Low-coordinate cobalt fluoride complexes: Synthesis, reactions, and production from C-F activation reactions

Article abstract of DOI:10.1021/om200991k

A cobalt(II) fluoride complex, [L^tBuCo(μ-F)]₂ [L^tBu = 2,2,6,6-tetramethyl-3,5-bis(2,6-diisopropylphenylimido)hept- 4-yl], was synthesized from L^tBuCo using Me₃SnF via homolytic cleavage of the Sn-F bond. L^tBuCo also performed the overall binuclear oxidative addition of fluorinated arenes to give [L ^tBuCo(μ-F)]₂ and a cobalt(II) aryl complex of the corresponding fluorobenzene substrate in a 1:2 molar ratio. The C-F activation reaction has a first-order rate dependence on both cobalt and fluorobenzene concentrations. The rate is increased by meta-fluoride substituents, and slowed by ortho-fluoride substituents, suggesting electronic and steric influences on the transition state, respectively. The data are most consistent with a mechanism beginning with rate-limiting oxidative addition of the aryl fluoride to cobalt(I), followed by rapid reduction of the cobalt(III) aryl fluoride intermediate by a second molecule of L^tBuCo. [L^tBuCo(μ- F)]₂ also reacts with Et₃SiH to give the hydride complex [L^tBuCo(μ-H)]₂. This hydride complex has low reactivity toward alkenes and N₂, in contrast to an earlier report.

Full text of DOI:10.1021/om200991k

Article
pubs.acs.org/Organometallics
Low-Coordinate Cobalt Fluoride Complexes: Synthesis, Reactions,
and Production from C−F Activation Reactions
Thomas R. Dugan, Jonathan M. Goldberg, William W. Brennessel, and Patrick L. Holland*
Department of Chemistry, University of Rochester, Rochester, New York 14627, United States
S
* Supporting Information
ABSTRACT: A cobalt(II) fluoride complex, [LtBuCo(μ-F)]2 [LtBu = 2,2,6,6-tetramethyl-3,5-bis(2,6-diisopropylphenylimido)-
hept-4-yl], was synthesized from LtBuCo using Me3SnF via homolytic cleavage of the Sn−F bond. LtBuCo also performed the
overall binuclear oxidative addition of fluorinated arenes to give [LtBuCo(μ-F)]2 and a cobalt(II) aryl complex of the
corresponding fluorobenzene substrate in a 1:2 molar ratio. The C−F activation reaction has a first-order rate dependence on
both cobalt and fluorobenzene concentrations. The rate is increased by meta-fluoride substituents, and slowed by ortho-fluoride
substituents, suggesting electronic and steric influences on the transition state, respectively. The data are most consistent with a
mechanism beginning with rate-limiting oxidative addition of the aryl fluoride to cobalt(I), followed by rapid reduction of the
cobalt(III) aryl fluoride intermediate by a second molecule of LtBuCo. [LtBuCo(μ-F)]2 also reacts with Et3SiH to give the hydride
complex [LtBuCo(μ-H)]2. This hydride complex has low reactivity toward alkenes and N2, in contrast to an earlier report.
metal−carbon bond formation via elimination of HF,15 Si−
INTRODUCTION
■
F,11a,16 B−F,17 and Al−F bonds;18 metal−fluorine bond
The C−F bond is one of the strongest bonds encountered in
formation via hydrodefluorination19 or defluorination;20 and
organic compounds.1 C−F bonds are significantly stronger than
nucleophilic attack by an electron-rich metal.21 However, in
complexes of precious metals, C−H activation is often
kinetically preferred over C−F activation.16a,22 Recently, the
oxidative addition of aryl C−F bonds to low-valent cobalt
phosphine complexes was described.23 These reports with
cobalt inspired us to investigate aryl C−F activation with
analogous C−H bonds.2 For example, the bond dissociation
energy (BDE) of F−CH3 is 5 kcal/mol stronger than that of
the C−H bond in methane, and the BDE of F−C6H5 is 13
kcal/mol stronger than that of the C−H bond in benzene.3 The
increased bond strength makes fluorocarbons more resistant to
degradation than analogous hydrocarbons. Therefore, mole-
cules containing C−F bonds are useful for applications where
resistance to oxidation or metabolism is beneficial.4 As a result,
nearly 20% of all pharmaceuticals and 40% of all agrochemicals
incorporate at least one C−F bond.5 In pharmaceuticals, the
strength of the C−F bond improves the metabolic stability, the
lipophilicity, and binding to drug interaction sites.6−8
L
tBuCo, a masked two-coordinate cobalt(I) complex [LtBu
=
2,2,6,6-tetramethyl-3,5-bis(2,6-diisopropylphenylimido)hept-4-
yl]. In a recent communication, we disclosed that LtBuCo reacts
with fluorobenzene to give a cobalt(II) fluoride complex,
[LtBuCo(μ-F)]2, and a cobalt(II) phenyl complex, LtBuCoPh.24
In the work described in this report, the scope of the reaction
between LtBuCo and fluorobenzenes was explored. Kinetic
studies show the rate law and substituent effects and elucidate
the mechanism of C−F activation. Although [LtBuCo(μ-F)]2
was previously reported, a new synthetic route to [LtBuCo(μ-
F)]2 from a tin reagent is provided here. This synthetic route
has now permitted [LtBuCo(μ-F)]2 to be isolated in high yield
free from impurities for the first time. The fluoride complex, in
turn, led to a low-coordinate cobalt hydride complex that is
compared to the literature.
The resistance of fluorocarbons to oxidation makes them
persist in the environment.9 For example, perfluorocyclopen-
tane and perfluorocyclohexane have half-lives of >3000 years in
the atmosphere.10 Therefore, the removal of fluorines from
fluorocarbon compounds (“C−F activation”) is an important
goal, in order to convert them into more biodegradable
fluorine-free analogues.11 The cleavage of C−F bonds often
uses transition-metal complexes.1a,12 The products of C−F
activation reactions are metal−fluoride complexes13 that are
also of interest because of their potentially reactive M−F
bonds.12m
Special Issue: Fluorine in Organometallic Chemistry
The metal-mediated cleavage of a C−F bond can occur by
different pathways, depending on the type of C−F bond and
the metal.12m These pathways include oxidative addition;14
Received: October 16, 2011
Published: January 19, 2012
© 2012 American Chemical Society
1349
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
RESULTS
Scheme 1. Proposed Mechanism for the Production of 1 and
Various Tin Products from a Cobalt(I) Source and Me3SnF
■
Reduction of Me3SnF by LtBuCo gives [LtBuCo(μ-F)]2 (1).
The synthesis of 1 was previously reported from C−F
activation of fluorobenzene by LtBuCo;24 however, because
this synthesis always gave 1 as part of a mixture with LtBuCoPh,
an alternative synthesis of 1 was desired. We have reported that
the cobalt(II) methyl complex LtBuCoCH3 reacts with
trimethyltin fluoride (Me3SnF) to give a cobalt fluoride
complex, but this product retained the tin in an unusual
trimetallic Co−Sn−Co core.25 Interestingly, LtBuCo reacted
with Me3SnF overnight in THF at 70 °C to give 1 in 73% yield.
In a modification that requires one less isolation step, it is
possible to reduce LtBuCoCl with KC8 in THF to give a solution
of LtBuCo(THF), which was then treated with Me3SnF in THF
to give 1 in a comparable yield. The characterization of
[LtBuCo(μ-F)]2 has been confirmed by several physical
methods (see the Experimental Section). Addition of pyridine
gave quantitative conversion to the previously characterized
25
1
L
tBuCoF(py), as judged by H NMR spectroscopy (see the
Supporting Information).
Compound 1 is formally a dicobalt(II) complex. The room-
temperature magnetic moment of 1 at 26 °C was 6.4 0.2 μB
in C6D6 and 6.1 0.1 μB in THF-d8. This was slightly higher
than the spin-only magnetic moment for two uncoupled S =
3/2 centers of 5.5 μB, suggesting a contribution from spin−
orbit coupling, as in other high-spin diketiminate-cobalt(II)
complexes.26 The signals in the H NMR spectrum of 1 in
1
Scheme 2
C6D6 and THF-d8 were not assigned to specific protons on
account of several broad and overlapped resonances. The
broadness was the result of a fluxional process with a
coalescence temperature near 0 °C (see the Supporting
Information).
The ability of Me3SnF to act as a net “F•” source suggested
that the tin fragment was reduced in the reaction. 119Sn{1H}
NMR spectroscopy was used to identify the tin product in the
crude product from the reaction of LtBuCo and Me3SnF. The
119Sn{1H} NMR spectrum showed a resonance at δ −108.82
ppm, which matched the literature chemical shift for Sn2Me6.27
Five additional resonances at δ −80.5, −89.9, −99.5, −261.7,
and −489.8 ppm in the 119Sn{1H} NMR spectrum were
assigned to (Me3Sn)2SnMe2, (Me3Sn)3SnMe, and
(Me3Sn)4Sn.28 The relative amounts of these four tin products
were not definitively quantified from the experiment, but their
formation is consistent with previous reports on the
decomposition of free Me3Sn• radicals through the intermedi-
acy of the carbenoid SnMe2.28 Thus, our data support the
reaction stoichiometry shown in Scheme 1.
2 was also solved in P1, but was a different crystal form.29 The
̅
Reaction of 1 with Et3SiH produces [LtBuCo(μ-H)]2 (2).
