Access to Metal Centers and Fluxional Hydride Coordination Integral for CO2Insertion into [Fe3(μ-H)3]3+Clusters

Article abstract of DOI:10.1021/acs.inorgchem.1c00244

CO2 insertion into tri(μ-hydrido)triiron(II) clusters ligated by a tris(β-diketiminate) cyclophane is demonstrated to be balanced by sterics for CO2 approach and hydride accessibility. Time-resolved NMR and UV-vis spectra for this reaction for a complex i

Full text of DOI:10.1021/acs.inorgchem.1c00244

pubs.acs.org/IC
Article
Access to Metal Centers and Fluxional Hydride Coordination
Integral for CO2 Insertion into [Fe3(μ-H)3]3+ Clusters
Dae Ho Hong, Ricardo B. Ferreira, Vincent J. Catalano, Ricardo García-Serres, Jason Shearer,*
and Leslie J. Murray*
Cite This: https://doi.org/10.1021/acs.inorgchem.1c00244
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: CO2 insertion into tri(μ-hydrido)triiron(II) clusters ligated by a tris(β-
diketiminate) cyclophane is demonstrated to be balanced by sterics for CO2 approach and
hydride accessibility. Time-resolved NMR and UV−vis spectra for this reaction for a complex in
which methoxy groups border the pocket of the hydride donor (Fe3H3L2, 4) result in a decreased
activation barrier and increased kinetic isotope effect consistent with the reduced sterics. For the
ethyl congener Fe3H3L1 (2), no correlation is found between rate and reaction solvent or added
Lewis acids, implying CO2 coordination to an Fe center in the mechanism. The estimated
hydricity (50 kcal/mol) based on observed H/D exchange with BD3 requires Fe−O bond
formation in the product to offset an endergonic CO2 insertion. μ3-hydride coordination is noted
to lower the activation barrier for the first CO2 insertion event in DFT calculations.
Scheme 1. Reported Multi-Iron Complexes with Bridging
Hydrides Coordinated by β-Diketiminate-Type Ligands
INTRODUCTION
■
Metal hydrides mediate various small molecule transformations
in chemical and biological systems, including proton, CO2, and
N2 reduction.1−7 Given such reactivity, metal hydrides have
gathered substantial interest, especially in the context of
renewable energy and carbon-neutral fuel synthesis as well as
fertilizer production.8−12 Mononuclear metal hydrides have
been extensively studied in the fields of energy storage,13
electrocatalysis,14,15 and chemical synthesis.16,17 Metal hydride
compounds have predominantly employed strong field ligands,
leading to well-defined acid−base reactivity profiles observed
in many compounds. Far fewer examples of metal-hydride-
supported weak-field donors are known, and even fewer in
which the hydrides adopt bridging coordination modes. Such
hydride-bridged multimetallic assemblies are proposed as key
intermediates in the activation of small molecules affected by
metalloenzymes and heterogeneous catalysts. Notable exam-
ples of important hydride-bridged transients include the
proposed structures of reduced intermediates of the FeMo
cofactor from molybdenum-dependent nitrogenases and of
[NiFe]- or [FeFe]-hydrogenases,7,18−20 H adatoms on the
surface of various heterogeneous catalysts,21−23 and copper
hydride clusters used to catalyze organic transformations (e.g.,
conjugate reduction reactions).24,25 Describing the factors
governing reactivity of weak field supported hydride-bridged
metal centers has broad relevance to chemical synthesis,
industrial processes, and biological systems.
(or BDI) supported di(μ-hydrido)diiron complexes in which
the H/D exchange with H2/D2 and between isotopologues,
bond metathesis with BEt3, insertion reactivity with CO2 and
various other possible electrophiles, and H2 reductive
elimination have been observed.35,36 In contrast to these
Received: February 1, 2021
Weak-field ligands in which steric bulk limits the
coordination number afford metal hydrides with diverse
reactivity profiles, including reductive elimination, insertion,
and H atom transfer reactivity (Scheme 1).26−34 For example,
Holland and co-workers synthesized a series of β-diketiminate
https://doi.org/10.1021/acs.inorgchem.1c00244
© XXXX American Chemical Society
Inorg. Chem. XXXX, XXX, XXX−XXX
A

Inorganic Chemistry
pubs.acs.org/IC
Article
organic phase dried under a vacuum to yield an orange-red oil (4.81 g,
14.4 mmol, yield 99%). H NMR (300 MHz, CDCl3, 298 K): δ 4.45
(s, 6 H), 3.92 (s, 9 H).
dimetallics which spontaneously assemble and exist in a
monomer−dimer equilibrium, the tri(μ-hydrido)triiron com-
plex templated by a tris(β-diketimine)cyclophane (i.e., Fe3(μ-
H)3L1 or 2) is remarkably specific for hydride transfer to CO2
with minimal or no hydride transfer reactivity to other
substrates.29 Similar hydride transfer reactivity was also
observed for the cobalt and the zinc congeners, leading to a
hypothesis that the ligand pocket surrounding the μ-hydride
donors strongly influences kinetics and selectivity for CO2
reduction to formate.29,34,37 In addition, we also observed
reductive elimination (re) of H2 upon the reaction of 2 with
CO. The resultant Fe3(μ3-H)(CO)2L1 product is competent
for the oxidative addition of H2 at slightly elevated temper-
atures to regenerate 2. Such reactivity suggested that the
hydrides in 2 are fluxional as the hydride−hydride distances
are substantially longer than that in H2.34 A detailed
understanding of the order of Fe−H bond breaking and C−
H bond formation and the factors governing the observed
substrate specificity are important for developing approaches
to controlling specificity and reactivity. Herein, we report the
effect of changes to the secondary coordination sphere
specifically, the consequence of a methoxy for ethyl
substitutionon the rate of CO2 insertion, effect of solvent
and exogeneous Lewis acids on reaction of 2 with CO2,
hydride exchange between and transfer from 2 to boranes, and
density functional theory simulations of the reaction
coordinate for transfer of the first hydride to CO2 to generate
the monoformate complex. The mechanistic picture that
emerges is one in which hydrides shift coordination modes
upon interaction with the substrate, an Fe−OCO interaction
precedes hydride transferalthough this interaction is
intimately related to the steric constraintsand sterics of the
pocket surrounding the hydride and the hydricity enforce the
observed substrate selectivity.
1
1,3,5-Tri(aminomethyl)-2,4,6-trimethoxybenzene. The title
compound was synthesized by a procedure analogous to that for
1,3,5-tri(aminomethyl)-2,4,6-triethylbenzene.41 1,3,5-Tri-
(azidomethyl)-2,4,6-trimethoxy-benzene (4.81 g, 14.4 mmol) and
Pd/C (10 wt % Pd, 0.200 g, 0.188 mmol, 1.3 mol %) were suspended
in ethanol (100 mL) and transferred to a Parr bomb. The system was
charged with H2 (700 psi) under stirring in an ice−water bath and
then allowed to warm to room temperature and then was stirred for 2
days. The resulting mixture was filtered over the ethanol-rinsed Celite
pad in a fine glass frit. Volatiles of the filtrate were removed under a
vacuum, resulting in a pale-yellow powder (3.54 g, 13.9 mmol, yield
96%). 1H NMR (300 MHz, CDCl3, 298 K): δ 3.84 (s, 6 H), 3.82 (s, 9
H).
H3L2. 1,3,5-tri(aminomethyl)-2,4,6-trimethoxybenzene (3.54 g,
13.9 mmol) and 2,4-pentanedione-2,2-(ethylene glycol) monoketal
(3.26 g, 22.2 mmol, 1.6 equiv) were dissolved in methanol (100 mL).
The mixture was brought to reflux under dinitrogen for 2 days, when a
pale-yellow suspension was formed. The mixture was then cooled to
room temperature and filtered through a fine fritted glass funnel. The
resulting pale-yellow powder was washed with methanol (3 × 30 mL)
and water (3 × 30 mL) and dried under a vacuum. Yield: 3.18 g
(65%). 1H NMR (300 MHz, CDCl3, 298 K): δ 10.48 (br s, 3 H), 4.56
(s, 3 H), 4.31 (s, 12 H), 3.64 (s, 18 H), 2.03 (s, 18 H). 13C{1H} NMR
(126 MHz, CDCl3, 298 K) δ: 160.03, 158.87, 123.34, 110.14, 93.79,
63.82, 40.85, 20.23. (+)ESI-MS ([M + H]+) m/z calcd. for
C39H54N6O6: 703.4183. Found: 703.4197.
