Effects of Tuning Intramolecular Proton Acidity on CO2Reduction by Mn Bipyridyl Species

Article abstract of DOI:10.1021/acs.organomet.0c00230

In order to understand the effect of intramolecular proton acidity on CO2 reduction by Mn bipyridyl species, three fac-Mn(CO)3 bipyridine complexes containing intramolecular phenol groups of varying acidities were synthesized and electrochemical, spectroscopic, and computational studies were performed. While the phenol group acidity has minimal influence on the metal center, the complex containing a fluoro-substituted (more acidic) phenol, MnBr(F-HOPh-bpy)(CO)3, exhibits a decreased catalytic to peak current ratio following the second reduction in comparison to the complexes with unsubstituted or methyl-substituted phenol groups (MnBr(HOPh-bpy)(CO)3 and MnBr(Me-HOPh-bpy)(CO)3, respectively). A second process is also present in the catalytic wave for MnBr(F-HOPh-bpy)(CO)3. Furthermore, MnBr(F-HOPh-bpy)(CO)3 exhibits decreased CO production and increased H2 production in comparison to MnBr(HOPh-bpy)(CO)3. Spectroelectrochemistry under an inert atmosphere in the presence of water shows that following the first reduction, for both MnBr(F-HOPh-bpy)(CO)3 and MnBr(HOPh-bpy)(CO)3, the major product is a phenoxide-coordinated fac-(CO)3 species formed from reductive deprotonation and the minor product is a six-coordinate Mn(I) hydride. For both species, the major species following the second reduction is a five-coordinate anion believed to be the active catalyst for CO2 reduction, but the Mn(I) hydride persists as a minor species. The IR assignments are supported by theoretical calculations. These findings show that changes to the acidity of an intramolecular substituent can have significant effects on the catalytic performance and product selectivity of Mn(CO)3 bipyridine catalysts despite having minimal effect on the metal center, with a more acidic intramolecular substituent increasing H2 production at the expense of CO2 reduction.

Full text of DOI:10.1021/acs.organomet.0c00230

pubs.acs.org/Organometallics
Article
Effects of Tuning Intramolecular Proton Acidity on CO2 Reduction by
Mn Bipyridyl Species
Sheri Lense,* Kyle A. Grice, Kara Gillette, Lucienna M. Wolf, Grace Robertson, Dylan McKeon,
Cesar Saucedo, Patrick J. Carroll, and Michael Gau
Cite This: https://dx.doi.org/10.1021/acs.organomet.0c00230
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: In order to understand the effect of intramolecular proton
acidity on CO2 reduction by Mn bipyridyl species, three fac-Mn(CO)3
bipyridine complexes containing intramolecular phenol groups of varying
acidities were synthesized and electrochemical, spectroscopic, and
computational studies were performed. While the phenol group acidity
has minimal influence on the metal center, the complex containing a
fluoro-substituted (more acidic) phenol, MnBr(F-HOPh-bpy)(CO)3,
exhibits a decreased catalytic to peak current ratio following the second
reduction in comparison to the complexes with unsubstituted or methyl-
substituted phenol groups (MnBr(HOPh-bpy)(CO)3 and MnBr(Me-HOPh-bpy)(CO)3, respectively). A second process is also
present in the catalytic wave for MnBr(F-HOPh-bpy)(CO)3. Furthermore, MnBr(F-HOPh-bpy)(CO)3 exhibits decreased CO
production and increased H2 production in comparison to MnBr(HOPh-bpy)(CO)3. Spectroelectrochemistry under an inert
atmosphere in the presence of water shows that following the first reduction, for both MnBr(F-HOPh-bpy)(CO)3 and
MnBr(HOPh-bpy)(CO)3, the major product is a phenoxide-coordinated fac-(CO)3 species formed from reductive deprotonation
and the minor product is a six-coordinate Mn(I) hydride. For both species, the major species following the second reduction is a
five-coordinate anion believed to be the active catalyst for CO2 reduction, but the Mn(I) hydride persists as a minor species. The IR
assignments are supported by theoretical calculations. These findings show that changes to the acidity of an intramolecular
substituent can have significant effects on the catalytic performance and product selectivity of Mn(CO)3 bipyridine catalysts despite
having minimal effect on the metal center, with a more acidic intramolecular substituent increasing H2 production at the expense of
CO2 reduction.
by MnBr(bipyridyl ligand)(CO)3-type catalysts.8 Furthermore,
mechanistic studies have shown that the rate-limiting step in
INTRODUCTION
■
Carbon dioxide (CO2) is the primary anthropogenic green-
the reduction of CO2 to CO by MnBr(2,2′-bipyridine)(CO)3
house gas in terms of total global radiative forcing and is also a
is proton-coupled C−O bond cleavage,9 indicating that the
addition of an intramolecular proton donor to the catalyst may
lead to reduced reaction barriers and higher catalytic turnover
frequencies. Several catalysts incorporating intramolecular
donors have been synthesized and have shown promising
contributor to ocean acidification.1 One promising way to
recycle CO2 is to convert it to carbon monoxide (CO), a
chemical commodity, via the addition of two electrons and two
protons, according to the reaction shown in eq 1.
CO2 + 2H+ + 2e− → CO + H2O
Franco et al. reported that MnBr-
catalytic properties.10−13
(1)
(dhbpy)(CO)3 (dhbpy = 4-phenyl-6-(1,3-dihydroxybenzen-2-
yl)-2,2′-bipyridine), in which the 2,2′-bipyridyl ligand contains
a diol substituent that situates two hydroxyl groups near the
metal center, can act as a CO2 reduction catalyst even in the
absence of intermolecular Brønsted acids.11 Further studies by
this group showed that this catalyst is more reactive to CO2
CO can then be combined with other molecules to produce
fuels and other value-added chemicals. Due to the environ-
mental and economic benefits of being able to carry out this
reaction effectively and inexpensively, there is a vast body of
research devoted to the catalysis of this reaction.2−5 In
particular, much research has focused on the class of catalysts
with the generic formula fac-MnBr(bipyridyl ligand)(CO)3,6,7
since the discovery that MnBr(2,2′-bipyridine)(CO)3 can act
as an electrocatalyst for the reduction of CO2 to CO in the
presence of weak Brønsted acids such as water.6 Equation 1
involves the transfer of both electrons and protons to CO2, and
addition of intermolecular Brønsted acids to the catalytic
solution has been found to be necessary for catalytic turnover
Received: April 6, 2020
https://dx.doi.org/10.1021/acs.organomet.0c00230
© XXXX American Chemical Society
Organometallics XXXX, XXX, XXX−XXX
A

Organometallics
pubs.acs.org/Organometallics
Article
than a catalyst in which the 2,2′-bipyridyl ligand contains a
triol substituent with the hydroxyl groups positioned away
from the metal center.12 Agarwal et al. reported that the
catalyst MnBr(HOPh-bpy)(CO)3 (Figure 1), in which the
RESULTS AND DISCUSSION
Synthesis and IR Spectroscopy. Complexes 1−3 were
prepared using the general synthetic scheme shown in Scheme
■
1. Briefly, the ligands were synthesized via a Suzuki coupling
Scheme 1. General Synthetic Scheme for the Synthesis of 1
(X = F, R = H), 2 (X = H, R = H), and 3 (X = H, R = CH3)
Figure 1. Molecular structures of the complexes in this study:
MnBr(F-HOPh-bpy)(CO)3 (1), MnBr(HOPh-bpy)(CO)3 (2), and
MnBr(Me-HOPh-bpy)(CO)3 (3).
2,2′-bipyridyl ligand contains a phenol substitituent, has more
favorable rate constants in comparison to either MnBr(2,2′-
bipyridine)(CO)3 or a similar catalyst in which the phenol
substituent is replaced by an anisole group, as determined by
the catalytic to peak current ratios (ic/ip) for each of the three
complexes at the catalytic reduction potential.10
In the present work, we studied the effect of varying
intramolecular proton donor strength on the properties and
performance of MnBr(bipyridyl ligand)(CO)3-type CO2
reduction catalysts. Previously, the effect of the intermolecular
Brønsted acid strength on catalytic performance has been
evaluated: Smieja et al. found that the ic/ip ratio and turnover
frequency for catalytic reduction of CO2 to CO by MnBr(bpy-
tBu)(CO)3 (bpy-tBu = 4,4′-di-tert-butyl-2,2′-bipyridine) were
higher in the presence of trifluoroethanol than in the presence
of methanol or water, using the optimal concentration of each
Brønsted acid, without compromising the Faradaic efficiency
or selectivity for CO production.8 However, computational
studies on the active catalysts [Re(2,2′-bipyridine)(CO)3]−
and [Mn(2,2′-bipyridine)(CO)3]− suggest that, if a strong
enough intermolecular Brønsted acid is used (e.g., HCl instead
of H2O), the barrier height for metal protonation can be
lowered enough to favor formation of the metal hydride and
subsequent H2 production over CO2 reduction.9 Thus, a
stronger intramolecular acid could both further lower the
energy barrier for the proton-coupled C−O bond cleavage
step, resulting in faster reduction of CO2 to CO, and lower the
energy barrier for protonation of the Mn center, resulting in
more H2 production at the expense of CO2 reduction. In order
to investigate the effect of intramolecular proton donor
strength on the catalytic reduction of CO2 to CO by
MnBr(bipyridyl ligand)(CO)3-type catalysts, we synthesized
three catalysts, which are shown in Figure 1: MnBr(F-HOPh-
bpy)(CO)3 (1), which contains an electronegative fluoro
substituent meta to the hydroxyl group of the phenol ring,
MnBr(HOPh-bpy)(CO)3 (2), previously synthesized and
investigated by Agarwal et al.,10 and MnBr(Me-HOPh-
bpy)(CO)3 (3), which contains an electron-donating methyl
substituent para to the hydroxyl group of the phenol ring. The
effects of the fluoro and methyl substituents on the proton
donor strength of phenol can be seen in the reported pKa
values of the corresponding phenols: 3-fluorophenol has a
reported pKa of 9.29, phenol 9.99, and p-cresol 10.26.14 The
same trend is seen in dimethyl sulfoxide (DMSO): 3-
fluorophenol, phenol, and p-cresol have pKa values of 15.8,
18.0, and 18.9, respectively.15
reaction between the appropriate pinacol boronic ester and 6-
bromo-2,2′-bipyridine. Each ligand was metalated by refluxing
equimolar amounts of the ligand and MnBr(CO)5 in methanol
for 2 h, following the procedure reported by Agarwal et al. for
the synthesis of complex 2.10 IR spectra of all complexes were
recorded in KBr matrices and showed three strong stretching
bands in the region 1800−2100 cm−1, as expected for carbonyl
ligands coordinated in a facial geometry. These stretches are
given in Table 1. These signals are similar between the three
Table 1. CO Stretching Frequencies for 1−3 Obtained as
Solids in KBr
complex
ν(CO) (cm−1
)
1
2
3
2025, 1942, 1921
2024, 1936, 1921
2024, 1936, 1914
complexes, though the middle carbonyl stretching frequency is
slightly higher for compound 1 and the lowest carbonyl
stretching frequency is slightly lower for 3, indicating slightly
less electron density at the metal center in the former and
slightly more electron density at the metal center in the latter.
