Facile and regioselective C-H bond activation of aromatic substrates by an fe(II) complex involving a spin-forbidden pathway

Article abstract of DOI:10.1021/om301219t

The Fe(II) complex Cp*Fe(CO)(NCMe)Ph (Cp* = η⁵-pentamethylcyclopentadienyl) is shown to mediate facile and highly regioselective C-H activation of aromatic substrates including benzene, furan, thiophene, thiazole, and 2-methylfuran. Experimental and computational evidence suggest a mechanism for C-H activation that involves NCMe dissociation, multiple spin intersystem crossings, C-H bond coordination, and C-H bond cleavage by a σ-bond metathesis reaction.

Full text of DOI:10.1021/om301219t

Article
pubs.acs.org/Organometallics
Facile and Regioselective C−H Bond Activation of Aromatic
Substrates by an Fe(II) Complex Involving a Spin-Forbidden Pathway
Steven E. Kalman,† Alban Petit,‡ T. Brent Gunnoe,*,† Daniel H. Ess,*,‡ Thomas R. Cundari,*,∥
and Michal Sabat§
†Department of Chemistry, University of Virginia, Charlottesville, Virginia 22904, United States
‡Department of Chemistry and Biochemistry, Brigham Young University, Provo, Utah 84602, United States
∥Department of Chemistry, Center for Advanced Scientific Computing and Modeling (CASCaM), University of North Texas,
Denton, Texas 76203, United States
§Nanoscale Materials Characterization Facility, Materials Science and Engineering Department, University of Virginia, Charlottesville,
Virginia 22904, United States
S
* Supporting Information
ABSTRACT: The Fe(II) complex Cp*Fe(CO)(NCMe)Ph
(Cp* = η5-pentamethylcyclopentadienyl) is shown to mediate
facile and highly regioselective C−H activation of aromatic
substrates including benzene, furan, thiophene, thiazole, and 2-
methylfuran. Experimental and computational evidence suggest
a mechanism for C−H activation that involves NCMe
dissociation, multiple spin intersystem crossings, C−H bond
coordination, and C−H bond cleavage by a σ-bond metathesis
reaction.
with developing catalysts based on the first row of the transition
metal series including typically weaker metal−carbon bond
strengths, propensity toward single-electron chemistry, compli-
cations due to the possibility of spin crossover, and access to
multiple oxidation states.11,12 These challenges have limited the
number of examples of nonoxo, nonradical stoichiometric C−H
activation reactions based on 3d transition metals.13−16 Some
examples of catalytic C−H activation and functionalization by
first-row transition metals have been disclosed.17−19
Two of our groups have previously reported that TpRu(L)-
(NCMe)Ph [Tp = hydridotrispyrazolylborate; L = CO, PMe3,
P(OCH2)3CEt, 2,6,7-trioxa-1-phosphabicyclo[2,2,1]heptane,
P(N-pyrrolyl)3] complexes readily activate aromatic and
olefinic C−H bonds.20−24 We envisioned that a similar
(Y)Fe(L)(NCMe)Ph (Y = monoanionic, six-electron donor
ligand) complex with an Fe(II) metal center might also provide
entry into C−H activation.25 Because of the similarities
between Tp and Cp*,26 we chose Cp*Fe(CO)(NCMe)Ph
(2) as the initial complex. Herein, we report experiments and
calculations that showcase the success of this Fe(II) complex
for regioselective aromatic C−H activation under mild
conditions.
INTRODUCTION
■
The majority of homogeneous catalysts for hydrocarbon C−H
bond activation and functionalization are based on second- or
third-row transition metals.1−3 Although first-row transition
metal oxo, alkoxide and amide complexes can functionalize C−
H bonds,4−10 they typically function via net H abstraction and
are often limited to relatively weak C−H bonds.4 While nature
has evolved highly selective enzymes for C−H functionalization
using net H abstraction by metal oxo complexes (including Fe
oxo complexes), extending the high reactivity and selectivity to
synthetic systems has been a substantial challenge.4 Thus, it is
desirable to develop nonradical, nonoxo metal C−H activation
catalysts based on inexpensive Earth-abundant transition metals
(Scheme 1). However, there are several challenges associated
Scheme 1. Oxo Mediated H Atom Abstraction and Fe
Mediated C−H Activation
Received: December 17, 2012
Published: February 27, 2013
© 2013 American Chemical Society
1797
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
calculate possible intermediates and transition states.31
Comparison of the M06 optimized structure of 2 (Figure 2)
shows highly similar bond lengths and angles compared to the
X-ray structure shown in Figure 1. Starting from complex 2, C−
H activation begins with NCMe dissociation to create a vacant
coordination site. The enthalpic (ΔH) penalty for complete
NCMe dissociation along the singlet energy surface is 27.1
kcal/mol. The resulting Cp*Fe(CO)(Ph) singlet complex (3s)
shows an unrestricted wave function with a spin contamination
<S2> value of 0.23, which suggests open-shell character and a
possible high-spin state of lower energy. Indeed, the triplet
Cp*Fe(CO)(Ph) complex (3t) is adiabatically 16.4 kcal/mol
more stable than 3s. The structures 3s and 3t show that after
complete loss of NCMe, the CO and Ph groups reorient
slightly, but the only major bond length change is a decrease in
the Fe−Ph bond length from 1.98 Å in 2 to 1.91 and 1.94 Å in
3s and 3t, respectively. The structures 3s and 3t are highly
similar.
RESULTS AND DISCUSSION
■
The target complex 2 was prepared by reaction of Cp*Fe-
(CO)2I27 with CuOTf and Bu3SnPh, which gave Cp*Fe-
(CO)2Ph (1) in 67% isolated yield. Complex 1 has been
previously reported,28 albeit in lower yields. The photolysis of
an NCMe solution of 1 produced Cp*Fe(CO)(NCMe)Ph (2)
in 87% isolated yield (Scheme 2). Complexes 1 and 2 were
both characterized by single crystal X-ray diffraction (Figure 1).
Scheme 2. Synthesis of Cp*Fe(CO)(NCMe)Ph (2)
Consideration of spin states and their interconversion is
important for first row transition metals.32−34 Direct conversion
of 3s to 3t can occur through a singlet−triplet intersystem
crossing, called a minimum energy crossing point (MECP).
The optimized structure of MECP-1 (Figure 3) connects 3s to
3t and has an energy and geometry nearly identical to 3t.
Although it is possible that NCMe completely dissociates
before singlet−triplet intersystem crossing, there is a lower
energy pathway for conversion of complex 2 to 3t. This
pathway involves singlet−triplet interconversion with partial
NCMe dissociation (Fe−N = 2.18 Å) via MECP-2 with a ΔH
value of 9.0 kcal/mol relative to 2. Optimization of the resulting
triplet structure after MECP-2 resulted in the triplet NCMe π
complex 4t with ΔH = 4.6 kcal/mol. Structure 3t is then
accessed upon NCMe dissociation. These structures suggest
that NCMe dissociation is facile through a dissociative
mechanism. An interchange coordination mechanism that
remains on the singlet energy surface is unlikely due to the
reluctance of the Fe metal center to increase its ligand
coordination number.
