Proton reduction by phosphinidene-capped triiron clusters

Article abstract of DOI:10.1016/j.jorganchem.2021.121816

Bis(phosphinidene)-capped triiron carbonyl clusters, including electron rich derivatives formed by substitution with chelating diphosphines, have been prepared and examined as proton reduction catalysts. Treatment of the known cluster [Fe₃(CO)₉(μ₃-PPh)₂] (1) with various diphosphines in refluxing THF (for 5, refluxing toluene) afforded the new clusters [Fe₃(CO)₇(μ₃-PPh)₂(κ²-dppb)] (2), [Fe₃(CO)₇(μ₃-PPh)₂(κ²-dppv)] (3), [Fe₃(CO)₇(μ₃-PPh)₂(κ²-dppe)] (4) and [Fe₃(CO)₇(μ₃-PPh)₂(μ-κ²-dppf)] (5) in moderate yields, together with small amounts of the corresponding [Fe₃(CO)₈(μ₃-PPh)₂(κ¹-Ph₂PxPPh₂)] cluster (x = -C₄H₆-, -C₂H₂-, -C₂H₄-, -C₃H₆-, -C₅H₄FeC₅H₄-). The molecular structures of complexes 3 and 5 have been established by X-ray crystallography. Complexes 1–5 have been examined as proton reduction catalysts in the presence of p-toluenesulfonic acid (p-TsOH) in CH₂Cl₂. Cluster 1 exhibits two one-electron quasi-reversible reduction waves at –1.39 V (ΔE = 195 mV) and at –1.66 V (ΔE = 168 mV; potentials vs. Fc⁺/Fc). Upon addition of p-TsOH the unsubstituted cluster 1 shows a first catalytic wave at –1.57 V and two further proton reduction processes at –1.75 and –2.29 V, each with a good current response. The diphosphine-substituted derivatives of 1 are reduced at more negative potentials than the parent cluster 1. Clusters 2–4 each exhibit an oxidation at ca. +0.1 V and a reduction at ca. –1.6 V; for 4 conversion to a redox active successor species is seen upon both oxidation and reduction. Clusters 2–4 show catalytic waves in the presence of p-TsOH, with cluster 4 exhibiting the highest relative catalytic current (i^cat/i⁰ ≈ 57) in the presence of acid, albeit at a new third reduction process not observed for 2 and 3. Addition of the dppf ligand to the parent diphosphinidene cluster 1 gave cluster 5 which exhibited a single reduction process at –1.95 V and three oxidation processes, all at positive values as compared to 2–4. Cluster 5 showed only weak catalytic activity for proton reduction with p-TsOH. The bonding in 4 was investigated by DFT calculations, and the nature of the radical anion and dianion is discussed with respect to the electrochemical data.

Full text of DOI:10.1016/j.jorganchem.2021.121816

Journal of Organometallic Chemistry 943 (2021) 121816
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Proton reduction by phosphinidene-capped triiron clusters✩
Ahibur Rahamana,∗, George C. Lisenskyb, Matti Haukkac, Derek A. Tocher d,
Michael G. Richmonde, Stephen B. Colbranf,∗, Ebbe Nordlandera,∗
a Chemical Physics, Department of Chemistry, Lund University, Box 120, SE-221 00 Lund, Sweden
b Department of Chemistry, Beloit College, 700 College Street, Beloit, WI 53511, USA
c Department of Chemistry, University of Jyväskylä, Box 111, FI-40014, Jyväskylä, Finland
d Department of Chemistry, University College London, 20 Gordon Street, London WC1H OAJ, UK
e Department of Chemistry, University of North Texas, Denton, TX 76203, USA
f School of Chemistry, University of New South Wales, Sydney, NSW 2052, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Bis(phosphinidene)-capped triiron carbonyl clusters, including electron rich derivatives formed by sub-
stitution with chelating diphosphines, have been prepared and examined as proton reduction cata-
lysts. Treatment of the known cluster [Fe3(CO)9(μ3-PPh)2] (1) with various diphosphines in refluxing
THF (for 5, refluxing toluene) afforded the new clusters [Fe3(CO)7(μ3-PPh)2(κ2-dppb)] (2), [Fe3(CO)7(μ3-
PPh)2(κ2-dppv)] (3), [Fe3(CO)7(μ3-PPh)2(κ2-dppe)] (4) and [Fe3(CO)7(μ3-PPh)2(μ-κ2-dppf)] (5) in moder-
