Thermochemical insight into the reduction of CO to CH3OH with [Re(CO)]+ and [Mn(CO)]+ complexes

Article abstract of DOI:10.1021/ja502316e

To gain insight into thermodynamic barriers for reduction of CO into CH₃OH, free energies for reduction of [CpRe(PPh₃)(NO)(CO)] ⁺ into CpRe(PPh₃)(NO)(CH₂OH) have been determined from experimental measurements. Using model complexes, the free energies for the transfer of H⁺, H^-, and e^- have been determined. A pK_a of 10.6 was estimated for [CpRe(PPh ₃)(NO)(CHOH)]⁺ by measuring the pK_a for the analogous [CpRe(PPh₃)(NO)(CMeOH)]⁺. The hydride donor ability (δG°_H^-) of CpRe(PPh₃)(NO) (CH₂OH) was estimated to be 58.0 kcal mol^-1, based on calorimetry measurements of the hydride-transfer reaction between CpRe(PPh ₃)(NO)(CHO) and [CpRe(PPh₃)(NO)(CHOMe)]⁺ to generate the methylated analogue, CpRe(PPh₃)(NO)(CH₂OMe). Cyclic voltammograms recorded on CpRe(PPh₃)(NO)(CMeO), CpRe(PPh ₃)(NO)(CH₂OMe), and [CpRe(PPh₃)(NO)(CHOMe)] ⁺ displayed either a quasireversible oxidation (neutral species) or reduction (cationic species). These potentials were used as estimates for the oxidation of CpRe(PPh₃)(NO)(CHO) or CpRe(PPh₃)(NO)(CH ₂OH) or the reduction of [CpRe(PPh₃)(NO)(CHOH)] ⁺. Combination of the thermodynamic data permits construction of three-dimensional free energy landscapes under varying conditions of pH and PH₂. The free energy for H₂ addition (δG°H ₂) to [CpRe(PPh₃)(NO)(CO)]⁺ (+15 kcal mol ^-1) was identified as the most significant thermodynamic impediment for the reduction of CO. DFT computations on a series of [Cp^XM(L)(NO) (CO)]⁺ (M = Re, Mn) complexes indicate that δG°H ₂ can be varied by 11 kcal mol^-1 through variation of both the ancillary ligands and the metal.

Full text of DOI:10.1021/ja502316e

Article
pubs.acs.org/JACS
Thermochemical Insight into the Reduction of CO to CH3OH with
[Re(CO)]+ and [Mn(CO)]+ Complexes
Eric S. Wiedner* and Aaron M. Appel
Physical Sciences Division, Pacific Northwest National Laboratory, P.O. Box 999, K2-57, Richland, Washington 99352, United States
S
* Supporting Information
ABSTRACT: To gain insight into thermodynamic barriers for
reduction of CO into CH3OH, free energies for reduction of
[CpRe(PPh3)(NO)(CO)]+ into CpRe(PPh3)(NO)(CH2OH) have
been determined from experimental measurements. Using model
complexes, the free energies for the transfer of H+, H−, and e− have
been determined. A pKa of 10.6 was estimated for [CpRe(PPh3)(NO)-
(CHOH)]+ by measuring the pKa for the analogous [CpRe(PPh3)-
+
−
(NO)(CMeOH)] . The hydride donor ability (ΔG°H ) of CpRe-
(PPh3)(NO)(CH2OH) was estimated to be 58.0 kcal mol−1, based on
calorimetry measurements of the hydride-transfer reaction between
CpRe(PPh3)(NO)(CHO) and [CpRe(PPh3)(NO)(CHOMe)]+ to
generate the methylated analogue, CpRe(PPh3)(NO)(CH2OMe). Cyclic voltammograms recorded on CpRe(PPh3)(NO)-
(CMeO), CpRe(PPh3)(NO)(CH2OMe), and [CpRe(PPh3)(NO)(CHOMe)]+ displayed either a quasireversible oxidation
(neutral species) or reduction (cationic species). These potentials were used as estimates for the oxidation of
CpRe(PPh3)(NO)(CHO) or CpRe(PPh3)(NO)(CH2OH) or the reduction of [CpRe(PPh3)(NO)(CHOH)]+. Combination
of the thermodynamic data permits construction of three-dimensional free energy landscapes under varying conditions of pH and
PH . The free energy for H2 addition (ΔG°H ) to [CpRe(PPh3)(NO)(CO)]+ (+15 kcal mol−1) was identified as the most
2
2
significant thermodynamic impediment for the reduction of CO. DFT computations on a series of [CpXM(L)(NO)(CO)]+ (M
= Re, Mn) complexes indicate that ΔG°H can be varied by 11 kcal mol−1 through variation of both the ancillary ligands and the
2
metal.
(CHO, step 1), hydroxycarbene (CHOH, step 2), and
hydroxymethyl (CH2OH, step 3) intermediates, as illustrated
in Scheme 1.20,21 Hydroxymethyl complexes have been shown
INTRODUCTION
■
The multiproton and multielectron reduction of CO2 has been
studied widely as a route for production of liquid fuels from
renewable energy sources such as sunlight.1−7 Many molecular
electrocatalysts for reduction of CO2 produce either CO or
formic acid in a 2e−/2H+ process instead of more reduced
products such as methanol.7−9 Industrial technologies currently
exist for the conversion of syngas (CO/H2) into methanol,
liquid hydrocarbons, or methane.10−12 However, industrial
syngas conversion is performed at elevated temperatures and
pressures on a large scale at centralized facilities, which may not
be economically viable for energy storage from disperse
renewable energy sources such as solar energy.7,13,14 In
contrast, a tandem electrocatalytic system composed of separate
electrocatalysts for reduction of CO2 and CO could lead to
decentralized production of renewable methanol or other liquid
fuels.9 Therefore, developing an understanding of how to
design electrocatalysts for the reduction of CO will be of
interest.