Figure 1. ORTEP drawings of Cp*Fe(CO)2Ph (1) and Cp*Fe(CO)-
(NCMe)Ph (2) (50% probability ellipsoids; H atoms omitted).
Complex 1 (left). Selected bond lengths (Å): Fe−C7/C7′ 1.756(1),
Fe−C8 2.002(2), C7−O1/C7′−O1′ 1.151(2). Selected bond angles
(°): C7−Fe−C7′ 95.87(8), C7/C7′−Fe−C8 91.30(5). Complex 2
(right). Selected bond lengths (Å): Fe−C14 1.990(1), Fe−C11
1.741(1), Fe−N1 1.903(1), C11−O1 1.154(2). Selected bond angles
(°): C11−Fe−N1 98.00(6), C11−Fe−C14 91.79(6), N1−Fe−C14
89.76(5).
For an initial test of C−H activation, 2 was heated to 50 °C
in C6D6. Cp*Fe(CO)(NCMe)(Ph-d5) (2-d5) forms in ∼80%
yield based on NMR spectroscopy (eq 1). The formation of 1
Scheme 3 outlines the lowest energy pathway calculated for
benzene C−H activation by complex 2. As discussed above, 3t
is generated via MECP-2 followed by NCMe loss. Although 3t
is a viable intermediate, Fe−Ph group exchange from this
species on the triplet energy surface showed barriers too high to
be reasonable, and it is therefore a dead-end intermediate. As a
result, a second intersystem crossing step to return to the
singlet energy surface is required during the C−H bond
coordination/activation mechanistic stage. There are two
possible pathways for this intersystem crossing. The first
pathway involves conversion of 3t to 3s via MECP-1 followed
and 1-d5 (∼10%) and other uncharacterized products make up
the remaining ∼20% of Fe complexes. Under pseudo-first-order
conditions in C6D6 at 49 °C, kobs is 4.6(5) × 10−4 s−1 for the
C−D activation of C6D6.
To explore the mechanism of C6D6 C−D (modeled as C−
H) bond activation by 2, we utilized M0629,30 density
functional calculations with SMD solvent corrections to
Figure 2. M06 ground-state structures. Bond lengths reported in Å and angles in degrees.
1798
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
Figure 3. MECPs and triplet π complex. Bond lengths reported in Å and angles in degrees.
Scheme 3. M06/6-311++G(3df,3dp)[LANL2TZ(f)]//M06/6-31G(d,p)[LANL2DZ] Calculated Enthalpies (Free Energies at
298 K) for C−H Activation of Benzene by Cp*Fe(CO)(NCMe)Ph (2) in Benzene Solvent (kcal/mol)
by benzene coordination on the singlet surface. The second
pathway, which is lower in energy, involves intersystem crossing
along with benzene coordination via MECP-3 to give singlet 5s
directly from 3t. The structure and energy of MECP-3 is nearly
identical to 5s (Figure 4). The enthalpy of triplet 5t is 24.0 kcal/
mol and confirms that the singlet−triplet crossing point occurs
just prior to structure 5s.
Structure 5s involves η2-C,H-benzene coordination. Because
of the orientation of the Fe−Ph group, no true η2-C,C-benzene
coordination complex could be located. For Ru(II) complexes,
η2-C,H-benzene coordination in preference to η2-C,C-benzene
coordination has been taken to imply steric congestion at the
metal center.22 From intermediate 5s, the lowest energy
pathway for C−H bond cleavage occurs via a four-centered
σ-bond metathesis type transition state 6-TS (Figure 4) with a
calculated ΔH‡ of 29.4 kcal/mol. In this transition structure,
the Fe−Ph bond length is stretched to 2.04 Å, and the benzene
C−H bond partial bond length is 1.49 Å. The transition state 6-
TS directly connects to another η2-C,H-benzene coordination
complex 5s. At this point NCMe can replace coordinated
benzene from 5s to regenerate 2. This last step is also
susceptible to spin intersystem crossings, but they are not
pictorially represented in Scheme 3.
We have ruled out several other C−H activation mechanisms
for Fe−Ph group exchange. The first involves the generation of
a so-called “tuck-in” type complex directly from 2. In this
mechanistic pathway a methyl C−H bond of the Cp* group
undergoes intramolecular σ-bond metathesis with the Fe−Ph
group to give a cyclometalated Cp* group and η2-CH-benzene
coordination. The calculated ΔH‡ for this process is 50.9 kcal/
mol. The ΔG‡ for forming the tuck-in complex after
dissociation of NCMe is 54.6 kcal/mol. We have observed no
Figure 4. M06 structures for benzene coordination and C−H bond
cleavage. Bond lengths reported in Å and angles in degrees.
1799
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
deuterium incorporation into the Cp* methyl resonances (1H
NMR spectroscopy) for the reaction of 2 and C6D6.
We have also considered the Fe(II) to Fe(IV) oxidative
addition from 3s and 3t to give the seven-coordinate diphenyl
hydride intermediate Cp*Fe(CO)(Ph)(Ph)(H). All optimiza-
tions starting with seven-coordinate Fe hydride structures
reverted back to 5s. Lastly, our calculations suggest several
hydrogen abstraction mechanisms from both 3s and 3t are
unlikely pathways. For example, the ΔH to give Cp*Fe(CO)-
(Ph)(H) and ·C6H5 is 48.4 kcal/mol relative to 3s. Eisenstein et
al. have found similar results for their study on the TpFe
motif.35
The ability of 2 to activate benzene prompted us to explore
the possibility of 2 activating heteroaromatic substrates under
similar mild conditions. Indeed, complex 2 regioselectively
activates furan, thiophene, thiazole and 2-methylfuran at or
below room temperature (Scheme 4). To optimize stability, the
Figure 5. 1H NMR spectra (Cp* protons) showing the decoalescence
of the Cp* peak for complex 8 (25 to −80 °C).
Scheme 4. C−H Activation of Heteroaromatic Substrates by
Cp*Fe(CO)(NCMe)Ph (2)
Figure 6. ORTEP drawing of Cp*Fe(CO)(PPh3)(2-furyl) (7) (50%
probability ellipsoids; H atoms omitted). Selected bond lengths (Å):
Fe−P1 2.2142(3), Fe−C15 1.730(1), C15−O2 1.159(2), Fe−C11
1.960(1). Selected bond angles (°): C11−Fe−C15 89.59(5), C15−
Fe−P1 89.65(4), C11−Fe−P1 92.89(4).
products from the C−H activation of furan, thiophene and 2-
methylfuran were isolated as their PPh3 adducts. Cp*Fe(CO)-
(PPh3)(2-furyl) (7) and Cp*Fe(CO)(PPh3)(2-thienyl) (8)
were isolated in 96 and 97% yield, respectively, from the
reaction of 2 with the furan or thiophene followed by the
addition of PPh3.