[LtBuCo(μ-H)]2 was previously synthesized from LtBuCoCl and
KBEt3H, but this route did not give a pure hydride product.29
We hypothesized that compound 1 could give the cobalt
hydride complex, based on the previously reported reaction of
positions of the bridging hydrides were located in the difference
Fourier map and refined. Because all four positions had similar
electron densities, the 50:50 ratio of disorder occupancies was
fixed in the refinement. The backbones of the two β-
diketiminate ligands are nearly perpendicular, with the angle
between the N11−Co1−N21 and N14−Co2−N24 planes
being 87.79(6)°. The distance between the Co atoms is
2.4763(4) Å, and the cobalt hydride distances are between 1.56
and 1.67 Å. All of these values are similar to the previous lower-
quality crystal structure of this molecule.29
L
tBuFeF with triethylsilane (Et3SiH) to give [LtBuFe(μ-
H)]2.16b,30 Addition of a slight excess of Et3SiH to a solution
of 1 in benzene-d6 resulted in the immediate darkening of the
solution to give a brown color. The 1H NMR spectrum showed
complete conversion of 1 to a new product. Compound 2 was
isolated in 61% yield from in situ generation of 1 from
Though the crystal for the crystal structure had incorporated
C6D6, bulk samples of compound 2 crystallized with two Et2O
molecules, as judged by the 1H NMR spectrum and combustion
analysis. The molecule has C1 symmetry in the solid state, but
L
tBuCoCl (Scheme 2).
The molecular structure of 2 is shown in Figure 1. The
structure was solved in the space group P1 and contained
̅
1
cocrystallized benzene-d6. The previously reported structure of
the H NMR spectrum of 2 indicated C2 symmetry with a C2
1350
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
or LtBuCo(n-hexyl), respectively), and most of the 2 remained
unchanged. LtBuCo(n-hexyl) was identified and characterized by
its independent synthesis from LtBuCoCl and hexylMgBr (see
the Experimental Section). Excess pyridine was added to a
solution of 2 in C6D6 in an attempt to reductively eliminate H2
to form LtBuCo(pyridine);25 however, no reaction was observed
after several hours at room temperature. Heating the solution at
80 °C overnight gave LtBuCo(pyridine) in 52% yield via the 1H
NMR integration standard. Heating 2 in C6D6 at 80 °C
overnight with a slight excess of tert-butylpyridine gave the
LtBuCo(tert-butylpyridine) complex in an improved 68% yield
via the 1H NMR integration standard. A quantitative yield of H2
was observed by GC when the reaction with tert-butylpyridine
was performed in a test tube. Overall, compound 2 was not very
reactive toward alkenes or Lewis bases.
Products from LtBuCo and Aryl Halides. We recently
reported that fluorobenzene reacted with LtBuCo at 60 °C for
18 h in cyclohexane-d12 to give LtBuCoPh and [LtBuCo(μ-F)]2
(1).24 The concentration of LtBuCoPh in the product mixture
had been determined by 1H NMR spectroscopy, but the
concentration of 1 in solution had not been determined directly
Figure 1. Molecular structure of [LtBuCo(μ-H)]2 (2) using 50%
probability thermal ellipsoids. All carbon-bound hydrogen atoms and
cocrystallized solvent have been omitted for clarity. Only one set of
hydride positions is shown for clarity. Selected bond distances [Å] and
bond angles [deg] for 2: Co1−Co2, 2.4763(4); Co−N, 1.961(2)−
1.977(2); N11−Co1−N21, 96.02(6); N14−Co−N24, 95.60(6).
1
due to its broad H NMR spectrum. We now report that it is
1
axis through the cobalt centers. The H NMR spectrum was
possible to quantify the production of 1 indirectly, by taking
advantage of its quantitative reaction with triethylsilane
(Et3SiH) to give [LtBuCo(μ-H)]2 (2). (Note that LtBuCo does
not react with Et3SiH under any conditions.) Thus, at the end
of the reaction, a slight excess of Et3SiH was added to produce
2, and the concentration could be determined relative to a
Cp2Ni integration standard. The ratio of [LtBuCoPh] to [2] was
2:1. Assuming quantitative production of 2 from 1, the ratio of
[LtBuCoPh] to [1] was therefore also 2:1, as illustrated in
Scheme 3.
similar in C6D6 and in THF-d8, indicating that the dimeric
structure is maintained in THF. The solution magnetic
moment of 2 at 27 °C was 6.1 0.1 μB in C6D6 and 6.0
0.1 μB in THF-d8., which indicated little magnetic coupling
between the cobalt(II) ions despite the short Co−Co distance.
Previous 1H NMR studies of the iron analogue of 2,
[LtBuFe(μ-H)]2, showed that it was in equilibrium with a three-
coordinate monomer in solution.31 To evaluate the potential
1
splitting of dimeric 2 into a C2v symmetry monomer, its H
NMR spectra in C6D6 were collected between 26 and 80 °C
(Figure 2). The only changes were in the chemical shift due to
Scheme 3. Net Binuclear Oxidative Addition of Aryl Halides
to LtBuCo Produces the Corresponding LtBuCoX and
LtBuCoPh Complexes (X = F, Cl, Br)
Figure 2. Variable-temperature 1H NMR spectra of 2 recorded
between 26 and 80 °C in C6D6.
the Curie Law, and the magnetic moment of 2 was the same
within error at all temperatures. Therefore, all of our evidence
suggests that compound 2 exists as a dimer in both the solid
state and in solution.
The reactivity of 2 was probed in order to compare it with
the analogous iron hydride complex.32 Heating excess cyclo-
hexene or 1-hexene with 2 in C6D6 overnight at 70 °C yielded
only trace amounts of the [1,2]-addition products (LtBuCoCy
The reaction of LtBuCo with other fluorobenzenes was
explored to discover the scope of the C−F activation reaction
(Table 1). In these reactions, excess (5−15 equiv) of the
fluorinated arene substrate was added to a solution of LtBuCo in
cyclohexane-d12. The reaction mixture was heated at 60 °C
1
from 1 to 18 h depending on the substrate. The H NMR
1
spectra showed that all LtBuCo disappeared over time. The H
1351
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
a
Table 1. Products from LtBuCo and Fluorobenzenes
were sometimes absent due to proximity to the paramagnetic
metal. The ratios of the cobalt aryl complexes to 1 were
determined to be roughly 2:1 using the same strategy described
above for fluorobenzene (addition of Et3SiH after completion
of the reaction to form easily quantified 2 from 1), and the
derived concentrations are given in Table 1.
Alkyl fluorides did not react with LtBuCo: heating excess
α,α,α-trifluorotoluene or perfluorocyclohexane (10 equiv) with
LtBuCo at 90 °C in cyclohexane-d12 yielded only slow
decomposition. However, LtBuCo did react with 1 equiv of
chlorobenzene or bromobenzene at room temperature in less
than 1 h to give a mixture of LtBuCoPh and LtBuCoX (X = Cl,
1
Br) via H NMR spectroscopy. A Cp2Ni internal integration
standard showed that LtBuCoPh and LtBuCoX were formed in a
1:1 ratio with quantitative yield. LtBuCoBr was identified based
1
on the similarity of its H NMR spectra to that of literature
L
tBuCoCl.33 Thus, the reaction of LtBuCo with each aryl halide
results in an overall binuclear oxidative addition,34 as shown in
Scheme 3.
Kinetic Studies of Aryl C−F Bond Cleavage by LtBuCo.
The novel C−F activation reaction and the quantitative
formation of products encouraged us to elucidate the
mechanism of the reaction of LtBuCo with fluorobenzenes
through kinetic studies. We used UV−vis spectrophotometry to
monitor the disappearance of a characteristic absorption band
of LtBuCo at 412 nm during reactions of LtBuCo with an excess
of fluorinated arene in hexane at 60 °C. The kinetic traces fit to
a single exponential equation with R ≥ 0.999 in each case; the
fits yielded pseudo-first-order rate constants (kobs). Plots of
ln(A − A∞) versus time were linear over 4−6 half-lives, and the
appropriate plots for half-order, three-halves-order, and second-
order dependences deviated significantly from linearity (see the
Supporting Information). The second-order rate constants (k)
were obtained by plotting the pseudo-first-order kobs versus
substrate concentration, which was corrected for the change in
density of hexane at 60 °C (see Figure 4 and the Supporting
Information).35
a
All reactions were performed in cyclohexane-d12. All reactions were
heated at 60 °C except for the reaction with hexafluorobenzene, which
was performed at room temperature.
NMR spectrum of the product mixture showed resonances for
[LtBuCo(μ-F)]2 and for a species that was consistent with the
expected aryl complex. The cobalt aryl products were identified
by 19F{1H} NMR spectroscopy (Table 2); meta- and para-
fluorine signals were always observed, but ortho-fluorine signals
Table 2. 19F{1H} NMR Chemical Shifts (ppm) and
Assignments of Cobalt Aryl Complexes
The spectral changes of the fluorobenzene reaction with
respect to time are shown in Figure 3. An isosbestic point at
390 nm suggested that no intermediates build up in the course
Figure 3. Spectral changes observed upon heating solutions of LtBuCo
(0.26 mM) and fluorobenzene (64 mM) at 60 °C. Concentrations are
reported after mixing.
1
of the reaction (consistent with observations from H NMR
spectroscopy). The fluoroarene concentrations were varied
1352
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
through concentrations up to approximately 1 M. As shown in
Figure 4, kobs increased linearly with increasing substrate
concentration, and the intercepts of the best-fit lines were close
bond), and the normalized rate constant relative to
fluorobenzene for easier comparison. Several trends are evident
when comparing the normalized rates for different difluor-
obenzene substrates: a para-fluorine substituent has no effect
on the rate constant, a meta-fluorine substituent speeds the rate
by a factor of 4, and an ortho substituent slows the rate by a
factor of 4. Though the effects of multiple fluorines were not
precisely additive, these general trends held throughout the
series. Thus, 1,3,5-trifluorobenzene, which contained all meta
C−F bonds, was ∼2.5 times faster than hexafluorobenzene,
which is impeded by ortho-F substituents. Finally, to define
inductive versus resonance contributions, we tested 4-
fluoroanisole, which has a methoxy substituent para to the
fluorine. As in 1,4-difluorobenzene, the effect of the substituent
was minimal: this result indicates that the electronic influence
of fluorines is primarily an inductive rather than a resonance
effect. This also explains why the meta-fluoride substituent is
more influential than the para-fluoride substituent. The ortho-
fluorine effect is opposite and is attributed to a steric effect.