Scheme 1. Proposed Cycle for Production of CH3OH from
[M(CO)]+
to react with further H+/H− equivalents to produce CH3OH
22,23
(step 4) or CH4
or to form new C−C bonds via CO
insertion.24−26 This reduction sequence could also be
envisioned as an electrochemical cycle in which the H−
equivalents are generated in a 2e−/1H+ process at a second
Many mechanistic studies have been conducted on the
reduction of homogeneous transition-metal carbonyl complexes
as models for catalytic processes.15−19 A reduction pathway that
is frequently considered in homogeneous systems is the
sequential transfer of hydride (H−) and proton (H+)
equivalents to a coordinated CO ligand to afford formyl
Received: March 6, 2014
Published: May 22, 2014
© 2014 American Chemical Society
8661
dx.doi.org/10.1021/ja502316e | J. Am. Chem. Soc. 2014, 136, 8661−8668

Journal of the American Chemical Society
Article
metal center such as [M(diphosphine)2]2+ (M = Ni, Pd, Pt).27
Previous studies have demonstrated intermolecular H− transfer
from [HM(diphosphine)2]+ (M = Ni, Pt) species to transition-
metal carbonyl complexes,27−30 thus providing a foundation for
an electrocatalytic CO reduction cycle.
acetyl complex, CpRe(PPh3)(NO)(CMeO),38 is not capable of
H− transfer. Therefore, the pKa of [CpRe(PPh3)(NO)-
(CMeOH)]+ in CD3CN (eq 1, R = Me) was determined
against o-toluidine (pKa = 10.50 for the conjugate acid)39 in
both the forward and reverse reactions. The 1H NMR spectrum
of these reactions displayed an averaged set of resonances for
CpRe(PPh3)(NO)(CMeO) and [CpRe(PPh3)(NO)-
(CMeOH)]+ due to fast chemical exchange, and an averaged
set of resonances was also observed for o-toluidine and o-
toluidinium (Figure S1). The ratios of all species were
determined from the weighted averages of the chemical shifts,
and an equilibrium constant of 0.8 0.3 was determined for
the reaction. This equilibrium constant, along with the pKa of o-
toluidinium in acetonitrile, can be used to calculate a pKa of
10.6 0.4 for [CpRe(PPh3)(NO)(CMeOH)]+.
Successful development of an electrocatalytic process for CO
reduction will be facilitated by a system in which the free
energy requirements of each step are carefully balanced.
Thermodynamic information on reduction of [M(CO)]+
complexes has been reported for several metal platforms.
Wayland has measured equilibrium constants for insertion of
CO into Rh−H bonds to afford [Rh(porphyrin)(CHO)]
complexes.31−35 DuBois measured hydride donor abilities
−
(ΔG°H ) and C−H bond dissociation free energies (BDFE)
for a series of CpXRe(L)(NO)(CHO) complexes.28,36 More
−
recently, Bercaw reported ΔG°H and pKa values for several
Measurements of the Hydride Donor Ability. The
hydride donor ability of CpRe(PPh3)(NO)(CH2OH) (eq 2, R
[Re(PBH)(CO)4(CHO)]+ complexes, where PBH is a
phosphine ligand containing a protonated pendant base.30
However, these studies have focused on the initial reduction
step to generate the formyl ligand and not the subsequent steps
required for further reduction of the formyl ligand with H+/H−
equivalents.
Significant challenges exist for performing experimental
measurements on the thermodynamics of [M(CO)]+ reduc-
tion. One challenge is that many M(CHO) and M(CH2OH)
species are thermally unstable at ambient temperature,17,26 thus
limiting the range of complexes for which useful measurements
can be made. A second challenge is that M(CHO) and
M(CH2OH) complexes react irreversibly with protons (or
other Lewis acids) to afford M(CH3) complexes due to
intermolecular H− transfer and C−O bond cleavage.20,21,37
Herein we report thermodynamic measurements for
reduction of [CpRe(PPh3)(NO)(CO)]+ to CpRe(PPh3)(NO)-
(CH2OH), completing the previous thermodynamic measure-
ments of this system28 to include steps for reduction past
CpRe(PPh3)(NO)(CHO). Key to this study is the use of
model complexes that do not undergo C−H and C−O bond
cleavage reactions that preclude measurement of the desired
thermodynamic properties. The thermochemical data allow us
to construct free-energy landscapes for reduction of [CpRe-
(PPh3)(NO)(CO)]+ and identify high-energy intermediates
that hinder development of an electrocatalytic system for
reduction of CO. Finally, DFT computations on a series of
[CpXM(L)(NO)(CO)]+ (M = Re, Mn) complexes are utilized
to investigate the effects of structural modification on
improving the energy matching of intermediates to facilitate
CO reduction.
= H) is the thermodynamic half-reaction that corresponds to
step 3 of the [M(CO)]+ reduction cycle in Scheme 1. A
previous report by Gladysz et al. indicated that the
hydroxymethyl complex CpRe(PPh3)(NO)(CH2OH) was not
isolable, unlike the analogous methylated complex CpRe-
(PPh3)(NO)(CH2OMe).20 In an attempt to measure an
equilibrium hydride transfer, a solution of CpRe(PPh3)(NO)-
(CH2OMe) in CD3CN was treated with [Cp*Re(NO)(CO)2]+
(Cp* = C5Me5), which has a hydride acceptor ability
28
−1
−
(−ΔG°H ) of −52.6 kcal mol . However, this reaction was
1
observed by H NMR spectroscopy to produce equal amounts
of the known complexes [CpRe(PPh3)(NO)(CHOMe)]+,
CpRe(PPh3)(NO)(CH3), and Cp*Re(NO)(CO)(COOMe)
(Scheme S1 and Figures S2−S3).20,40 These reaction products
are consistent with transfer of MeO− from CpRe(PPh3)(NO)-
(CH2OMe) to [Cp*Re(NO)(CO)2]+, followed by a rapid and
irreversible H− transfer from CpRe(PPh3)(NO)(CH2OMe) to
transient [CpRe(PPh3)(NO)(CH2)]+.20 No resonances could
be observed for Cp*Re(NO)(CO)(CHO), which would be the
product of H− transfer from CpRe(PPh3)(NO)(CH2OMe) to
[Cp*Re(NO)(CO)2]+. These results suggested that it is not
−
feasible to obtain ΔG°H for CpRe(PPh3)(NO)(CH2OMe) by
measuring an equilibrium H− transfer to an acceptor due to
preferential MeO− loss.