Scheme 5. Calculated Relative Enthalpies for the Two
Possible Coordination Isomers of Cp*Fe(CO)(thiazole)(2-
thiazolyl) (9)
The 1H and 13C NMR spectra for complexes 7 and 8 exhibit
broadened resonances in the aromatic region. A variable-
1
temperature H NMR experiment was performed on 8, and
decoalescence of the Cp* resonance was observed (Figure 5).
When free thiophene (0.5 equiv) was added to a THF-d8
solution of 8, no changes to the peak shape or chemical shifts
1
for 8 in the H NMR spectrum were observed. These data,
taken in account with only one carbonyl stretching frequency in
the IR spectrum and one resonance in the 31P NMR spectrum,
implicates that the fluxionality observed is not due to reversible
phosphine cyclometalation but possibly a result of hindered
rotation of PPh3. The structure of 7 has been determined by
single-crystal X-ray diffraction (Figure 6), which confirms the
regioselectivity of C−H activation at the 2-position.
methylfuran followed by the addition of PPh3 produced
Cp*Fe(CO)(PPh3)[2-(5-methylfuryl)] (10) in 82% isolated
yield. Similar to complexes 7 and 8, complex 10 also exhibits
1
broadened H and 13C NMR resonances in the downfield
region of the spectra, which may also be attributed to slow
rotation around the Fe−PPh3 bond. This selectivity, in which
Fe activates a stronger aromatic C−H bond in preference to a
weaker CH3 bond, suggests that C−H activation proceeds via a
pathway that does not involve the formation of free radicals in
accordance with the M06 density functional calculation
predictions (see above).
To experimentally probe these C−H activation reactions we
studied several kinetic features for the reaction of 2 with furan.
Under pseudo-first-order conditions in 2 at 3 °C, a plot of [2]
vs time shows a first-order exponential decay (Figure 7).
Varying the equivalents of furan relative to 2 revealed that the
The reaction of 2 with thiazole produces Cp*Fe(CO)-
(thiazole)(2-thiazolyl) (9) in 36% isolated yield. Calculations
suggest the N-bound thiazole complex is more favorable than
the S-bound isomer by 13.5 kcal/mol (Scheme 5, see
Supporting Information). The formation of 9 proceeds via an
1
intermediate (observed by H NMR spectroscopy), which is
likely Cp*Fe(CO)(thiazole)Ph.
The C−H activation of 2-methylfuran by 2 is selective for the
5-position over the methyl group. The reaction of 2 with 2-
1800
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
[furan]) gave activation parameters of ΔH‡ = 23.5(4) kcal/mol
and ΔS‡ = 12(2) cal/mol·K (Figure 10).
Figure 7. Sample first-order decay plot for the reaction of 2 and furan
(10 equiv) at 3 °C (R2 = 0.99).
reaction rate has a first-order dependence on the concentration
of furan at low concentrations with saturation kinetics at higher
concentrations (Figure 8). The rate of reaction is dependent on
Figure 10. Eyring plot for furan C−H activation (10 equiv) by
Cp*Fe(CO)(NCMe)Ph (2) (R2 = 0.99; −12 to 13 °C).
We computed the enthalpy profile for reaction of 2 with
furan in THF solvent (Scheme 6). We have found a nearly
identical mechanism for benzene and furan C−H activation by
2. The initial steps of NCMe dissociation and intersystem
crossing are identical with the benzene mechanism shown in
Scheme 3. The energies of MECP-2, 4t, and 3t in THF solvent
are a few tenths of a kcal/mol lower than in benzene solvent.
From 3t furan coordinates via MECP-4 (Figure 11, ΔH = 15.9
kcal/mol) to give singlet 11s, which involves the formation of a
true η2-C,C-complex Cp*Fe(CO)(η2-C,C-furan) with a ΔH of
13.0 kcal/mol.
The η2-C,C-complex 11s is less endergonic than the benzene
complex 5s because it involves π forward- and back-bonding in
contrast to mainly σ type interactions in 5s. For Lewis basic
metals, furan is known to bind stronger than benzene in a
dihapto-coordination mode.36 The calculated ΔH‡ for C−H
bond cleavage of furan via the σ-bond metathesis transition
state 13-TS is 22.2 kcal/mol, which is in good agreement with
the experimental value for ΔH‡ of 23.5(4) kcal/mol. Despite
the 8 kcal/mol computed stronger C−H bond strength of furan
(ΔH = 118 kcal/mol) versus benzene (ΔH = 110 kcal/mol),
the ΔH‡ value for furan C−H activation is ∼7 kcal/mol lower
than the benzene ΔH‡ value. The lower ΔH‡ for C−H
activation of furan could be interpreted as a result of the lower
energy η2-C,C-complex 11s. However, prior to 13-TS the η2-
C,C-complex rearranges to an η2-C,H-complex 12s with
coordination to the furan σ C−H bond rather than the furan
π bond. An alternative explanation for the lower ΔH‡ value for
furan C−H activation versus benzene C−H activation is the
result from a more stable Cp*Fe(CO)(2-furyl)(η2-C,H-
benzene) intermediate 14s (ΔH = 12.7 kcal/mol) compared
with 5s. The thermodynamic stability of 14s is manifested in 13-
TS as stability gained from formation of a more stable Fe−furyl
bond compared with the formation of a Fe−Ph bond in 6-TS
(see Supporting Information for details).37
Figure 8. Plot of pseudo-first-order rate constants (kobs) versus [furan]
for the reaction of 2 with excess furan at 3 °C.
Figure 9. Plot of 1/kobs versus [NCMe] (R2 = 0.99) for the reaction of
2 with excess furan.
the concentration of free NCMe (Figure 9). Treating complex
2 with 10 equiv of a 1:1 molar solution of furan and furan-d4
and analyzing the relative quantities of C6H6 and C6H5D by
GC/MS revealed a kinetic isotope effect of 5.0(4). Additionally,
we have examined the kinetic isotope effect with a large excess
of furan (50 equiv), and kH/kD = 4.8(1), which is statistically
equivalent with that determined with 10 equiv of furan. Kinetic
data for the C−H activation of furan (10 equiv) were obtained
from −12 to 13 °C. An Eyring plot (in which the
experimentally determined kobs were corrected to remove
On the enthalpy surface the furan C−H bond cleavage
transition state 13-TS is higher in energy than MECP-2 and
MECP-4 points. This suggests that the rate of Fe−Ph group
transformation into a Fe−furyl group is controlled by 13-TS,
which is in accordance with the relatively large KIE value (∼5)
observed experimentally.
On the basis of calculations and experimental results, a
proposed mechanism and corresponding rate law for the C−H
1801
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
Scheme 6. Calculated Enthalpies (Free Energies at 298 K) for C−H Activation of Furan by Cp*Fe(CO)(NCMe)Ph (2) in THF
Solvent (kcal/mol)
Figure 11. Furan C−H activation structures. Bond lengths reported in Å and angles in degrees.
activation of furan by 2 is shown in Scheme 7. Reversible
NCMe dissociation via a spin-forbidden pathway yields the
coordinatively unsaturated intermediate. After furan coordina-
tion and subsequent C−H bond cleavage with concomitant
release of benzene, acetonitrile recoordinates completing the
transformation. The rate law predicts 1/[NCMe] inhibition
(Figure 8) and saturation kinetics in concentration of furan
(Figure 9).