DISCUSSION
■
Formation of [LtBuCo(μ-F)]2. Transition-metal fluoride
complexes have not been reported as frequently as other
transition-metal halide complexes because of obstacles to their
preparation.13a Most cobalt fluoride complexes have been
formed via fluoride abstraction from BF4− and PF6− salts.36
Other fluoride sources for cobalt fluoride complexes are KF37
and organofluorine compounds.23 This paper demonstrates a
new method for the synthesis of transition-metal fluorides, the
use of Me3SnF as a source of formal F•. The balanced reaction
is shown in Schemes 1 and 3 above. The one-electron
reduction of the Sn−F reagent by cobalt(I) produces the
cobalt(II) fluoride complex and an unstable trimethyltin radical.
Trimethyltin radicals dimerize to give hexamethylditin38 and
also form dimethylstannylene and methyl radicals under
heating or photolytic conditions.39 These known byproducts
from Me3Sn• account for the species that we observed by
119Sn{1H} NMR spectroscopy (Scheme 1 above). To our
knowledge, this is the first report of the formation of a simple
transition-metal fluoride complex by the one-electron reduction
of the Sn−F bond in Me3SnF. Me3SnF has been used as a F−
source in the synthesis of main group and transition-metal
fluorides.40 The new synthetic method shown here may be
useful more generally for making transition-metal fluoride
complexes from a low-valent metal precursor.
Figure 4. Linear dependences of knorm on the concentration of aryl
fluorides at 60 °C are shown. A plot detailing the low-concentration
region is in the Supporting Information.
to zero. These observations indicate the rate law
tBu
rate = k[L Co][fluorobenzene]
(1)
These results show that C−F bond cleavage by LtBuCo was the
rate-determining step of the reaction.
Table 3 provides the second-order rate constant for each
substrate, the rate constant normalized for the number of C−F
bonds (knorm, which represents the reactivity of each C−F
Table 3. Rate Constants for Fluorobenzene Substrates at 60
°C in Hexane
Reactivity of [LtBuCo(μ-H)]2 Compared to Other Late
Transition-Metal β-Diketiminate Hydride Complexes.
The reaction of 1 with silanes yielded pure samples of 2 that
were suitable for reactivity studies. Compound 2 showed
limited reactivity toward small-molecule substrates, such as
alkenes, N2, and pyridines. This behavior was distinctly
different than the closely related iron analogues [LtBuFe(μ-
H)]2 and [LMeFe(μ-H)]2, which were very reactive toward a
variety of small molecules. The iron hydride complexes
exhibited three general types of reactivity: first, they reacted
as Brønsted bases with Brønsted acids to release H2 and to give
an Fe complex of the conjugate base; second, the Fe−H bond
added across multiple bonds to give a [1,2]-addition product;
third, the complexes underwent reductive elimination of H2,
followed by coordination of the donor ligand, to give an FeI
product.32 [LMeNi(μ-H)]2 was not as reactive as the Fe hydride
complexes as it only exhibited the third reactivity trend.41
Hydride complexes of [LtBuFeH]2 did not react thermally with
1353
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
Scheme 4. Potential Pathways for the Activation of Aryl C−F Bonds by 2: (a) Radical Pathway, (b) Oxidative Addition,
Followed by Reduction, (c) Coordination of Aryl Fluoride, Followed by Bond Cleavage, and (d) Termolecular Pathway
N2.42 We show here that the [LtBuCoH]2 analogue 2 also does
not react with N2. Interestingly, 2 also failed to show the
reactivity trends exhibited by the other hydride complexes. The
reaction of alkenes with 2 only gave a trace amount of the
expected Co alkyl complex after prolonged heating. Displace-
ment of H2 by pyridine required heating for the reaction to
proceed, whereas the iron analogue eliminated H2 at room
temperature.
resonances, with no simple correspondence to the crystallo-
graphically determined structure.29 This mixture will be
subsequently designated as “2”. Subsequent work indicated
that the products of this reaction did not reliably give the same
crude 1H NMR spectrum. In the present work, analytically pure
2 was obtained through an alternative route, by reacting Et3SiH
with [LtBuCo(μ-F)]2. The new technique gave pure 2 that
allowed us to re-evaluate the properties of impure “2”.
What is the reason for the reduced reactivity of [LtBuCo(μ-
H)]2 relative to its iron analogue? Kinetic studies on [LtBuFe(μ-
H)]2 with alkynes indicated that the dimer must break into
monomers in order to react.31 With [LMeFe(μ-H)]2, kinetic
studies on its reaction with boranes suggested that the Fe2(μ-
H)2 core must partially dissociate into an intermediate with one
bridging hydride ligand and one terminal hydride ligand.43
Thus, our working model is that the reactivity of iron hydride
complexes is dependent on the ability to dissociate bridging
hydrides in the dimeric core. This model can also explain why
the cobalt hydrides are less reactive than the iron analogues,
since the more electronegative cobalt binds its ligands more
tightly.44 We thus suggest that breaking up the Co2(μ-H)2 core
has a significantly higher barrier with Co than with Fe, which
accounts for its reduced reactivity.
Surprisingly, the 1H NMR spectrum of analytically pure 2 did
not have any peaks in common with the reported spectrum of
“2”.29 This indicates that very little (if any) 2 was present in the
samples of “2” used in the previous study. This observation
brought into question the reported reactivity of the cobalt(II)
hydride complex.29 Compound “2” was reported to react with
N2 thermally to give LtBuCoNNCoLtBu and H2. However,
purified 2 did not react with N2 in C6D6 under any conditions
to give LtBuCoNNCoLtBu. In the previous paper, “2” was
reported to react with cyclohexene at room temperature to give
L
tBuCoCy. However, the addition of cyclohexene to pure 2
yielded only traces of LtBuCoCy after heating overnight at 70
°C. Similarly, no [1,2]-addition was observed in the reaction of
purified 2 with 1-hexene. K2[LtBuCoH]2 was previously
reported from the addition of 2 equiv of KBEt3H to
29
Comparison of 2 to Previously Reported [LtBuCo(μ-
H)]2. A previous report described the reaction of LtBuCoCl with
KBEt3H, which was proposed to give [LtBuCo(μ-H)]2;29
however, measurements on this species indicated a mixture.
L
tBuCoCl;29 however, K2[LtBuCoH]2 was not observed when
2 was mixed with KBEt3H or with KC8. However, LtBuCo did
react with KBEt3H to give K2[LtBuCoH]2 in quantitative yield
by the 1H NMR integration standard. Also, pure 2 reacted with
pyridine or tert-butylpyridine to give the corresponding
1
For example, the reported H NMR spectrum contained many
1354
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
cobalt(I) pyridine complex25 and H2 upon heating to 80 °C.
This was the same product previously reported, but a higher
temperature was required with the purified 2.
(1) and a CoII aryl complex of the corresponding aryl fluoride.
Four plausible pathways to the observed products are shown in
Scheme 4. The first is a radical pathway (Scheme 4a), where
LtBuCo reduces the aryl C−F bond by one electron to form the
Co−F bond in LtBuCoF and to produce a phenyl radical. The
second molecule of LtBuCo could then react with Ph• to form
the Co−C bond in LtBuCoPh. In the second pathway (Scheme
4b), LtBuCo oxidatively adds the aryl C−F bond to give a CoIII
complex LtBuCo(F)(Ph). The CoIII complex then oxidizes a
second molecule of LtBuCo in a rapid step to give the two CoII
products. The third pathway (Scheme 4c) starts with
coordination of the fluorobenzene to LtBuCo to give a cobalt
arene complex. The cobalt arene complex oxidatively adds the
aryl C−F bond to give a CoIII complex, which proceeds to the
products like Scheme 4b. The fourth pathway (Scheme 4d) has
two molecules of LtBuCo simultaneously attacking fluoroben-
zene, and the Co centers reduce the aryl C−F bond by one
electron each. This forms the Co−F and Co−aryl bonds in one
concerted step. These pathways are similar to those that were
proposed for the formation of [LtBuFe(μ-H)]2 in an overall
binuclear oxidative addition of H2 to an FeI intermediate.30
Can the pathway be distinguished based on our experimental
results? One important piece of evidence is that the Co aryl
complexes accounted for 50% of the Co in the product, and the
amount of 1 accounted for the other 50% of the Co. No
biphenyl or benzene was detected in the product mixture by 1H
NMR spectroscopy or GC-MS. These results argue against the
radical pathway (a), which would be expected to give some
biphenyl, benzene, other coupled side products, and/or a
nonstatistical product ratio, as in the literature results on aryl
chlorides described above.46
Thus, the current studies show the earlier description29 of the
preparation of “2” from hydride reagents, and the description of its
reactivity with cyclohexene, pyridine, and N2, to be in error. It
should be noted that the preparation and reactivity of the
cobalt(I) hydride complex K2[LtBuCoH]2, also reported in that
paper, has proven reliable in continued studies on its properties.