Hydride transfer from CpRe(PPh3)(NO)(CHO) to [CpRe-
(PPh3)(NO)(CHOMe)]+ (eq 3) was previously reported to be
RESULTS
■
pKa Measurement. The pKa of [CpRe(PPh3)(NO)-
(CHOH)]+ (eq 1, R = H) is the thermodynamic half-reaction
that corresponds to step 2 of the [M(CO)]+ reduction cycle in
Scheme 1. Hydride transfer from CpRe(PPh3)(NO)(CHO) to
[CpRe(PPh3)(NO)(CHOH)]+ is known to be rapid and
irreversible,20 which prevents direct measurement of this pKa
value due to decomposition into [CpRe(PPh3)(NO)(CO)]+
and CpRe(PPh3)(NO)(CH3). In contrast, the closely related
rapid and quantitative at −70 °C in CD2Cl2 solution,20 and we
found the same to be true when the reaction was performed in
CD3CN at 25 °C. Therefore calorimetric studies were
performed to measure the enthalpy of this reaction. Using an
isothermal titration calorimeter, 16 aliquots of an acetonitrile
8662
dx.doi.org/10.1021/ja502316e | J. Am. Chem. Soc. 2014, 136, 8661−8668

Journal of the American Chemical Society
Article
solution of [CpRe(PPh3)(NO)(CHOMe)]+ were injected into
an acetonitrile solution of CpRe(PPh3)(NO)(CHO) such that
<10% of the formyl complex was consumed at the end of the
titration. A reaction enthalpy (ΔH°rxn) of −11.5
2.6 kcal
mol−1 was determined from six independent measurements.
The size and charge of the reactants and products of eq 3 are
similar, so ΔS°rxn is expected to be negligible, and therefore
ΔH°rxn is expected to be approximately equal to ΔG°rxn, within
−
the experimental error. On this basis, ΔG°H of CpRe(PPh3)-
(NO)(CH2OMe) was calculated to be 58.0 kcal mol−1 by using
ΔG°rxn for eq 3 (−11.5 kcal mol−1) and the previously
−
determined ΔG°H of CpRe(PPh3)(NO)(CHO) (46.5 kcal
mol−1).28
As a control experiment to cross-check the accuracy of the
above calorimetric measurement, a reaction enthalpy of −7.0
kcal mol−1 was measured by calorimetry for H− transfer from
CpRe(PPh3)(NO)(CHO) to [CpRe(NO)(CO)2]+, shown in
eq 4. The entropy change of this reaction is expected to be
larger than that for eq 3, since the size and charge of the
reactants and products in eq 4 are dissimilar. Hydride-transfer
reactions have been proposed to be similar to electron-transfer
reactions,36 so the entropy change associated with eq 4 can be
estimated using a semiempirical equation developed by Weaver
for electron-transfer reactions.41 By using estimated radii of 3.0
and 4.5 Å for [CpRe(NO)(CO)2]+ and [CpRe(PPh3)(NO)-
(CO)]+, an entropy change of +2.3 cal mol−1 K−1 can be
calculated for eq 4, which contributes −0.7 kcal mol−1 to the
reaction free energy. The combined ΔH°rxn and ΔS°rxn values
afford ΔG°rxn = −7.7 kcal mol−1 for eq 4, which is in good
agreement with the free energy change of −8.5 kcal mol−1
Figure 1. Cyclic voltammograms showing the Re(II/I) couple of
CpRe(PPh3)(NO)(CMeO) (A) and CpRe(PPh3)(NO)(CH2OMe)
(B), and the Re(I/0) couple of [CpRe(PPh3)(NO)(CHOMe)]+ (C).
Conditions: 1 mM analyte, 0.2 M NBu4PF6 in MeCN solution, υ = 10
V s−1(A,B) or 50 V s−1 (C), 1 mm diameter glassy carbon electrode.
−
calculated from the experimentally measured ΔG°H values of
CpRe(PPh3)(NO)(CHO) and CpRe(NO)(CO)(CHO).28
Electrochemistry. A cyclic voltammogram showing
oxidation of CpRe(PPh3)(NO)(CMeO) in acetonitrile solution
at a scan rate (υ) of 10 V s−1 is presented in Figure 1a and is
consistent with a one-electron oxidation from Re(I) to a
chemically unstable Re(II) species. Partial chemical reversibility
is observed for this oxidation as measured by the ipc/ipa value of
0.26, where ipc/ipa is the ratio of the cathodic peak current (ipc)
to the anodic peak current (ipa). An E1/2 value of +0.20 V vs
Cp2Fe+/0 (the reference couple for all potentials in this paper)
was measured for the oxidation of CpRe(PPh3)(NO)(CMeO)
at υ = 10 V s−1 with a peak-to-peak separation (ΔEp) of 100
mV. The irreversible chemical reaction is not expected to affect
the measured peak potentials under conditions in which ipc/ipa
can be measured,42,43 thus the measured E1/2 is expected to be a
reasonable estimate for E° of the Re(II/I) couple. A similar
oxidation wave was observed for CpRe(PPh3)(NO)(CH2OMe)
(Figure 1b) with ipc/ipa = 0.51 and E1/2 = +0.03 V (ΔEp = 84
mV) at υ = 10 V s−1. We assign an error of 0.03 V to these
E1/2 values, which reflects variability in the measured peak
potentials and possible differences in resistance losses (iR drop)
at higher scan rates for a wave with ipc/ipa < 1. The oxidation
wave for CpRe(PPh3)(NO)(CHO) was completely irreversible
at υ = 10 V s−1 (Figure S4), though the measured peak
potential of +0.17 V is similar to the E1/2 of +0.20 V that was
measured for CpRe(PPh3)(NO)(CMeO).
A quasireversible reduction wave with ipa/ipc ≈ 0.20 was
observed at E1/2 = −1.93 V (ΔEp = 144 mV) in cyclic
voltammograms of [CpRe(PPh3)(NO)(CHOMe)]+ recorded
at υ = 50 V s−1 (Figure 1c). This reduction wave is assigned to
the Re(I/0) couple for simplicity, though it is likely that the
unpaired electron of the reduced product is delocalized onto
the NO ligand. The Re(I/0) couple of [CpRe(PPh3)(NO)-
(CHOMe)]+ is 590 mV more negative than the Re(I/0) of
[CpRe(PPh3)(NO)(CO)]+ (E1/2 = −1.34 V),28 consistent with
the CO ligand being more electron withdrawing than the
CHOMe ligand. A reduction wave was observed for [CpRe-
(PPh3)(NO)(CHOH)]+ at a similar potential but was
completely irreversible, even at scan rates of 50 V s−1 (Figure
S5).