Fe3Br3L2 (3). H3L2 (217 mg, 0.308 mmol) and lithium LiNiPr2
(LDA, 134 mg, 1.22 mmol, 4.0 equiv) were combined with THF (4
mL) and stirred for 10 min to afford a purple solution. All volatiles
were removed from the solution, and the resulting dark purple residue
was combined with FeBr2 (315 mg, 1.46 mmol, 4.7 equiv) and
toluene (20 mL). The mixture was stirred at 50 °C for 20 h, then the
dark red mixture was filtered over the toluene-rinsed Celite pad, and
the filtrate was cooled down to −34 °C for 2 days, where dark red
crystals were obtained (149 mg, 0.127 mmol, yield 42%). Crystals
suitable for X-ray diffraction were grown by slow evaporation of a
saturated toluene solution of the complex. 1H NMR (500 MHz,
benzene-d6, 298 K) δ: 250.06 (2 H), 208.67 (2 H), 128.91 (4 H),
41.20 (4 H), 17.99 (6 H), 3.62 (1 H), −5.26 (12 H), −32.55 (18 H),
−82.29 (2 H). FT-IR ν (cm−1): 2920, 1579, 1512, 1390, 1338, 1100,
EXPERIMENTAL METHODS
■
General Considerations. All manipulations except ligand
synthesis were performed inside an N2-filled Innovative Technologies
glovebox unless otherwise stated. Tetrahydrofuran (THF), benzene,
toluene, and n-hexane were purchased from Sigma-Aldrich, then dried
using an Innovative Technologies solvent purification system,
transferred under an inert atmosphere to the anaerobic chamber,
and stored over activated 3 Å molecular sieves for at least 24 h prior
to use. Benzene-d6, toluene-d8, and THF-d8 were purchased from
Cambridge Isotope Laboratories, dried over CaH2 or Na/
benzophenone under reflux, then distilled, degassed, and stored
over 3 Å molecular sieves.
̈
1004, 736, 572. Moßbauer parameters (80 K, zero-applied field, mm/
s): δ/ΔEQ = 0.90/2.47 and 0.93/1.72. Combustion Anal. exp.(calc.)
for C44.25H57Br3Fe3N6O6 (Fe3Br3L2·0.75C7H8): C, 45.27(45.19); H,
4.80(4.88); N, 7.02(7.15).
Fe3H3L2 (4). 3 (134 mg, 0.121 mmol) was combined with toluene
(4 mL), resulting in a red suspension at room temperature. To this
suspension, a solution of KBEt3H (49.2 mg, 0.357 mmol, 3.0 equiv) in
toluene (2 mL) was added dropwise under vigorous stirring. The
system turned red-brown upon addition, and it was kept under 25 °C
for 10 min. The resulting dark red-orange mixture was filtered over a
Nylon filter paper, resulting in a black residue that was washed with
toluene (2 mL). The combined filtrate was evaporated under reduced
pressure to afford a dark brown solid. This dark brown solid was
recrystallized in cold diethyl ether at −35 °C (72.5 mg, 0.077 mmol,
yield 64%). 1H NMR (500 MHz, benzene-d6, 298 K): δ (ppm): 73.70
(12 H), 2.28 (3 H), −16.81 (18 H), −29.15 (18 H). FT-IR ν (cm−1):
2918, 1592, 1522, 1397, 1348, 1343, 1228, 1112, 1008, 741, 499.
1H nuclear magnetic resonance (1H NMR) spectra were recorded
on a 500 MHz Varian Inova spectrometer or a 300 MHz Mercury
spectrometer equipped with a three-channel 5 mm indirect detection
probe with z-axis gradients. Chemical shifts were reported in δ (ppm)
and were referenced to solvent resonances of δH = 7.16 ppm for
benzene-d6, 7.01 ppm for toluene-d8, and 3.58 ppm for THF-d8.
Fourier transform infrared (FT-IR) spectra were recorded as solids on
a Thermo Fisher iS5 instrument equipped with an ATR diamond iD7
stage and operated by the OMNIC software package. H3L1, Fe3Br3L1
(1),38 Fe3H3L1 (2),29 and 1,3,5-tri(bromomethyl)-2,4,6-trimethox-
ybenzene39 were prepared following the previous literature.
̈
Moßbauer parameters (80 K, zero-applied field, mm/s): δ/ΔEQ
=
0.78/2.27. Combustion anal. exp. (calcd.) for C45H69Fe3N6O7.5
(Fe3H3L2·1.5Et2O): C, 55.18(55.06); H, 6.58(7.09); N, 8.62(8.56).
Fe3(CO2H)3L2 (6c). A solution of 4 in a benzene-d6 was reacted
with CO2 as described in NMR kinetic experiment below except the
temperature was maintained at 50 °C. After 2 days, the resulting
1,3,5-Tri(azidomethyl)-2,4,6-trimethoxybenzene. The title
compound was synthesized using the protocol described for 1,3,5-
tri(azidomethyl)-2,4,6-triethyl-benzene.40 Briefly, 1,3,5-tri-
(bromomethyl)-2,4,6-trimethoxybenzene (6.51 g, 14.6 mmol) was
dissolved in acetone (150 mL) and treated with NaN3 (3.80 g, 58.4
mmol, 4.0 equiv), and the white suspension was heated to reflux. H2O
(40 mL) was then added dropwise, and the mixture was stirred at 50
°C for 12 h. The product was extracted with dichloromethane and the
1
product was subjected to analysis. H NMR (500 MHz, benzene-d6,
298 K) δ (ppm): 216.64, 100.38, −2.66, −4.92, −7.08, −33.01,
−49.84. FT-IR ν (cm−1): 2916, 1938, 1690 (CO of formate), 1520,
B
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
1401, 1340, 1105, 1005, 798. (+)ESI-MS ([M+2H]+) m/z calcd. for
C42H56Fe3N6O12: 1004.2005. Found: 1004.2347.
l
o exp(−k1t) − exp(−k3t)
k1k2
o
m
C(t) = C
C k2 − k1
k3 − k1
o
o
n
X-ray Crystallography. Single crystal X-ray diffraction was
performed on a Bruker SMART Apex II CCD instrument at 100 K
using graphite-monochromated Mo Kα radiation. The crystals were
covered in Paratone oil and mounted on glass fibers. Lorentz and
polarization effects were corrected using SAINT, and the absorption
corrections were applied using SADABS. The structures were solved
using direct methods or Patterson syntheses using SHELXS and
SHELXL.
|
exp(−k2t) − exp(−k3t)o
o
−
}
o
o
k3 − k2
~
l
o
k2(k3 exp(−k1t) − k1 exp(−k3t))
o
D(t) = CDm1 −
o
o
(k3 − k1)(k2 − k1)
n
|
k1(k3 exp(−k2t) − k2 exp(−k3t))o
o
}
+
o
o
(k3 − k2)(k2 − k1)
~
̈
Moßbauer Spectroscopy. In a N2-filled glovebox, samples were
ground into fine powders, placed in Delrin sample containers, and
where A, B, C, and D are integrals as a function of time for resonance
corresponding to the starting, intermediate, and product kinetic
models for two- or three-step irreversible reactions; CA, CB, CC, and
CD are arbitrary constants; and k1, k2, and k3 are the rate constants for
each reaction step.
̈
sealed. Moßbauer spectra were measured either on a low-field
̈
Moßbauer spectrometer equipped with a closed-cycle SHI-850-5
cryostat from Janis and SHI or an Oxford Instruments Spectromag
4000 cryostat containing an 8 T split-pair superconducting magnet.
Both spectrometers were operated in constant acceleration mode in
transmission geometry. The isomer shifts are referenced against a
room temperature metallic iron foil. Analysis of the data was
performed using the program WMOSS (WEB Research).
NMR Kinetic Experiments: Reaction of 4 with CO2. A solution
of 4 in a benzene-d6 and a capillary charged with a hexamethylben-
zene standard solution were transferred to a J-Young NMR tube, and
the tube was degassed by five cycles of freeze−pump−thaw.
Concurrently, a 50 mL Schlenk flask was evacuated and charged
with CO2, and CO2 was transferred from this flask to the headspace of
an NMR tube using a three-way valve. The NMR tube valve was then
closed, and the tube was quickly shaken several times. The
equilibrated CO2 concentration was determined by the following
equation:
UV−Visible Absorption Kinetic Experiments: Reaction of 4
with CO2. A Cary 50 spectrophotometer, equipped with Unisoku
single-cell cryostat accessory to maintain the temperature within 0.1
°C, was used for all kinetics experiments. All measurements were
performed in Schlenk-adapted cuvettes with a 1 cm optical path
length. Baselines of all spectra were corrected by subtracting the
spectrum of neat toluene at the same temperature. The typical
procedure for sample preparation at variable temperatures is described
below.