These differences are small, given the data spacing of 1.928
cm−1, but are consistent within 1 cm−1 across multiple
spectra collected for each complex.
X-ray Crystallography. X-ray-quality crystals of complexes
1 and 3 were grown by slow diffusion of pentane into THF
solutions of the complexes. In the crystal of 1 that was
analyzed, both atropisomers of the complex had cocrystallized
(Figure 2). In the major component (70%), the phenol
hydroxyl group is pointed toward the axial bromide ligand. In
the minor component (30%), the phenol hydroxyl group is
pointed toward the axial carbonyl ligand and forms a hydrogen
bond with the carbonyl ligand; the distance between the
phenolic hydroxyl group hydrogen, H1*, and the oxygen of the
carbonyl group, O2, is 2.138 Å, and the O1*−H1*−O2 bond
B
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
Figure 2. Crystal structure of 1 (left, major component, and right,
minor component) with ellipsoids set at 50% probability.
angle (where O1* is the phenolic hydroxyl group oxygen) is
158.6°. The crystal of 3 that was analyzed contained only a
single atropisomer (Figure 3). The p-methyl phenol group is
1
Figure 4. H and 19F NMR spectra of 1 recorded in CD3CN.
(Figure 6) by analysis of the chemical shift and comparison to
similar complexes;6 each of the four signals would be expected
to be split into a doublet due to three-bond coupling to a
proton on the adjacent carbon. In the spectra, the pairs of the
doublets that overlap (a-Br and a′-Br, a-S and a′-S) are of
approximately equal intensity to each other, making each
appear as a pseudotriplet. The two pseudotriplets have very
different integrations. The presence of four species in solution
can be seen more clearly in the 19F{1H} spectrum of 1, which
shows four separate resonances (Figure 4). CD3CN solutions
of complexes 1−3 were treated with stoichiometric quantities
of silver salt (either silver triflate or silver hexafluorophos-
phate) and then filtered to remove the resulting AgBr
precipitate, leaving the acetonitrile-solvento adducts, [Mn-
(CD3CN)(F-HOPh-bpy)(CO)3]+ (1-ACN), [Mn(CD3CN)-
(HOPh-bpy)(CO)3]+ (2-ACN), and [Mn(CD3CN)(Me-
HOPh-bpy)(CO)3]+ (3-ACN), respectively, in solution. In
the NMR spectra of the solutions following treatment with the
silver salts, two sets of complex peaks disappear and two
remain (Figure 5 and Figures S5b and S6b), indicating that the
Figure 3. Crystal structure of 3·THF with ellipsoids set at 50%
probability.
positioned such that the phenol hydroxyl group is pointed
toward the axial carbonyl ligand rather than the bromide ligand
and is hydrogen-bonded to a THF molecule that cocrystallizes
with the complex. Selected bond lengths and angles are
compared in Table 2. Metal−ligand bond distances are similar
Table 2. Selected Bond Lengths and Angles for 1 and 3·THF
1
3·THF
Mn−N (Å)
2.030(3), 2.099(3)
2.0391(14),
2.0870(13)
Mn−Cequatorial (Å)
1.819(4), 1.799(4)
1.8213(17),
1.7916(18)
Mn−Caxial (Å)
C−Oequatorial (Å)
C−Oaxial (Å)
1.799(4)
1.149(5), 1.149(5)
1.143(4)
1.8028(18)
1.150(2)
1.362(2),
1.150(2)
Mn−Br (Å)
2.544(3) (major),
2.485(7) (minor)
2.5271(3)
dihedral angle between phenol and 64.72
2,2′-bipyridyl groups (deg)
72.45
for the two complexes, which is consistent with the lack of
influence of substituents on the phenol ring on the electronics
of the metal center. The phenolic C−O bond would be
expected to be shorter in the phenol ring with the electron-
withdrawing fluoro substituent than in the phenol ring with the
electron-donating methyl substituent, but an analysis of bond
lengths in the structure of 1 is complicated by the disorder
resulting from cocrystallization of the two atropisomers.
Figure 5. 1H and 19F NMR spectra of 1-ACN (top) in CD3CN,
prepared by treatment of complex 1 with AgCF3SO3 in CD3CN
followed by filtration of the AgBr precipitate. The 19F signal from the
−
CF3SO3 anion is outside the window shown.
former are from two atropisomers of the bromo adduct and the
latter are from two atropisomers of the acetonitrile-solvento
adduct, which forms upon solvolysis of the bromo adduct in
acetonitrile (Figure 6). The atropisomers differ with regard to
the position of the phenol hydroxyl group, which can be either
parallel or antiparallel to the axial plane, as seen in the crystal
structure of complex 1.
When water (3% by volume) was added to 1 in CD3CN, a
third set of peaks appeared (Figure S7). However, a second set
of peaks did not appear when water (also 3% by volume) was
1
NMR Spectroscopy. H NMR spectroscopy of complexes
1−3, performed in CD3CN, was similar for all complexes and
showed that each complex exists as a mixture of four species in
acetonitrile, on the basis of the number of peaks and their
relative integrations (Figure 4 and Figures S4−S6). While
analysis of much of the aromatic region is impeded by
overlapping peaks, the four most downfield peaks (two sets of
overlapping doublets labeled a-Br, a′-Br, a-S, and a′-S in Figure
4) can be assigned to four protons ortho to a bipyridyl nitrogen
C
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
Figure 6. Proposed species of 1 in acetonitrile solution (in CD3CN,
this deuterated solvent would replace CH3CN in the solvent adduct).
added to 1-ACN (Figure S8). The chemical shifts of the
phenol hydroxyl proton peaks are sensitive to water
concentration, shifting almost 1 ppm downfield upon the
addition of the water to 1-ACN, showing significant
intermolecular hydrogen bonding between the phenol
substituent and H2O. Similarly, in CD3CN samples of 1 and
2 without added H2O, chemical shifts of the phenol hydroxyl
proton signals generally shifted upfield with increasing
temperature due to a diminished effect of intermolecular
hydrogen bonding on chemical shift. Together, these NMR
studies indicate that these complexes are sensitive to their
solution environment.
Variable-temperature NMR was performed in order to
determine whether the two atropisomers exchanged on an
NMR time scale at elevated temperatures, but neither line
broadening nor coalescence was observed at temperatures up
to 65 °C for 1 or 2 (Figure S9 in the Supporting Information).
Additionally, attempts to determine the pKa of 1-ACN and 2-
ACN in CD3CN were unsuccessful due to the decomposition
and precipitation of the complexes from acetonitrile following
treatment with base (Figures S10−S12 in the Supporting
Information).
Figure 8. CV of complex 3 under Ar (ν = 100 mV/s) containing 1
mM complex in acetonitrile with 0.2 M TBAPF6 as the supporting
electrolyte without added water.
Table 3. Peak Reduction Potentials Referenced to Fc/Fc+
a
for 1−3
complex
R1 (V)
R2 (V)
R3 (V)
1
2
3
−1.54
−1.54
−1.55
−1.82, −1.78 (sh)
−1.84 (br)
−1.86 (br)
−2.21
−2.24
not obsd
a
Data were collected in acetonitrile with 0.2 M TBAPF6 as the
supporting electrolyte without added water; ν = 100 mV/s.
Abbreviations: sh, shoulder; br, broadened peak.
Voltammetry. Cyclic voltammetry of 2 in acetonitrile with
0.1 M tetrabutylammonium perchlorate as the supporting
electrolyte and 5% water under both Ar and CO2 was reported
by Agarwal et al.10 Here, CVs of complexes 1−3 were first
recorded in acetonitrile containing 0.2 M tetrabutylammonium
hexafluorophosphate (TBAPF6) under an argon atmosphere
without the addition of water using a glassy-carbon working
electrode (Figures 7 and 8 and Figure S13). The cathodic peak
waves, resulting in either a broadening of the R2 peak in
comparison to R1 or a resolved shoulder at the R2 peak.
Agarwal et al. also reported this finding for complex 2;10
Franco et al. proposed that the two overlapping waves are due
to varying protonation states of the catalyst.12 In this work, the
presence of broadening or a resolved shoulder at R2 is noted in
Table 3. Complexes 1 and 2 also undergo quasi-reversible
reductions at −2.21 and −2.25 V vs Fc/Fc+, respectively,
termed R3. A third reduction event at more negative potentials
has been observed for MnBr(2,2′-bipyridine)(CO)36 and other
Mn(2,2′-diimine ligand)(CO)3Br species.16 Computational
studies by McKinnon et al. suggest it is a further ligand-
based reduction that produces a 19-electron 5-coordinate
dianion.16 The CV of complex 1 was also recorded under the
same conditions using a platinum working electrode and did
not show any significant differences in comparison to the CVs
collected using a glassy-carbon working electrode (Figure
S14).
The effect of the fluoro and methyl phenol substituents on
the reduction potentials is small, with a slightly greater effect
for R2 than for R1. This is consistent with a more metal based
nature for R1 and more ligand based nature for R2,6 with the
effect on R2 here expected to be small, given the dihedral
angles between the phenol and 2,2′-bipyridyl groups. This is in
contrast with the pronounced electronic effects observed for
electron-withdrawing and -donating substituents directly on
the 2,2-bipyridyl ligand.17
Complexes 1−3 each have complex anodic scans, with
several smaller peaks in addition to the oxidative peak
attributed to quasi-reversibility of the R2 process. This
complexity may be due to the presence of several different
species in solution, including multiple protonation states of the
catalysts. An oxidative peak associated with oxidative cleavage
of the Mn0-Mn0 dimer formed during R1 was reported at −0.4
Figure 7. CV of complex 1 under Ar (ν = 100 mV/s) containing 1
mM complex in acetonitrile with 0.2 M TBAPF6 as the supporting
electrolyte without added water.
potentials observed under noncatalytic conditions are
summarized for the three complexes in Table 3. All three
complexes show an irreversible first reduction event, termed
R1. Additionally, all complexes exhibit a second reduction
event, termed R2, that appears to be quasi-reversible, similar to
that observed for MnBr(2,2′-bipyridine)(CO)3.6 However, for
complexes 1−3 the R2 peaks are composed of two overlapping
D
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
V vs SCE for 210 and −0.3 V vs SCE for MnBr(2,2′-
bipyridyl)(CO)3.6 When the values are corrected for the use of
Fc/Fc+ as the reference, an oxidative feature would be expected
at −0.7 to −0.8 V. Interestingly, we observed no significant
peaks in this region for 1 or 3; therefore, if dimer formation
occurs, it is only to a small extent on this time scale. In order to
determine whether this is due to the subsequent R3 process in
1, a CV was recorded with a turnaround after R2, but only a
small, broad oxidative feature was observed at ∼−0.4 V (Figure
S15). Lack of evidence of dimer formation has been observed
for other MnBr(2,2′-bipyridyl)(CO)3-type complexes as well:
Sampson et al. reported that MnBr(mesbpy)(CO)3, in which
mesbpy is the bulky bipyridine ligand 6,6′-dimesityl-2,2′-
bipyridine, does not form a dimer due to steric hindrance.18
However, this complex undergoes a single reversible two-
electron reduction rather than the 2−3 nonreversible reduction
events observed for the complexes reported in this study.