Scheme 7. Proposed Mechanism and Corresponding Rate
Law for the C−H Activation of Furan by
Cp*Fe(CO)(NCMe)Ph (2)
Using the proposed rate law, k1 can be independently
determined from the data in Figure 8 and Figure 9. Under
saturation conditions (Figure 8), the rate law is reduced to rate
= k1[2], and using the data in Figure 8, k1 = 8.8 × 10−4 s−1. The
value of k1 can also be determined from the y-intercept of
Figure 9, which yields k1 = 7.1 × 10−4 s−1. Thus, the two
independently determined values of k1 are in good agreement.
1802
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
Under saturation conditions in furan, kobs = k1, which
provides the rate of NCMe dissociation from 2 (see above).
Thus, an Eyring analysis was performed under saturation
conditions (30 equiv of furan, −22 to 3 °C) in order to extract
the activation parameters for the NCMe dissociation sequence
(Figure 12). For NCMe dissociation, ΔH‡ = 20.2(3) kcal/mol
have also suggested thermodynamic influence on the rates of
metal-mediated C−H activation.39−41
SUMMARY
■
We have reported an Fe(II) complex, Cp*Fe(CO)(NCMe)Ph,
that undergoes regioselective C−H activation with a range of
aromatic substrates under mild conditions. The furan 2-CH
bond (calculated bond dissociation enthalpy = 118 kcal/mol) is
readily broken below 0 °C. In addition, we have disclosed a
combined experimental and computational mechanistic study
for the regioselective C−H activation of furan. The results
herein demonstrate a relatively rare example of Fe(II)-mediated
C−H activation by a nonradical pathway and demonstrates the
potential viability of Fe-based catalysts for the functionalization
of inert hydrocarbons and other substrates with strong C−H
bonds.
EXPERIMENTAL SECTION
■
General Considerations. Unless otherwise noted, all synthetic
procedures were performed under anaerobic conditions in a nitrogen-
filled glovebox or by using standard Schlenk techniques. Glovebox
purity was maintained by periodic nitrogen purges and was monitored
by an oxygen analyzer (O2 < 15 ppm for all reactions).
Tetrahydrofuran and n-pentane were dried by distillation from
sodium/benzophenone and P2O5, respectively. Diethyl ether was
distilled over CaH2. Benzene was purified by passage through a
column of activated alumina. Benzene-d6, acetone-d6, CD3CN, and 1,4-
dioxane-d8 were used as received and stored under a N2 atmosphere
over 4 Å molecular sieves. For kinetic experiments, THF-d8 was
degassed by two conventional freeze−pump−thaw cycles and stored
over 4 Å molecular sieves. 1H NMR spectra were recorded on a Varian
Mercury 300 or Varian Inova 500 MHz spectrometer, and 13C NMR
spectra were recorded on a Varian Inova 500 MHz spectrometer
(operating frequency 125 MHz) or Bruker Avance III 800 MHz
spectrometer (operating frequency 201 MHz). All 1H and 13C spectra
are referenced against residual proton signals (1H NMR) or 13C
resonances (13C NMR) of the deuterated solvents and are reported in
ppm. 31P NMR spectra were obtained on a Varian Mercury 300 MHz
(operating frequency 121 MHz) spectrometer and referenced against
an external standard of H3PO4 (δ = 0). GC/MS was performed using a
Shimadzu GCMS-QP2010 Plus system with a 30 m × 0.25 mm RTx-
Qbond column with 8 μm thickness using electron impact ionization.
IR spectra were obtained on a Shimadzu IRAffinity-1 Fourier
transform infrared spectrometer. Samples were prepared in solution
flow cells. Photolysis experiments were performed using UV−vis
radiation generated by a 450 W power supply (Model #l7830, Ace
Glass, Inc.) equipped with a water-cooled 450 W 5 in. arc IMMER
UV−vis lamp (Model #7825-34, Ace Glass, Inc.). Furan-d4 was
purchased from Aldrich and distilled prior to use. All other chemicals
were used as purchased from commercial sources. Elemental analyses
were performed by Atlantic Microlabs, Inc. Cp*Fe(CO)2I27 was
prepared according to the literature procedure.
Figure 12. Eyring plot for furan (30 equiv relative to 2) C−H
activation by Cp*Fe(CO)(NCMe)Ph (2) (R2 = 0.99; −22 to 3 °C).
Under these conditions, kobs = k1 (see Scheme 7), which gives the rate
of NCMe dissociation.
and ΔS‡ = 0(2) cal/mol·K. Scheme 6 shows that the lowest
energy pathway for NCMe dissociation results in 3t.
Comparison of the enthalpy of 3t with the measured ΔH‡
shows that the predicted value is ∼10 kcal/mol too low. We
examined the possibility that this discrepancy is due to the M06
density functional. However, all other functionals tested predict
lower enthalpy values for 3t. We note that k1 values have been
determined using an indirect method, and despite the
discrepancy in experimental and calculated values, the
comparison of overall energetics between theory and experi-
ment are a good fit, and the predicted rate limiting C−H
activation is in accord with the observed kinetic isotope effects.
We have also explored the kinetics of thiophene activation by
2. Using 20 equiv of thiophene at 3 °C the kobs is 3.2(4) × 10−4
s−1, which is approximately 2.5 times slower than the reaction
with furan. Because no intermediates were observed through
the course of the reaction (1H NMR spectroscopy), we
speculate that the decrease in rate from the reaction with furan
might be explained from the smaller energy stabilization gained
from the formation of the Fe−thienyl from the breaking C−H
bond versus the analogous transformation with furan.
The regioselective activation of furan13,17,38 prompted us to
investigate the underlying reasons for this observed selectivity.
Calculations show a 3.8 kcal/mol ΔΔH‡ value between the 13-
TS and the regioisomeric transition state at the 3-position of
furan. Again, the regioselectivity can be rationalized on the basis
of the relative stability gained by formation of an Fe−C2(furyl)
bond versus an Fe−C3(furyl) bond in the transition state. The
Fe−C2(furyl) intermediate 14s generated from C−H bond
cleavage has a ΔH value of 12.7 kcal/mol, while the Fe−
C3(furyl) intermediate generated from C−H bond activation
has a ΔG value of 17.1 kcal/mol. Quantitative analysis of so-
called “transition state bond energies”37 showed that in 13-TS
the Fe−C2(furyl) bond energy is 8 kcal/mol more stable than
the Fe−C3(furyl) bond energy in the alternative regioisomeric
C−H activation transition state. Eisenstein, Perutz, and Jones
Cp*Fe(CO)2Ph42 (1). A mixture of Cp*Fe(CO)2I (0.533 g, 1.43
mmol), CuOTf (0.480 g, 1.91 mmol), Bu3SnPh (0.610 mL, 1.87
mmol) and 1,4-dioxane (∼6 mL) was prepared. The mixture was
stirred at 60 °C for 4 h during which time the mixture turned from
dark brown to orange-beige. After cooling to room temperature, the
mixture was filtered through a short plug of silica gel on a fine porosity
frit followed by the in vacuo removal of the volatiles from the filtrate.