A stock solution of 2 or 4 and a magnetic stir bar were transferred
to a solid addition flask, which was connected to a Schlenk cuvette in
the glovebox. After five freeze−pump−thaw cycles, the solution was
transferred to the cuvette, and the cuvette was placed in the Unisoku
cryostat. UV−visible spectra were recorded until no further change
was observed. CO2 gas was introduced to the evacuated headspace
immediately prior to starting the kinetic measurement, and the
solution was agitated for 1 min to saturate the solution with CO2
under atmospheric pressure. The concentration of saturated CO2 was
calculated or referred from the previous literature.42,43
[CO2]eqm = nCO ,total
γ /V
soln
soln
2
γ
= 1/(1 + /(Vhead + Vflask)/nsolvRT)
soln
UV−visible spectra were recorded from 400−1000 nm with a 600
nm/min scan rate at a logarithmic time interval. The first point was
discarded as the temperature, and the concentration of CO2 had not
been equilibrated. Kinetic traces for each reaction were generated by
plotting absorbance (y) versus reaction time (t) at the peak
wavelengths, which were analyzed using Origin v.8.5. Absorptions at
10 distinct wavelengths in the range of 405−495 nm with a 10 nm
interval were chosen, and the curves were fitted to the first-order
kinetic equations for two- or three-step irreversible reactions with
shared parameters of rate constants:
where R, γsoln, Vsoln, Vhead, VFlask, and nsolv are the ideal gas constant, the
ratio of CO2 remained in solution, the volume of the solution, the
headspace volume of the J-young NMR tube, the volume of the
Schlenk flask, and moles of solvent, respectively. The Henry
constants,
/
, of benzene and toluene are 10.370 × 106 and 9.705
× 106 Pa, respectively, at 298.15 K.42
NMR spectra were measured up to 7 days, with each spectrum
measured using the same power, gain, and shimming parameters.
After phase correction and reference peak assignment, baseline
correction was performed using a Whittaker smoother method (filter:
50, 28.57 Hz; smooth factor: 106) provided by MestReNova 8.1.
Common integration ranges were used for all spectra, and absolute
integral values were recorded and used for kinetic analysis. For
resonances with small integrals, the baseline correction routine above
resulted in negative integration values in some cases; therefore, a
parallel baseline correction was performed with different parameters
(filter: 50, 28.57 Hz; smooth factor: 3 × 104) to account for these
features. Absolute integration values (y) over time (t) were plotted for
each integration range and fitted using one of the following equations:
For the A → B → C model:
For the two-step A → B → C model:44
Abstotal(t) = AbsA(t) + AbsB(t) + AbsC(t)
k1
= (ϵA exp(−k1t) + ϵ
[exp(−k1t)
B k2 − k1
l
o
1
o
− exp(−k2t)]+ϵCm1 −
[k2 exp(−k1t)
o
o
k2 − k1
n
|
o
o
− k1 exp(−k2t)]}) × [A]0
o
o
~
where [A]0, εA, εB, and εC are the initial concentration of the starting
material and absorption constants of each A, B, and C species at a
specified wavelength, and k1 and k2 are the pseudo-first-order rate
constants.
A(t) = CA exp(−k1t)
k1
B(t) = C
[exp(−k1t) − exp(−k2t)]
This global analysis approach wherein the rate constants k1,obs
,
B k2 − k1
k
2,obs, and k3,obs are shared over the wavelength range is critical
because the rate constants tend to diverge to fit the absorption curves
with convex and concave regions found in different reaction time
courses at different wavelengths.
Electronic Structure Calculations. All electronic structure
calculations were performed using ORCA v 4.1. Geometry
optimizations and transition state searches were performed using
l
o
|
o
1
o
o
C(t) = CCm1 −
[k2 exp(−k1t) − k1 exp(−k2t)]}
o
o
o
o
k2 − k1
n
~
For the A → B → C → D model where only C(t) and D(t) are
different from the previous model:
C
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
the BP86 level of theory using a broken-symmetry wave function, with
the def2-tzvp basis set on all atoms, and Becke-Johnson damping
function to account for dispersive forces (Grimme’s D3). Geometry
scans to identify starting structures for transition states and
intermediates were performed by scanning the H−C bond for the
insertion reaction and the Fe−O bond for the final rearrangement of
the formate ligand to the monoformate product prior to transition
state optimizations and intermediate geometry optimizations. Sta-
tionary points and transition states on the potential energy were
verified by the absence of imaginary frequencies (stationary points) or
the presence of one imaginary mode (transition state). Reported free
energies were determined in a similar manner but used the PBE0
hybrid density functional as opposed to the BP86 density functional.
Semiclassical room temperature (T = 298.15 K) H/D kinetic isotope
effects (KIEs) were estimated by employing an Eyring model:
Scheme 2. Ligand and Complex Synthesis
kH/kD = exp[(ΔGD‡ − ΔGH‡ )/RT]
To account for tunneling effects, a Wigner correction45 to the
semiclassical KIEs was applied by multiplying the Eyring KIEs with an
H/D tunneling ratio correction (QH/QD), where
Q = 1 + ui2/24
i
and:
ui = hν/kBT
i
where h is the Planck constant, kB is the Boltzmann constant, and νi is
the transition state imaginary frequency.
a
b
Benzyl potassium for H3L1, LDA for H3L2, r.t. in THF, 10 min. 80
RESULTS AND DISCUSSION
■
c
°C for 1, 50 °C for 2 in C7H8, 20 h. r.t. in C7H8, 10 min.
Steric and Secondary Coordination Sphere Ligand
Effect for CO2 Insertion into Bridging Hydrides.
Previously, we reported the significant through-space coupling
between bridging the hydride and −CH2− of ethyl substituents
in the Zn isomer of Fe3H3L1 (2).46 To tune the sterics and
electrostatics of the pocket surrounding the hydride donor,
ligand H3L2 and the corresponding triiron complexes, Fe3Br3L2
(3) and Fe3H3L2 (4), were prepared in which the ethyl
substituents on the two arene caps are replaced with methoxy
groups (Scheme 2).29,38 Substituting the ethyl groups for a
smaller methyl group or a proton yield ligands with poor
solubility; thus, we opted for an ethyl for methoxy substitution
to probe the role of sterics on reactivity. The crystallographic
structure of 3 shows successful metalation of the three β-
diketiminate arms (Figure S16).38 The structures of the
[Fe3Br3N6] cores in 1 and 3 are isostructural with a minimal
RMSD value of 0.34 between the two compounds (Figure 1a).
The position of the terminal bromide contributes the most to
this calculated RMSD as it resides much closer to the adjacent
iron center in 3 than in 1 (4.329 vs 5.224 Å, Figure 1a).
Complexes 1 and 3 support our assertion that variations to the
ligand architecture can be readily accomplished without
sacrificing access to isostructural complexes.
Figure 1. (a) Structure comparison of the [Fe3Br3N6] core of 1 (red)
and 3 (blue). Two-headed arrow highlights the narrowed distance
̈
between iron and terminal bromide of 3. (b) Moßbauer spectrum of
Fe3Br3L1 (1, top) and Fe3Br3L2 (3, bottom). The black noisy lines
represent the experimental data. The colored lines are simulated
quadrupole doublets, as described in the text (see Table S1 for
simulation parameters), whereas the black solid line is a composite
spectrum obtained by combining individual doublets.
̈
The Moßbauer spectrum for a solid sample of 3 recorded
under a zero-applied field is well-simulated with a pair of
quadrupole doublets centered at 0.90 and 0.93 mm·s−1 with
ΔEQ values of 2.47 and 1.72 mm·s−1, respectively, which are
comparable to those obtained for 1 (0.95−1.02 mm·s−1; Figure
and one μ3). Zero-applied field spectra recorded on a solid
sample of 4 were best fit with a single quadrupole doublet
centered at 0.78 mm·s−1 with a quadrupole splitting of 2.27
mm·s−1, consistent with the anticipated D3h molecular
symmetry (Figure S15, Table S1).29 These parameters are in
excellent agreement with those obtained for 2 (viz. δ = 0.79
mm·s−1, ΔEQ = 2.34 mm·s−1) and are consistent with
tetrahedral-ligated high-spin iron(II) centers.29,47−49 Gratify-
ingly, the similar parameters of 4 and 2 support that there is
minimal, if any, interaction between the Fe centers and the
methoxy O atoms in 4. We conclude then that our ligand
design is an effective secondary coordination sphere
̈
1b). The 2:1 ratio for the integrals of these Moßbauer
absorptions suggests that Fe1 and Fe3 are equivalent with Fe2
as unique (Figure 1a). The difference between the isomer
shifts for 1 and 3 are attributed to greater fluxional behavior of
the bromide donors and flexibility afforded by the OMe-for-Et
substitution in 3 as compared to 1, which allows for
interconversion between Cs and C2v structures (the latter
being a ladder-like configuration with the bromides as two μ-
D
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
perturbation, allowing targeted evaluation of pocket electro-
static or steric effects on hydride reactivity.
An initial survey of the reactivity of 4 toward CO2 evidenced
the faster reaction rate as compared to 2 with the same
substrate under a dynamic atmosphere of CO2 (Scheme 3 and
recorded spectra over the time course were integrated and
plotted vs time. Grouping resonances by their time constants,
we infer the presence of three kinetically related species after
the addition of CO2 (Figure 2). The four resonances
associated with the D3h symmetric 4 decreased exponentially
with the concomitant increase in intensity of 11 new signals of
the first intermediate (6a). The concentration of 6a maximized
at ∼30 min, then decreased over the subsequent 15 h with the
concomitant appearance of 10 new resonances of the second
intermediate (6b). The intensities of 6b maximized at ∼15 h
and decreased over the subsequent 7 days to afford seven new
signals corresponding to a third species. From the independent
reaction under a dynamic CO2 atmosphere, the resonances for
this final product are those for the C3v symmetric triformate
complex 6c. These data imply that the reaction follows a four-
component three sequential step reaction model (i.e., A → B
→ C → D), with the two intermediates 6a and 6b
corresponding to the mono- and diformato-triiron complexes,
respectively.