ate yields, together with small amounts of the corresponding [Fe3(CO)8(μ3-PPh)2(κ1-Ph2PxPPh2)] cluster
(x = -C4H6-, -C2H2-, -C2H4-, -C3H6-, -C5H4FeC5H4-). The molecular structures of complexes 3 and 5 have
been established by X-ray crystallography. Complexes 1–5 have been examined as proton reduction cat-
alysts in the presence of p-toluenesulfonic acid (p-TsOH) in CH2Cl2. Cluster 1 exhibits two one-electron
quasi-reversible reduction waves at –1.39 V (ꢀE = 195 mV) and at –1.66 V (ꢀE = 168 mV; potentials
vs. Fc+/Fc). Upon addition of p-TsOH the unsubstituted cluster 1 shows a first catalytic wave at –1.57 V
and two further proton reduction processes at –1.75 and –2.29 V, each with a good current response. The
diphosphine-substituted derivatives of 1 are reduced at more negative potentials than the parent cluster
1. Clusters 2–4 each exhibit an oxidation at ca. +0.1 V and a reduction at ca. –1.6 V; for 4 conversion
to a redox active successor species is seen upon both oxidation and reduction. Clusters 2–4 show cat-
alytic waves in the presence of p-TsOH, with cluster 4 exhibiting the highest relative catalytic current
(icat/i0 ≈ 57) in the presence of acid, albeit at a new third reduction process not observed for 2 and 3.
Addition of the dppf ligand to the parent diphosphinidene cluster 1 gave cluster 5 which exhibited a
single reduction process at –1.95 V and three oxidation processes, all at positive values as compared to
2–4. Cluster 5 showed only weak catalytic activity for proton reduction with p-TsOH. The bonding in 4
was investigated by DFT calculations, and the nature of the radical anion and dianion is discussed with
respect to the electrochemical data.
Received 7 February 2021
Revised 29 March 2021
Accepted 5 April 2021
Available online 20 April 2021
Keywords:
Triiron
Phosphinidine
Electrocatalysis
Proton reduction
DFT
© 2021 The Author(s). Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
1. Introduction
Considerable attention has been paid to the fact that hydrogen
is a promising energy vector that may function as an alternative
to fossil fuels in the future. In nature, hydrogenase enzymes cat-
alyze the reduction of protons to form dihydrogen and/or the ox-
idation of hydrogen according to the reaction H2 ꢀ 2H+ + 2e−
[1,2]; both the forward and backward reactions are of direct rel-
evance to the use of hydrogen as an energy vector. There are
three types of hydrogenases with different metal-containing ac-
tive sites [3,4], of which the most widely studied are the [Fe-Fe]
✩
Dedicated to Professor Shariff Enamul Kabir on the occasion of this 65th birth-
day, and in recognition of his outstanding contributions to organometallic chemistry
and the development of chemistry in Bangladesh.
∗
Corresponding authors.
E-mail
addresses:
ahibur_rahaman@yahoo.com
(A.
Rahaman),
s.colbran@unsw.edu.au (S.B. Colbran), Ebbe.Nordlander@chemphys.lu.se (E. Nord-
lander).
https://doi.org/10.1016/j.jorganchem.2021.121816
0022-328X/© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
hydrogenases that contain a diiron center supported by CO/CN−
ligands and a dithiolate bridge (the so called H-cluster). This ac-
tive site is connected through one of the iron atoms, via a bridg-
ing cysteinyl sulfur, to an Fe4S4 cluster that acts as an electron-
donor/acceptor and relays electrons via a chain of closely situated
Fe-S clusters [5–7]. A plethora of diiron biomimetics of this ac-
tive site have been prepared and studied [8–10], and more recently
researchers have turned their attention to related tri- and tetra-
iron sulfur-containing clusters as potential bioinspired functional
proton-reduction catalysts [11,12].