Franco et al. observed two nonreversible reductions as well as a
third quasi-reversible reduction in the CV of MnBr(dhbpy)-
(CO)3 but no oxidative feature associated with dimer cleavage.
On the basis of infrared spectroelectrochemistry (IR-SEC)
data and DFT computations, they conclude that this is due to
reductive deprotonation of the intramolecular resorcinol group
upon the first one-electron reduction, which leads to the
formation of a six-coordinate species with an intramolecular
Mn−O bond as the main R1 product rather than the Mn0-Mn0
dimer.12 Reductive deprotonation at R1 to form an entropi-
cally favored six-coordinate Mn-phenoxide species also
explains the lack of dimer formation and reoxidation for
complexes 1−3 in the present study, and this mechanism was
investigated further for complexes 1 and 2 through IR-SEC
and computational methods (vide infra).
Figure 9. CVs of 1 (1 mM) under Ar (ν = 100 mV/s) with 0% (red
trace), 3% (blue trace), and 5% (green trace) added water (pH of
water adjusted to 3.7 with HClO4) by volume in acetonitrile with 0.2
M TBAPF6 as the supporting electrolyte.
in any of the cases, and so the complex is not operating under
purely kinetic catalytic conditions.
In the presence of CO2 and water, both 1 and 3 show
current enhancements at R2 and R3, as was previously
observed for 2.10 The CVs for all three complexes were
recorded in the presence of CO2 with varying water
concentrations (Figure 10 and Figures S18 and S19). For
Scan rate dependence studies for 1 under argon show that
the peak current of R1 is proportional to the square root of the
scan rate, indicating a diffusion-controlled process but that,
while the R2 and R3 currents increase with increasing scan
rate, they deviate from ipα ν (Figure S16). This may be due
to the closeness in potentials of R2 and R3 to R1 and R2,
respectively, or to the existence of multiple reduction processes
at the R2 and R3 potentials, discussed in more detail below.
To observe the effect of the intermolecular Brønsted acid
strength and concentration on the voltammetry under an inert
atmosphere, CVs were recorded under Ar with the addition of
3% and 5% (v/v) H2O, with the pH adjusted to 3.7 using
HClO4(aq), as well as with the addition of 1% trifluoroethanol
(TFE) instead of H2O (Figure 9 and Figure S17, respectively).
Whereas the R1 peak potential is not affected by increasing
[H2O] or the addition of TFE, the R2 peak potential becomes
progressively more anodic; in comparison to the main peak
potential in the absence of an added Brønsted acid, the R2
potential shifts +0.06 V with 3% H2O, +0.09 V with 5% H2O,
and +0.12 V with 1% TFE. The anodic shift of the R2 peak
potential with increasing concentration and strength of the
added Brønsted acid may indicate proton-coupled electron
transfer (PCET).19 PCET at R2 is further supported by
infrared spectroelectrochemistry (IR-SEC) data and DFT
calculations (vide infra).
Figure 10. CVs of 1 (1 mM) under CO2 with varying water
concentration (ν = 100 mV/s) in acetonitrile with 0.2 M TBAPF6 as
the supporting electrolyte. Only the first segment is shown for clarity.
both 1 and 3, the catalytic current at R2 was greatest at 3%
H2O (v/v) (ic/ip = 10 and 13.2, respectively, with ic taken at
the cusp of the R2 catalytic wave). While Agarwal et al.
reported electrochemical results in the presence of 5% H2O,
we also found that the catalytic current for 2 was greatest in the
presence of 3% H2O, with ic/ip = 13.6 when ic was taken at the
cusp of the catalytic wave. For all three complexes, the catalytic
current at R2 continues to shift to slightly more positive
potentials as [H2O] increases, consistent with PCET at this
reduction. For example, the cusp of the catalytic wave for 2
shifts from −1.92 V with 1% H2O (v/v) to −1.82 V with 5%
H2O.
The reduction current at R3 increases with increasing
concentration and strength of the added Brønsted acid. In
comparison to the current without added acid, the current
increases 1.8-fold with 3% H2O, 2.4-fold with 5% H2O, and
12.8-fold with 1% TFE. These findings suggest catalytic proton
reduction at this potential, though the current is not S-shaped
For all complexes, the current at R3 is also enhanced in the
presence of CO2 and water, though this current lacks an S
shape, and so the complex is not operating under purely kinetic
catalytic conditions at this potential. The current at R3 was
E
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
studied in more detail for 1. While the peak current increases
with increasing water concentration under argon (Figure 9), a
much greater current increase is seen when both CO2 and H2O
are present. In the presence of CO2, the peak current continues
to increase through a water concentration of 10%, unlike the
R2 peak current. Additionally, while the R2 peak current is
independent of scan rate in the presence of CO2 and 5% H2O,
the current at R3 increases with increasing scan rate, again
indicating that the catalyst is not operating under purely kinetic
catalytic conditions (Figure S20). Bulk electrolysis studies,
discussed in greater detail below, show that this current is due
to a combination of CO and H2 production and catalyst
decomposition.
For 1, the current at R2 does not plateau but rather slopes
downward due to overlap with the current at R3. In order to
determine where to measure the current at R2, we took the
first derivative (dI/dV) of the CV under catalytic conditions
(Figure 11). Interestingly, derivatization revealed an inflection
Figure 12. SWVs (amplitude 50 mV, increment 10 mV, sampling
width 0.001 s, and period 0.1 s) for 1 with CO2 and various
concentrations of water. SWVs were collected in acetonitrile with 0.2
M TBAPF6 as the supporting electrolyte.
catalytic process at or near the R2 potential may be due to
either the existence of multiple protonation states of the
catalyst or to the existence of competing processes of CO2
reduction and metal hydride formation and/or H2 production
at this potential. These possibilities were investigated further
through infrared spectroelectrochemistry (IR-SEC) and DFT
calculations (vide infra). Interestingly, under CO2 with the
addition of TFE rather than H2O, the potential of C2′
becomes much more negative, suggesting a different
mechanism or different protonation states of the catalyst are
at play in the presence of a stronger acid (Figure S24).
IR-SEC and DFT Calculations. To better understand the
complex electrochemical behavior described above, IR-SEC
experiments on complexes 1 and 2 were performed in an
Optically Transparent Thin Layer Electrochemical (OTTLE)
cell in acetonitrile containing 0.1 M TBAPF6 and 1% water (v/
v) adjusted to either pH 3.8 (Figure 13) or 9.9 (Figures S25
Figure 11. One segment of the CV and first derivative (dI/dV) for 1
under CO2 with 3% H2O (ν = 100 mV/s) in acetonitrile with 0.2 M
TBAPF6 as the supporting electrolyte.
in the catalytic wave at R2 approximately midway through the
wave, indicating that there are in fact multiple processes
occurring at this potential, which we have termed C2 (more
positive) and C2′ (more negative). (An inflection was not
observed in the R2 catalytic wave for the other two complexes
in this study.) Furthermore, for 1 the shape of the first
derivative curve at R2 changes with varying water concen-
tration (Figure S21). With no water added, there is only one
peak at R2, likely C2. When water is added, a second feature
(C2′) becomes visible. The C2 conductance continually
decreases as more water is added, plateauing at 7% and 10%
H2O, whereas C2′ has the highest conductance at 3% H2O,
though the conductances at 2% and 5% H2O are close. Both
C2 and C2′ shift to more positive potentials with greater water
concentration, consistent with PCET.
To resolve the C2 and C2′ processes, square wave
voltammetry (SWV) was performed on complex 1. Under
catalytic conditions (CO2 + 3% H2O) the wave at R2 resolves
slightly in the SW voltammograms with a period of 0.1 s, with
the current at R2 showing a shoulder approximately −0.1 V
more negative than the main peak (Figure S23). In SWVs
collected at varying [H2O] (Figure 12), the C2 current peaks
at 1% H2O, whereas C2′ peaks at 3−5% H2O. The second
Figure 13. IR-SEC of compound 2 under Ar with acidified water.
and S26). The observed IR-SEC peaks from this study are
summarized in Table 4. Density functional theory (DFT)
calculations were performed to model several possible
compounds, and the scaled calculated ν(CO) values are
given in Table 5. Further DFT details are included in the
Supporting Information. The proposed reduction mechanism
F
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
a
Table 4. Experimental ν(CO) (cm−1) Observed in the IR-SEC of Complexes 1 and 2
complex 1
complex 2
acidic
basic
acidic
basic
resting potential 2045, 2025*, 1960, 1944, 1918* 2045, 2025*, 1960, 1945,
1917*
2045, 2025*, 1959, 1945, 1918*
2045, 2025*, 1959, 1945,
1921*
following R1
2024, 1988*, 1930, 1912, 1882* 2025, 1931, 1913
2025, 1989*, 1928, 1914, 1878*,
1845
2026, 1988*, 1931, 1915,
1878*
following R2
1989*, 1906, 1878*, 1809
(broad)
1990, 1906, 1809
1989*, 1906, 1807 (broad)
1989*, 1899, 1878*, 1809
a
When multiple species are present, frequencies distinguishable from the minor species are noted with an asterisk.
for 1 and 2 is shown in Scheme 2. For notational simplicity, in
Table 5. Calculated Scaled ν(CO) for Potential
Electrochemical Reduction Products of Complex 2
a
the discussion below LOH refers to the HOPh-bpy ligand and
−
LO refers to the deprotonated ligand OPh-bpy.