Scheme 3. Reaction Product of Trihydride Complexes (2 or
4) with CO2
To simplify our kinetic analysis, we flooded the reaction with
CO2 and fit the integrals of the resonances for each of the four
observed species as a function of time using a pseudo-first
order A → B → C → D model (Figure 2 and Table 1). From
the spectra recorded at 25 °C, the integrals for resonances
corresponding to 4 decrease exponentially and are well fit with
the proposed model to afford k1,obs = 2.1(2) × 10−3 s−1, which
is 20 times faster than that of 2 (viz. 1.1(1) × 10−4 s−1, Table
Figure S11).29 To investigate the reaction kinetics and the
intermediates formed during the reaction with CO2, we
exposed a degassed solution of 4 in benzene-d6 to CO2 at
293−308 K, agitated the solution to minimize the saturation
time, and monitored the reaction by in situ 1H NMR
spectroscopy using hexamethylbenzene as an internal standard.
All well-resolved peaks between −200 to 250 ppm in the
1
Figure 2. Changes of H NMR integration values grouped by species 4 (a), 6a (b), 6b (c), and 6c (d) for reaction of 4 with CO2 (95 mM) in
benzene-d6 at 25 °C. Resonances are specified in the legend for which the integrations were fit using an A → B → C → D as described in the
experimental section. Fitted rate constants from 4, 6a, and 6b were used to simulate the data for 6c. Each set of points represents the integration for
the corresponding chemical shift as described in the legend. Peaks with minor contributions are omitted for clarity.
E
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
experiments. Nonetheless, the first reaction rate constant
(k1,obs) determined from the NMR and UV/visible data is in
Table 1. Rate Constants Determined for Conversion of 4 to
6a by CO2 in Benzene
excellent agreement; however, the second rate constant (k2,obs
)
a
[CO2]
(mM)
T
(K)
fitting
target
k1,obs (× 10−3
k1 (× 10−2
appears overestimated as we cannot separate this rate constant
from contributions of those for an unproductive decom-
position side reaction. Consequently, our variable-temperature
data allow determination of ΔH‡, ΔS‡, and ΔG298K‡ values for
the first CO2 insertion event only; these values are 12.1(5)
kcal·mol−1, −26(2) e.u., and 19.7(7) kcal·mol−1, respectively.
Using the previously reported thermodynamic activation
parameters for CO2 insertion into 2,50 we surmise that the
difference in the free energies of activation for CO2 insertion
method
NMR
s−1
)
M−1·s−1
)
105
105
105
110
298
298
298
293
4
6a
6b
2.1(2)
2.0(2)
1.2(2)
1.6(1)
2.0(2)
1.9(2)
1.1(2)
1.4(1)
UV−
405−495
vis
nm
105
100
96
298
303
308
2.16(7)
2.82(7)
4.0(1)
2.05(7)
2.82(7)
4.2(1)
‡
a
for 2 and 4 (ΔΔG298K = 1.7 kcal/mol) arises almost
Calculated. See the Experimental Section for detail.
exclusively from the larger enthalpy of activation with ΔΔH‡
= 2.6 kcal/mol for 2 vs 4. These data then suggest comparable
order within both transition states, as one might expect given
that 1 and 3 are isostructural, and that the steric or electrostatic
changes resulting from the OMe for Et substitution reflect a
decrease in the energetic cost to assemble the transition state.
The methoxy for ethyl substitution can influence the sterics
and the electrostatics of the pocket surrounding the bridging
hydride. To dissect which effect likely dominates here, we
employed SambVca to estimate the buried volume around the
bridging hydride in 2 and 4.51 In the absence of a crystal
structure for 4 but armed with spectroscopic data equivalent to
2, we modeled 4 as replacing the CH2 units in the ethyl
substituents of 2 to O atoms. The resultant buried volumes
calculated for these structures were 82.8% for 2 and 80.1% for
4 (Figure 3).52,53 We note that the 20-fold increase in k1,obs
correlates with a ∼3% reduction of the buried volume. A series
of nickel pincer complexes are reported to have a 500- to 647-
fold rate enhancement for inner-sphere CO2 insertion
including the interaction between a metal center and CO2 in
the transition stateupon reducing the buried volume around
the terminal hydride by 6−7%,43 suggesting that the modest
acceleration observed for 4 compared to 2 is consistent with
primarily a steric effect. Decreasing the sterics of the pocket
also serves to increase access to the metal centers, and we
cannot discriminate between the possible contributions of
steric accessibility to the hydride vs metal ion; that is, rate
enhancements as a function of sterics may be similar for both
inner and outer sphere pathways.
S4). From the kinetic model employed, we can also extract
experimental values for k1,obs from kinetic profiles of 6a and 6b;
the determined values of 2.0(2) × 10−3 s−1 and 1.2(2) × 10−3
s−1 for 6a and 6b, respectively, are comparable to that
determined from the decay of 4. The small k1,obs value of 6b is
attributed to underestimation of the integrals for resonances of
6b at early reaction times (<2 h), as these peaks are broad and
poorly resolved (Figure S17). An identical value of k2,obs of
1.2(1) × 10−4 s−1 was determined from the decay and
formation of 6a and 6b. The formation of 6c is too slow for
reliable determination or modeling of the rate constants.
Changing one hydride in 6a to a formate in 6b results in a 10-
fold decrease in the rate of CO2 insertion, implying that the
electronic environment of the iron center plays an important
role in the reaction.
We also employed variable temperature UV/visible spec-
trophotometry of reaction mixtures to measure the rate
constants for these sequential reactions independently (and
validate our NMR methodology) as well as to determine the
activation parameters for the reaction of 4 with CO2. The
spectra were modeled using the sequential four-component
kinetic model as above and globally fitted the absorption vs
time data. Unlike the NMR kinetic analysis, deleterious
reactions which introduce minor quantities of as-yet-identified
side products were observed for all steps, but common rate
constants for the 4-to-6c pathway were consistent between
methods. This difference likely arises from trace impurities in
the gas stream coupled to the differences in the scale and static
vs dynamic atmospheres used for the NMR and UV/visible
Solvent and Lewis Acid Effect for CO2 Insertion.
Choice of the reaction solvent or addition of cations can
Figure 3. Affected area for the topographic steric maps around the bridging hydride of 2 (left, green circle) and the maps of 2 (middle, %VBur
=
82.8%) and 4 (right, %VBur = 80.1%). C, N, H, and Fe are represented as gray, blue, white, and orange spheres with atomic radii, respectively.
F
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Figure 4. (a) 1H NMR spectrum of 2 in toluene-d8. Peaks are marked as follows: −CH2− of aminomethyl A, −CH− of beta-ketiminate B, CH3−
1
of β-ketiminate C, CH3− of ethyl D, −CH2− of ethyl E, and solvent and solvent impurities S. (b) H NMR spectrum from mixing equimolar
amounts of 2 and deuterated borane THF adduct (BD3·THF) in toluene-d8 for 20 h. Peaks from the minor species are denoted as *. All four
isotopologues (H3, H2D, HD2, and D3) are visible as two groups of four nearby peaks at A (c) in a statistical 1:3:3:1 ratio, which are located in
spatial vicinity to the bridging hydrides/deuterides. Relative intensities were determined through the deconvolution of the peak with Lorentzian
functions (green, 1H NMR spectrum; red, fitted Lorentzian functions for each component; black, sum of all component functions). Only the parts
of the spectra from δ 160 to −60 ppm are shown for clarity.
enhance the rate of the CO2 insertion into terminal hydride
complexes when a charge is developed for the reaction
intermediate.8,43,54,55 Dielectric constants (ε) of the solvents
have been employed to correlate with the rate enhancement by
the stabilization of the polar intermediate;56 however,
ANs43,54,57,58 and solvatochromic constants (ET(30))55 have
also been considered for transition metal mediated trans-
formations. We examined diethyl ether, benzene, 1,4-dioxane,
tetrahydrofuran, and acetonitrile as solvents for the reaction of
2 with CO2 (Figure S31 and Table S5). We observe the
changes to the rate constant k1,obs as a function of reaction
solvent, but there is no discernible correlation between k1,obs
and any of the parameters ε, AN, or ET(30). Likewise, the
addition of alkali cations has a negligible effect on the rate
(Table S6). We also compared the effect of anions and ionic
strength and, again, observed rate constants that are within
error of reactions lacking such additives. We conclude that the
reaction must either be proceeding with minimal charge
accumulation on the O atom of the CO2 substrate molecule
inconsistent with an outer-sphere pathwayor there is limited
steric access of solvents and exogenous cations to the reaction
center.
reaction to proceed by an outer sphere hydride transfer
pathway. We selected the following boranes as these
compounds cover a broad range of hydricities: BEt3 (24),
BPh3 (36), BD3 (50), and tris(pentafluorophenyl)borane or
BCF (65). The values in parentheses are the calculated
hydricities of conjugate borohydrides in kcal·mol−1 determined
by DFT-level simulations.59 Whereas complex 2 is unreactive
toward BEt3 (as expected given the synthetic route to 2) and
1
BPh3, complete consumption of 2 is observed by H NMR
spectroscopy upon reaction with BCF (Figure S12). We then
examined the reaction with BD3 insofar as the hydricity of
borane is intermediate between BPh3 and BCF. Here, we
elected to use BD3 rather than BH3 to evaluate the reversibility
of hydride abstraction.35,60 The reaction of 2 with BD3 in
toluene resulted in broad distribution of isotopologues of 2
based on the number of satellite peaks, which appear for each
resonance corresponding to 2 (Figure 4). For example, the
peaks at 80 ppm and −20 ppm are each well resolved into four
peaks with the ratio of the integrals being 1:3:3:1. This ratio
suggests a statistical mixture of available isotopically labeled
complexes (viz. [Fe3H3]3+, [Fe3H2D]3+, [Fe3D2H]3+, and
[Fe3D3]3+) results from the reversible hydride−deuteride
exchange. In addition, new resonances are also observed in
the region of 150 to −50 ppm, which could arise from complex
products with reduced symmetry, as one might anticipate for a
cationic triiron dihydride compound or a borohydride adduct.