Dinuclear iron complexes with thiolate-, sulfide-, selenide- and
telluride bridges have been investigated as electrocatalysts for pro-
ton reduction in organic solvents in the presence of suitable pro-
ton sources [13–17]. The chemistry of electron-poor diphosphido-
bridged diiron carbonyl complexes has been reported [18–27], but
only a few diiron diphosphido complexes have been investigated
as proton reduction catalysts [28–32].
In a recent study, Colbran and co-workers investigated elec-
trocatalytic proton reduction effected by diiron complexes of the
general formula [Fe2(CO)6{μ-P((CH2-η5-C5H4)Fe(η5-C5H5))H}](η5-
C5H4)Fe(η5-C5H5)
=
the ferrocenyl radical), derived from the
Scheme 1. Schematic depiction of the syntheses of clusters 2–5 from the starting
corresponding primary phosphine P((CH2-η5-C5H4)Fe(η5-C5H5))H2
[32]. One of the motivations for this work was the desire to in-
corporate a redox active moiety (a ferrocenyl group) into the lig-
and(s) in order to emulate the electronic communication between
the H-cluster and the iron-sulfur cluster in [FeFe] hydrogenase.
It was found that an important aspect of the proton reduction
catalyzed by these complexes was the likely formation of hydro-
gen via the combination of a metal-bound hydride and a bridge-
head (phosphido-bound) proton, but that the electrocatalytic be-
havior was probably not directly dependent on the presence of the
potentially redox-active ferrocenyl substituents. In further related
studies, we have synthesized and investigated the electrocatalytic
proton reduction properties of dinuclear diphosphido-bridged iron
clusters where the phosphido and phosphinidene units are derived
from the two relatively stable primary phosphines PRH2 (R = –Ph,
material [Fe3(CO)9(μ3-PPh)2] (1).
derivative [Fe3(CO)7(μ3-PPh)(μ3-κ-O,2,3-κ2P-(OPPh)(κ2-dppb)] (2’)
(Scheme 1).
2. Experimental
2.1. General procedures
Unless otherwise stated, solvent purification, reactions, and ma-
nipulations of compounds were carried out under a nitrogen atmo-
sphere using standard Schlenk techniques. Reagent grade solvents
were dried by standard procedures and were freshly distilled prior
to use. All chromatographic separations and ensuing work-up were
carried out in air. Thin layer chromatography was carried out on
glass plates pre-coated with 0.5 mm Merck 60 silica gel. All phos-
phines were purchased from Acros Organics Chemicals Inc. and
used as received. Infrared spectra were recorded on a Nicolet 6700
FT-IR or a Nicolet Avatar 360 FT-IR-spectrometer in solution cells
fitted with calcium fluoride or sodium chloride plates; subtraction
of the solvent absorptions was achieved by computation. Fast atom
bombardment (FAB) mass spectra were obtained on a JEOL SX-102
spectrometer using 3-nitrobenzyl alcohol as matrix and CsI as cal-
ibrant. Proton and 31P{1H}NMR spectra were recorded on a Varian
Unity 500 MHz or a Bruker AMX400 spectrometer. The 1H chemi-
cal shifts were referenced to residual solvent resonances, while the
31P chemical shifts were referenced to external 85% H3PO4. Ele-
mental analyses were performed by the microanalytical laborato-
ries, University College London.
2-methoxy-1,1-binaphthyl) [33].