Thus, the cobalt(I) hydride complex K2[LtBuCoH]2 indeed
reacts with N2, and the mechanistic studies reported on the N2
reaction using this material29 are reliable. An alternative
synthesis of K2[LtBuCoH]2 is reported here, from the reaction
of LtBuCo with KBEt3H; a related transformation is probably
involved in the previous isolation of the cobalt(I) hydride
compound K2[LtBuCoH]2.
What factors can account for the differences between 2 and
the previously reported “2”? Closer inspection of the reported
1H NMR spectrum of “2” in C6D6 (aided by the subsequent
identification of key LtBuCo compounds) reveals that “2” was a
crude mixture of LtBuCo and LtBuCo(benzene). Thus, KBEt3H
acted as a reducing agent rather than a hydride donor in the
production of “2”. In addition, the reverse reaction of LtBuCo
and H2 did not produce 2 under any conditions, including high
temperature, high pressure, and long reaction times. This
contrasts with the iron analogues, where reduction of LtBuFeCl
under an atmosphere of H2 rapidly produced [LtBuFe(μ-H)]2.30
These observations point toward either LtBuCo being lower in
energy than 2 + H2 or LtBuCo having a much higher barrier to
oxidative addition of H2 than Fe. In addition, the prevalence of
L
tBuCo in previously reported samples of “2” explains its
reactivity with N2. It was recently demonstrated that LtBuCo
reacts with N2 to give LtBuCoNNCoLtBu in high yield.24 Thus,
the resonances that disappeared upon addition of N2 to “2”
were not [LtBuCo(μ-H)]2, but rather this reaction consumed
The observed experimental rate law, expressed in eq 1, is
first-order in both cobalt concentration and fluorobenzene
concentration. Pathway (d) is expected to have a rate law that is
second-order with respect to cobalt concentration and first-
order in fluorobenzene concentration. This does not agree with
the experimental rate law, so pathway (d) may be eliminated
from consideration.
L
tBuCo.
Aryl C−F Activation by LtBuCo. The activation of C−F
bonds by cobalt complexes is rare.1a,23,45 Recent reports utilized
Co0 and CoI phosphine complexes to perform the activation of
aryl C−F bonds.23 These complexes often required a directing
group, such as a ketone or an imine, ortho to the C−F bond for
reactivity, though in one case, no directing group was
required.23c Also, some substrates were observed to give C−
H activation instead of C−F activation in competition
experiments.23b,d The overall binuclear oxidative addition of
aryl C−F bonds by LtBuCo reported here exhibited none of
these limitations. LtBuCo does not attack C−H bonds in arenes,
in contrast to complexes of Rh, Pt, and Pd, where the activation
of aryl C−H bonds is preferred kinetically to the activation of
aryl C−F bonds.22 As a result, LtBuCo is useful for
understanding the selectivity of aryl C−F bond cleavage in
partially fluorinated substrates.
The rate law for mechanism (b) with the first step rate-
limiting is shown in eq 2.
tBu
d[L Co]
tBu
−
= k [L Co][PhF]
1
(2)
dt
The steady-state approximation was used to derive the rate law
for mechanism (c), as no intermediates were observed during
the reaction. The rate law predicted for mechanism (c) is
shown (eq 3).
2k1k2[LtBuCo]2[PhF]
k−1[PhF] + k2[LtBuCo]
d[LtBuCo]
−
=
dt
(3)
The rate is expected to saturate in [PhF]. When k−1[PhF] ≪
k2[LtBuCo], it reduces to the rate law observed in eq 4, and
when k−1[PhF] ≫ k2[LtBuCo], it gives the rate law observed in
eq 5.
Zhu and Budzelaar recently reported a Co system that
accomplishes the overall binuclear oxidative addition of aryl
chlorides.46 In their system, aryl C−X bond cleavage resulted
from the hydrogenolysis of a bis(imino)pyridine Co alkyl
complex in the presence of the substrate. This resulted in the
formation of a Co−X complex and the corresponding Co−aryl
complex. The product ratio varied greatly depending on the
substrate, suggesting a pathway that involves aryl radicals. The
complex did not react with C−F bonds.
tBu
d[L Co]
tBu
−
−
= 2k [L Co][PhF]
1
(4)
(5)
dt
tBu
2k k
k
−1
d[L Co]
1 2 tBu
2
=
[L Co]
LtBuCo undergoes the overall binuclear oxidative addition of
aryl fluorides to give two CoII products: a CoII fluoride complex
dt
1355
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
Equation 4 agrees with the experimental rate law, and no
curvature of the plot of kobs versus [PhF] was observed up to 1
M.
Thus, the kinetic data admit mechanism (c) only if k−1[PhF]
≪ k2[LtBuCo], conditions where eq 4 holds. In a relevant
observation, we recently described the binding of other arenes
(benzene, toluene) to LtBuCo.24 The binding equilibria were
established very rapidly, within a minute at room temperature,
and the equilibrium constants were near unity. In contrast,
Figure 5. Proposed transition state for C−F cleavage. This model is
consistent with the rate law, the steric and electronic substituent
effects, and the lack of radical products. The binding mode of LtBu to
cobalt in the transition state is purposely unspecified: it may be κN,η6-
arene24 or κ2.
1
LtBuCo was not observed to bind fluorobenzenes by H NMR
spectroscopy (no changes in the chemical shifts of the
resonances, or appearance of new signals up to [PhF] = 2
M). Thus, considering the high concentration of [PhF] and
higher temperature used in the reaction with fluoroarenes, and
the analogy to rapid coordination/decoordination of other
unreactive arenes, the term k−1[PhF] must be large. As a result,
it is very unlikely that the rate of PhF coordination/
decoordination could be slow enough to fulfill the requirement
that k−1[PhF] ≪ k2[LtBuCo]. A final argument against arene
precoordination with a slow k−1 comes in the substituent
effects: mechanism (c) would not predict 1,3,5-trifluoroben-
zene to be faster than the better π-acceptor hexafluorobenzene.
Therefore, our data argue against mechanism (c) in multiple
ways.
By process of elimination, our data are most consistent with
mechanism (b) in Scheme 4. To agree with our data, the
oxidative addition of PhF to a single cobalt center must be the
rate-limiting step. (We cannot exclude irreversible precoordina-
tion of the PhF in the rate-limiting step, nor can our kinetic
data resolve the details of the mechanism after the rate-limiting
step.) Mechanism (b) is further supported by its ability to
explain the relative rates of C−F activation for a variety of
fluorobenzene substrates using steric and electronic effects. A
meta-fluorine substituent has a σm value of +0.34, whereas a
para-fluorine substituent has a σp value of +0.06.47 Therefore, a
meta-fluorine substituent has a greater electron-withdrawing
effect than a para substituent, as observed in the kinetic results.
The normalized rate constants for fluorobenzene and 1,4-
difluorobenzene were the same within error, whereas the
normalized rate constant of 1,3-difluorobenzene is 3.8 times
greater than that of fluorobenzene. The inductive effect was
most clearly visible in 1,3,5-trifluorobenzene, which was the
fastest substrate. Finally, an electron-donating para-methoxy
substituent (σp = −0.27)47 gave a rate that is only slightly
slower than fluorobenzene, and the small impact implies that
the resonance contribution is minor. This suggests a transition-
state structure as shown in Figure 5, with negative charge
density on the carbon of the C−F bond. Increased rates from
an inductive effect have been observed in the oxidative
additions of aryl halides to other metal complexes.48
difluorobenzene, indicating that the electronic meta influence
is dominant over the steric ortho influence.
CONCLUSIONS
■
Diketiminate-bound cobalt(I) complexes homolytically cleaved
the strong Sn−F bond in Me3SnF to give a cobalt(II) fluoride
complex, [LtBuCo(μ-F)]2. The availability of [LtBuCo(μ-F)]2 led
to the synthesis of pure [LtBuCo(μ-H)]2. [LtBuCo(μ-H)]2 was
shown to be relatively unreactive, which is potentially due to a
kinetic barrier to cleavage of the dimer core. The reactivity of
this hydride complex contradicts previously published work,
showing that the previously reported synthesis and reactivity of
“[LtBuCo(μ-H)]2” was incorrect and that the previous samples
were primarily composed of LtBuCo and unknown impurities.
Cobalt(I) sources LtBuCo and LtBuCo(THF) reacted with aryl
fluorides in a net binuclear oxidative addition reaction, to give
cobalt(II) fluoride and cobalt(II) aryl complexes. Kinetic and
other mechanistic studies showed that the aryl C−F activation
occurred via oxidative addition of the aryl fluoride to LtBuCo,
followed by rapid reduction with a second molecule of LtBuCo.
Overall, these results demonstrate that low-coordinate cobalt
systems have great promise for breaking strong bonds through
well-defined mechanisms.
EXPERIMENTAL SECTION
■
General Considerations. All manipulations were performed
under an argon atmosphere (or nitrogen atmosphere where specified)
by Schlenk techniques or in an M. Braun glovebox maintained at or
below 1 ppm of O2 and H2O. Glassware was dried at 150 °C
overnight, and Celite was dried overnight at 200 °C under vacuum.