Given the equilibrium mixture observed and the near
Estimating Thermodynamic Hydricity of 2. To isolate
the hydride insertion event from the thermodynamic
compensation afforded by formate coordination, we estimated
the hydricity of the Fe−H−Fe units by the reaction of 2 with a
series of boranes. We envisioned this hydride abstraction
G
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
equivalent distribution of the various istopologues, we estimate
We redetermined the KIE for the reaction in benzene to nullify
any potential effect from solvent and to isolate the effect of the
OMe-for-Et substitution on KIE. A value of 1.0(1) was
obtained for 2/2-D3 within the limits of error on our previous
measurement. One anticipates that an outer-sphere mechanism
is likely to have a substantial kinetic isotope effect arising from
formation and cleavage of C−H and M−H bonds, respectively,
in the transition state. Indeed, inverse KIE values were
reported for CO2 insertion into terminal hydride complexes of
rhenium or nickel, and the lack of a KIE for 2 and a small
normal KIE for 4 imply that an outer sphere mechanism is
unlikely (Scheme 4).43,56
The KIE near to unity is intriguing since it suggests two
plausible scenarios: (1) hydride transfer from iron to carbon
minimally contributes to the transition state63 or (2) the late
transition state has a similar zero-point energy gap as the
starting material and is reflected in the bond dissociation free
energies of the C−H being formed and the Fe−H bond(s)
being broken. The comparable hydricities of 2 and CO2 are
consistent with the second scenario, although one might
anticipate a larger primary KIE. Consistent with appreciable
hydride transfer in the transition state, we observe an increase
of the KIE to 1.50(4) upon reducing the steric crowding of the
pocket as in 4/4-D3. These KIE results and the observed first
order dependence on [CO2] can be rationalized based on the
relative rates for an initial association of the CO2 with the iron
center as compared to the hydride transfer step, although any
changes in hydride coordination during the reaction will
complicate this analysis. Specifically, reduced steric pressure
around the Fe center enhances the rate of CO2 binding relative
to hydride transfer, resulting in a greater observed contribution
of the Fe−H bond cleavage and C−H bond forming to the
overall activation barrier. Having exhausted the kinetic
the hydricity of 2 as comparable to that of borane and
consistent with precedent.61 Given that ΔGH (HCOO ) = 44
−
−
kcal·mol−1 and the estimated hydricity for 2, the reaction of 2
with CO2 requires formate coordination to compensate for an
otherwise endergonic CO2 insertion in the Fe−H bonds in 2.
Previously, the bond dissociation free energy for an Fe−
OC(O)H bond in a mononuclear iron complex was estimated
at −8 kcal/mol, and a similar bond strength would be sufficient
to favor the observed CO2 to formate conversion here.62
Kinetic Isotope Effect for CO2 Insertion. We measured
the kinetic isotope effect (KIE) for the first CO2 insertion step
to investigate the contribution of the hydride transfer in
transition state. Deuteride complexes, 2-D3 and 4-D3, were
prepared by the reaction of LiBEt3D with the bromide
complexes, 1 and 3. Previously, we communicated the KIE
of 1.08(9) for reaction 2 with CO2 in toluene (Scheme 4).50
Scheme 4. Hydride Complexes with Reported KIE Values
for CO2 Insertion
Figure 5. Hydride insertion mechanism for the conversion of 2·CO2 to 5a showing the rearrangement of the Fe3H3 core. Free energies are given in
kcal mol−1. Inset figures display ball-and-stick configuration of metal clusters. C, N, O, H, and Fe are represented as gray, blue, red, white, and
orange spheres. Bonds formed during the reaction are represented as sky-blue sticks.
H
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Scheme 5. Proposed Associative Pathway for H2/CO Exchange
methodologies at our disposal, we sought to complement and
elucidate the mechanistic details using density functional
methods.
differing in energy by 11.0 kcal mol−1. The higher energy
intermediate (ΔG = 7.1 kcal mol−1 relative to 2·CO2)
comprises a monodentate terminal formate, whereas the
lower energy intermediate (ΔG = −3.9 kcal mol−1) results
from a rotation of the formate with the formate bound in a μ-
1κO,2κO′ or μ-1,3 mode. These two intermediates evolve to
virtually indistinguishable transition states (ΔG‡ = 8.5 kcal
mol−1), resulting in the formation of the product 5a (ΔG =
−4.5 kcal mol−1 relative to 2·CO2).
Although the higher energy reaction pathway was not
explored in detail, both mechanisms involve association of an
oxygen atom from CO2 with one of the iron atoms in the
Fe3H3 core. Furthermore, the calculated Fe−O interaction at
TS1 is consistent with the observed insensitivity of the reaction
to exogenous Lewis acids insofar as an intramolecular one is
present. We surmise that the sterics of the pocket which
influence the Fe−O interactions in TS1 and TS2 lead to the
differences between the KIE for reaction of 2 vs 4 with CO2.
The reduced steric bulk of the hydride pocket in 4 as
compared to 2 results in a greater contribution of Fe−H
cleavage and C−H bond formation to TS1 as compared to the
association of CO2 with the cluster and reduces the barrier to
reorientation of the formate to a bridging coordination mode
as it progresses through TS2.
Implications for Reactivity of Polynuclear Iron
Hydrides. From our computational studies, the Fe3H3 cluster
readily distorts to a μ3 mode with minimal energy input,
implying that these weak field ligated metal hydrides are quite
fluxional. We posit that the hydride coordination modes
change in response to the local electronic environment of any
one metal center within the cluster. Focusing on the structure
of TS1, a comparison to our previous report of reaction of 2
highlights this idea. Two hydrides shift coordination modes
upon interaction with the Lewis acidic C atom of the CO2
substrate. In TS1, the lengthening of one Fe−H bond and
formation of an Fe−O interaction occur together with the shift
of a second hydride to the central μ3 position. One can
envision that the interaction with CO2 serves to effectively trap
the μ3-hydride, allowing for the downstream reaction. Such an
interpretation would be consistent with the reaction of 2 with
CO, wherein H2 is reductively eliminated and the final product
contains a μ3 hydride (Scheme 5). One possibility is the
capture of a transient short-lived (μ3-hydrido)-di(μ-hydrido)-
triiron(II) species by either CO2 or CO. Coordination to an Fe
center is envisioned in the case of CO, whereas the interaction
of the Lewis acidic C of CO2 with one μ-hydride leads to Fe−
H bond elongation and ultimate cleavage. Alternatively, the
coordination modes change upon interaction with the
substrate; that is, electronic changes at the Fe centers upon
substrate binding effect the shift. In either model, the substrate
must present a Lewis basic site to stabilize the μ3-hydride
isomer. Indeed, the observed hydricity and the near
concomitant need for a donor is consistent with the lack of
Computationally Derived Reaction Pathway. To
improve our understanding of CO2 insertion by these
trihydride complexes, we computationally explored the hydride
transfer reaction between 2 and CO2 to yield 5a to construct
viable mechanisms for formate formation. Geometry opti-
mizations of 2 and 5a using a broken symmetry wave function
(PBE0/def2-tzvp/D3) yielded structures that were consistent
with the crystallographic data obtained for the two complexes.
Both compounds were found to be weakly antiferromagneti-
cally coupled with J = −116.23 (2) and −27.15 cm−1 (5a). We
therefore utilized a broken symmetry approximation for
transition state searches and the optimization of intermediate
structures.
Attempts to construct reaction pathways for hydride
insertion into CO2 by 2 yielded two viable mechanisms
involving the direct insertion of a hydride into the CO2
molecule (Figure 5). In both viable mechanisms, we find
that CO2 forms a weak van der Waals complex with 2, 2·CO2,
with the CO2 molecule associating within a pocket formed by
two of the ethyl arms of the aromatic rings. There is no
bonding interaction between the two iron atoms and CO2 from
analysis of the Laplacian of the electron density of 2·CO2.