Thermochemical Cycles. By utilizing established thermo-
dynamic cycles,44−50 the experimental measurements described
above can be used to construct the thermochemical diagram in
Scheme 2. This diagram illustrates the energetics in acetonitrile
solution of the CHO, CHOH, and CH2OH ligands in terms of
8663
dx.doi.org/10.1021/ja502316e | J. Am. Chem. Soc. 2014, 136, 8661−8668

Journal of the American Chemical Society
Article
Scheme 2. Experimental Thermochemical Data for [CpRe(PPh3)(NO)(CO)]+ and Related Species in Acetonitrile Solution,
a
−
Showing the Relationships Between E1/2, pKa, BDFE, ΔG°H , and ΔG°H Values
2
a
Values in parentheses were estimated from experimental measurements of model complexes, and values in brackets were determined using the
measured values and thermodynamic cycles.
their Brønsted−Lowry acidities (pKa), homolytic bond
Density functional theory (DFT) was employed to
determine the effect of changing the ancillary ligands or the
identity of the metal on the thermodynamics for reduction of
the CO ligand. Using an isodesmic protocol that has been
demonstrated to have a mean absolute error of 1.4 kcal mol−1
for computation of M−H bond strengths,51 a series of
isodesmic reactions were constructed in which H2 is transferred
from [CpXM(L)(NO)(CHOH)]+ (M = Re, Mn) to a reference
species, [CpRe(PPh3)(NO)(CO)]+, as shown in eq 5. The
dissociation free energies (BDFE), and hydride donor abilities
−
−
(ΔG°H ). In construction of the diagram, the pKa, ΔG°H , and
E1/2 values measured for the model, methyl substituted
complexes are assumed to be equivalent to the values of the
Re(CHO), [Re(CHOH)]+, and Re(CH2OH) complexes, and
the values of the surrogates are listed in parentheses in Scheme
2. When only two sides of a triangle within this scheme were
known, the third side was calculated by completion of a
thermochemical cycle to give the values that were not directly
measured, which are listed in brackets in Scheme 2. Each
thermochemical cycle used to complete this diagram is
tabulated in the Supporting Information with estimates of
experimental uncertainty.
values of ΔG°H for the [CpXM(L)(NO)(CHOH)]+ complexes
2
were then determined from the computed free energy for eq 5
and the experimentally determined ΔG°H value of [CpRe-
2
(PPh3)(NO)(CO)]+. Values of ΔG°H have been reported for
−
In Scheme 2, [CpRe(PPh3)(NO)(CHOH)]+ can be
described as resulting from addition of H2 to the CO ligand
of [CpRe(PPh3)(NO)(CO)]+. The free energy of H2 addition
the analogous series of CpXRe(L)(NO)(CHO) complexes28,36
but not for CpXMn(L)(NO)(CHO). Therefore, a second
−
isodesmic reaction was utilized to compute values of ΔG°H for
(ΔG°H ) to [CpRe(PPh3)(NO)(CO)]+, denoted by a purple
2
CpXMn(L)(NO)(CHO) relative to CpRe(PMe3)(NO)(CHO)
(eq 6). The resulting values of these computations are provided
in Table 1, along with pKa values of [CpXM(L)(NO)-
arrow in Scheme 2, was calculated to be +15.0 0.7 kcal mol−1
−
using a thermochemical cycle composed of ΔG°H of
CpRe(PPh3)(NO)(CHO), the pKa of [CpRe(PPh3)(NO)-
(CHOH)]+, and the free energy for heterolytic cleavage of H2
Table 1. Thermodynamic Properties for
[CpXM(L)(NO)(CO)]+ and Related Species in Acetonitrile
Solution
(Scheme 3). Values of ΔG°H were also calculated for
2
Scheme 3. Thermochemical Cycle for Determination of the
ΔG°H of
Free Energy of H2 Addition (ΔG°H ) to
−
ΔG°H of
pKa of
2
2
b
c
d
[M(CO)]+
M(CHO)
[M(CHOH)]+
[CpRe(PPh3)(NO)(CO)]+
a
CpX/L
Re
Mn
Re
Mn
Re
Mn
Cp/CO
17.5
16.2
13.2
10.4
8.9
55.0
52.6
46.5
44.1
42.1
56.7
51.6
44.7
42.3
39.6
2.6
5.3
4.5
Cp*/CO
Cp/PPh3
Cp/PMe3
Cp*/PMe3
10.3
16.4
18.3
21.8
e
15.0
15.4
13.2
10.6
12.1
15.2
8.8
6.7
a
b
Cp* = C5Me5. Computed by DFT using eq 5 and listed in units of
CpRe(PPh3)(NO)(CO) (+28 3 kcal mol−1), CpRe(PPh3)-
(NO)(CHO) (+4 4 kcal mol−1), and [CpRe(PPh3)(NO)-
(CHO)]+ (−1 6 kcal mol−1) using thermochemical cycles
similar to the one in Scheme 3. These results indicate that
reduction of the CO ligand by H2 is significantly more
challenging than reduction of the CHO ligand.
c
kcal mol−1. Experimental values of Re from refs 28 and 36, Mn values
d
computed by DFT using eq 6, in units of kcal mol−1. Determined
−
from a thermochemical cycle using ΔG°H and ΔG°H (see Scheme 3).
2
e
This value was measured experimentally and was used as the
reference for the DFT isodesmic reactions (and therefore matches
experiment by construction).
8664
dx.doi.org/10.1021/ja502316e | J. Am. Chem. Soc. 2014, 136, 8661−8668

Journal of the American Chemical Society
Article
Figure 2. Free energy landscapes in acetonitrile, showing the relative free energies of [Re(CO)]+/0, [Re(CHO)]+/0, [Re(CHOH)]+/0, and
[Re(CH2OH)]+/0 species under varying conditions of pH and PH . Right: Reduction of [Re(CO)]+ to Re(CH2OH) via a H−/H+/H− pathway.
2
(CHOH)]+ calculated from the thermodynamic relationships
in Scheme 3.
The utility of the free energy landscapes becomes evident
when considering pathways by which [Re(CO)]+ might be
reduced to Re(CH2OH), as the energetically accessible species
can be identified visually. Pathways involving reduction by
either H• or e− lead to the radical species [Re(CHO)]+,
Re(CHOH), or [Re(CH2OH)]+, which are all high-energy
intermediates (>30 kcal mol−1); thus reaction steps that
produce radical intermediates are not likely to be viable for a
catalytic system based upon these complexes. A sequential H−/
H+/H− transfer pathway (Figure 2, right-hand side) also leads
to one or more high-energy intermediates, though these
intermediates are still lower in energy than the radical species
[M(CHOH)]+ + [Re(CO)]+
ref
→ [M(CO)]+ + [Re(CHOH)]+
(5)
ref
Mn(CHO) + [Re(CO)]+
ref
→ [Mn(CO)]+ + Re(CHO)ref
(6)
produced by H• or e− transfer. At PH = 1 atm, the most
DISCUSSION
2
■
difficult step of a H−/H+/H− pathway depends on the solution
pH. For example, formation of Re(CHO) (step 1) is the most
challenging step at low pH (+15.9 kcal mol−1 at pH 10, Figure
2a). At a higher solution pH of 20 (Figure 2b), formation of
Re(CHO) is only slightly uphill (+2.2 kcal mol−1), while
conversion of Re(CHO) to [Re(CHOH)]+ (step 2) is much
more difficult (+12.8 kcal mol−1).