Pentane, hexane, benzene, diethyl ether, and toluene were purified by
passage through activated alumina and Q5 columns from Glass
Contour Co. (Laguna Beach, CA). THF was distilled under N2 from a
potassium benzophenone ketyl solution. All solvents were degassed to
remove N2 before bringing them into the glovebox. All solvents were
stored over 3 Å molecular sieves. Benzene-d6 was dried and stored
over flame-activated alumina. Toluene-d8, cyclohexane-d12, THF-d8,
and hexamethyldisiloxane were vacuum-transferred from sodium
benzophenone ketyl solutions and were stored over 3 Å molecular
sieves. Before use, an aliquot of each solvent was tested with a drop of
sodium benzophenone ketyl in THF solution. Triethylsilane was
degassed and was stored over 3 Å molecular sieves. BEt3 (1.0 M in
hexanes) and hexylmagnesium bromide (2.0 M in Et2O) were
purchased from Aldrich and were used as received. KH was purchased
from Aldrich as a suspension in mineral oil; it was washed with hexane
and dried prior to use. Hexafluorobenzene, 1,2,4,5-tetrafluorobenzene,
1,3,5-trifluorobenzene, 1,2-difluorobenzene, 1,3-difluorobenzene, 1,4-
difluorobenzene, fluorobenzene, 4-fluoroanisole, chlorobenzene, bro-
mobenzene, α,α,α-trifluorotoluene, pyridine, and tert-butylpyridine
were dried by distillation from calcium hydride and were stored over 3
Å molecular sieves. Potassium triethylborohydride,49 potassium
graphite,42a trimethyltin fluoride,16b LtBuCoCl,33 and LtBuCo24 were
prepared by published procedures.
The addition of fluorine ortho to the cleaved C−F
significantly slowed the rate of C−F activation due to steric
effects; for example, the addition of ortho-fluorines in
hexafluorobenzene retarded the rate when compared with
1,3,5-trifluorobenzene. This observation is inconsistent with a
mechanism that begins with rate-limiting electron transfer from
the metal complex to the fluoroarene, as observed in rhodium
systems.11a An interesting test case is 1,2,4,5-tetrafluoroben-
zene, in which all C−F bonds are equivalent, and each has one
ortho-, one meta-, and one para-fluorine. The C−F activation
rate for 1,2,4,5-tetrafluorobenzene was greater than that for
fluorobenzene and nearly the same as that for 1,3-
1356
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
1H and 119Sn{1H} NMR data were recorded on a Bruker Avance
500 spectrometer (500 and 186.50 MHz, respectively). 19F{1H} NMR
data were recorded on a Bruker Avance 400 spectrometer or a Bruker
DRX-300 spectrometer (376 and 282.40 MHz, respectively). All
resonances in the 1H NMR spectra are referenced to residual protiated
solvents: benzene (δ 7.16 ppm), toluene (δ 2.09 ppm), THF (δ 3.58
or 1.73 ppm), cyclohexane (δ 1.38 ppm). All resonances in the
119Sn{1H} NMR spectra were referenced to an internal standard of
tetramethyltin (δ 0 ppm). Resonances in the 19F{1H} NMR spectra
were referenced to hexafluorobenzene (δ −164.9 ppm) and were
reported against CFCl3 (δ 0 ppm). Resonances were singlets unless
otherwise noted. The NMR probe temperature was calibrated using
either ethylene glycol or methanol.50 IR data were recorded on a
Shimadzu FTIR spectrophotometer (FTIR-8400S) using a KBr pellet.
UV−vis spectra were recorded on a Cary 50 spectrophotometer using
Schlenk-adapted quartz cuvettes with a 1 mm optical path length.
Solution magnetic susceptibilities were determined by the Evans
method.51 Elemental analyses were determined by the CENTC
Elemental Analysis Facility at the University of Rochester. Micro-
analysis samples were weighed with a PerkinElmer model AD-6
Autobalance, and their compositions were determined with a
PerkinElemer 2400 Series II Analyzer. Air-sensitive samples were
handled in a VAC Atmospheres glovebox under argon.
brought the total volume to 25 mL. The mixture was heated at 70 °C
for 20 h to give an apple-green solution. The volatile components were
removed under reduced pressure to give a red residue. The residue was
extracted with toluene (25 mL) and filtered through Celite into a small
Schlenk flask. Triethylsilane (200 μL, 1.25 mmol) was added. The
solution slowly turned color from red to brown-red over 15 min. The
volatile components were removed under reduced pressure after 3.5 h
to give a brown residue. The residue was extracted with THF (7 mL)
and filtered through Celite to give a brown solution. The solution was
concentrated to 2 mL, and Et2O (6 mL) was added. The solution was
cooled to −45 °C, which produced 131 mg of a semicrystalline brown
solid. A second crop of product was collected by concentrating the
mother liquor to 1 mL, adding more Et2O, and cooling to −45 °C.
This gave an additional 30 mg of product for a total yield of 60.5%.
Single crystals for X-ray crystallography were grown from a saturated
1
solution in benzene-d6. H NMR (C6D6, 25 °C): δ 93.3 (2H), 28.0
t
t
(18H, Bu), 24.7 (6H), 23.3 (2H), 12.4 (18H, Bu), 6.4 (6H), 4.9
(2H), 3.2 (8H, Et2O CH2), 1.1 (12H, Et2O CH3), −1.7 (2H), −3.3
(6H), −5.9 (6H), −11.8 (6H), −24.6, −26.1, −39.8, −40.1, −42.0,
−57.1, −59.6 ppm. A few of the resonances were broad and
overlapping, which prevented reliable integrations for identification
(see the Supporting Information). μeff (C6D6, 27 °C) 6.0(1) μB. IR
(KBr): 3060 (w), 3025 (w), 2962 (s), 2932 (m), 2870 (m), 1535 (m),
1484 (s), 1466 (m), 1430 (m), 1381 (s), 1358 (s), 1312 (s), 1271 (w),
1255 (m), 1214 (w), 1180 (w), 1159 (w), 1125 (w), 1100 (w), 1022
(w), 782 (s), 769 (w), 759 (w), 712 (s) cm−1. UV−vis (THF): 315
(22.5 mM−1 cm−1), 351 (21.5 mM−1 cm−1), 480 (br sh, ∼4 mM−1
cm−1), 560 (br sh, ∼1 mM−1 cm−1), 690 (sh, ∼0.5 mM−1 cm−1), 760
(0.56 mM−1 cm−1) nm. Anal. Calcd for C70H108N4Co·2C4H10O2: C,
73.66; H, 10.17; N, 4.41. Found: C, 74.06; H, 9.95; N, 4.56.
Synthesis of [LtBuCo(μ-F)]2 (1). LtBuCo (162 mg, 0.289 mmol)
and Me3SnF (52.7 mg, 0.288 mmol) were added to a resealable flask.
THF (10 mL) and toluene (15 mL) were added to produce a dark
green mixture. The mixture was mixed and heated at 70 °C for 21 h,
which gave an apple-green mixture. The volatile components were
removed under reduced pressure to give a red residue. The residue was
dissolved in 3 mL of toluene/pentane (1:1 v/v) and was filtered
through Celite. Cooling the solution to −45 °C gave 53 mg of a
cherry-red crystalline solid. Three subsequent crystallizations from the
mother liquor produced an additional 70 mg of product. The total
yield was 73.2%.
Synthesis of LtBuCo(n-hexyl). LtBuCoCl (508 mg, 0.852 mmol)
was dissolved in toluene (50 mL) to give a brown solution.
HexylMgBr (0.43 mL, 0.86 mmol) was added to the solution, which
immediately produced a red-orange mixture. After 1 h, the volatile
components were removed under reduced pressure to give an orange
residue. The residue was extracted with pentane (50 mL) and was
filtered through Celite to remove the insoluble material. The resulting
orange solution was concentrated to 3 mL, and hexamethyldisiloxane
(3 mL) was added. This solution was concentrated to ∼3 mL and was
stored at −45 °C, which produced a crystalline red product (330 mg,
Alternative synthesis: LtBuCoCl (227 mg, 0.381 mmol) was
dissolved in THF (15 mL) to give an olive-green solution. KC8
(56.0 mg, 0.414 mmol) was added, giving LtBuCo(THF)24 as a green
mixture, which was stirred for 1.5 h. The precipitate was allowed to
settle, and the solution was filtered through Celite. The resulting green
solution was transferred to a resealable flask. Trimethyltin fluoride,
Me3SnF (70.9 mg, 0.388 mmol), was added to the flask. Additional
THF was used to rinse in the tin reagent, which brought the total
volume to 30 mL. The mixture was heated at 70 °C for 18 h to give an
apple-green solution. The volatile components were removed under
reduced pressure to give a red residue. The product was extracted with
toluene (15 mL) and was filtered through Celite to give a red solution.
The product was crystallized by the method in the first preparation to
1
60% yield). H NMR (C6D6, 25 °C): δ 242 (2H), 73.3 (4H), 27.0
t
i
(18H, Bu), 18.0 (2H), 11.3 (2H), 9.8 (2H), 4.2 (2H), 3.8 (12H, Pr-
i
CH3), −28.3 (2H), −50.9 (4H), −52.9 (1H,), −120 (12H, Pr-CH3)
ppm. μeff (C6D6, 26 °C) 4.9(1) μB. IR (KBr): 2959 (s), 2925 (m),
2868 (m), 1508 (m), 1460 (m), 1431 (m), 1384 (s), 1363 (s), 1317
(s), 1253 (w), 1220 (m), 1197 (m), 1154 (m), 1129 (w), 1098 (m),
1055 (w), 1030 (w), 933 (w), 893 (w), 845 (w), 803 (w), 777 (m),
756 (m) cm−1. UV−vis (pentane): 278 (14.6 mM−1 cm−1), 332 (12.9
mM−1 cm−1), 370 (sh, ∼9 mM−1 cm−1), 468 (1.21 mM−1 cm−1), 560
(sh, ∼0.1 mM−1 cm−1), 736 (0.12 mM−1 cm−1) nm. Anal. Calcd for
C41H66N2Co: C, 76.24; H, 10.32; N, 4.34. Found: C, 76.07; H, 10.52;
N, 4.13.