The resulting residue was dissolved in a minimal amount of THF and
chromatographed on silica gel eluting with 1:10 (v/v) diethyl ether/
hexanes. A yellow band was collected and dried in vacuo. The resulting
solid was triturated with a minimal amount of pentane to yield a
yellow solid (0.311 g, 67%). A crystal suitable for single crystal X-ray
diffraction was grown by the slow evaporation of a pentane solution of
1
3
1. H NMR (300 MHz, C6D6): 7.58 (2H, d, JHH = 6 Hz, phenyl
3
3
ortho), 7.16 (2H, t, JHH = 6 Hz, phenyl meta), 7.07 (1H, t, JHH = 6
1803
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
Hz, phenyl para), 1.32 (15H, s, C5Me5). 13C NMR (75 MHz, C6D6):
218.3 (CO), 128.6, 143.4, 123.2 (Ph), 96.2 (C5Me5), 9.3 (C5Me5)
(Note: 1 resonance of phenyl is missing presumably from coincidental
overlap). IR (NCMe solution): νCO = 1994, 1937 cm−1. Anal. Calcd
for C18H20O2Fe: C 66.69, H 6.22; found C 66.66, H 6.39.
C5Me5). 13C NMR (201 MHz, acetone-d6): 223.1 (s, CO), 205.1 (s,
thiazolyl ipso), 159.3 (s, thiazolyl/thiazole), 149.9 (s, thiazolyl/
thiazole), 144.3 (s, thiazolyl/thiazole), 120.9 (s, thiazolyl/thiazole),
120.7 (s, thiazolyl/thiazole), 92.1 (s, C5Me5), 9.3 (C5Me5). IR (THF
solution): νCO = 1910 cm−1. Anal. Calcd for C17H17N2S2OFe: C 52.58,
H 5.19, N 7.21; found C 52.16, H 5.06, N 7.04.
Cp*Fe(CO)(NCMe)Ph (2). A solution of 1 (0.536 g, 1.67 mmol) in
acetonitrile (∼50 mL) was irradiated in an ice bath with stirring for a
total of 3 h. After the first and second hour, photolysis was ceased and
2 conventional freeze−pump−thaw cycles were performed on the
reaction flask. After 3 h, the volatiles were removed in vacuo. The
resulting residue was extracted with diethyl ether (∼50 mL) and
filtered through Celite. Removal of volatiles produced a red solid,
which was washed with pentane (∼10 mL in portions) to yield a red-
orange solid (0.450 g, 87% yield). This compound is moderately stable
at room temperature in the solid state, but was typically stored at −35
°C. A crystal suitable for single crystal X-ray diffraction was grown by
the slow evaporation of a pentane solution of 2. In order to obtain
satisfactory elemental analysis, 2 was recrystallized from diethyl ether
Cp*Fe(CO)(PPh3)[2-(5-methylfuryl)] (10). To a THF (∼4 mL)
solution of 2 (0.041 g, 0.12 mmol) was added 2-methylfuran (0.11 mL,
1.2 mmol). After stirring the red solution for 1 h, PPh3 was dissolved
in THF (0.032 g, 0.12 mmol) and added to the reaction mixture. The
volatiles were removed in vacuo. The subsequent solid was washed
with pentane (∼3 mL) and dried in vacuo to obtain a red solid of 5
(0.056 g, 82% yield). 1H NMR (300 MHz, acetone-d6): 7.36 (15H, br
m, PPh3), 5.52 (2H, overlapping, methylfuryl 3 and 4), 2.12 (3H, s,
methyl), 1.43 (15H, s, C5Me5). 13C NMR (201 MHz, acetone-d6)
2
2
224.8 (d, JCP = 27 Hz, CO), 173.5 (d, JCP = 32 Hz, thienyl ipso),
156.9 (s, PPh3), 135.0 (br s, PPh3), 130.3 (s, PPh3), 128.5 (s,
methylfuryl), 122.6 (s, methylfuryl), 108.3 (s, methylfuryl), 94.0 (s,
C5Me5), 14.2 (s, methylfuryl), 10.1 (s, C5Me5). Note: The ipso carbon
for PPh3 could not be located likely due to coincidental overlap.
31P{1H} NMR (121 MHz, acetone-d6): 77.7. IR (THF solution): νCO
= 1913 cm−1. Anal. Calcd for C31H37O2PFe: C 72.60, H 6.27; found C
72.61, H 6.26.
1
3
at −35 °C. H NMR (300 MHz, dioxane-d8): 7.34 (2H, d, JHH = 7
Hz, phenyl ortho), 6.82 (2H, t, 3JHH = 7 Hz, phenyl meta), 6.72 (1H, t,
3JHH = 7 Hz, phenyl para), 2.52 (3H, s, NCCH3), 1.46 (15H, s,
C5Me5). 13C NMR (125 MHz, THF-d8): 223.9 (CO), 172.6, 143.0,
128.9, 126.5 (Ph), 121.4 (CH3CN), 91.4 (C5Me5), 9.6 (C5Me5), 3.5
(CH3CN). IR (NCMe solution): νCO = 1903 cm−1. Anal. Calcd for
C19H23NOFe: C 67.67, H 6.87, N 4.15; found C 67.52, H 6.85, N
4.07.
Reaction of Cp*Fe(CO)(NCMe)Ph and C6D6. In a screw-cap
NMR tube, 2 (0.005 g, 0.02 mmol) and hexamethyldisilane (HMDS,
internal standard, ∼1 μL) were dissolved in C6D6 (0.25 mL). The
NMR tube was heated to 50 °C in a temperature-controlled oil bath.
Cp*Fe(CO)(PPh3)(2-furyl) (7). To a THF solution (∼5 mL) of 2
(0.067 g, 0.20 mmol) was added furan (0.29 mL, 4.0 mmol). After
stirring at room temperature for 1 h, PPh3 (0.054 g, 0.21 mmol)
dissolved in ∼2 mL of THF was added. After stirring for an additional
1 h, the volatiles were removed under reduced pressure to leave a red
residue. After transferring the residue to a vial with pentane (∼3 mL)
and subsequent removal of the volatiles under reduced pressure, a low
density beige solid of 3 was obtained (0.105 g, 96% yield). A single
crystal suitable for X-ray diffraction was grown from a saturated
pentane solution of 3. 1H NMR (300 MHz, acetone-d6): 7.62 (1H, m,
furyl 3), 7.35 (15H, br, PPh3), 5.99 (1H, m, furyl 5), 5.64 (1H, m, furyl
4), 1.42(15H, s, C5Me5). 13C NMR (201 MHz, acetone-d6): 224.6 (d,
2JCP = 28 Hz, CO), 177.3 (d, 2JCP = 40 Hz, furyl ipso), 148.1 (s, PPh3),
134.6 (br, PPh3), 130.3 (br s, PPh3),128.6 (s, furyl), 121.7 (s, furyl),
112.1 (s, furyl), 94.0 (s, C5Me5), 9.9 (s, C5Me5). Note: The ipso carbon
for PPh3 could not be located and may be obscured by the broad peaks
for the remaining PPh3 signals. 31P{1H} NMR (121 MHz, acetone-d6):
77.8. IR (C6H6 solution): νCO = 1913 cm−1. Anal. Calcd for
C30H34O2PFe: C 72.27, H 6.06; found C 72.42, H 6.19.