Alternate mechanisms, such as a CO2 adduct formed with the
central C atom of a BDI arm, were not apparent from our
calculations. CO2 insertion with minimal rearrangement of the
Fe3H3 core affords a first transition state that is substantially
higher in energy than that for a mechanism involving a
coordination mode change for one hydride donor (23.8 kcal
mol−1 vs 8.6 kcal mol−1 relative to 2·CO2). There, a ∼7 kcal
mol−1 difference between the experimentally derived activation
free energy and the activation free-energy of the rate-
determining step in the low-energy pathway is a reasonable
deviation from the experiment considering the approximations
used in the computational model.64 H/D KIEs for the first step
of the reaction were estimated by replacing the hydride H with
D in the calculated structures and then employing an Eyring
model to which the tunneling correction of Wigner was
applied.45 The higher energy pathway yielded a calculated H/
D KIE = 5.0, while the lower energy pathway yielded a
calculated H/D KIE = 1.8. Given both its activation energy
and calculated KIEs, we chose to focus on the lower-energy
pathway for further analysis; however, we note that neither
pathway can be ruled out based on the available data and the
approximations/assumptions made in both the computational
model and KIE calculations.
As the CO2 molecule approaches a bridging hydride in the
lower energy pathway, one of the oxygen atoms of CO2
associates with an iron atom with a shift in coordination
mode of a hydride from a μ to a μ3 mode. After traversing TS1,
the reaction branches with two possible intermediates formed,
I
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Accession Codes
reactivity of 2 with other substrates (e.g., Me3SiCCH, C2H2,
and CH3CN), despite evidence that ligand sterics in other
complexes of this ligand (viz. intermediate formed from
Cu3(N2)L1 with O2) do not preclude reaction with
substantially larger substrates (e.g., dihydroanthracene).29
Although we cannot rigorously exclude a rapid pre-equilibrium
between two conformers of the Fe3H3 core (viz. all μ-H and a
less symmetric structure), such a scenario would be challenging
to reconcile with the results reported here and that in prior
reactivity studies of 2. Indeed, the absence of scrambling of the
hydride ligands between 2 and its tri(deuteride) isotopologue
suggests that similar fluxional coordination is not accessible in
this system as compared to the related tri(μ-sulfide)
compound.65 In the case of CO2, hydride transfer is the
ultimate outcome to generate formate, whereas the π-acidity of
the CO donor favors a reductive elimination pathway. Thus,
the coordinative plasticity of the hydrides in this polyiron
polyhydride leads to common structure types from which the
reaction outcome can bifurcate. Such plasticity is likely a
common theme with weak field ligated polyiron sys-
tems28,33,35,36 and underpins the reactivity of the various
reduced states of FeMoco.4 Similar rearrangement of the
hydrides at the E4 state could access a hydride-decorated
dinitrogen bound form of FeMoco to liberate H2.7 In addition,
the observed plasticity in this polyiron system is reminiscent of
the facile migration of H adatoms on metal surfaces, such as
those competent for dinitrogen fixation,66 implying that such
molecular systems are useful models for catalysis at metal
surfaces.
CCDC 2043492 contains the supplementary crystallographic
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 Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Authors
■
Leslie J. Murray − Center for Catalysis and Florida Center for
Heterocyclic Chemistry, Department of Chemistry, University
of Florida, Gainesville, Florida 32611, United States;
orcid.org/0000-0002-1568-958X; Email: murray@
chem.ufl.edu
Jason Shearer − Department of Chemistry, Trinity University,
San Antonio, Texas 78212, United States; orcid.org/
0000-0001-7469-7304; Email: jshearer@trinity.edu
Authors
Dae Ho Hong − Center for Catalysis and Florida Center for
Heterocyclic Chemistry, Department of Chemistry, University
of Florida, Gainesville, Florida 32611, United States
Ricardo B. Ferreira − Center for Catalysis and Florida Center
for Heterocyclic Chemistry, Department of Chemistry,
University of Florida, Gainesville, Florida 32611, United
States
Vincent J. Catalano − Department of Chemistry, University of
Nevada, Reno, Nevada 89557, United States; orcid.org/
0000-0003-2151-2892
Ricardo García-Serres − Université Grenoble Alpes, CNRS,
CEA, BIG, LCBM (UMR 5249), F-38054 Grenoble,
France; orcid.org/0000-0001-5203-0568
CONCLUSION
■
In summary, we demonstrated that the marked effect of metal
ion access toward hydride transfer in an iron hydride cluster
and, notably, the role of minor changes in the sterics from ethyl
in 2 to methoxy in 4 within the secondary coordination sphere
of the metal centers on the specificity toward CO2 over other
substrates. Indeed, reaction of 4 with the [Fe3H3]3+ cluster,
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.inorgchem.1c00244
Notes
The authors declare no competing financial interest.
which shares the D3h-symmetric structure with Fe3H3L1 (2) as
1
ACKNOWLEDGMENTS
̈
confirmed by H NMR and Moßbauer spectra, presented a 20-
■
D. H. H., R. B. F., and L. J. M. acknowledge the National
Institutes of Health (R01-GM123241). Mass spectrometry
data were collected by the University of Florida Mass
Spectrometry Research and Education Center on instrumenta-
tion purchased with an award from the National Institutes of
Health (S10 OD021758-01A1). The content is solely the
responsibility of the authors and does not necessarily represent
the official views of the National Institutes of Health. J. S.
acknowledges National Science Foundation (CHE-1854854).
R. G. S. acknowledges Labex ARCANE (ANR-11-LABX-0003-
01).
fold enhanced rate for the reaction with CO2.
In the reaction for the hydride transfer from 2 to CO2, the
critical role of CO2 binding to an iron center is supported by
marginal KIE and insensitivity of added salts. The donor-free
hydricity of 2, estimated as 50 kcal·mol−1 by the hydride-
deuteride exchange with BD3, is larger than that of formate (44
kcal·mol−1) and supports the key role of Fe−O bond
formation on the overall reaction progress. This experimental
result is confirmed by computational methods, in which the
structure of TS1 arises from formation of a μ3 hydride with the
coordination of O to Fe. These results point to similar effects
in other hydride-bridged polynuclear metal species and offer a
strategy for the designed control of reactivity in such species.
REFERENCES
■
(1) Schilter, D.; Camara, J. M.; Huynh, M. T.; Hammes-Schiffer, S.;
Rauchfuss, T. B. Hydrogenase Enzymes and Their Synthetic Models:
The Role of Metal Hydrides. Chem. Rev. 2016, 116 (15), 8693−8749.
(2) Reijerse, E. J.; Pham, C. C.; Pelmenschikov, V.; Gilbert-Wilson,
R.; Adamska-Venkatesh, A.; Siebel, J. F.; Gee, L. B.; Yoda, Y.;
Tamasaku, K.; Lubitz, W.; Rauchfuss, T. B.; Cramer, S. P. Direct
Observation of an Iron-Bound Terminal Hydride in [FeFe]-
Hydrogenase by Nuclear Resonance Vibrational Spectroscopy. J.
Am. Chem. Soc. 2017, 139 (12), 4306−4309.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.inorgchem.1c00244.
FT-IR, 1H NMR, 13C{1H} NMR, and ESI-MS spectra of
the products; kinetic analysis; and crystal structures
(PDF)
(3) Appel, A. M.; Bercaw, J. E.; Bocarsly, A. B.; Dobbek, H.; DuBois,
D. L.; Dupuis, M.; Ferry, J. G.; Fujita, E.; Hille, R.; Kenis, P. J. A.;
J
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Kerfeld, C. A.; Morris, R. H.; Peden, C. H. F.; Portis, A. R.; Ragsdale,
S. W.; Rauchfuss, T. B.; Reek, J. N. H.; Seefeldt, L. C.; Thauer, R. K.;
Waldrop, G. L. Frontiers, Opportunities, and Challenges in
Biochemical and Chemical Catalysis of CO2 Fixation. Chem. Rev.
2013, 113 (8), 6621−6658.
(21) Honkala, K.; Hellman, A.; Remediakis, I. N.; Logadottir, A.;
Carlsson, A.; Dahl, S.; Christensen, C. H.; Nørskov, J. K. Ammonia
Synthesis from First-Principles Calculations. Science 2005, 307
(5709), 555−558.
(22) Ertl, G. Reactions at Surfaces: From Atoms to Complexity
(Nobel Lecture). Angew. Chem., Int. Ed. 2008, 47 (19), 3524−3535.
(23) Mager-Maury, C.; Bonnard, G.; Chizallet, C.; Sautet, P.;
Raybaud, P. H2-Induced Reconstruction of Supported Pt Clusters:
Metal-Support Interaction versus Surface Hydride. ChemCatChem
2011, 3 (1), 200−207.
(24) Churchill, M. R.; Bezman, S. A.; Osborn, J. A.; Wormald, J.
Synthesis and Molecular Geometry of Hexameric
Triphenylphosphinocopper(I) Hydride and the Crystal Structure of
H6Cu6(PPh3)6. HCONMe2 [Hexameric Triphenylphosphino
Copper(I) Hydride Dimethylformamide]. Inorg. Chem. 1972, 11
(8), 1818−1825.