1
give the product (164 mg, 74.2% yield). H NMR (C6D6, 25 °C): δ
71.1, 39.6, 38.4, 29.4, 23.3, 8.8 8.2, 8.1, 6.3, 3.2, 1.4, −16.0, −20.3,
−27.5, −39.5, −88.8, −138 ppm. Many of the resonances were broad
and overlapping, which prevented reliable integration and assignment
(see the Supporting Information). 119Sn{1H} (THF-d8): δ −80.5
((Me3Sn)4Sn), −89.9 ((Me3Sn)3SnMe), −99.5 ((Me3Sn)2SnMe2),
−108.82 (Me3SnSnMe3), −261.7 ((Me3Sn)2SnMe2), −489.8
((Me3Sn)3SnMe) ppm (spectra in the Supporting Information). μeff
(C6D6, 26 °C) 6.4(2) μB. IR (KBr): 3053 (w), 2961 (s), 2927 (m),
2907 (m), 2868 (s), 1539 (m), 1491 (s), 1465 (m), 1433 (m), 1384
(s), 1361 (s), 1310 (s), 1274 (w), 1253 (m), 1215 (m), 1183 (w),
1156 (w), 1122 (w), 1100 (m), 1056 (w), 1023 (w), 934 (w), 782
(m), 757 (m), 712 (w) cm−1. UV−vis (toluene): 350 (29.9 mM−1
cm−1), 550 (sh, ∼0.5 mM−1 cm−1) nm. Anal. Calcd for
C70H106N4F2Co: C, 72.51; H, 9.23; N, 4.83. Found: C, 72.58; H,
9.35; N, 4.96.
X-ray Crystallography. The crystal structure of [LtBuCo(μ-F)]2
was previously reported.24 A crystal structure of [LtBuCo(μ-H)]2 has
also been previously reported;29 however, a higher-quality structure is
reported here, with different packing.
Crystals were placed onto the tip of a 0.1 mm diameter glass
capillary tube or fiber and mounted on a Bruker SMART APEX II
CCD Platform diffractometer52 for a data collection at 100.0(1) K
using Mo Kα radiation and a graphite monochromator. A randomly
oriented region of reciprocal space was surveyed: six major sections of
frames were collected with 0.50° steps in ω at six different φ settings
and a detector position of −38° in 2θ. The intensity data were
corrected for absorption.53 Final cell constants were calculated from
the xyz centroids of >3700 strong reflections from the actual data
collection after integration.54
Synthesis of [LtBuCo(μ-H)]2 (2). LtBuCoCl (250 mg, 0.419 mmol)
was dissolved in THF (12 mL) to give a dark olive solution. KC8 (61.9
mg, 0.458 mmol) was added, giving LtBuCo(THF) as a green mixture
that was stirred for 3 h. The mixture was allowed to settle and was
filtered through Celite. The resulting green solution was transferred to
a resealable flask. Me3SnF (78.0 mg, 0.427 mmol) was added to the
flask. Additional THF was used to rinse in the tin reagent, which
Structure 2 was solved using SIR9755 and refined using SHELXL-
97.56 The space group P1 was determined based on intensity statistics.
̅
A direct-methods solution was calculated, which provided most non-
hydrogen atoms from the E-map. Full-matrix least-squares/difference
1357
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
Fourier cycles were performed, which located the remaining non-
hydrogen atoms. The bridging hydride positions were located in the
difference Fourier map at low resolution (∼1.2 Å). Their positions,
once established, were refined relative to those of atom Co2. Their
isotropic displacement parameters were also refined relative to those of
Co2: Uiso[H] = 2Ueq[Co]. The 50:50 ratio of disorder occupancies
was fixed once it was observed that all four positions had very similar
peak heights. All other hydrogen atoms were placed in ideal positions
and refined as riding atoms with relative isotropic displacement
parameters. Also, two diketiminate ligand arms, two isopropyl groups,
and one tert-butyl group are modeled as disordered over two positions
each. All except the latter required 50:50 ratios of disorder
components due to nearby crystallographic inversion centers. The
tert-butyl group is rotationally disordered, with a refined component
ratio of 82:18. Per dicobalt molecule are two cocrystallized half-
occupancy deuterated benzene solvent molecules, which lie on
crystallographic inversion centers. The final full-matrix least-squares
refinement converged to R1 = 0.0621 (F2, I > 2I) and wR2 = 0.1576
(F2, all data).
REFERENCES
■
(1) (a) Kiplinger, J. L.; Richmond, T. G.; Osterberg, C. E. Chem. Rev.
1994, 94, 373. (b) Barthazy, P.; Hintermann, L.; Stoop, R. M.; Worle,
M.; Mezzetti, A.; Togni, A. Helv. Chim. Acta 1999, 82, 2448. (c) Smart,
B. E. J. Fluorine Chem. 2001, 109, 3. (d) O’Hagan, D. Chem. Soc. Rev.
2008, 37, 308.
(2) Clot, E.; Eisenstein, O.; Jasim, N.; MacGregor, S. A.; McGrady, J.
E.; Perutz, R. N. Acc. Chem. Res. 2011, 44, 333.
(3) Luo, Y.-R. Comprehensive Handbook of Chemical Bond Energies;
CRC Press: Boca Raton, FL, 2007.
(4) Begue, J.-P.; Bonnet-Delpon, D. J. Fluorine Chem. 2006, 127, 992.
(5) (a) Thayer, A. M. Chem. Eng. News 2006, 84, 15. (b) Isanbor, C.;
O’Hagan, D. J. Fluorine Chem. 2006, 127, 303.
(6) Kirk, K. L. J. Fluorine Chem. 2006, 127, 1013.
(7) Park, B. K.; Kitteringham, N. R.; O’Neill, P. M. Annu. Rev.
Pharmacol. Toxicol. 2001, 41, 443.
(8) Ojima, I., Ed. Fluorine In Medicinal Chemistry and Chemical
Biology; Wiley: Chichester, U.K., 2009.
(9) McCulloch, A. J. Fluorine Chem. 2003, 123, 21.
(10) Ravishankara, A. R.; Solomon, S.; Turnipseed, A. A.; Warren, R.
F. Science 1993, 259, 194.
The structure of LtBuCo(n-hexyl) was solved using SHELXS-9756
and refined using SHELXL-97.56 The space group Pbca was
determined based on systematic absences and intensity statistics. A
direct-methods solution was calculated, which provided all non-
hydrogen atoms from the E-map. All non-hydrogen atoms were
refined with anisotropic displacement parameters. All hydrogen atoms
were placed in ideal positions and refined as riding atoms with relative
isotropic displacement parameters. The final full-matrix least-squares
refinement converged to R1 = 0.0546 (F2, I > 2I) and wR2 = 0.1433
(F2, all data).
UV−vis Kinetics. A Cary 50 spectrophotometer, equipped with a
Cary single-cell Peltier accessory to maintain the temperature within
0.1 °C, was used for all kinetics experiments. All measurements were
performed in Schlenk-adapted cuvettes with a 1 cm optical path
length. The standard procedure for sample preparation in the glovebox
is described below. A 5 mM stock solution of LtBuCo in hexane (0.25
mL) was added to a 5 mL volumetric flask. The desired amount of
fluorinated substrate was added to the flask, which was then diluted to
5 mL with hexane. The yellow solution was transferred to a Schlenk
cuvette, which was sealed with a glass stopper. The cuvette was taken
from the glovebox and was placed in the UV−vis apparatus. UV−vis
spectra were recorded between 900 and 300 nm with a 600 nm/min
scan rate at 3 min intervals. The interval time was varied depending on
the substrate concentration. The first 1−2 points were typically
discarded as the temperature had not equilibrated.
(11) Selected examples of hydrodefluorination: (a) Aizenberg, M.;
Milstein, D. Science 1994, 265, 359. (b) Aizenberg, M.; Milstein, D. J.
Am. Chem. Soc. 1995, 117, 8674. (c) Burch, R. R.; Calabrese, J. C.;
́
Ittel, S. D. Organometallics 1988, 7, 1642. (d) Clot, E.; Megret, C.;
Kraft, B. M.; Eisenstein, O.; Jones, W. D. J. Am. Chem. Soc. 2004, 126,
5647. (e) Douvris, C.; Nagaraja, C. M.; Chen, C.-H.; Foxman, B. M.;
Ozerov, O. V. J. Am. Chem. Soc. 2010, 132, 4946. (f) Edelbach, B. L.;
Jones, W. D. J. Am. Chem. Soc. 1997, 119, 7734. (g) Young, R. J.;
Grushin, V. V. Organometallics 1999, 18, 294. (h) Fuchibe, K.;
Ohshima, Y.; Mitomi, K.; Akiyama, T. J. Fluorine Chem. 2007, 128,
1158. (i) Amii, H.; Uneyama, K. Chem. Rev. 2009, 109, 2119.
(12) (a) Brothers, P. J.; Roper, W. R. Chem. Rev. 1988, 88, 1293.
(b) Hughes, R. P. Adv. Organomet. Chem. 1990, 31, 183. (c) Morrison,
J. A. Adv. Organomet. Chem. 1993, 35, 211. (d) Burdeniuc, J.; Jedlicka,
B.; Crabtree, R. H. Chem. Ber. 1997, 130, 145. (e) Richmond, T. G.
Top. Organomet. Chem. 1999, 3, 243. (f) Richmond, T. G. Angew.
Chem., Int. Ed. 2000, 39, 3241. (g) Alonso, F.; Beletskaya, I. P.; Yus, M.
Chem. Rev. 2002, 102, 4009. (h) Braun, T.; Perutz, R. N. Chem.
Commun. 2002, 2749. (i) Mazurek, U.; Schwarz, H. Chem. Commun.