1
The reaction was periodically monitored by H NMR spectroscopy
until completion using a delay time of 5 s. During that time, the phenyl
resonances decreased in intensity relative to HMDS. Using the
integration of the Cp* methyl peaks versus the integration of HMDS,
an approximate yield of 80% was determined for the formation of 2-d5.
Determination of the Rate of Benzene C−D Activation. A
stock solution of 2 (0.024 g, 0.071 mmol) and HMDS (∼3 μL) was
prepared in 1.5 mL of C6D6. Three 0.4 mL aliquots of this stock
solution were added to three different screw-cap NMR tubes. The
samples were frozen until it was time to collect data. Each sample was
1
subsequently monitored by H NMR spectroscopy in a temperature-
regulated probe (calibrated at 49 °C) through 2 half-lives, collecting
spectra every 5 min (5 s delay). The reaction was monitored through
only 2 half-lives due to decomposition at longer reaction times. A plot
of [2] vs time was created and fitted to an exponential decay curve.
The rate constants were extracted from these plots to yield kobs
=
4.6(5) × 10−4 s−1 (see Supporting Information).
Dependence on Furan Concentration of the C−H Activation
of Furan by Cp*Fe(CO)(NCMe)Ph (2). A representative experiment
follows. Two stock solutions were prepared in 2 separate 1 mL
volumetric flasks. In the first stock solution, complex 2 (0.027 g, 0.080
mmol) was dissolved in 1 mL of THF-d8. In the second stock solution,
furan (128 μL) and HMDS (8 μL) were added and diluted to 1 mL
with THF-d8. From the first stock solution, 275 μL (0.022 mmol of 2)
aliquots were transferred to 3 separate screw-cap NMR tubes equipped
with Teflon-lined septa. The second stock solution was transferred to a
1 dram vial with a Teflon-lined septum cap. Outside the glovebox, one
NMR tube was cooled in an ice−water bath. Using a microsyringe, a
125 μL (10 equiv of furan) aliquot of the second stock solution was
injected through the cap of the NMR tube. The tube was vigorously
shaken and placed in a temperature calibrated NMR probe (3 °C). 1H
NMR spectra (5 s delay, every 2.5 min) were acquired through at least
3 half-lives. By monitoring the disappearance of the ortho-phenyl
protons of 2 versus HMDS, a plot of [2] vs time was created (see
Supporting Information pp. S2−S5). Fitting the data to an exponential
decay curve allowed the rate constant to be extracted. This was
repeated for the two remaining NMR tubes. The whole procedure was
performed for 7, 10, 15, 20, 25, 30, 35 equiv of furan.
Cp*Fe(CO)(PPh3)(2-thienyl) (8). Thiophene (0.31 mL, 3.9
mmol) was added to a THF solution (∼7 mL) of 2 (0.065 g, 0.19
mmol). After stirring at room temperature for 1 h, PPh3 (0.051 g, 0.19
mmol) dissolved in ∼2 mL of THF was added. After stirring for an
additional 30 min, the volatiles were removed under reduced pressure
to leave a light brown residue. After transferring the residue to a vial
with a small amount of pentane and diethyl ether and subsequent
removal of the volatiles under reduced pressure, a low-density red solid
of 4 was obtained (0.109 g, 97% yield). 1H NMR (300 MHz, acetone-
d6): 7.39 (15H, br m, PPh3), 7.19 (1H, d, 3JHH = 5 Hz, thienyl 3), 6.77
3
(dd, JHH = 5, 3 Hz, thienyl 5), 6.33 (br s, thienyl 4), 1.43 (C5Me5).
Due to fluxionality, a clean 13C NMR spectrum could not be acquired.
31P{1H} NMR (121 MHz, acetone-d6): 74.0. IR (C6H6 solution): νCO
= 1913 cm−1. Anal. Calcd for C30H34OSPFe: C 70.21, H 5.89; found C
69.92, H 6.05.
Cp*Fe(CO)(N-thiazole)(2-thiazolyl) (9). Thiazole (0.20 mL, 2.9
mmol) was added to a THF (∼5 mL) solution of 2 (0.053 g, 0.16
mmol). The solution immediately became dark red and was stirred for
18 h, after which time the volatiles were removed in vacuo. The
residue was washed with 3 mL of pentane and collected on a fine
1
porosity frit to give a red-brown solid of 6 (0.026 g, 36% yield). H
Dependence on NCMe Concentration for the C−H
Activation of Furan by Cp*Fe(CO)(NCMe)Ph (2). A representative
experiment follows. Two stock solutions were prepared in 2 separate 1
mL volumetric flasks. In the first stock solution, complex 2 (0.027 g,
0.080 mmol) was dissolved in 1 mL of THF-d8. In the second stock
3
NMR (300 MHz, acetone-d6): 9.89 (1H, d, JHH = 3 Hz, thiazole 2),
3
3
8.04 (1H, d, JHH = 3 Hz, thiazole/thiazolyl 4), 7.91 (1H, d, JHH = 3
3
3
Hz, thiazole/thiazolyl 4) 7.65 (1H, dd, JHH = 3 Hz, JHH = 3 Hz,
thiazole 5) 7.08 (1H, d, JHH = 3 Hz, thiazolyl 5), 1.42 (15H, s,
3
1804
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
solution, furan (256 μL), NCMe (18.3 μL, 5% v:v in THF-d8), and
HMDS (8 μL) were diluted to 1 mL with THF-d8. From the first stock
solution, 275 μL (0.022 mmol of 2) aliquots were transferred to 3
separate screw-cap NMR tubes equipped with Teflon-lined septa. The
second stock solution was transferred to a 1-dram vial with a Teflon-
lined septum cap. Outside the glovebox, one NMR tube was cooled in
an ice−water bath. Using a microsyringe, a 125 μL (20 equiv of furan,
0.1 equiv of NCMe) aliquot of the second stock solution was injected
through the cap of the NMR tube. The tube was vigorously shaken
with a Teflon-lined screw cap. Following the acquisition of an initial
1H NMR spectrum, thiophene (0.8 μL, 0.010 mmol) was added. A 1H
NMR spectrum of the resulting solution was acquired and revealed the
presence of free thiophene and identical chemical shifts and peak
shapes of the resonances assigned to 4 as observed in the initial
spectrum.
Experimental Evidence against the Formation of a “Tuck-In”
Complex during the Reaction of Cp*Fe(CO)(NCMe)Ph (2) and
C6D6. While monitoring the reaction of 2 and C6D6, the total
integration for the Cp* peaks relative to HMDS remained constant
(within deviation), suggesting there is no H/D exchange into the
methyl resonances of the Cp* ligand during the course of the reaction.