Free Energy Landscapes for Reduction of a CO
Ligand. The thermochemical diagram in Scheme 2 contains
valuable information for prediction of chemical reactivity.
However, reaction pathways under non-standard state con-
ditions can be difficult to envision from Scheme 2, in which the
energetics of individual reaction steps are depicted in a
numerical format under standard state conditions (pH = 0
Taken together, the first H− (step 1) and H+ (step 2)
transfers are equivalent to a net addition of H2 across the CO
ligand. At 1 atm of H2, [Re(CHOH)]+ will always be 15.0 kcal
mol−1 higher in energy than [Re(CO)]+, regardless of the
solution pH. In principal the thermodynamic barrier for H2
addition can be eliminated by increasing the pressure of H2.
and PH = 1 atm). Presentation of the thermochemical
2
information in a three-dimensional free energy landscape
(also called a free energy map) permits more facile visualization
of possible reaction pathways under varying conditions of pH
and PH .47
2
To construct a free energy landscape, the relative free energy
for each species is determined under a specified set of
However, the large ΔG°H value of 15.0 kcal mol−1 for
2
[Re(CO)]+ means that an experimentally insurmountable H2
pressure of 1 × 1011 atm would be required for [Re(CO)]+ and
[Re(CHOH)]+ to be isoenergetic. At an experimentally
attainable H2 pressure of 500 atm, [Re(CHOH)]+ would still
be 11.4 kcal mol−1 higher in free energy than [Re(CO)]+.
conditions (e.g., pH = 10 and PH = 1 atm in acetonitrile).
2
The layout for the base is similar to Scheme 2, where the charge
decreases from left to right and the number of C−H and O−H
bonds increases from back to front. The vertical height of each
bar corresponds to the free energy of the species relative to the
most stable species under the specified conditions. For example,
Selection of a solution pH of 20 at PH = 500 atm would result
2
[Re(CO)]+ is the most stable species at pH = 10 and PH = 1
in Re(CHO) being the lowest-energy species, and protonation
of Re(CHO) to afford [Re(CHOH)]+ would have an energetic
penalty of 12.8 kcal mol−1 (Figure 2c).
2
atm in acetonitrile, and its free energy is set to 0 kcal mol−1
under these conditions (Figure 2a). Each species differs from
the adjacent species by transfer of a H+, H•, H−, or e− either to
or from H+ or H2 in solution. Therefore, the free energies for
interconversion between adjacent species are affected by either
The free energy landscapes also illustrate the necessity of
using model complexes in order to measure thermodynamic
properties related to reduction of [CpRe(PPh3)(NO)(CO)]+.
For example, all of the open-shell species containing C−H and
O−H bonds are high in energy, consistent with the irreversible
electrochemistry due to favorable hydrogen atom transfer to a
second equivalent of the same complex to generate two closed-
shell species that are each 20−30 kcal mol−1 lower in energy.
Similarly, at pH 10 (Figure 2a), [CpRe(PPh3)(NO)(CHOH)]+
the solution pH (H+ transfers) or PH (H• transfers) or by both
2
the solution pH and P (H− and e− transfers).47 An interactive
H2
worksheet for generating free energy landscapes for reduction
of [CpRe(PPh3)(NO)(CO)]+ is included as Supporting
Information.
8665
dx.doi.org/10.1021/ja502316e | J. Am. Chem. Soc. 2014, 136, 8661−8668

Journal of the American Chemical Society
Article
and CpRe(PPh3)(NO)(CHO) are nearly isoenergetic and are
interchangeable by proton transfer, however, hydride-transfer
results in the formation of the equally energetic CpRe(PPh3)-
(NO)(CH2OH) and the much lower in energy species
[CpRe(PPh3)(NO)(CO)]+. The hydride-transfer reaction is
favorable by 15 kcal mol−1, which prevents the direct
determination of the pKa value unless the analogous methylated
complex is used as a surrogate.
which means that [Re(CO)]+ becomes a poorer H− acceptor as
it becomes a better H2 acceptor. However, the decrease in
ΔG°H of CpXRe(L)(NO)(CHO) is accompanied by an
−
increase in the basicity of the formyl ligand, indicated by the
increase in free energy for proton dissociation from [CpXRe-
(L)(NO)(CHOH)]+ (ΔG°H = 1.364 × pKa, Figure 3a). The
+
−
+
trends for ΔG°H and ΔG°H have an opposing effect on
ΔG°H , leading to the modest decrease in ΔG°H across the
2
2
Tuning the Thermodynamic Properties. Since large
series of complexes.
Similar ancillary ligand effects are observed for the
values of ΔG°H for [Re(CO)]+ represent a thermodynamic
2
impediment for catalysis, it is desirable to control the
corresponding Mn complexes. However, ΔG°H of each
2
[CpXMn(L)(NO)(CO)]+ complex is 4−6 kcal mol−1 smaller
magnitude of ΔG°H . As seen in Figure 3a, DFT computations
2
than its Re congener (Figure 3b). Thus, by varying both the
ancillary ligands and metal identity, ΔG°H can be reduced by
2
approximately 11 kcal mol−1 across this series of complexes.
For [Cp*Mn(PMe3)(NO)(CO)]+, the complex with the
lowest ΔG°H value (6.7 kcal mol−1) in this series, [Mn-
2
(CHOH)]+ would be only 3.0 kcal mol−1 higher in energy than
[Mn(CO)]+ at a pressure of H2 of 500 atm. Therefore, the
ΔG°H value of [Cp*Mn(PMe3)(NO)(CO)]+ represents a
2
significant improvement in the thermodynamics for reduction
of the CO ligand compared to the Re analogues.
For the electrocatalytic cycle proposed in Scheme 1,
M(CHO) is formed by intermolecular H− transfer from
[HM(diphosphine)2]+ (M = Ni, Pd, Pt). Therefore, the H−
acceptor ability of [M(CO)]+ must be properly tuned to match
the H− donor ability of [HM(diphosphine)2]+. As described
above, ancillary ligand variations that make H2 addition more
favorable also make the CO ligand a poorer H− acceptor, which
severely hinders the ability to tune the thermodynamics for
catalytic CO reduction. In contrast, the [Mn(CHOH)]+
complexes are less acidic than their [Re(CHOH)]+ counter-
parts by 3−9 kcal mol−1, while analogous Mn(CHO) and
Re(CHO) complexes are comparable hydride donors (within
2.5 kcal mol−1, Figure 3c). Thus, switching from Re to Mn
allows the H2 acceptor ability of the CO ligand to be changed
independently of its H− acceptor ability.