(4) Khadka, N.; Dean, D. R.; Smith, D.; Hoffman, B. M.; Raugei, S.;
Seefeldt, L. C. CO2 Reduction Catalyzed by Nitrogenase: Pathways to
Formate, Carbon Monoxide, and Methane. Inorg. Chem. 2016, 55
(17), 8321−8330.
(5) Nishibayashi, Y. Recent Progress in Transition-Metal-Catalyzed
Reduction of Molecular Dinitrogen under Ambient Reaction
Conditions. Inorg. Chem. 2015, 54 (19), 9234−9247.
(6) Hoffman, B. M.; Lukoyanov, D.; Yang, Z.-Y.; Dean, D. R.;
Seefeldt, L. C. Mechanism of Nitrogen Fixation by Nitrogenase: The
Next Stage. Chem. Rev. 2014, 114 (8), 4041−4062.
(7) Rohde, M.; Sippel, D.; Trncik, C.; Andrade, S. L. A.; Einsle, O.
The Critical E4 State of Nitrogenase Catalysis. Biochemistry 2018, 57
(38), 5497−5504.
(25) Dhayal, R. S.; van Zyl, W. E.; Liu, C. W. Polyhydrido Copper
Clusters: Synthetic Advances, Structural Diversity, and Nanocluster-
to-Nanoparticle Conversion. Acc. Chem. Res. 2016, 49 (1), 86−95.
(26) Andres, H.; Bominaar, E. L.; Smith, J. M.; Eckert, N. A.;
(8) Bernskoetter, W. H.; Hazari, N. Reversible Hydrogenation of
Carbon Dioxide to Formic Acid and Methanol: Lewis Acid
Enhancement of Base Metal Catalysts. Acc. Chem. Res. 2017, 50
(4), 1049−1058.
Holland, P. L.; Mu
̈
nck, E. Planar Three-Coordinate High-Spin FeII
̈
Complexes with Large Orbital Angular Momentum: Mossbauer,
Electron Paramagnetic Resonance, and Electronic Structure Studies. J.
Am. Chem. Soc. 2002, 124 (12), 3012−3025.
(9) Sordakis, K.; Tang, C.; Vogt, L. K.; Junge, H.; Dyson, P. J.;
Beller, M.; Laurenczy, G. Homogeneous Catalysis for Sustainable
Hydrogen Storage in Formic Acid and Alcohols. Chem. Rev. 2018, 118
(2), 372−433.
(10) Boddien, A.; Mellmann, D.; Gärtner, F.; Jackstell, R.; Junge, H.;
Dyson, P. J.; Laurenczy, G.; Ludwig, R.; Beller, M. Efficient
Dehydrogenation of Formic Acid Using an Iron Catalyst. Science
2011, 333 (6050), 1733−1736.
(11) Nielsen, M.; Alberico, E.; Baumann, W.; Drexler, H.-J.; Junge,
H.; Gladiali, S.; Beller, M. Low-Temperature Aqueous-Phase
Methanol Dehydrogenation to Hydrogen and Carbon Dioxide.
Nature 2013, 495 (7439), 85−89.
(12) Jia, H.-P.; Quadrelli, E. A. Mechanistic Aspects of Dinitrogen
Cleavage and Hydrogenation to Produce Ammonia in Catalysis and
Organometallic Chemistry: Relevance of Metal Hydride Bonds and
Dihydrogen. Chem. Soc. Rev. 2014, 43 (2), 547−564.
(13) Klankermayer, J.; Wesselbaum, S.; Beydoun, K.; Leitner, W.
Selective Catalytic Synthesis Using the Combination of Carbon
Dioxide and Hydrogen: Catalytic Chess at the Interface of Energy and
Chemistry. Angew. Chem., Int. Ed. 2016, 55 (26), 7296−7343.
(14) Qiao, J.; Liu, Y.; Hong, F.; Zhang, J. A Review of Catalysts for
the Electroreduction of Carbon Dioxide to Produce Low-Carbon
Fuels. Chem. Soc. Rev. 2014, 43 (2), 631−675.
(15) Ceballos, B. M.; Yang, J. Y. Directing the Reactivity of Metal
Hydrides for Selective CO2 Reduction. Proc. Natl. Acad. Sci. U. S. A.
2018, 115 (50), 12686−12691.
(16) Deutsch, C.; Krause, N.; Lipshutz, B. H. CuH-Catalyzed
Reactions. Chem. Rev. 2008, 108 (8), 2916−2927.
(17) Ye, Y.; Kim, S.-T.; Jeong, J.; Baik, M.-H.; Buchwald, S. L. CuH-
Catalyzed Enantioselective Alkylation of Indole Derivatives with
Ligand-Controlled Regiodivergence. J. Am. Chem. Soc. 2019, 141 (9),
3901−3909.
(18) Lukoyanov, D. A.; Khadka, N.; Yang, Z.-Y.; Dean, D. R.;
Seefeldt, L. C.; Hoffman, B. M. Hydride Conformers of the
Nitrogenase FeMo-Cofactor Two-Electron Reduced State E2(2H),
Assigned Using Cryogenic Intra Electron Paramagnetic Resonance
Cavity Photolysis. Inorg. Chem. 2018, 57 (12), 6847−6852.
(19) Brecht, M.; van Gastel, M.; Buhrke, T.; Friedrich, B.; Lubitz, W.
Direct Detection of a Hydrogen Ligand in the [NiFe] Center of the
Regulatory H2 -Sensing Hydrogenase from Ralstonia eutropha in Its
Reduced State by HYSCORE and ENDOR Spectroscopy. J. Am.
Chem. Soc. 2003, 125 (43), 13075−13083.
(27) DeRosha, D. E.; Mercado, B. Q.; Lukat-Rodgers, G.; Rodgers,
K. R.; Holland, P. L. Enhancement of C-H Oxidizing Ability in Co-O2
Complexes through an Isolated Heterobimetallic Oxo Intermediate.
Angew. Chem., Int. Ed. 2017, 56 (12), 3211−3215.
(28) Yu, Y.; Sadique, A. R.; Smith, J. M.; Dugan, T. R.; Cowley, R.
E.; Brennessel, W. W.; Flaschenriem, C. J.; Bill, E.; Cundari, T. R.;
Holland, P. L. The Reactivity Patterns of Low-Coordinate Iron-
Hydride Complexes. J. Am. Chem. Soc. 2008, 130 (20), 6624−6638.
(29) Lee, Y.; Anderton, K. J.; Sloane, F. T.; Ermert, D. M.; Abboud,
K. A.; García-Serres, R.; Murray, L. J. Reactivity of Hydride Bridges in
High-Spin [3M-3(μ-H)] Clusters (M = FeII, CoII). J. Am. Chem. Soc.
2015, 137 (33), 10610−10617.
(30) Gehring, H.; Metzinger, R.; Braun, B.; Herwig, C.; Harder, S.;
Ray, K.; Limberg, C. An Iron(II) Hydride Complex of a Ligand with
Two Adjacent β-Diketiminate Binding Sites and Its Reactivity. Dalton
Trans. 2016, 45 (7), 2989−2996.
(31) Hein, N. M.; Pick, F. S.; Fryzuk, M. D. Synthesis and Reactivity
of a Low-Coordinate Iron(II) Hydride Complex: Applications in
Catalytic Hydrodefluorination. Inorg. Chem. 2017, 56 (23), 14513−
14523.
(32) Thompson, C. V.; Arman, H. D.; Tonzetich, Z. J. A Pyrrole-
Based Pincer Ligand Permits Access to Three Oxidation States of Iron
in Organometallic Complexes. Organometallics 2017, 36 (9), 1795−
1802.
(33) Ott, J. C.; Blasius, C. K.; Wadepohl, H.; Gade, L. H. Synthesis,
Characterization, and Reactivity of a High-Spin Iron(II) Hydrido
Complex Supported by a PNP Pincer Ligand and Its Application as a
Homogenous Catalyst for the Hydrogenation of Alkenes. Inorg. Chem.
2018, 57 (6), 3183−3191.
(34) Anderton, K. J.; Knight, B. J.; Rheingold, A. L.; Abboud, K. A.;
García-Serres, R.; Murray, L. J. Reactivity of Hydride Bridges in a
High-Spin [Fe3(μ-H)3]3+ Cluster: Reversible H2/CO Exchange and
Fe-H/B-F Bond Metathesis. Chem. Sci. 2017, 8 (5), 4123−4129.
(35) Dugan, T. R.; Bill, E.; MacLeod, K. C.; Brennessel, W. W.;
Holland, P. L. Synthesis, Spectroscopy, and Hydrogen/Deuterium
Exchange in High-Spin Iron(II) Hydride Complexes. Inorg. Chem.
2014, 53 (5), 2370−2380.
(36) Yu, Y.; Brennessel, W. W.; Holland, P. L. Borane B-C Bond
Cleavage by a Low-Coordinate Iron Hydride Complex and N-N Bond
Cleavage by the Hydridoborate Product. Organometallics 2007, 26
(13), 3217−3226.
(37) Hong, D. H.; Knight, B. J.; Catalano, V. J.; Murray, L. J.