The total deviation for the integrations of the Cp* region is ∼4%.
Eyring Plot for the C−H Activation of Furan by Cp*Fe(CO)-
(NCMe)Ph (2) Under Saturation Conditions. Two stock solutions
were prepared in 2 separate 1 mL volumetric flasks. In the first stock
solution, complex 2 (0.027 g, 0.080 mmol) was dissolved in 1 mL of
THF-d8. In the second stock solution, furan (384 μL) and HMDS (8
μL) were added and diluted to 1 mL with THF-d8. From the first stock
solution, 275 μL (0.022 mmol of 2) aliquots were transferred to 2
separate screw-cap NMR tubes equipped with Teflon-lined septa. The
second stock solution was transferred to a 1 dram vial with a Teflon-
lined septum cap. Outside the glovebox, one NMR tube was cooled in
an ice−water bath. Using a microsyringe, a 125 μL (30 equiv of furan)
aliquot of the second stock solution was injected through the cap of
the NMR tube. The tube was vigorously shaken and placed in a
temperature calibrated NMR probe (−22 or −12 °C). 1H NMR
spectra (5 s delay, every 15 or 25 min) were acquired through at least
2.5 half-lives. By monitoring the disappearance of the ortho-phenyl
protons of 2 versus HMDS, a plot of [2] vs time was created (see
Supporting Information p. S10). Fitting the data to an exponential
1
and placed in a temperature calibrated NMR probe (3 °C). H NMR
spectra (5 s delay, every 5 min) were acquired through at least 3 half-
lives. By monitoring the disappearance of the ortho-phenyl protons of
2 versus HMDS, a plot of [2] vs time was created (see Supporting
Information pp. S6−S7). Fitting the data to an exponential decay
curve allowed the rate constant to be extracted. This was repeated for
the two remaining NMR tubes. The whole procedure was performed
for 0.05, 0.1, and 0.2 equiv of NCMe. A plot of 1/kobs vs [NCMe]
showed an excellent linear correlation.
Determination of Kinetic Isotope Effect for Furan C−H(D)
Activation. A representative experiment follows. In a 4-dram vial with
a magnetic stir bar was prepared a THF solution (1.5 mL) of 2 (0.015
g, 0.045 mmol). To this solution was added a 1:1 (molar) mixture of
furan and furan-d4 (35 μL, ∼10 equiv arene). After ∼20 min, an
aliquot was removed and analyzed by GC/MS. The average mass
spectrum for the peak representing benzene was analyzed. Using the
relative ratios of m/z = 78 and 79 (adjusted for the natural abundance
of 2H), the relative quantities of C6H6 and C6H5D were determined to
give a kinetic isotope effect of 5.43. The reported 5.0(4) is the average
of 3 independent reactions as described above. Additionally, the above
procedure was performed using 50 equiv of arene, which gave a kinetic
isotope effect of 4.8(1). To ensure irreversibility of the reaction,
multiple aliquots were analyzed throughout the course of the reaction
(over ∼2 h) with no significant deviation in the relative amounts of
C6H6 and C6H5D.
decay curve allowed the rate constant to be extracted. Plotting ln(kobs
/
T) vs 1/T gave a very good linear fit and allowed the determination of
the activation parameters.
Computational Details. All stationary points were optimized in
the gas phase using either restricted or unrestricted M06 density
functional theory with the 6-31G(d,p) basis set for all atoms except Fe.
The LANL2DZ basis set and pseudopotential was utilized for Fe
during optimization. Single point energies were further refined using
the M06 functional with the 6-311++G(3df,3dp) basis set for light
atoms and LANL2TZ(f) with an f exponent of 2.462 for Fe.29,30
Solvent energy corrections were calculated using the SMD solvent
model of benzene and furan. Solvation calculations were performed on
the gas-phase optimized structures.31 Optimization, single point, and
solvent calculations were all carried out in Gaussian 09.43 Location of
singlet−triplet intersystem crossing points, also called minimum
energy crossing points (MECPs), was done using the algorithm of
Harvey44,45 in conjunction with Gaussian 09. Although MECPs are not
stationary points, frequency calculations were carried out on these
structures to obtain approximate enthalpy and free energy corrections.
Eyring Plot for the C−H Activation of Furan by Cp*Fe(CO)-
(NCMe)Ph (2). The procedure described above for measuring the
dependence of rate of furan activation on furan concentration was
repeated at −12, −6, and 13 °C using 10 equiv of furan (3 runs each,
see Supporting Information). Plotting ln(kobs/T) vs 1/T gave an
excellent linear fit and allowed the determination of the activation
parameters. The plotted kobs value is the value determined after
dividing the experimentally determined kobs value by [furan].
Determination of the Rate of Thiophene C−H Activation.
Two stock solutions were prepared in 2 separate 1 mL volumetric
flasks. In the first stock solution, complex 2 (0.027 g, 0.080 mmol) was
dissolved in 1 mL of THF-d8. In the second stock solution, thiophene
(281 μL) and HMDS (8 μL) were added and diluted to 1 mL with
THF-d8. From the first stock solution, 275 μL (0.022 mmol 2) aliquots
were transferred to 3 separate screw-cap NMR tubes equipped with
Teflon-lined septa. The second stock solution was transferred to a 1
dram vial with a Teflon-lined septum cap. Outside the glovebox, one
NMR tube was cooled in an ice−water bath. Using a microsyringe, a
125 μL (20 equiv of thiophene) aliquot of the second stock solution
was injected through the cap of the NMR tube. The tube was
vigorously shaken and placed in a temperature calibrated NMR probe
ASSOCIATED CONTENT
* Supporting Information
■
S
Computational details, supporting figures, full reference 43, and
crystallographic data (CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
1
(3 °C). H NMR spectra (5 s delay, every 2.5 min) were acquired
through at least 3 half-lives. By monitoring the disappearance of the
ortho-phenyl protons of 2 versus HMDS, a plot of [2] vs time was
created (see Supporting Information p. S9). Fitting the data to an
exponential decay curve allowed the rate constant to be extracted. This
was repeated for the two remaining NMR tubes.
Variable Temperature NMR for Cp*Fe(CO)(PPh3)(2-thienyl)
(8). An NMR sample of 4 dissolved in THF-d8 was incrementally
cooled in a 600 MHz NMR probe. Upon cooling, the resonance
assigned to the Cp* methyl protons decoalesced into 2 singlets.
Spectra were acquired at the following temperatures (not calibrated):
25 °C, 0 °C, −20 °C, −40 °C, −50 °C, −60 °C, −80 °C.
Monitoring Cp*Fe(CO)(PPh3)(2-thienyl) (8) in the Presence
of Excess Thiophene. Complex 4 (0.011 mg, 0.020 mmol) was
dissolved in THF-d8 (0.4 mL) and transferred to an NMR tube sealed
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: dhe@chem.byu.edu (D.H.E.), t@unt.edu (T.R.C.),
tbg7h@virginia.edu (T.B.G.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge the Office of Basic Energy Science, U.S.