The rhenium and manganese complexes all possess a large
free energy for H2 addition to the CO ligand and thus are
incapable of effectively catalyzing CO reduction. However, the
significant decrease in free energy for H2 addition upon shifting
from Re to Mn suggests that carbonyl complexes of other
metals may not have the same thermodynamic limitations. We
are currently exploring whether additional periodic effects can
be utilized to further improve the thermodynamics of
[M(CO)]+ reduction.
SUMMARY AND CONCLUSIONS
■
An extended thermochemical scheme was developed for
reduction of [CpRe(PPh3)(NO)(CO)]+ to CpRe(PPh3)(NO)-
(CH2OH) by using model complexes that do not undergo
undesirable side reactions. Insights into chemical reactivity were
obtained by converting the thermochemical data into free
energy landscapes, in which the relative free energies of reaction
intermediates can be illustrated under varying conditions of pH
−
+
Figure 3. Plot of (A) ΔG°H , ΔG°H , and ΔG°H for varying
2
[CpXRe(L)(NO)(CO)]+ complexes, (B) ΔG°H for varying [CpXM-
2
(L)(NO)(CO)]+ complexes, and (C) ΔG°H versus ΔG°H for varying
+
−
[CpXM(L)(NO)(CO)]+ complexes.
on a series of [CpXRe(L)(NO)(CO)]+ complexes indicate that
ΔG°H decreases by 4 kcal mol−1 upon progressing from
and PH . For a reduction mechanism involving sequential H−
2
2
[CpRe(NO)(CO)2]+ to [Cp*Re(PMe3)(NO)(CO)]+, i.e., as
the ancillary ligands become stronger σ-donors and weaker π-
and H+ transfers to the CO ligand, [CpRe(PPh3)(NO)-
(CHOH)]+ was found to be a high energy intermediate due
to a large free energy requirement for the net addition of H2 to
[CpRe(PPh3)(NO)(CO)]+. DFT computations predict that
acceptors. The trend of decreasing ΔG°H parallels the decrease
2
−
of ΔG°H values that were measured experimentally for the
analogous series of CpXRe(L)(NO)(CHO) complexes,28,36
ΔG°H can be tuned by ∼11 kcal mol−1 by varying both the
2
8666
dx.doi.org/10.1021/ja502316e | J. Am. Chem. Soc. 2014, 136, 8661−8668

Journal of the American Chemical Society
Article
ancillary ligands and metal in a series of [CpXM(L)(NO)-
(CO)]+ complexes. Notably, switching from Re to Mn led to a
significant improvement in the H2 acceptor ability of the CO
complexes without appreciable detriment to the H− acceptor
ability. Through this comparison of Re and Mn complexes, it is
apparent that the choice of metal can provide substantial
control over the relative energies of catalytic intermediates,
thereby allowing tuning of critical factors such as the free
energy for the addition of H2.
of 1 μL [CpRe(PPh3)(NO)(CMeOH)](OTf) solution were added to
the cell with a 4 min delay between injections. The first data point was
discarded, and the average enthalpy per mol for the subsequent 14
injections was −11.3 0.2 kcal mol−1. This procedure was performed
a total of 6 times using 3 independently prepared sets of stock
solutions, resulting in an average enthalpy of −11.5 2.6 kcal mol−1
(error listed at 2σ).
Electrochemistry. Cyclic voltammograms were recorded in a
glovebox at ambient temperature, 23 2 °C, using a 5 mL conical
glass vial fitted with a polysilicone cap having openings sized to closely
accept each electrode. The working electrode (1 mm PEEK-encased
glassy carbon disc, ALS) was polished using diamond paste (Buehler,
0.25 μm) on a polishing pad wet with purified H2O, then the electrode
was rinsed with neat acetonitrile. A glassy carbon rod (Structure Probe,
Inc.) was used as the counterelectrode, and a silver wire suspended in a
solution of Bu4NPF6 (0.2 M) in acetonitrile and separated from the
analyte solution by a Vycor frit (CH Instruments 112) was used as a
pseudoreference electrode. Ferrocene or cobaltocenium hexafluor-
ophosphate (E1/2 = −1.33 V) was used as an internal standard, and all
potentials are referenced to the ferrocenium/ferrocene couple at 0 V.
Computational Details. All calculations were carried out with the
program Gaussian 09.53 Structures were optimized without symmetry
constraints using the B3P8654,55 functional. The Stuttgart basis set
with effective core potential (ECP)56 was used for the Re and Mn
atoms, whereas the 6-31G* basis set57,58 was used for all other atoms
with one additional p polarization function [ξ(p) = 1.1] for the
hydrogens located on the CHOH and CHO ligands. Each structure
was confirmed by a frequency calculation at the same level of theory to
be a real local minimum on the potential energy surface without
imaginary frequency. For each species, the gas-phase free energy was
calculated at 298 K in the harmonic approximation. Solvation free
energies in acetonitrile were calculated by using the polarizable
continuum model (C-PCM)59,60 using Bondi61 atomic radii. Several
rotational isomers were considered for each complex containing a
CHOH or CHO ligand, and the lowest-energy conformations were
used in the thermodynamic calculations (see Supporting Information
for details). This computational setup has been shown to accurately
predict thermodynamic properties of [HNi(diphosphine)2]2+ com-
plexes with a mean absolute error of <2 kcal mol−1.51
EXPERIMENTAL SECTION
■
Methods and Materials. All manipulations were carried out
under N2 using standard vacuum line, Schlenk, and inert-atmosphere
glovebox techniques. Acetonitrile (Alfa-Aesar, anhydrous, amine-free)
was purified by sparging with N2 and passage through neutral alumina
using an Innovative Technology, Inc., Pure Solv solvent purification
system. Acetonitrile-d3 (Cambridge Isotope Laboratories) was
vacuum-distilled from P2O5. Water was dispensed from a Millipore
Milli-Q purifier and sparged with nitrogen. Tetrabutylammonium
hexafluorophosphate (Fluka, ≥99%) was used as received. o-Toluidine
(Fluka, ≥99.5%) was degassed via three consecutive freeze−pump−
thaw cycles and dried over neutral alumina. o-Toluidinium
tetrafluoroborate was prepared by reaction of o-toluidine with 1.5
equiv of HBF4·Et2O followed by recrystallization from CH3CN/Et2O.