complex
DFT ν(CO) (cm−1
)
Complexes 1 and 2 show similar behavior in the IR-SEC
under both acidic and basic conditions. In addition, the
wavenumbers for the species formed are nearly identical,
indicating that the fluoro substituent does not significantly
modulate the electron density at the Mn center during
reduction. Two sets of peaks are seen in the IR-SEC for both
complexes at resting potential under both acidic and basic
conditions. This is consistent with the autosolvolysis observed
by NMR spectroscopy. The calculated ν(CO) values for both
the acetonitrile-bound ([Mn(LOH)(CO)3(NCCH3)]+) and
water-bound ([Mn(LOH)(CO)3(H2O)]+) complexes are very
close to the experimental ν(CO) values observed for the major
species. The ν(CO) values for the minor product are very close
[Mn(LOH)(CO)3(NCCH3)]+
[Mn(LOH)(CO)3(H2O)]+
Mn(LO)(CO)3(H2O)
Mn(LOH)(CO)3(OH)
Mn(LO)(CO)3
Mn(LOH)(CO)3(NCCH3)
Mn(LOH)(CO)3H
[Mn(LO)(CO)3H]−
[Mn(LOH)(CO)3]−
2041, 1961, 1954
2044, 1960, 1944
2030, 1944, 1932
2005, 1915, 1893
2018, 1925, 1917
2009, 1926, 1918
1987, 1898, 1890
1982, 1893, 1878
1899, 1820, 1806
a
LOH refers to the ligand with the phenol O−H intact, whereas LO
indicates a deprotonated ligand.
a
Scheme 2. Species Observed in IR-SEC Studies
a
The calculated DFT ν(CO) values are similar for Mn(LOH)(CO)3H and [Mn(LO)(CO)3H]−, and both are consistent with the minor product
observed in the IR-SEC. The conversion Mn(LO)(CO)3 → Mn(LOH)(CO)3H would be accompanied by the addition of two protons and two
electrons; the conversion between Mn(LOH)(CO)3H and [Mn(LOH)(CO)3]− would involve addition or loss of a proton.
G
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
to those observed for 1 and 2 obtained as solids in KBr (Table
1); thus, these peaks are likely from the bromide-bound
complexes (Mn(LOH)(CO)3Br). The calculated ν(CO)
stretches for neither the phenol-deprotonated species (Mn-
(LO)(CO)3(H2O)) nor the aqua ligand deprotonated species
(Mn(LOH)(CO)3(OH)) match as well with the the major or
minor experimental species observed.
Supporting Information). On the basis of these values, it is
likely that the complexes reprotonate at or after R2, which is
consistent with the anodic shifts observed in the R2 peak
potentials and catalytic wave with increasing water concen-
tration or with the addition of trifluoroethanol. It should be
noted that an intermediate reduction product, corresponding
to the C2′ process in CVs of 1, was not observed in IR-SEC,
but this may be due to the difference in electrode material and
cell setup between the cyclic voltammetry and the IR-SEC
cells.
Under acidic conditions for both 1 and 2, the minor product
assigned to the Mn hydride persists through the second
reduction. Thus, protonation of the metal center is detrimental
to the ability of the catalyst to reduce CO2 to CO, since
formation of the Mn hydride competes with formation of
[Mn(LOH)(CO)3]−.7 While analysis of the ic/ip ratios in the
present study is complicated by the multiple processes in the
electrochemistry, the smaller ic/ip ratio of complex 1 in
comparison to those of complexes 2 and 3 may be due to the
more acidic pendant phenol group in complex 1 favoring
intramolecular protonation of the metal center more than in
complexes with less acidic pendant phenol groups. In a
previous study, we found that a complex containing a benzoic
acid substituent rather than a phenol substituent showed no
catalytic activity at R2 in the presence of CO2 and water.20
Benzoic acid is significantly more acidic than phenol, with a
pKa of 11.1 in DMSO,21 and the lack of catalytic activity at R2
may be due to Mn(I) hydride formation outcompeting
formation of the five-coordinate anion.
In all of the IR-SEC experiments, the peaks observed at
resting potential decay following the first reduction and two
new sets of peaks, one major and one minor on the basis of
absorbance values, appear. For both 1 and 2, the major
products following R1 have three CO stretches (2024, 1930,
and 1912 cm−1 for 1 and 2026, 1931, and 1915 cm−1 for 2),
typical patterns for fac-tricarbonyl complexes rather the Mn0-
Mn0 dimer typically formed by MnBr(2,2′-bipyridyl)(CO)3Br-
type complexes following R1.7 Our results are consistent with
IR-SEC data reported by Franco et al. for MnBr(dhbpy)-
(CO)3, in which the major product following R1 contained
three ν(CO) bands.12 By comparison with ν(CO) from DFT
calculations, Franco et al. assigned this species as a reductive
deprotonation product, fac-Mn(dhbpy-H+)(CO)3, in which
the deprotonated phenoxide is the sixth ligand. In the present
study, the major species following R1 also align well with the
calculated ν(CO) values for the reductive deprotonation
product fac-Mn(LO)(CO)3, in which the deprotonated
phenoxide is similarly coordinated to the Mn (Figure S34
and Scheme 2). This assignment was confirmed by performing
IR-SEC of both 1 and 2 in the presence of excess
tetramethylguanidine; at the resting potential, the ν(CO)
peaks observed in the presence of base closely matched the
ν(CO) peaks observed following R1 (Figure S27). A minor
product also appears for both complexes 1 and 2 after the first
one-electron reduction and persists even through the second
reduction. The experimental ν(CO) bands for this product for
both complexes 1 and 2 match well with calculated frequencies
for either the protonated or deprotonated Mn(I) hydride,
(Mn(LOH)(CO)3H) or [Mn(LO)(CO)3H]−. The Mn−H
stretch for this species was calculated at 1572 cm−1, but in the
IR-SEC that region is obscured by solvent. Franco et al.
synthesized and characterized by IR-SEC the Mn(I) hydride
Mn(2,2′-bipyidine)(CO)3H in situ from MnBr(2,2′-
bipyridine)(CO)3 and 0.1 M phenol in CH3CN; ν(CO)
bands were observed at 1991, 1892, and 1888 (sh) cm−1,
which also supports the assignment of the minor species to the
metal hydride in the present study with bands observed at
around 1988 and 1882 cm−1.12
Following R2, for both complexes 1 and 2 the minor species,
assigned as the metal hydride, persists, but the major species is
replaced by a new product with ν(CO) stretches at 1906 cm−1
(1899 cm−1 for 2 under basic conditions) and broad peaks
centered at 1807−1809 cm−1. DFT calculations support the
assignment of the new major product as the five-coordinate
Mn(0) anion [Mn(LOH)(CO)3]−, which could form
following the two-electron reduction of Mn(LO)(CO)3
(Scheme 2). This species is consistent with the catalytically
active species of other fac-Mn(2,2′-bipyridyl)(CO)3Br-type
complexes.7 The pKa values for the [Mn(LOH)(CO)3]−
complexes would be expected to be high, given that these
complexes are anionic and the zerovalent Mn center would
preclude coordination of the deprotonated phenoxide to the
metal center. DFT calculations suggest that the O−H on
[Mn(LOH)(CO)3]− has a pKa of approximately 31 in
acetonitrile, which is less acidic than phenol (see the
Bulk Electrolysis. Bulk electrolysis was performed on
complexes 1 and 2 in the presence of CO2 and 3% H2O. The
results are summarized in Table 6. When electrolysis was run
Table 6. Bulk Electrolysis Data for Compounds 1 and 2
complex 1
complex 2
potential (V)
experiment time (min)
j (mA/cm2)
FE(CO) (%)
FE(H2) (%)
−1.9
−2.2
−1.9
60
−2.2
60
1.9
67
11
60
5.1
49
13
56
2.0
110
5
2.4
103
1
7
7
7
7
7
7
7
7
at the R2 potential (−1.9 V), complex 2 produced CO in
approximately quantitative yield, whereas the Faradaic
efficiency (FE) for 1 was only 67 7% for CO production
and 11 7% for H2 production. No formate production was
observed for either catalyst. Mechanistic studies have shown
that selectivity for CO production over H2 production for both
MnBr(2,2′-bipyidine)(CO)3 and ReCl(2,2′-bipyidine)(CO)3
is due to a ∼10 kcal/mol higher reaction barrier for the
protonation of the catalytic intermediate [M (2,2′-bipyidine)-
(CO)3]− (M = Mn, Re) in comparison to the barrier for CO2
addition to this intermediate.9 For ReCl(2,2′-bipyidine)(CO)3,
calculations suggest that H2 production becomes favorable
over CO reduction if a strong enough Brønsted acid is used.
We believe H2 production is higher and catalyst selectivity
lower for 1 in comparison to 2 due to the greater acidity of the
intramolecular proton donor in the former, which could lower
the reaction barrier for intramolecular protonation of the metal
center. In the CVs run under Ar under 3% and 5% H2O, in
which the pH of the H2O was acidified to 3.7, and under Ar
with the addition of 1% TFE, no current enhancement was
H
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
observed at R2, indicating that catalytic proton reduction does
not occur at R2 under these conditions and suggesting that
either dissolved CO2 plays a role in H2 production, as has been
previously reported for a bis(imino)pyridine Mn electro-
catalyst,22 or that the complex must undergo further reduction
before significant H2 production occurs.
at this potential. Mn hydride peaks were observed in the IR-
SEC of complex 2 as well, but it is possible that for complex 2
the potential of the process involving the Mn hydride is close
enough to the potential for CO2 reduction by the anion that
separate processes cannot be distinguished.
The voltammetry, IR-SEC data, and DFT computations in
this study indicate that the R2 ic/ip ratios for these catalysts
must be viewed cautiously as a benchmark of catalytic turnover
frequency. For all three catalysts, the R2 peaks under
noncatalytic conditions are composed of multiple overlapping
waves, and for complex 1, multiple processes are also evident
in the R2 catalytic wave. Additionally, formation of the Mn
hydride would be expected to decrease the catalytic current;
since it persists through R2, it will form at the expense of
[Mn(LOH)(CO)3]−. However, IR-SEC data for complexes 1
and 2 suggest that the Mn hydride forms following the first
reduction event (R1), in which case both the ic and ip values at
R2 would both be diminished due to Mn hydride formation, so
that Mn hydride formation may not be fully reflected in the ic/
ip ratio at R2. On the other hand, the anion [Mn(LOH)-
(CO)3]− and either Mn(LOH)(CO)3H or [Mn(LO)-
(CO)3H]− may be in dynamic equilibrium following R2, as
these hydridic species can be formed by either proton transfer
to or proton transfer within the anion, respectively. In this case,
Mn hydride formation would be expected to result in a lower
ic/ip ratio. The lower ic/ip value for 1 in comparison to those
for 2 and 3 may be due to a protonation of the metal center
being more favored for 1, leading to greater tendency to form
the Mn(I) hydride. This species may be either noncatalytically
or minimally catalytically active until it undergoes further
reduction, resulting in a smaller ic value at R2 for 1 as well as
more H2 production. Mn hydride formation may also explain
why the ic/ip for all catalysts in this study decreases when more
than 3% H2O is added; while a Brønsted acid is necessary for
the reduction of CO2 to CO, the benefit of additional water
may be counteracted by the tendency of water to facilitate
hydride production. Overall, these results illustrate the
challenge in minimizing the activation barrier for protonation
of catalyst-bound CO2 while avoiding the competing pathway
of metal hydride formation and the need to precisely tune the
acidity of intramolecular substituents in order to optimize CO2
reduction by fac-Mn(CO)3 bipyridine complexes.