Department of Energy, for financial support (DE-SC0000776
1805
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806

Organometallics
Article
(34) Schroder, D.; Shaik, S.; Schwarz, H. Acc. Chem. Res. 2000, 33,
139−145.
(T.B.G.) and DE-FG02-03ER15387 (T.R.C.)) and the Na-
tional Science Foundation for the purchase of X-ray diffraction
instrumentation at the University of Virginia (CHE-1126602).
D.H.E. thanks Brigham Young University for financial support
and access to the Fulton Supercomputing Lab.
(35) Lam, W. H.; Jia, G.; Lin, Z.; Lau, C. P.; Eisenstein, O. Chem.
Eur. J. 2003, 9, 2775−2782.
(36) Brooks, B. C.; Gunnoe, T. B.; Harman, W. D. Coord. Chem. Rev.
2000, 206−207, 3−61.
(37) Petit, A.; Flygare, J.; Miller, A. T.; Winkel, G.; Ess, D. H. Org.
Lett. 2012, 14, 3680−3683.
REFERENCES
■
(38) Pittard, K. A.; Lee, J. P.; Cundari, T. R.; Gunnoe, T. B.;
Petersen, J. L. Organometallics 2004, 23, 5514−5523.
(39) Balcells, D.; Clot, E.; Eisenstein, O. Chem. Rev. 2010, 110, 749−
823.
(1) Goj, L. A.; Gunnoe, T. B. Curr. Org. Chem. 2005, 9, 671−685.
(2) Goldberg, K. I.; Goldman, A. S. Activation and Functionalization of
C−H Bonds; American Chemical Society: Washington, D.C., 2004;
Vol. 885.
(40) Jones, W. D.; Feher, F. J. Acc. Chem. Res. 1989, 22, 91−100.
(41) Clot, E.; Megret, C.; Eisenstein, O.; Perutz, R. N. J. Am. Chem.
Soc. 2006, 128, 8350−8357.
(3) Kuhl, N.; Hopkinson, M. N.; Wencel-Delord, J.; Glorius, F.
Angew. Chem., Int. Ed. 2012, 51, 10236−10254.
(4) Gunay, A.; Theopold, K. H. Chem. Rev. 2010, 110, 1060−1081.
(5) Wang, D.; Farquhar, E. R.; Stubna, A.; Munck, E.; Que, L. Nat.
Chem. 2009, 1, 145−150.
(42) Yasuda, S.; Yorimitsu, H.; Oshima, K. Organometallics 2010, 29,
273−274.
(43) Frisch, M. J., et al. Gaussian 09, Revision B.01; Gaussian, Inc.:
Wallingford, CT, 2009.
(6) Gamez, P.; Aubel, P. G.; Driessen, W. L.; Reedijk, J. Chem. Soc.
Rev. 2001, 30, 376−385.
(44) Harvey, J. N.; Aschi, M.; Schwarz, H.; Koch, W. Theor. Chem.
Acc. 1998, 99, 95−99.
(45) Harvey, J. N.; Aschi, M. Phys. Chem. Chem. Phys. 1999, 1, 5555−
5563.
(7) Goldsmith, C. R.; Cole, A. P.; Stack, T. D. P. J. Am. Chem. Soc.
2005, 127, 9904−9912.
(8) Gephart, R. T.; Huang, D. L.; Aguila, M. J. B.; Schmidt, G.;
Shahu, A.; Warren, T. H. Angew. Chem., Int. Ed. 2012, 51, 6488−6492.
(9) Mayer, J. M. Acc. Chem. Res. 1998, 31, 441−450.
(10) Wiese, S.; McAfee, J. L.; Pahls, D. R.; McMullin, C. L.; Cundari,
T. R.; Warren, T. H. J. Am. Chem. Soc. 2012, 134, 10114−10121.
(11) Bullock, R. M. Catalysis without Precious Metals; Wiley-VCH
Verlag & Co.: Weinheim, Germany, 2010.
(12) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals, 5th ed.; John Wiley and Sons: New York, 2009.
(13) Ohki, Y.; Hatanaka, T.; Tatsumi, K. J. Am. Chem. Soc. 2008, 130,
17174−17186.
(14) Ittel, S. D.; Tolman, C. A.; English, A. D.; Jesson, J. P. J. Am.
Chem. Soc. 1978, 100, 7577−7585.
(15) Tolman, C. A.; Ittel, S. D.; English, A. D.; Jesson, J. P. J. Am.
Chem. Soc. 1978, 100, 4080−4089.
(16) Baker, M. V.; Field, L. D. J. Am. Chem. Soc. 1986, 108, 7433−
7434.
(17) Hatanaka, T.; Ohki, Y.; Tatsumi, K. Chem.Asian J. 2010, 5,
1657−1666.
(18) Nakao, Y.; Kashihara, N.; Kanyiva, K. S.; Hiyama, T. J. Am.
Chem. Soc. 2008, 130, 16170−16171.
(19) Jones, W. D.; Foster, G. P.; Putinas, J. M. J. Am. Chem. Soc.
1987, 109, 5047−5048.
(20) Lail, M.; Arrowood, B. N.; Gunnoe, T. B. J. Am. Chem. Soc.
2003, 125, 7506−7507.
(21) Foley, N. A.; Gunnoe, T. B.; Cundari, T. R.; Boyle, P. D.;
Petersen, J. L. Angew. Chem., Int. Ed. 2007, 47, 726−730.
(22) Foley, N. A.; Lee, J. P.; Ke, Z.; Gunnoe, T. B.; Cundari, T. R.
Acc. Chem. Res. 2009, 42, 585−597.
(23) Foley, N. A.; Lail, M.; Lee, J. P.; Gunnoe, T. B.; Cundari, T. R.;
Petersen, J. L. J. Am. Chem. Soc. 2007, 129, 6765−6781.
(24) Joslin, E. E.; McMullin, C. L.; Gunnoe, T. B.; Cundari, T. R.;
Sabat, M.; Myers, W. H. Organometallics 2012, 31, 6851−6860.
(25) Morello, G. Ph.D. Thesis. Modeling transition metal chemistry
for catalytic functionalization of molecules. University of North Texas,
2011.
(26) Tellers, D. M.; Skoog, S. J.; Bergman, R. G.; Gunnoe, T. B.;
Harman, W. D. Organometallics 2000, 19, 2428−2432.
(27) Diaz, C.; Cabezas, N.; Mendizabal, F. Bol. Soc. Chil. Quim. 2002,
47, 213−220.
(28) Jacobson, S. E.; Wojcicki, A. J. Am. Chem. Soc. 1973, 95, 6962−
6970.
(29) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157−167.
(30) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−241.
(31) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
2009, 113, 4538−4543.
(32) Poli, R. Chem. Rev. 1996, 96, 2135−2204.
(33) Poli, R.; Harvey, J. N. Chem. Soc. Rev. 2003, 32, 1−8.
1806
dx.doi.org/10.1021/om301219t | Organometallics 2013, 32, 1797−1806