2003, 1321. (j) Jones, W. D. Dalton Trans. 2003, 3991. (k) Torrens,
H. Coord. Chem. Rev. 2005, 249, 1957. (l) Braun, T.; Wehmeier, F.
Eur. J. Inorg. Chem. 2011, 613. (m) Braun, T.; Perutz, R. N. Transition-
Metal Mediated C−F Bond Activation. In Comprehensive Organo-
metallic Chemistry III; Crabtree, R. H., Mingos, D. M. P., Eds.; Elsevier:
Amsterdam, 2006.
A kinetic trace for each reaction was generated by plotting
absorbance (y) versus reaction time (t), which was analyzed using
Kaleidagraph v. 3.51.57 These data were fit to the general first-order
integrated kinetic equation, Abs = a + [b exp(−kobst)], where a and b
are constants and kobs is the pseudo-first-order rate constant. The R
values for fitting the kinetic traces were at least 0.999.
(13) (a) Doherty, N. M.; Hoffmann, N. W. Chem. Rev. 1991, 91, 553.
(b) Murphy, E. F.; Murugavel, R.; Roesky, H. W. Chem. Rev. 1997, 97,
3425. (c) Plenio, H. Chem. Rev. 1997, 97, 3363.
(14) Osterberg, C. E.; Richmond, T. G. Activation of Carbon−
Fluorine Bonds by Oxidative Addition to Low-Valent Transition
Metals. In Inorganic Fluorine Chemistry; Oxford University Press:
London, 1994.
(15) (a) Amii, H.; Uneyama, K. Chem. Rev. 2009, 109, 2119.
(b) Meier, G.; Braun, T. Angew. Chem., Int. Ed. 2009, 48, 1546.
(c) Hughes, R. P. Eur. J. Inorg. Chem. 2009, 4591. (d) Driver, T. G.
Angew. Chem., Int. Ed. 2009, 48, 7974. (e) Sun, A. D.; Love, J. A.
Dalton Trans. 2010, 39, 10362. (f) Richmond, T. G. Angew. Chem., Int.
Ed. 2000, 39, 3241.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional kinetic, spectroscopic, and crystallographic data.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: holland@chem.rochester.edu.
(16) (a) Vela, J.; Smith, J. M.; Yu, Y.; Ketterer, N. A.; Flaschenriem,
C. J.; Lachicotte, R. J.; Holland, P. L. J. Am. Chem. Soc. 2005, 127,
7857. (b) Noveski, D.; Braun, T.; Schulte, M.; Neumann, B.;
Stammler, H.-G. Dalton Trans. 2003, 4075. (c) Braun, T.;
ACKNOWLEDGMENTS
■
Wehmeier, F.; Altenhoner, K. Angew. Chem., Int. Ed. 2007, 46, 5321.
̈
Funding was provided by the U.S. Department of Energy,
Office of Basic Energy Sciences, Grant No. DE-FG02-
09ER16089. The authors thank Dr. Ryan Cowley and Prof.
Daniel Weix for helpful discussions.
(d) Lindup, R. J.; Marder, T. B.; Perutz, R. N.; Whitwood, A. C. Chem.
Commun. 2007, 3664. (e) Yang, J.; Brookhart, M. J. Am. Chem. Soc.
2007, 129, 12656. (f) Reade, S. P.; Mahon, M. F.; Whittlesey, M. K. J.
Am. Chem. Soc. 2009, 131, 1847.
1358
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
(17) (a) Braun, T.; Ahijado-Salomon, M.; Altenhoner, K.;
(26) Holland, P. L.; Cundari, T. R.; Perez, L. L.; Eckert, N. A.;
Lachicotte, R. J. J. Am. Chem. Soc. 2002, 124, 14416.
(27) Wakasa, M.; Kugita, T. Organometallics 1999, 18, 2941.
(28) Davies, A. G., Gielen, M., Pannell, K. H., Tiekink, E. R. T., Eds.
Tin Chemistry: Fundamentals, Frontiers, and Applications; Wiley: New
York, 2008; p 36.
(29) Ding, K.; Brennessel, W. W.; Holland, P. L. J. Am. Chem. Soc.
2009, 131, 10804.
(30) Dugan, T. R.; Holland, P. L. J. Organomet. Chem. 2009, 694,
2825.
(31) Smith, J. M.; Lachicotte, R. J.; Holland, P. L. J. Am. Chem. Soc.
̈
Teltewskoi, M.; Hinze, S. Angew. Chem., Int. Ed. 2009, 48, 1818.
(b) Teltewskoi, M.; Panetier, J. A.; Macgregor, S. A.; Braun, T. Angew.
Chem., Int. Ed. 2010, 49, 3947.
(18) (a) Jager-Fiedler, U.; Klahn, M.; Arndt, P.; Baumann, W.;
̈
Spannenberg, A.; Burlakov, V. V.; Rosenthal, U. J. Mol. Catal. A 2007,
261, 184. (b) Ali, M.; Liu, L.-P.; Hammond, G. B.; Xu, B. Tetrahedron
Lett. 2009, 50, 4078. (c) Klahn, M.; Fischer, C.; Spannenberg, A.;
Rosenthal, U.; Krossing, I. Tetrahedron Lett. 2007, 48, 8900. (d) Terao,
J.; Begum, S. A.; Shinohara, Y.; Tomita, M.; Naitoh, Y.; Kambe, N.
Chem. Commun. 2007, 855.
2003, 125, 15752.
(19) (a) Douvris, C.; Ozerov, O. V. Science 2008, 321, 1188.
(b) Kraft, B. M.; Lachicotte, R. J.; Jones, W. D. J. Am. Chem. Soc. 2000,
122, 8559. (c) Kraft, B. M.; Lachicotte, R. J.; Jones, W. D. J. Am. Chem.
Soc. 2001, 123, 10973. (d) Hughes, R. P.; Laritchev, R. B.; Williamson,
A.; Incarvito, C. D.; Zakharov, L. N.; Rheingold, A. L. Organometallics
2003, 22, 2134. (e) Edelbach, B. L.; Fazlur-Rahman, A. K.; Lachicotte,
R. J.; Jones, W. D. Organometallics 1999, 18, 3170. (f) Maron, L.;
Werkema, E. L.; Perrin, L.; Eisenstein, O.; Andersen, R. A. J. Am.
Chem. Soc. 2005, 127, 279. (g) Adonin, N. Y.; Starichenko, V. F.
Mendeleev Commun. 2000, 10, 60. (h) Kraft, B. M.; Lachicotte, R. J.;
Jones, W. D. Organometallics 2002, 21, 727. (i) Johnson, S. A.; Taylor,
E. T.; Cruise, S. J. Organometallics 2009, 28, 3842. (j) Prikhod’ko, S.
A.; Adonin, N. Y.; Babushkin, D. E.; Parmon, V. N. Mendeleev
Commun. 2008, 18, 211. (k) Reade, S. P.; Mahon, M. F.; Whittlesey,
M. K. J. Am. Chem. Soc. 2009, 131, 1847. (l) Fout, A. R.; Scott, J.;
Miller, D. L.; Bailey, B. C.; Pink, M.; Mindiola, D. J. Organometallics
2009, 28, 331. (m) Bailey, B. C.; Huffman, J. C.; Mindiola, D. H. J.
Am. Chem. Soc. 2007, 129, 5302. (n) Krasnov, V. I.; Platonov, V. E.;
Beregoyava, I. V.; Shohegolava, L. N. Tetrahedron 1997, 53, 1797.
(32) Yu, Y.; Sadique, A. R.; Smith, J. M.; Dugan, T. R.; Cowley, R. E.;
Brennessel, W. W.; Flaschenriem, C. J.; Bill, E.; Cundari, T. R.;
Holland, P. L. J. Am. Chem. Soc. 2008, 130, 6624.
(33) Ding, K.; Dugan, T. R.; Holland, P. L.; Adhikari, D.; Mindiola,
D. J. Inorg. Synth. 2010, 35, 43.
(34) (a) Crabtree, R. H. The Organometallic Chemistry of the
Transition Metals, 4th ed.; Wiley: Hoboken, NJ, 2005; p 161.
(b) Hartwig, J. Organotransition Metal Chemistry From Bonding to
Catalysis; University Science Books: Sausalito, CA, 2010; p 262.
(35) Beg, S. A.; Tukur, N. M.; Al-Harbi, D. K.; Hamad, E. Z. J. Chem.
Eng. Data 1993, 38, 461.
(36) (a) van Driel, G. J.; Driessen, W. L.; Reedijk, J. Inorg. Chem.
1985, 24, 2919. (b) Rietmeijer, F. J.; van Albada, G. A.; De Graaff, R.
A. G.; Hasnoot, J. G.; Reedijk, J. Inorg. Chem. 1985, 24, 3597.
(c) Rietmeijer, F. J.; Haasnoot, J. G.; Den Hartog, A. J.; Reedijk, J.
Inorg. Chim. Acta 1986, 113, 147. (d) Jansen, J. C.; van Koningsveld,
H.; Reedijk, J. Nature 1977, 269, 318. (e) van Kralingen, C. G.;
Reedijk, J. Chem. Commun. 1976, 533. (f) Meyer, F.; Heinze, K.;
Nuber, B.; Zsolnai, L. J. Chem. Soc., Dalton Trans. 1998, 207.
(g) Handley, D. A.; Hitchcock, P. B.; Leigh, G. J. Inorg. Chim. Acta
2001, 314, 1. (h) Gorrell, I. B.; Parkin, G. Inorg. Chem. 1990, 29, 2452.
(i) Lee, S. C.; Holm, R. H. Inorg. Chem. 1993, 32, 4745.
(37) Chang, W.-K.; Chiang, R.-K.; Jiang, Y.-C.; Wang, S.-L.; Lee, S.-
F.; Lii, K.-H. Inorg. Chem. 2004, 43, 2564.