Bulk electrolysis was also performed at −2.2 V, the potential
at which R3 occurs. For complex 1, the current started off
higher than that for the −1.9 V run, but when the experiment
was stopped after 56 min, the current was 4 times lower than
that for the −1.9 V run. Dark solids were observed to have
precipitated on the electrode, and the solution turned brown.
Similarly, a brown color change and precipitate formation was
observed for the solution of 2 after 1 h of electrolysis at −2.2
V. These results show the instability of the catalysts at more
reducing potentials, suggesting that careful control of potential
is necessary for the successful operation of these catalysts. CO
production was also higher for 2 and H2 production higher for
1 at this potential, which may again be due to the stronger
intramolecular proton donor in the latter facilitating proton
reduction at the expense of CO2 reduction. No formate
production was observed for either catalyst. The FE of over
110% for compound 2 may be due to the observed
decomposition of the catalyst.
CONCLUSION
■
In order to evaluate the effect of intramolecular acid strength
on CO2 reduction by Mn(CO)3 bipyridine complexes,
complexes 1−3 were synthesized and characterized spectro-
scopically and electrochemically. At resting potential in
acetonitrile, the complexes were all found to exist as a mixture
of four species in solution due to autosolvolysis and the
existence of atropisomers and were shown to be sensitive to
their solution environment. IR-SEC and DFT computations on
complexes 1 and 2 indicate that reductive deprotonation
occurs at the first reduction (R1), giving a phenoxide-
coordinated fac-Mn(CO)3 species, Mn(LO)(CO)3, as the
major R1 product. Additionally, both complexes contain a
minor product following R1 that we assign as the Mn(I)
hydride (either Mn(LOH)(CO)3H or [Mn(LO)(CO)3H]−).
At the second reduction, we propose that the phenoxide-
coordinated fac-Mn(CO)3 species undergoes a two-electron
reduction and reprotonation to form a five-coordinate anion,
[Mn(LOH)(CO)3]−, the active catalyst for CO2 reduction.
The Mn(I) hydride persists through the second reduction
potential.
EXPERIMENTAL SECTION
■
General Procedures. Reagents and solvents were purchased
commercially and used without further purification unless otherwise
specified. Manganese complexes were protected from light as much as
possible: The reaction flasks were covered with aluminum foil during
synthesis, and the complexes were stored in flasks or vials covered
with aluminum foil. The fume hood and glovebox lights were turned
off when handling the complexes and manipulation was performed
behind an aluminum foil shield whenever possible. The balance was
covered with a box during weighing. MnBr(HOPh-bpy)(CO)3 was
synthesized using the previously published procedure.10
The addition of electron-withdrawing (fluoro) or electron-
donating (methyl) substituents to the intramolecular phenol
groups has a minimal effect on the Mn center, as evidenced by
the close reduction potentials of the three complexes and the
similarity in the CO stretching frequencies in the IR spectra.
However, this ancillary change has a measurable effect on the
catalytic activity and product selectivity, with complex 1
producing more H2 and less CO than complex 2.
The electrochemistry of complex 1 under catalytic
conditions is complex, with two processes occurring at the
R2 catalytic wave that responded differently to water
concentration, whereas only one process was observed at the
R2 catalytic wave for complexes 2 and 3. For complex 1, the
two processes at R2 may be due to the existence of multiple
protonation states of the catalyst or to the existence of
competing processes of CO2 reduction and metal hydride
formation and/or protonation of the metal hydride to form H2
Instrumentation. Nuclear magnetic resonance spectroscopy
(NMR) was performed on a Bruker Avance III 400 MHz FT-
NMR. Infrared spectroscopy (IR) was performed on a Thermo
Nicolet Nexus 670 FT-IR or Thermo-Nicolet iS5 FT-IR, both in
transmission mode. Liquid chromatography−mass spectrometry (LC-
MS) was performed with an Agilent 1100 MSD. Gas chromatog-
raphy−mass spectrometer (GC-MS) was performed using a Thermo
Trace DSQ. Elemental analysis was performed by Atlantic Microlabs,
Inc. (Norcross, GA).
Synthesis. 4-Fluoro-2-methoxyphenylboronic Acid Pinacol
Ester. This procedure was repeated several different times on different
I
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
scales. One example is given here: the reagent 4-fluoro-2-
methoxyphenylboronic acid (1.50 g, 8.83 mmol) was combined
with pinacol (1.57 g, 13.2 mmol), and sodium sulfate (Na2SO4) (1.88
g, 13.2 mmol) in THF (15.0 mL) in a flame-dried flask. The reaction
mixture was stirred under a positive pressure of argon for 48 h. The
mixture was then gravity filtered and the solvent removed from the
resulting solution in vacuo. The crude product was redissolved in 15
mL of ethyl acetate. The resulting ethyl acetate solution was washed
with deionized water (6.6 mL × 3) and brine (4.5 mL × 1). The ethyl
acetate layer was dried over magnesium sulfate (MgSO4) and then
gravity-filtered and the solvent removed in vacuo, giving 4-fluoro-2-
methoxyphenylboronic acid pinacol ester as a colorless solid with a
crystalline appearance (yield 45%).
Hz, J = 2.2 Hz), 6.91 (1H, d, J = 8.4 Hz), 3.84 (3H, s), 2.38 (3H, s),
1.58 (H2O), 1.34 (impurity). 13C{1H} NMR (CDCl3, 100.6 MHz): δ
(ppm) 156.63, 155.61, 155.26, 149.02, 136.79, 136.49, 131.85,
130.35, 130.28, 128.89, 125.23, 123.52, 121.34, 118.87, 111.72, 55.88,
24.82 (impurity), 20.61. ES-API MS: calcd m/z for M + 1 277.1,
found m/z 277.1 (100%).
F-HOPh-bpy. F-MeOPh-bpy (0.550 g, 1.96 mmol) was combined
with pyridine hydrochloride (9.07 g, 78.5 mmol) in a Schlenk tube.
The tube was then heated with stirring at 180 °C for 5 h under a
positive pressure of nitrogen. After it was cooled to room temperature,
the reaction mixture was slowly dissolved in an aqueous solution of
saturated sodium bicarbonate (NaHCO3) (69 mL) and then
extracted with ethyl acetate (69 mL × 3). The ethyl acetate layers
were combined, washed with brine (69 mL), and then dried over
sodium sulfate (Na2SO4). After gravity filtration, the solvent was
removed in vacuo. The crude product was purified by column
chromatography on silica gel 60 with diethyl ether as eluent. Fractions
containing the product were collected and combined in a flask and
then placed under vacuum to remove the solvent (yield 78%).
1H NMR (CDCl3, 400 MHz): δ (ppm) 14.94 (1H, s), 8.67 (1H, d,
J = 4.7 Hz), 8.23 (1H, d, J = 8.23 Hz), 8.08 (1H, d, J = 8.0 Hz), 7.91
(1H, t, J = 8.0 Hz), 7.82−7.73 (3H, m), 7.30 ppm (1H, ddd (non first
order), J = 7.5 Hz, J = 5.8 Hz, J = 0.9 Hz), 6.68 (1H, dd, J = 10.4 Hz, J
= 2.6 Hz), 6.59 (1H, ddd (non first order), J = 2.6 Hz, J = 8.1 Hz, J =
8.1 Hz). 13C{1H} NMR (CDCl3, 100.6 MHz): δ (ppm) 164.7 (d, JCF
= 249.8 Hz), 161.8 Hz (d, JCF = 13.0 Hz), 156.7, 154.4, 153.3, 149.7,
138.9, 137.3, 127.9 (d, JCF = 10.9 Hz), 124.2, 120.7, 119.4, 119.0,
115.5 (d, JCF = 3.0 Hz), 106.5 (d, JCF = 22.3 Hz). 105.1 (d, JCF = 23.4
Hz). ES-API MS: calcd m/z for M + 1 267.1, found m/z 267.0
(100%).
1H NMR (CDCl3, 400 MHz): δ (ppm) 7.65 (1H, t, J = 7.9 Hz),
6.63 (1H, ddd (nonfirst order), J = 8.3 Hz, J = 8.3 Hz, J = 2.2 Hz),
6.56 (1H, dd, J = 11.6 Hz, 2.2 Hz), 3.82 (3H, s), 1.34 (12H, s).
13C{1H} NMR (CDCl3, 100.6 MHz): δ (ppm) 166.16 (JCF= 249 Hz),
166.11 Hz (JCF = 9.8 Hz), 138.30 (JCF = 10.4 Hz), 106.94 (JCF = 20.2
Hz), 98.89 (JCF = 24.7 Hz), 83.5 (s), 55.95 (s), 24.84 (s) (C-B not
seen). EI-MS: m/z 253.1 (M + 1), 251.7, 250.7, 238.0, 236.8, 235.7,
220.8.
5-Methyl-2-methoxyphenylboronic Acid Pinacol Ester. The
synthesis and purification of 5-methyl-2-methoxyphenylboronic acid
pinacol ester used the same process described above with 5-methyl-2-
methoxyphenylboronic acid in place of 4-fluoro-2-methoxyphenylbor-
onic acid. This reaction was repeated several times on different scales,
producing a white solid with yields ranging from 53% to 73%.
1H NMR (CDCl3, 400 MHz): δ (ppm) 7.47 (1H, d, J = 2.0 Hz),
7.18 (1H, dd, J = 8.4 Hz, J = 2.1 Hz), 6.76 (1H, d, J = 8.4 Hz), 3.80
(3H, s), 2.27 (3H, s), 1.35 (12H, s). 13C{1H} NMR (CDCl3, 100.6
MHz): δ (ppm) 162.35, 137.10, 132.88, 129.28, 110.80, 83.43, 56.12,
24.84, 20.27 (C-B not seen). EI-MS: m/z 249.1 (M + 1), 247.8,
246.7, 232.8, 216.5.