All Re complexes were synthesized according to literature
procedures.20,37,38,40,52
1
Instrumentation. H NMR spectra were recorded on a Varian
1
1
NMR S spectrometer (500 MHz for H) at 25 °C. All H chemical
shifts have been internally calibrated using the monoprotio impurity of
the deuterated solvent. Calorimetry experiments were performed using
a GE MicroCal iTC200. Voltammetry measurements were performed
using a CH Instruments 620D potentiostat equipped with a standard
three-electrode cell.
pKa Determination. [CpRe(PPh3)(NO)(CMeOH)](OTf) (15
mM) was equilibrated with four different concentrations of o-toluidine
(8−23 mM) in CD3CN solution and monitored by 1H NMR
spectroscopy. Averaged resonances were observed due to rapid
exchange between the protonated and deprotonated forms of each
species, and the ratio of each species was determined from the
weighted average of the chemical shifts. The equilibrium was also
measured in the reverse direction using o-toluidinium tetrafluoroborate
(11 mM) and three separate concentrations of CpRe(PPh3)(NO)-
(CMeO) (5−19 mM). The measured equilibrium constants and the
known pKa value of o-toluidinium (10.50)39 were used to calculate a
pKa 10.6 0.4 for [CpRe(PPh3)(NO)(CMeOH)](OTf), where the
listed error is twice the standard deviation of the measurements.
Attempted Equilibrium H− Transfer. CpRe(PPh3)(NO)-
(CH2OMe) (10.9 mg, 0.019 mmol, 1.0 equiv) and [Cp*Re-
(CO)2(NO)](BF4) (7.2 mg, 0.015 mmol, 0.8 equiv) were combined
in 0.7 mL CD3CN, and the reaction conversion to [CpRe(PPh3)-
(NO)(CHOMe)]+ (0.5 equiv), CpRe(PPh3)(NO)(CH3) (0.5 equiv),
and [Cp*Re(CO)(NO)(COOMe)] (0.5 equiv) was monitored
periodically by 1H NMR spectroscopy. A reaction conversion of
80% was observed at 45 min, and a reaction conversion of 100% was
observed at 2 h 15 min. The identity of each species was confirmed by
comparison to the 1H NMR chemical shifts recorded on
independently prepared samples.
ASSOCIATED CONTENT
■
S
* Supporting Information
NMR spectra, cyclic voltammograms, tabulation of thermo-
chemical cycles, description of computational conformational
analysis, energies from DFT calculations, complete refs 7, 13,
and 53, a text file of all computed molecule Cartesian
coordinates in a format for convenient visualization, and an
interactive spreadsheet for construction of free energy land-
scapes for [CpRe(PPh3)(NO)(CO)]+. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
eric.wiedner@pnnl.gov
Notes
The authors declare no competing financial interest.
Calorimetry. Calorimetry measurements were recorded under
inert atmosphere in a glovebox. Acetonitrile stock solutions of
CpRe(PPh3)(NO)(CHO) (6.2 mM) and [CpRe(PPh3)(NO)-
(CHOMe)](OTf) (10.7 mM) were prepared, and the concentrations
ACKNOWLEDGMENTS
■
This work was supported by the US Department of Energy,
Office of Basic Energy Sciences, Division of Chemical Sciences,
Geosciences & Biosciences. Computational resources were
provided at the National Energy Research Scientific Computing
Center (NERSC) at Lawrence Berkeley National Laboratory.
Pacific Northwest National Laboratory (PNNL) is a multipro-
gram national laboratory operated for DOE by Battelle.
1
were determined by H NMR spectroscopy using hexamethyldisilox-
ane as an integration standard. The reaction cell of the calorimeter was
loaded by syringe with 230 μL of the CpRe(PPh3)(NO)(CHO)
solution, the injection syringe was loaded with the [CpRe(PPh3)-
(NO)(CHOMe)](OTf) solution, and the reference cell was loaded
with neat acetonitrile. Stirring was commenced at 1000 rpm, and the
system was equilibrated at 25.0 °C. Upon equilibration, 15 injections
8667
dx.doi.org/10.1021/ja502316e | J. Am. Chem. Soc. 2014, 136, 8661−8668

Journal of the American Chemical Society
Article
(39) Kaljurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.;
Leito, I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019−1028.
(40) Tetrick, S. M.; Tham, F. S.; Cutler, A. R. J. Am. Chem. Soc. 1997,
119, 6193−6194.
(41) Hupp, J. T.; Weaver, M. J. Inorg. Chem. 1984, 23, 3639−3644.
(42) Nicholson, R. S.; Shain, I. Anal. Chem. 1964, 36, 706−723.
(43) Bard, A. J.; Faulkner, L. R. Electrochemical Methods:
Fundamentals and Applications, 2nd ed.; John Wiley & Sons: Hoboken,
NJ, 2001.
(44) Wayner, D. D. M.; Parker, V. D. Acc. Chem. Res. 1993, 26, 287−
294.
(45) Parker, V. D.; Handoo, K. L.; Roness, F.; Tilset, M. J. Am. Chem.
Soc. 1991, 113, 7493−7498.
(46) Tilset, M.; Parker, V. D. J. Am. Chem. Soc. 1989, 111, 6711−
6717.
REFERENCES
■
(1) Costentin, C.; Robert, M.; Saveant, J.-M. Chem. Soc. Rev. 2013,
42, 2423−2436.
(2) Windle, C. D.; Perutz, R. N. Coord. Chem. Rev. 2012, 256, 2562−
2570.
(3) Kondratenko, E. V.; Mul, G.; Baltrusaitis, J.; Larrazabal, G. O.;
Perez-Ramirez, J. Energy Environ. Sci. 2013, 6, 3112−3135.
(4) DuBois, D. L. Inorg. Chem. 2014, 53, 3921−4268.
(5) Schneider, J.; Jia, H. F.; Muckerman, J. T.; Fujita, E. Chem. Soc.
Rev. 2012, 41, 2036−2051.
(6) Oh, Y.; Hu, X. Chem. Soc. Rev. 2013, 42, 2253−2261.
(7) Appel, A. M.; et al. Chem. Rev. 2013, 113, 6621−6658.
́
(8) Saveant, J.-M. Chem. Rev. 2008, 108, 2348−2378.
(9) Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Chem.
Soc. Rev. 2009, 38, 89−99.
