
(b) Davies, K. W.; Maivald, D.; Grabowski, J. J. J. Photochem. Photobiol.
A 2008, 197, 335.
(24) The distance between the departing CO and the metal center in
TS2 is 2.875 Å, significantly longer than the average Re−CO bond
distances of 1.975 and 1.916 Å in the reactant and product,
respectively.
13049
dx.doi.org/10.1021/ic302182b | Inorg. Chem. 2012, 51, 13041−13049