F-MeOPh-bpy. Prior to the reaction, toluene and ethanol were
degassed by sparging with N2 gas. The reagents 6-bromo-2,2′-
bipyridine (0.366 g, 1.56 mmol), 4-fluoro-2-methoxyphenylboronic
acid pinacol ester (0.471 g, 1.87 mmol), and tetrakis-
(triphenylphosphine)palladium (Pd(PPh3)4) (0.081 g, 0.070 mmol;
handled in a glovebox) were combined in a flask, which was
subsequently placed under vacuum and then refilled with N2(g).
Toluene (19.2 mL), ethanol (2 mL), and a 2.0 M aqueous solution of
potassium carbonate (K2CO3) (1.68 mL) were transferred to the
reaction flask via syringe. The reaction was then stirred and heated to
95 °C for 24 h under N2(g). After the reaction mixture was cooled to
room temperature, aqueous saturated ammonium chloride solution
(19 mL) and DI water (19 mL) were added to the reaction mixture.
The mixture was extracted with dichloromethane (CH2Cl2) (44.5 mL
× 3). The dichloromethane layers were combined and dried over
MgSO4 and then gravity-filtered, and the solvent was removed under
vacuum. The crude product was then purified by column
chromatography using silica gel 60 with diethyl ether as eluent
(yield 89.3%).
Me-HOPh-bpy. The synthesis of Me-HOPh-bpy used the same
procedure as that for F-HOPh-bp, with Me-MeOPh-bpy in place of F-
MeOPh-bpy. The synthesis was performed multiple times with
product yields ranging from 68% to 88%.
1H NMR (CDCl3, 400 MHz): δ (ppm) 14.29 (1H, s), 8.72 (1H,
ddd (nonfirst order), J = 4.8 Hz, J = 1.7 Hz, J = 0.9 Hz), 8.30 (1H, dd,
J = 7.0 Hz, J = 1.7 Hz), 8.18 (1H, ddd, J = 8.0 Hz, J = 0.9 Hz, J = 0.9
Hz), 7.99−7.93 (2H, m), 7.85 (1H, ddd, J = 7.7 Hz, J = 7.7 Hz, J =
1.8 Hz), 7.64 (1H, d, J = 1.6 Hz), 7.35 (1H, ddd, J = 7.5 Hz, J = 4.8
Hz, J = 1.1 Hz), 7.14 (1H, dd, J = 8.3 Hz, J = 1.8 Hz), 6.96 (1H, d, J =
8.3 Hz), 2.35 (3H, s). 13C{1H} NMR (CDCl3, 100.6 MHz): δ (ppm)
157.51, 157.46, 154.72, 153.42, 149.60, 138.70, 137.29, 132.44,
127.92, 126.66, 124.14, 120.78, 119.33, 119.22, 118.59, 118.22, 20.75.
ES-API MS: calcd m/z for M + 1 263.1, found m/z 263.1 (100%).
MnBr(HOPh-bpy)(CO)3. While MnBr(HOPh-bpy)(CO)3 was
synthesized using the previously published procedure,10 previously
NMR spectroscopy was performed in DMSO-d6. In this study, NMR
spectroscopy was performed in CD3CN as well as DMSO-d6. The 1H
NMR spectrum in CD3CN is tabulated below. The peak integrations
of several of the peaks are approximate due to the broad and
overlapping nature of these peaks. The chemical shifts of the phenolic
peaks were extremely sensitive to small amounts of water in solution.
1H NMR (CD3CN, 400 MHz): δ (ppm) 9.25 (2.0H, two
overlapping doublets, J ≈ 6 Hz, Br), 9.17 (0.7H, broad, S), 8.46−
8.35 (5.4H, multiplet with broad shoulder, Br + S), 8.24 (1.4H, broad,
S), 8.15−8.10 (4.1H, multiplet, Br), 7.65 (poorly resolved, 0.3H),
7.71−7.60 (4.8H, multiplet, Br + S), 7.55−7.53 (1.1H, two peaks,
Br), 7.47−7.39 (4.7H, multiplet, Br + S), 7.36−7.30 (1.9H, multiplet,
Br + S), 7.18 (poorly resolved, 0.4H) 7.10−7.02 (6H, multiplet, Br +
S), 2.21 (H2O).
1H NMR (CDCl3, 400 MHz): δ (ppm) 8.61 (1H, d, J = 4.7 Hz),
8.47 (1H, d, J = 8.0 Hz), 8.26 (1H, d, J = 7.5 Hz), 7.96−7.92 (1H,
m), 7.81−7.72 (3H, m), 7.24−7.21 (1H, m), 6.76 (1H, ddd (nonfirst
order), J = 8.2 Hz, J = 8.2 Hz, J = 2.4 Hz), 6.68 (1H, dd, J = 10.9 Hz, J
= 2.3 Hz), 3.81 (3H, s). 13C{1H} NMR (CDCl3, 100.6 MHz): δ
(ppm) 163.98 (d, JCF = 247.8 Hz), 158.47 (d, JCF = 9.7 Hz), 156.48,
155.62, 154.20, 149.05, 136.81, 136.67, 132.58 (d, JCF = 10.0 Hz),
125.18 (d, JCF = 3.3 Hz), 124.88, 123.60, 121.22, 118.91, 107.62 (d,
JCF = 21.1 Hz), 99.51 (d, JCF = 25.7 Hz), 55.85. ESI-MS: calcd m/z
for M + 1 281.1, found m/z 281.1 (100%).
MnBr(F-HOPh-bpy)(CO)3. The synthesis of MnBr(F-HOPh-bpy)-
(CO)3 was adapted from the procedure described by Bocarsly et al.
for the synthesis of MnBr(HOPh-bpy)(CO)3.10 The ligand, F-HOPh-
bpy (0.1000 g, 0.3756 mmol), was transferred to a three-neck flask
containing a reflux condenser and stir bar. The flask was placed under
three alternative cycles of vacuum and argon, ending under argon.
Methanol (31.2 mL, degassed) was transferred to the flask via syringe.
The reaction flask was covered with aluminum foil to keep out light.
Manganese(I) pentacarbonyl bromide (0.1032 g, 0.3756 mmol) was
added to the reaction flask and allowed to dissolve. The solution was
Me-MeOPh-bpy. The synthesis of Me-MeOPh-bpy used the same
procedure as that for F-MeOPh-bpy, with 5-methyl-2-methoxyphe-
nylboronic acid pinacol ester in place of 4-fluoro-2-methoxyphenyl-
boronic acid pinacol ester. The synthesis was performed multiple
times with product yields ranging from 68% to 88%.
1H NMR (CDCl3, 400 MHz): δ (ppm) 8.67 (1H, d, J = 4.7 Hz),
8.55 (1H, d, J = 8.0 Hz), 8.30 (1H, d, J = 7.6 Hz), 7.89−7.76 (4H,
m), 7.29 (1H, ddd, J = 7.6 Hz, 4.8 Hz, 0.9 Hz), 7.18 (1H, dd, J = 8.3
J
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
then stirred, heated to reflux for 2 h, and then cooled to room
temperature.
(AgCF3SO3) or silver hexafluorophosphate (AgPF6) in CD3CN or
CH3CN in a 1:1 mol equiv ratio with the complex was added to the
complex in solution, producing a fine white precipitate. The mixture
was then passed through a syringe filter (Whatmann, 1.6 μm grade
GF/A, Lot 10985) producing a clear, light yellow solution that was
collected in a dry NMR tube, which was then capped with a septum.
The NMR tube was shielded from light using aluminum foil or by
placement in the NMR sample holder or magnet.
A second flask was flame-dried, wrapped in foil, purged, and left
under a positive pressure of nitrogen. A cannula needle with an
attached syringe filter (0.45 μm pore size) was flushed with nitrogen
and left as a vent in the septum of the second flask. After cooling, the
brown reaction solution was drawn into a foil-wrapped syringe and
transferred to the second flask via the syringe filter and cannula
needle. Black particulate was observed on the filter, and the bright
orange filtrate was placed under vacuum to remove the solvent. The
orange solid was washed with degassed diethyl ether (∼5 mL × 3)
and the residual solvent removed under vacuum (yield 49%). Crystals
for single-crystal X-ray diffraction analysis were grown by slow
diffusion of pentane into a THF solution of the complex.
Voltammetry Experiments. Voltammetry experiments were
performed using a Pine Wavedriver 10 potentiostat with AfterMath
Data Organizer software. All electrochemical experiments were
performed in acetonitrile, containing 0.2 M TBAPF6 as the electrolyte.
All acetonitrile used in electrochemistry experiments was HPLC grade
and was dried over activated alumina before use. The TBAPF6 was
purchased commercially and recrystallized from hot 200 proof
ethanol. The electrochemistry experiments were performed in a 15
mL three-neck flask, which was dried in an oven and cooled under
argon prior to use. Each neck was capped with a septum. The auxiliary
electrode, a platinum wire (0.5 mm diameter), was fed through one
septum. Holes were drilled in the other two septa, such that the
working electrode and reference electrode (a nonaqueous Ag/Ag +
reference electrode w/porous Teflon tip filled with 10 mM AgNO3
dissolved in the acetonitrile electrolyte solution) were fitted snuggly.
The 15 mL flask was connected to an acetonitrile bubbler using a
cannula needle, and argon or carbon dioxide gases used to sparge the
electrolyte solution were first bubbled through the acetonitrile to
prevent the solution volume in the electrochemical flask from
decreasing. The electrochemical flask also contained a stir bar so that
the solution could be stirred during degassing. A glassy-carbon
working electrode (disk with 3 mm diameter, purchased from CH
Instruments) was used unless otherwise noted. The Pt electrode used
was a 2 mm diameter disk, also purchased from CH Instruments.
All electrochemical experiments were performed under air-free
conditions, using either argon (MnBr(F-HOPh-bpy)(CO)3 and
MnBr(HOPh-bpy)(CO)3) or nitrogen (MnBr(Me-HOPh-bpy)-
(CO)3) and carbon dioxide as noted. Before the complexes were
added to the electrolyte solution, the electrolyte solution (10.0 mL)
was bubbled with argon for 30 min to ensure a featureless background
over the region to be scanned. (This region was adjusted depending
on the potential of the reference electrode vs Fc/Fc+.) In general, a
potential range of approximately +0.5 to −2.5 V vs Fc/Fc+ was used.