(47) Appel, A. M.; Lee, S. J.; Franz, J. A.; DuBois, D. L.; Rakowski
DuBois, M. J. Am. Chem. Soc. 2009, 131, 5224−5232.
(48) Appel, A. M.; Lee, S. J.; Franz, J. A.; DuBois, D. L.; Rakowski
DuBois, M.; Twamley, B. Organometallics 2009, 28, 749−754.
(49) Roberts, J. A. S.; Appel, A. M.; DuBois, D. L.; Bullock, R. M. J.
Am. Chem. Soc. 2011, 133, 14604−14613.
(10) Bartholomew, C. H.; Farrauto, R. J. Fundamentals of Industrial
Catalytic Processes, 2nd ed.; Wiley-VCH: Hoboken, NJ, 2006.
(11) Olah, G. A.; Goeppert, A.; Prakash, G. K. S. Beyond Oil and Gas:
the Methanol Economy, Wiley-VCH: Weinheim, 2006.
(12) Khodakov, A. Y.; Chu, W.; Fongarland, P. Chem. Rev. 2007, 107,
1692−1744.
(50) Ciancanelli, R.; Noll, B. C.; DuBois, D. L.; Rakowski DuBois, M.
J. Am. Chem. Soc. 2002, 124, 2984−2992.
(13) Arakawa, H.; et al. Chem. Rev. 2001, 101, 953−996.
(14) Cook, T. R.; Dogutan, D. K.; Reece, S. Y.; Surendranath, Y.;
Teets, T. S.; Nocera, D. G. Chem. Rev. 2010, 110, 6474−6502.
(15) Cutler, A. R.; Hanna, P. K.; Vites, J. C. Chem. Rev. 1988, 88,
1363−1403.
(51) Chen, S.; Rousseau, R.; Raugei, S.; Dupuis, M.; DuBois, D. L.;
Bullock, R. M. Organometallics 2011, 30, 6108−6118.
́
(52) Agbossou, F.; O’Connor, E. J.; Garner, C. M.; Mendez, N. Q.;
́
Fernandez, J. M.; Patton, A. T.; Ramsden, J. A.; Gladysz, J. A.;
(16) West, N. M.; Miller, A. J. M.; Labinger, J. A.; Bercaw, J. E. Coord.
Chem. Rev. 2011, 255, 881−898.
O’Connor, J. M.; Tajima, T.; Gable, K. P. Cyclopentadienyl Rhenium
Complexes. In Inorganic Syntheses; John Wiley & Sons, Inc.: Hoboken,
NJ, 1992; pp 211−225.
(17) Gladysz, J. A. Adv. Organomet. Chem. 1982, 20, 1−38.
̀
(18) Torrent, M.; Sola, M.; Frenking, G. Chem. Rev. 2000, 100, 439−
(53) Frisch, M. J. et al. Gaussian 09; revision B.01; Gaussian, Inc.:
Wallingford, CT, 2010.
494.
(19) Marko, L. Transition Met. Chem. 1992, 17, 474−480.
(20) Tam, W.; Lin, G. Y.; Wong, W. K.; Kiel, W. A.; Wong, V. K.;
Gladysz, J. A. J. Am. Chem. Soc. 1982, 104, 141−152.
(21) Gibson, D. H.; Owens, K.; Mandal, S. K.; Sattich, W. E.; Franco,
J. O. Organometallics 1991, 10, 1203−1206.
(54) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(55) Perdew, J. P. Phys. Rev. B 1986, 33, 8822−8824.
(56) Andrae, D.; Haussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chem. Acc. 1990, 77, 123−141.
(57) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon,
M. S.; Defrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77, 3654−3665.
(58) Rassolov, V. A.; Ratner, M. A.; Pople, J. A.; Redfern, P. C.;
Curtiss, L. A. J. Comput. Chem. 2001, 22, 976−984.
(59) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995−2001.
(60) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem.
2003, 24, 669−681.
(22) Lapinte, C.; Catheline, D.; Astruc, D. Organometallics 1988, 7,
1683−1691.
(23) Tanaka, K.; Ooyama, D. Coord. Chem. Rev. 2002, 226, 211−218.
(24) Berke, H.; Huttner, G.; Weiler, G.; Zsolnai, L. J. Organomet.
Chem. 1981, 219, 353−362.
(25) Vaughn, G. D.; Gladysz, J. A. Organometallics 1984, 3, 1596−
1598.
(61) Bondi, A. J. Phys. Chem. 1964, 68, 441−451.
(26) Vaughn, G. D.; Gladysz, J. A. J. Am. Chem. Soc. 1986, 108,
1473−1480.
(27) Miedaner, A.; DuBois, D. L.; Curtis, C. J.; Haltiwanger, R. C.
Organometallics 1993, 12, 299−303.
(28) Ellis, W. W.; Miedaner, A.; Curtis, C. J.; Gibson, D. H.; DuBois,
D. L. J. Am. Chem. Soc. 2002, 124, 1926−1932.
(29) Elowe, P. R.; West, N. M.; Labinger, J. A.; Bercaw, J. E.
Organometallics 2009, 28, 6218−6227.
(30) Teets, T. S.; Labinger, J. A.; Bercaw, J. E. Organometallics 2013,
32, 5530−5545.
(31) Imler, G. H.; Zdilla, M. J.; Wayland, B. B. J. Am. Chem. Soc.
2014, 136, 5856−5859.
(32) Fu, X.; Li, S.; Wayland, B. B. Inorg. Chem. 2006, 45, 9884−9889.
(33) Fu, X.; Wayland, B. B. J. Am. Chem. Soc. 2005, 127, 16460−
16467.
(34) Wayland, B. B.; Van Voorhees, S. L.; Wilker, C. Inorg. Chem.
1986, 25, 4039−4042.
(35) Farnos, M. D.; Woods, B. A.; Wayland, B. B. J. Am. Chem. Soc.
1986, 108, 3659−3663.
(36) Ellis, W. W.; Ciancanelli, R.; Miller, S. M.; Raebiger, J. W.;
Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc. 2003, 125,
12230−12236.
(37) Sweet, J. R.; Graham, W. A. G. J. Am. Chem. Soc. 1982, 104,
2811−2815.
(38) Buhro, W. E.; Wong, A.; Merrifield, J. H.; Lin, G. Y.; Constable,
A. C.; Gladysz, J. A. Organometallics 1983, 2, 1852−1859.
8668
dx.doi.org/10.1021/ja502316e | J. Am. Chem. Soc. 2014, 136, 8661−8668