Isolation of Chloride- and Hydride-Bridged Tri-Iron and -Zinc
Clusters in a Tris(β-Oxo-δ-Diimine) Cyclophane Ligand. Dalton
Trans. 2019, 48, 9570−9575.
(20) Mebs, S.; Senger, M.; Duan, J.; Wittkamp, F.; Apfel, U.-P.;
Happe, T.; Winkler, M.; Stripp, S. T.; Haumann, M. Bridging Hydride
at Reduced H-Cluster Species in [FeFe]-Hydrogenases Revealed by
Infrared Spectroscopy, Isotope Editing, and Quantum Chemistry. J.
Am. Chem. Soc. 2017, 139 (35), 12157−12160.
K
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
(38) Guillet, G. L.; Sloane, F. T.; Ermert, D. M.; Calkins, M. W.;
H and MeO), and Their Reactions with CO2. Inorg. Chim. Acta 2000,
299 (2), 155−163.
(58) Kang, P.; Cheng, C.; Chen, Z.; Schauer, C. K.; Meyer, T. J.;
Brookhart, M. Selective Electrocatalytic Reduction of CO2 to Formate
by Water-Stable Iridium Dihydride Pincer Complexes. J. Am. Chem.
Soc. 2012, 134 (12), 5500−5503.
̌
̌
Peprah, M. K.; Knowles, E. S.; Cizmár, E.; Abboud, K. A.; Meisel, M.
W.; Murray, L. J. Preorganized Assembly of Three Iron(II) or
Manganese(II) β-Diketiminate Complexes Using a Cyclophane
Ligand. Chem. Commun. 2013, 49 (59), 6635−6637.
(39) Li, H.; Homan, E. A.; Lampkins, A. J.; Ghiviriga, I.; Castellano,
R. K. Synthesis and Self-Assembly of Functionalized Donor-σ-
Acceptor Molecules. Org. Lett. 2005, 7 (3), 443−446.
(40) Ruppel, S.; Schulte to Brinke, C.; Hahn, F. E. Synthesis and
Coordination Chemistry of a Tris(Benzene-o-Dithiolato)- Function-
alized Ligand as a Siderophore Analog. Z. Naturforsch., B: J. Chem. Sci.
2013, 68 (5−6), 551−556.
(41) Wallace, K. J.; Hanes, R.; Anslyn, E.; Morey, J.; Kilway, K. V.;
Siegel, J. Preparation of 1,3,5-Tris(Aminomethyl)-2,4,6-Triethylben-
zene from Two Versatile 1,3,5-Tri(Halosubstituted) 2,4,6-Triethyl-
benzene Derivatives. Synthesis 2005, 2005 (12), 2080−2083.
(42) Fogg, P. G. T. Carbon Dioxide in Non-Aqueous Solvents At
Pressures Less Than 200 KPA; Elsevier, 1992, DOI: 10.1016/C2009-0-
00247-5.
(43) Heimann, J. E.; Bernskoetter, W. H.; Hazari, N.; Mayer, J. M.
Acceleration of CO2 Insertion into Metal Hydrides: Ligand, Lewis
Acid, and Solvent Effects on Reaction Kinetics. Chem. Sci. 2018, 9
(32), 6629−6638.
(44) Espenson, J. Chemical Kinetics and Reaction Mechanisms, 2 ed.;
McGraw-Hill Education: New York, 2002.
(59) Heiden, Z. M.; Lathem, A. P. Establishing the Hydride Donor
Abilities of Main Group Hydrides. Organometallics 2015, 34 (10),
1818−1827.
(60) Heintz, R. A.; Theopold, K. H.; Neiss, T. G. Unusually Large
Isotope Effects on NMR Chemical Shifts of Paramagnetic Organo-
metallic Compounds. Angew. Chem., Int. Ed. Engl. 1994, 33 (22),
2326−2328.
(61) Mock, M. T.; Potter, R. G.; Camaioni, D. M.; Li, J.; Dougherty,
W. G.; Kassel, W. S.; Twamley, B.; DuBois, D. L. Thermodynamic
Studies and Hydride Transfer Reactions from a Rhodium Complex to
BX3 Compounds. J. Am. Chem. Soc. 2009, 131 (40), 14454−14465.
(62) Fong, H.; Peters, J. C. Hydricity of an Fe-H Species and
Catalytic CO2 Hydrogenation. Inorg. Chem. 2015, 54 (11), 5124−
5135.
(63) Gómez-Gallego, M.; Sierra, M. A. Kinetic Isotope Effects in the
Study of Organometallic Reaction Mechanisms. Chem. Rev. 2011, 111
(8), 4857−4963.
(64) Liu, Z.; Patel, C.; Harvey, J. N.; Sunoj, R. B. Mechanism and
Reactivity in the Morita-Baylis-Hillman Reaction: The Challenge of
Accurate Computations. Phys. Chem. Chem. Phys. 2017, 19 (45),
30647−30657.
(65) Buratto, W. R.; Ferreira, R. B.; Catalano, V. J.; García-Serres,
R.; Murray, L. J. Cleavage of cluster iron-sulfide bonds in cyclophane-
coordinated FenSm complexes. Dalton Trans. 2021, 50, 816−821.
(66) Qian, J.; An, Q.; Fortunelli, A.; Nielsen, R. J.; Goddard, W. A.
Reaction Mechanism and Kinetics for Ammonia Synthesis on the
Fe(111) Surface. J. Am. Chem. Soc. 2018, 140 (20), 6288−6297.
(45) Lars, M.; William, H. S., Jr In Reaction Rates of Isotopic
Molecules; Robert E. Krieger Publishing Company: Malabar, FL, 1987.
(46) Ermert, D. M.; Ghiviriga, I.; Catalano, V. J.; Shearer, J.; Murray,
L. J. An Air- and Water-Tolerant Zinc Hydride Cluster That Reacts
Selectively With CO2. Angew. Chem., Int. Ed. 2015, 54 (24), 7047−
7050.
(47) Beinert, H.; Holm, R. H.; Munck, E. Iron-Sulfur Clusters:
̈
Nature’s Modular, Multipurpose Structures. Science 1997, 277 (5326),
653−659.
(48) Dugan, T. R.; Holland, P. L. New Routes to Low-Coordinate
Iron Hydride Complexes: The Binuclear Oxidative Addition of H2. J.
Organomet. Chem. 2009, 694 (17), 2825−2830.
(49) Albers, A.; Demeshko, S.; Pröpper, K.; Dechert, S.; Bill, E.;
Meyer, F. A Super-Reduced Diferrous [2Fe-2S] Cluster. J. Am. Chem.
Soc. 2013, 135 (5), 1704−1707.
(50) Hong, D. H.; Murray, L. J. Carbon Dioxide Insertion into
Bridging Iron Hydrides: Kinetic and Mechanistic Studies. Eur. J. Inorg.
Chem. 2019, 2019 (15), 2146−2153.
(51) Poater, A.; Cosenza, B.; Correa, A.; Giudice, S.; Ragone, F.;
Scarano, V.; Cavallo, L. SambVca: A Web Application for the
Calculation of the Buried Volume of N-Heterocyclic Carbene
Ligands. Eur. J. Inorg. Chem. 2009, 2009 (13), 1759−1766.
(52) Poater, A.; Ragone, F.; Giudice, S.; Costabile, C.; Dorta, R.;
Nolan, S. P.; Cavallo, L. Thermodynamics of N-Heterocyclic Carbene
Dimerization: The Balance of Sterics and Electronics. Organometallics
2008, 27 (12), 2679−2681.
(53) Poater, A.; Ragone, F.; Mariz, R.; Dorta, R.; Cavallo, L.
Comparing the Enantioselective Power of Steric and Electrostatic
Effects in Transition-Metal-Catalyzed Asymmetric Synthesis. Chem. -
Eur. J. 2010, 16 (48), 14348−14353.
(54) Creutz, C.; Chou, M. H. Rapid Transfer of Hydride Ion from a
Ruthenium Complex to C1 Species in Water. J. Am. Chem. Soc. 2007,
129 (33), 10108−10109.
(55) Heimann, J. E.; Bernskoetter, W. H.; Hazari, N. Understanding
the Individual and Combined Effects of Solvent and Lewis Acid on
CO2 Insertion into a Metal Hydride. J. Am. Chem. Soc. 2019.
14110520.
(56) Sullivan, B. P.; Meyer, T. J. Kinetics and Mechanism of Carbon
Dioxide Insertion into a Metal-Hydride Bond. A Large Solvent Effect
and an Inverse Kinetic Isotope Effect. Organometallics 1986, 5 (7),
1500−1502.
(57) Konno, H.; Kobayashi, A.; Sakamoto, K.; Fagalde, F.; Katz, N.
E.; Saitoh, H.; Ishitani, O. Synthesis and Properties of [Ru(Tpy)(4,4′-
X2bpy)H]+ (Tpy = 2,2′:6′,2″-Terpyridine, Bpy = 2,2′-Bipyridine, X =
L
https://doi.org/10.1021/acs.inorgchem.1c00244
Inorg. Chem. XXXX, XXX, XXX−XXX