Following the background scan, ferrocene and the complex to be
measured, both at a concentration of 1.0 mM, were added to the
electrolyte solution. After the complex was added, the solution was
bubbled with argon for 15 min. Before data collection, the argon
needle was pulled into the headspace of the flask. Experiments were
performed in triplicate under argon in order to ensure reproducibility.
Between each scan, the working electrode was repolished and the
solution was repurged with either CO2 or Ar for 5−10 min. Water or
acid used in the voltammetry experiments was degassed prior to use.
CO2 was bubbled through the solution for 30 min before
commencing data collection under CO2. The electrochemical cell
was kept covered in aluminum foil when the complex was present.
IR-SEC. Infrared spectroelectrochemistry (IR-SEC) was performed
using an optically transparent thin-layer electrochemical (OTTLE)
cell purchased from Dr. Frantisek Hartl at the University of
Reading.23 The airtight demountable OTTLE cell is comprised of
CaF2 windows and a Teflon spacer (ca. 0.2 mm optical path)
equipped with a platinum-minigrid working electrode, a platinum
minigrid counter electrode, and a silver-wire pseudoreference. An
eDaq ER466 potentiostat was used to control the potential, and IR
spectra were obtained using an ABB FLTA2000 IR spectrometer.
Samples were prepared at 5.0 mM in 0.1 M TBAPF6/acetonitrile with
added 3% (v/v) aqueous solutions (of pH 9.9 and 3.8 generated using
KOH and HBF4, respectively). Solutions were sparged with argon for
15 min prior to analysis. The solutions were protected from light to
prevent unwanted photochemical reactivity during both sample
preparation and experimentation. For IR-SEC data collection, a
potential was applied to the working electrode, and spectra were
collected regularly, with each measurement consisting of the average
IR (KBr, cm−1): 3369 (br) (OH), 2025 (s) (CO), 1942 (s) (CO).
Anal. Found: C, 47.27; H, 2.31; N, 5.74. Calcd: C, 47.04; H, 2.29; N,
5.77.
1H NMR: in a CD3CN solution, this complex exists as a mixture of
MnBr(F-HOPh-bpy)(CO)3 and [Mn(CD3CN)(F-HOPh-bpy)-
(CO)3]+, in which the bromide ligand is replaced by an acetonitrile
ligand. Both the original complex and the solvolysis product exist as
pairs of atropisomers, with the hydroxyphenol group oriented parallel
or antiparallel to the axial plane, so that there are four forms of the
complex in an acetonitrile solution. NMR peaks from MnBr(F-
HOPh-bpy)(CO)3 versus peaks from [Mn(CD3CN)(F-HOPh-bpy)-
(CO)3]+ could be distinguished by comparing the spectra before and
after treatment with a silver salt such as AgPF6 or AgCF3SO3. In the
tabulation below, peaks from [Mn(CD3CN)(F-HOPh-bpy)(CO)3]+
are specified with “S” and peaks from MnBr(F-HOPh-bpy)(CO)3 are
specified with “Br”. From the peaks farthest downfield, which on the
basis of chemical shift can be assigned to the bipyridyl proton ortho to
the nitrogen, the ratio of MnBr(F-HOPh-bpy)(CO)3 atropisomers to
[Mn(CD3CN)(F-HOPh-bpy)(CO)3]+ atropisomers is 1:0.3. The
peak integrations of several of the peaks are approximate due to the
broad and overlapping nature of these peaks. The chemical shifts of
the phenolic peaks were extremely sensitive to small amounts of water
in solution.
1H NMR (CD3CN, 400 MHz): δ (ppm) 9.26 (2.0H, Br), 9.17
(0.6H, S), 8.45 (sh, S) + 8.39 (Br) (5.4H total), 8.25 (1.4H, S), 8.14
(4.3H, Br), 7.82 (1.1H, Br), 7.71 (sh, S) + 7.63 (Br) (5.2H total),
7.54 (1.3H, two peaks, Br), 7.41 (Br) + 7.35 (sh, S) (2.9H total), 7.1
(0.3H, broad, unknown), 7.0 (0.3H, broad, unknown), 6.90−6.83
(4.7H, multiple peaks, S + Br), 2.23 (H2O, chemical shift influenced
by hydrogen bonding with complex).
MnBr(Me-HOPh-bpy)(CO)3. The synthesis of MnBr(Me-HOPh-
bpy)(CO)3 used the same procedure as that for MnBr(F-HOPh-
bpy)(CO)3, with Me-HOPh-bpy in place of F-HOPh-bpy. The
purification differed, as MnBr(Me-HOPh-bpy)(CO)3 was soluble in
diethyl ether and hence could not be washed with this solvent. After
the solvent was removed from the filtered reaction product, the
product was further purified by slow diffusion of pentane into a
solution of the complex in degassed THF. A small amount of the
crystals was set aside for single-crystal X-ray diffraction analysis. The
remaining crystals were collected and washed with degassed pentane,
and residual solvent was removed in vacuo. The complex cocrystal-
lized with THF in a 1:1 ratio (yield 40%).
IR (KBr, cm−1): 3417 (br) (OH), 2024 (s) (CO), 1914 (s) (CO).
Anal. Found: C, 52.36; H, 3.94; N, 5.13. Calcd for MnBr(Me-HOPh-
bpy)(CO)3 + THF: C, 52.10; H, 4.01; N, 5.06. 1H NMR (400 MHz,
CD3CN): δ 9.27−9.24 (1.0H, Br), 9.19−9.16 (0.25H, S), 8.47−8.34
(Br + S, 2.6H), 8.28−8.20 (0.51, S), 8.16−8.08 (2.1H, Br), 7.73−7.59
(2.2H, Br + S), 7.54 + 7.52 (two peaks, 0.46H), 7.43 (0.12H), 7.29−
7.23 (2.3H, Br + S), 7.16 (0.13H, S), 7.12 (0.12H, S), 7.09−7.06
(two peaks, 0.61H, Br), 7.01−6.97 (three peaks, 0.69H), 6.93−6.91
(two peaks, 1.01H), 3.68 (t, 4.6H, THF), 2.36 (s, 0.3H, S), 2.34 (s,
3.4H, Br + S), 2.16 (H2O), 1.83 (t, 4.3H, THF).
General Procedure for Synthesis of Acetonitrile Adducts in
Solution. The preparation was carried out in the dark under an inert
atmosphere. A known mass of the complex was dissolved in
deuterated acetonitrile (CD3CN). A solution of silver triflate
K
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
of four scans. The potential was held constant until the spectra no
longer exhibited changes in activity, at which point the potential was
changed to the next value of interest. TBAPF6 was recrystallized from
HPLC-grade methanol and dried in a vacuum oven prior to use.
HPLC-grade acetonitrile was purified using an Innovative Technology
PureSolv system under N2 prior to use. Argon was purchased from
American Gases Corporation.
AUTHOR INFORMATION
■
Corresponding Author
Sheri Lense − Department of Chemistry, University of Wisconsin
Oshkosh, Oshkosh, Wisconsin 54902, United States;
orcid.org/0000-0002-9970-3675; Email: lenses@
uwosh.edu
Density Functional Theory (DFT) Calculations. DFT calcu-
lations were carried out using Gaussian 1624 and visualized using
GaussView 6.0.25 The methods involved Truhlar’s MN15-L func-
tional26 with the def2-TZVP basis set.27 Acetonitrile solvation was
modeled using the polarized continuum model. Species were
optimized to stationary points, and frequency calculations were
used to ensure that no imaginary frequencies were present. Calculated
energies were converted from hartrees to kcal/mol using the
conversion of 1 hartree = 627.095 kcal/mol. Energies of reactions
(ΔG298 K) were found using G298 K(products) − G298 K(reactants). The
energy of a proton in acetonitrile was set as the published value of
−260.2 kcal/mol.28
X-ray Crystallography. X-ray intensity data were collected on a
Bruker D8QUEST (for complex 129 and for complex 330) CMOS
area detector employing graphite-monochromated Mo Kα radiation
(λ = 0.71073 Å) at a temperature of 100 K. Preliminary indexing was
performed from a series of 24 0.5° rotation frames with exposures of
10 s. A total of 912 frames (1) or 3389 frames (3) was collected with
a crystal to detector distance of 33.0 mm, rotation width of 0.5°, and
exposure of 10 s.
Authors
Kyle A. Grice − Department of Chemistry and Biochemistry,
DePaul University, Chicago, Illinois 60614, United States
Kara Gillette − Department of Chemistry, University of
Wisconsin Oshkosh, Oshkosh, Wisconsin 54902, United States
Lucienna M. Wolf − Department of Chemistry and
Biochemistry, DePaul University, Chicago, Illinois 60614,
United States
Grace Robertson − Department of Chemistry, University of
Wisconsin Oshkosh, Oshkosh, Wisconsin 54902, United States
Dylan McKeon − Department of Chemistry, University of
Wisconsin Oshkosh, Oshkosh, Wisconsin 54902, United States
Cesar Saucedo − Department of Chemistry and Biochemistry,
DePaul University, Chicago, Illinois 60614, United States
Patrick J. Carroll − P. Roy and Diana T. Vagelos Laboratories,
Department of Chemistry, University of Pennsylvania,
Philadelphia, Pennsylvania 19104, United States; orcid.org/
0000-0002-8142-7211
Rotation frames were integrated using SAINT,31 producing a listing
of unaveraged F2 and σ(F2) values. For 1, a total of 34724 reflections
was measured over the ranges 6.368 ≤ 2θ ≤ 55.05°, −16 ≤ h ≤ 16,
−13 ≤ k ≤ 13, and −19 ≤ l ≤ 19, yielding 4313 unique reflections
(Rint = 0.0433). For 3, a total of 37299 reflections was measured over
the ranges 5.978 ≤ 2θ ≤ 55.06°, −20 ≤ h ≤ 20, −11 ≤ k ≤ 13, and
−17 ≤ l ≤ 18, yielding 5430 unique reflections (Rint = 0.0299). For
both complexes, the intensity data were corrected for Lorentz and
polarization effects and for absorption using SADABS32 (for 1,
minimum and maximum transmissions were 0.5314 and 0.7456,
respectively; for 3, minimum and maximum transmissions were
0.6299 and 0.7456, respectively). Both structures were solved by
direct methods using ShelXT.33 Refinement was by full-matrix least
squares based on F2 using SHELXL-201834 (complex 1) or SHELXL-
201735 (complex 3). All reflections were used during refinement. For
Michael Gau − P. Roy and Diana T. Vagelos Laboratories,
Department of Chemistry, University of Pennsylvania,
Philadelphia, Pennsylvania 19104, United States; orcid.org/
0000-0002-4790-6980
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.organomet.0c00230
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
S.L., K.G., and G.R. thank the American Chemical Society
Petroleum Research Fund (Grant No. 54833- UNI3 to Sheri
Lense) for funding. S.L. also thanks Dr. Timothy Paschkewitz
of Pine Research for help in designing the SWV experiments
and Dr. Eric Wiedner of Pacific Northwest National
Laboratory for helpful conversations regarding the cyclic
voltammetry experiments. NMR spectroscopy was enabled
by NSF-MRI Award No. 1625483, “MRI: Acquisition of a
Nuclear Magnetic Resonance Spectrometer”.