(o) Kuhnel, M. F.; Lentz, D. Angew. Chem., Int. Ed. 2010, 49, 2933.
̈
(p) Rieth, R. D.; Brennessel, W. W.; Jones, W. D. Eur. J. Inorg. Chem.
2007, 2839. (q) Kraft, B. M.; Jones, W. D. J. Am. Chem. Soc. 2002, 124,
8681. (r) Watson, L. A.; Yandulov, D. V.; Caulton, K. G. J. Am. Chem.
́
Soc. 2001, 123, 603. (s) Deck, P. A.; Konate, M. M.; Kelly, B. V.;
Slebodnick, C. Organometallics 2004, 23, 1089. (t) Bouwkamp, M. W.;
deWolf, J.; Morales, I. D. H.; Gercama, J.; Meetsma, A.; Troyanov, S.
I.; Hessen, B.; Teuben, J. H. J. Am. Chem. Soc. 2002, 124, 12956.
(u) Santamaría, C.; Beckhaus, R.; Haase, D.; Saak, W.; Koch, R. Chem
Eur. J. 2001, 7, 622. (v) Bareille, L.; Becht, S.; Cui, J. L.; Le Gendre, P.;
Moïse, C. Organometallics 2005, 24, 5802.
(38) Dong, J.; Devi, P. S.; Mendoza, E.; Gafney, H. D. Inorg. Chem.
2002, 41, 11.
(39) (a) Omae, I. Organotin Chemistry; Elsevier: Dordrecht, The
Netherlands, 1989; p 181. (b) Bowman, W. R.; Bridge, C. F.; Brookes,
P.; Cloonan, M. O.; Leach, D. C. J. Chem. Soc., Perkin Trans. 2002, 1,
58.
(40) A few selected references that illustrate the reaction of Me3SnF
with metal chlorides: (a) Herzog, A.; Liu, F.-Q.; Roesky, H. W.;
Demsar, A.; Keller, K.; Noltemeyer, M.; Pauer, F. Organometallics
1994, 13, 1251. (b) Murphy, E. F.; Lubben, T.; Herzog, A.; Roesky, H.
W.; Demsar, A.; Noltemeyer, M.; Schmidt, H.-G. Inorg. Chem. 1996,
́
35, 23. (c) Liu, F.-Q.; Herzog, A.; Roesky, H. W.; Uson, I. Inorg. Chem.
1996, 35, 741. (d) Shah, S. A. A.; Dorn, H.; Voigt, A.; Roesky, H. W.;
Parisini, E.; Schmidt, H.-G.; Noltemeyer, M. Organometallics 1996, 15,
3176. (e) Ding, Y.; Hao, H.; Roesky, H. W.; Noltemeyer, M.; Schmidt,
H.-G. Organometallics 2001, 20, 4806. (f) Singh, S.; Ahn, H.-J.; Stasch,
A.; Jancik, V.; Roesky, H. W.; Pal, A.; Biadene, M.; Herbst-Irmer, R.;
Noltemeyer, M.; Schmidt, H.-G. Inorg. Chem. 2006, 45, 1853.
(g) Yang, Y.; Schulz, T.; John, M.; Ringe, A.; Roesky, H. W.; Stalke,
D.; Magull, J.; Ye, H. Inorg. Chem. 2008, 47, 2585. (h) Eger, S.; Immel,
T. A.; Claffey, J.; Muller-Bunz, H.; Tacke, M.; Groth, U.; Huhn, T.
Inorg. Chem. 2010, 49, 1292.
(41) Pfirrmann, S.; Limberg, C.; Ziemer, B. Dalton Trans. 2008,
6689.
(20) (a) Burdeniuc, J.; Chupka, W.; Crabtree, R. H. J. Am. Chem. Soc.
1995, 117, 10119. (b) Burdeniuc, J.; Crabtree, R. H. J. Am. Chem. Soc.
1996, 118, 2525. (c) Burdeniuc, J.; Crabtree, R. H. Organometallics
1998, 17, 1582.
(21) (a) King, R. B.; Bisnette, M. B. J. Organomet. Chem. 1964, 2, 38.
(b) Jolly, P. W.; Bruce, M. I.; Stone, F. G. A. J. Chem. Soc. 1965, 5830.
(c) Edelbach, B. L.; Jones, W. D. J. Am. Chem. Soc. 1997, 119, 7734.
(d) Doig, S. J.; Hughes, R. P.; Davis, R. E.; Gadol, S. M.; Holland, K.
D. Organometallics 1984, 3, 1921. (e) Hughes, R. P.; Doig, S. J.;
Hemond, R. C.; Smith, W. L.; Davis, R. E.; Gadol, S. M.; Holland, K.
D. Organometallics 1990, 9, 2745. (f) Rausch, M.; Bruhn, C.;
Steinborn, D. J. Organomet. Chem. 2001, 622, 172.
̈
(22) (a) Bosque, R.; Clot, E.; Fantacci, S.; Maseras, F.; Eisenstein, O.;
Perutz, R. N.; Renkema, K. B.; Caulton, K. G. J. Am. Chem. Soc. 1998,
120, 12634. (b) Aizenberg, M.; Milstein, D. J. Am. Chem. Soc. 1995,
117, 8674.
̈
(23) (a) Li, X.; Sun, H.; Yu, F.; Florke, U.; Klein, H.-F.
̈
Organometallics 2006, 25, 4695. (b) Camadanli, S.; Beck, R.; Florke,
̈
U.; Klein, H.-F. Dalton Trans. 2008, 5701. (c) Zheng, T.; Sun, H.;
Chen, Y.; Li, X.; Durr, S.; Radius, U.; Harms, K. Organometallics 2009,
28, 5771. (d) Zheng, T.; Sun, H.; Ding, J.; Zhang, Y.; Li, X. J.
Organomet. Chem. 2010, 695, 1873. (e) Lian, X.; Xu, X.; Sun, H.;
Chen, Y.; Zheng, T.; Li, X. Dalton Trans. 2010, 39, 9523.
(42) (a) Smith, J. M.; Sadique, A. R.; Cundari, T. R.; Rodgers, K. R.;
Lukat-Rodgers, G.; Lachicotte, R. J.; Flaschenriem, C. J.; Vela, J.;
Holland, P. L. J. Am. Chem. Soc. 2006, 128, 756. (b) The smaller LMeNi
hydride complex also does not react with N2: Pfirrmann, S.; Yao, S.;
Ziemer, B.; Stosser, R.; Driess, M.; Limberg, C. Organometallics 2009,
28, 6855.
(24) Dugan, T. R.; Sun, X.-R.; Rybak-Akimova, E. V.; Olatunji-Ojo,
O.; Cundari, T. R.; Holland, P. L. J. Am. Chem. Soc. 2011, 133, 12418.
(25) Ding, K.; Dugan, T. R.; Brennessel, W. W.; Bill, E.; Holland, P.
L. Organometallics 2009, 28, 6650.
(43) Yu, Y.; Brennessel, W. W.; Holland, P. L. Organometallics 2007,
26, 3217.
1359
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360

Organometallics
Article
(44) Housecroft, C. E.; Sharpe, A. G. Inorg. Chem., 2nd ed.; Pearson:
London, 2001.
(45) Korn, T. J.; Schade, M. A.; Wirth, S.; Knochel, P. Org. Lett.
2006, 8, 725.
(46) Zhu, D.; Budzelaar, P. H. M. Organometallics 2010, 29, 5759.
(47) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
(48) (a) Portnoy, M.; Milstein, D. Organometallics 1993, 12, 1665.
(b) Jutand, A.; Mosleh, A. Organometallics 1995, 14, 1810. (c) Biscoe,
M. R.; Fors, B. P.; Buchwald, S. L. J. Am. Chem. Soc. 2008, 130, 6686.
(d) Kleimark, J.; Hedstrom, A.; Larsson, P.-F.; Johansson, C.; Norrby,
̈
P.-O. ChemCatChem 2009, 1, 152. (e) Guijarro, A.; Rosenberg, D. M.;
Rieke, R. D. J. Am. Chem. Soc. 1999, 121, 4155. (f) Ahlquist, M.;
Norrby, P.-O. Organometallics 2007, 26, 550. (g) Bohm, V. P. W;
̈
Herrmann, W. A. Chem. Eur. J. 2001, 7, 4191. (h) Tsou, T. T.; Kochi,
J. K. J. Am. Chem. Soc. 1979, 101, 6319.
(49) Fryzuk, M. D.; Lloyd, B. R.; Clentsmith, G. K. B.; Rettig, S. J. J.
Am. Chem. Soc. 1994, 116, 3804.
(50) (a) Kaplan, M. L.; Bovey, F. A.; Cheng, H. N. Anal. Chem. 1975,
47, 1703. (b) Ammann, C.; Meier, P.; Merbach, A. E. J. Magn. Reson.
1982, 46, 319.
(51) (a) Baker, M. V.; Field, L. D.; Hambley, T. W. Inorg. Chem.
1988, 27, 2872. (b) Schubert, E. M. J. Chem. Educ. 1992, 69, 62.
(52) APEX2, version 2010.7-0; Bruker AXS: Madison, WI, 2010.
(53) Sheldrick, G. M. SADABS, version 2008/1; University of
Gottingen: Gottingen, Germany, 2008.
̈
̈
(54) SAINT, version 7.68A; Bruker AXS: Madison, WI, 2009.
(55) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. SIR97: A new program for solving and refining crystal
structures; Istituto di Cristallografia, CNR: Bari, Italy, 1999.
(56) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(57) KaleidaGraph 3.51; Synergy Software: Reading, PA, 2000.
1360
dx.doi.org/10.1021/om200991k | Organometallics 2012, 31, 1349−1360