2
1, the weighting scheme used was w = 1/[σ2(Fo ) + 7.1422P], where
2
P = (Fo + 2Fc2)/3. For 3, the weighting scheme used was w = 1/
2
[σ2(Fo )+ (0.0197P)2 + 1.8951P], where P = (Fo2 + 2Fc2)/3. For both
complexes, non-hydrogen atoms were refined anisotropically and
hydrogen atoms were refined using a riding model. Figures of the
crystal structures were created using Olex2.36
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.organomet.0c00230.
REFERENCES
■
(1) Doney, S. C.; Fabry, V. J.; Feely, R. A.; Kleypas, J. A. Ocean
Acidification: The Other CO2 Problem. Annu. Rev. Mar. Sci. 2009, 1
(1), 169−192.
NMR spectra, details of experiments and data from pKa
measurement attempts, additional electrochemistry data,
additional infrared spectroelectrochemistry data, and
density functional theory data (PDF)
(2) Elgrishi, N.; Chambers, M. B.; Wang, X.; Fontecave, M.
Molecular polypyridine-based metal complexes as catalysts for the
reduction of CO2. Chem. Soc. Rev. 2017, 46, 761−796.
(3) Francke, R.; Schille, B.; Roemelt, M. Homogeneously Catalyzed
Electroreduction of Carbon DioxideMethods, Mechanisms, and
Catalysts. Chem. Rev. 2018, 118 (9), 4631−4701.
(4) Franco, F.; Fernandez, S.; Lloret-Fillol, J. Advances in the
electrochemical catalytic reduction of CO2 with metal complexes.
Curr. Opin. Electrochem. 2019, 15, 109−117.
(5) Elouarzaki, K.; Kannan, V.; Jose, V.; Sabharwal, H. S.; Lee, J. M.
Recent Trends, Benchmarking, and Challenges of Electrochemical
Reduction of CO2 by Molecular Catalysts. Adv. Energy Mater. 2019, 9
(24), 1900090.
Cartesian coordinates for the optimized structures
(XYZ)
Accession Codes
CCDC 1990924−1990925 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
L
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
(6) Bourrez, M.; Molton, F.; Chardon-Noblat, S.; Deronzier, A.
[Mn(bipyridyl)(CO)3Br]: An Abundant Metal Carbonyl Complex as
Efficient Electrocatalyst for CO2 Reduction. Angew. Chem., Int. Ed.
2011, 50 (42), 9903−6.
(24) Frisch, M.; Trucks, G.; Schlegel, H.; Scuseria, G.; Robb, M.;
Cheeseman, J.; Scalmani, G.; Barone, V.; Petersson, G.; Nakatsuji, H.,
et al. Gaussian 16, Rev. A.03; Gaussian. Inc.: Wallingford, CT, 2016.
(25) Dennington, T. R.; Millam, J. M. GaussView, Ver. 6; Semichem
Inc.: Shawnee Mission, KS, 2016.
(26) Yu, H. S.; He, X.; Truhlar, D. G. MN15-L: A New Local
Exchange-Correlation Functional for Kohn−Sham Density Functional
Theory with Broad Accuracy for Atoms, Molecules, and Solids. J.
Chem. Theory Comput. 2016, 12 (3), 1280−1293.
(27) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence,
triple zeta valence and quadruple zeta valence quality for H to Rn:
Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7
(18), 3297−3305.
(28) Kelly, C. P.; Cramer, C. J.; Truhlar, D. G. Single-Ion Solvation
Free Energies and the Normal Hydrogen Electrode Potential in
Methanol, Acetonitrile, and Dimethyl Sulfoxide. J. Phys. Chem. B
2007, 111 (2), 408−422.
(29) APEX3 2018.7-2; Bruker-AXS: Madison, WI, USA, 2016.
(30) APEX3 2016.1-0; Bruker-AXS: Madison, WI, USA, 2016.
(31) SAINT v8.38A; Bruker-AXS: Madison, WI, USA, 2014.
(32) Krause, L.; Herbst-Irmer, R.; Sheldrick, G. M.; Stalke, D.
SADABS v2016/2. J. Appl. Crystallogr. 2015, 48, 3−10.
(33) Sheldrick, G. M. SHELXT v2014/4. Acta Crystallogr., Sect. A:
Found. Adv. 2015, 71, 3−8.
(34) Sheldrick, G. M. SHELXL-2018/3. Acta Crystallogr., Sect. A:
Found. Adv. 2015, 71, 3−8.
(35) Sheldrick, G. M. SHELXL-2017/1. Acta Crystallogr., Sect. A:
Found. Adv. 2015, 71, 3−8.
(7) Sinopoli, A.; Porte, N. T. L.; Martinez, J. F.; Wasielewski, M. R.;
Sohaila, M. Manganese carbonyl complexes for CO2 reduction. Coord.
Chem. Rev. 2018, 365, 60−74.
(8) Smieja, J. M.; Sampson, M. D.; Grice, K. A.; Benson, E. E.;
Froehlich, J. D.; Kubiak, C. P. Manganese as a substitute for rhenium
in CO2 reduction catalysts: the importance of acids. Inorg. Chem.
2013, 52 (5), 2484−91.
(9) Riplinger, C.; Sampson, M. D.; Ritzmann, A. M.; Kubiak, C. P.;
Carter, E. A. Mechanistic Contrasts between Manganese and
Rhenium Bipyridine Electrocatalysts for the Reduction of Carbon
Dioxide. J. Am. Chem. Soc. 2014, 136, 16285−16298.
(10) Agarwal, J.; Shaw, T. W.; Schaefer, H. F.; Bocarsly, A. B. Design
of a Catalytic Active Site for Electrochemical CO2 Reduction with
Mn(I)-Tricarbonyl Species. Inorg. Chem. 2015, 54, 5285−5294.
(11) Franco, F.; Cometto, C.; Vallana, F. F.; Sordello, F.; Priola, E.;
Minero, C.; Nervi, C.; Gobetto, R. A local proton source in a
[Mn(bpy-R)(CO)3Br]-type redox catalyst enables CO2 reduction
even in the absence of Brønsted acids. Chem. Commun. 2014, 50 (93),
14670−14673.
(12) Franco, F.; Cometto, C.; Nencini, L.; Barolo, C.; Sordello, F.;
Minero, C.; Fiedler, J.; Robert, M.; Gobetto, R.; Nervi, C. Local
Proton Source in Electrocatalytic CO2 Reduction with [Mn(bpy−
R)(CO)3Br] Complexes. Chem. - Eur. J. 2017, 23 (20), 4782−4793.
(13) Tignor, S. E.; Shaw, T. W.; Bocarsly, A. B. Elucidating the
origins of enhanced CO2 reduction in manganese electrocatalysts
bearing pendant hydrogen-bond donors. Dalton Trans. 2019, 48 (33),
12730−12737.
(36) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.
K.; Puschmann, H. OLEX2: a complete structure solution, refinement
and analysis program (pages 339−341). J. Appl. Crystallogr. 2009, 42,
339−341.
(14) Lide, D. R. CRC Handbook of Chemistry and Physics, 84th ed.;
CRC Press: Boca Raton, FL, 2003; p 51.
(15) Bordwell, F. G.; McCallum, R. J.; Olmstead, W. N. Acidities
and hydrogen bonding of phenols in dimethyl sulfoxide. J. Org. Chem.
1984, 49 (8), 1424−1427.
(16) McKinnon, M.; Belkina, V.; Ngo, K. T.; Ertem, M. Z.; Grills, D.
C.; Rochford, J. An Investigation of Electrocatalytic CO2 Reduction
Using a Manganese Tricarbonyl Biquinoline Complex. Front. Chem.
2019, 7 (628), 1−14.
(17) Rotundo, L.; Azzi, E.; Deagostino, A.; Garino, C.; Nencini, L.;
Priola, E.; Quagliotto, P.; Rocca, R.; Gobetto, R.; Nervi, C. Electronic
Effects of Substituents on fac-M(bpy-R)(CO)3 (M = Mn, Re)
Complexes for Homogeneous CO2 Electroreduction. Front. Chem.
2019, 7, 1−13.
(18) Sampson, M. D.; Nguyen, A. D.; Grice, K. A.; Moore, C. E.;
Rheingold, A. L.; Kubiak, C. P. Manganese Catalysts with Bulky
Bipyridine Ligands for the Electrocatalytic Reduction of Carbon
Dioxide: Eliminating Dimerization and Altering Catalysis. J. Am.
Chem. Soc. 2014, 136 (14), 5460−71.
́
(19) Costentin, C.; Robert, M.; Saveant, J.-M. Concerted Proton-
Electron Transfers: Electrochemical and Related Approaches. Acc.
Chem. Res. 2010, 43 (7), 1019−1029.
(20) Lense, S.; Guzei, I. A.; Andersen, J.; Thao, K. C. Crystal
structures of a manganese(I) and a rhenium(I) complex of a
bipyridine ligand with a non-coordinating benzoic acid moiety. Acta
Cryst. E 2018, 74 (5), 731−736.
(21) Olmstead, W. N.; Bordwell, F. G. Ion-Pair Association
Constants in Dimethyl Sulfoxide. J. Org. Chem. 1980, 45, 3299−3305.
(22) Mukhopadhyay, T. K.; MacLean, N. L.; Gan, L.; Ashley, D. C.;
Groy, T. L.; Baik, M.-H.; Jones, A. K.; Trovitch, R. J. Carbon Dioxide
Promoted H+ Reduction Using a Bis(imino)pyridine Manganese
Electrocatalyst. Inorg. Chem. 2015, 54 (9), 4475−4482.
̌
̌
(23) Krejcik, M.; Danek, M.; Hartl, F. Simple construction of an
infrared optically transparent thin-layer electrochemical cell: Applica-
tions to the redox reactions of ferrocene, Mn2(CO)10 and Mn-
(CO)3(3,5-di-t-butyl-catecholate)−. J. Electroanal. Chem. Interfacial
Electrochem. 1991, 317 (1−2), 179−187.
M
https://dx.doi.org/10.1021/acs.organomet.0c00230
Organometallics XXXX, XXX, XXX−XXX