´
´
Considering the ability of phosphido-bridged diiron com-
plexes to catalyze proton reduction, we were interested in
studying the ability of related iron phosphinidene clusters to
function as electrocatalysts for the same reaction. The chem-
istry of phosphinidene-capped triiron clusters has been relatively
thoroughly studied [34–46], but their electrochemical and electro-
catalytic behaviors remain relatively poorly investigated. Kochi and
co-workers reported the radical anions and dianions of the parent
triiron cluster [Fe3(CO)9(μ3-PPh)2] (1) and its derivatives [39,43].
These authors revealed that 1 undergoes successive, reversible,
one-electron reductions in THF to produce the anion radical [1]•
–
and the corresponding dianion [1]2− at the potentials -0.79 and
-1.30 V vs SCE, respectively, without substantial changes in the
cluster framework. The anion radical [1]• is a labile species that
–
undergoes a series of interesting transformations, especially during
electron-transfer catalysis. In
a related study, Perkinson et al.
2.2. Synthesis of [Fe3(CO)9(μ3-PPh)2] (1)
[41] reported that the diphosphinidene clusters [Fe3(CO)9(μ3-P-p-
´
C6H4X)(μ3-P-p-C6H4X) (X = X´ = NMe2, OCH3, Cl, CF3, CN)] exhibit
Cluster 1 was synthesized by minor modification of previously
reported syntheses [22,33]. A toluene solution (40 mL) of PPhH2
(0.28 ml, 2.5 μmol) was added to iron pentacarbonyl (0.34 ml,
1.7 μmol), and the mixture was heated to 120 ˚C under a ni-
trogen atmosphere for 72 h. The solvent was removed under re-
duced pressure and the residue was chromatographed by TLC on
silica gel. Elution with hexane/CH2Cl2 (9:1 v/v) gave two bands.
The faster moving band gave a small amount of orange microcrys-
tals as a mixture of anti/syn-[Fe2(CO)6{(μ2-P(PhH}2] and the sec-
ond band gave the major product [Fe3(CO)9(μ3-PPh)2](1) (350 mg,
32%) as red crystals after recrystallization from hexane/CH2Cl2 at
4 °C. Characterization data for 1: Anal. Calcd for C21H10Fe3O9P2:
two quasi-reversible one-electron reduction processes. However,
the electrocatalytic activity of these complexes towards proton re-
duction has not been investigated. Here we present the synthesis,
characterization and electrochemical behavior of [Fe3(CO)9(μ3-
PPh)2] (1), and the new electron-rich phosphine derivatives
[Fe3(CO)7(μ3-PPh)2(κ2-dppb)] (2), [Fe3(CO)7(μ3-PPh)2(κ2-dppv)]
(3), [Fe3(CO)7)(μ3-PPh)2(κ2-dppe)] (4) and [Fe3(CO)7)(μ3-PPh)2(μ-
κ2-dppf)] (5). The abilities of clusters 1–5 to act as catalysts
for the electrochemical reduction of protons provided from p-
toluenesulfonic acid (p-TsOH) have been studied. In an attempt
to crystallize cluster 2, we also obtained the phosphinidene-oxido
2

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
2.6. Synthesis of [Fe3(CO)7)(μ3-PPh2)(κ2-dppf)] (5)
C, 39.67 H, 1.59 Found: C, 39.85; H, 1.64 %. IR (ν(CO), CH2Cl2):
.
2073 br, 2042 vs, 2021vs, 2003s, 1990 br and 1957 br cm−1 1H
NMR (CDCl3): δ 7.72 (d, J = 5 Hz, 4H), 7.52 (m, 6H). 31P{1H} NMR
A toluene solution (20 mL) of 1 (58 mg, 0.092 mmol), 1,1ʹ-
bis(diphenylphosphino)ferrocene (dppf, 51 mg, 0.092 mmol) and
a solution of Me3NO (14 mg, 0.182 mmol) in the same solvent
(5 ml) was added dropwise through an addition funnel and un-
der nitrogen flow. Upon addition of Me3NO, the color of the so-
lution gradually turned darker and the reaction mixture was re-
fluxed for 24 h. The solvent was removed under reduced pressure
and the residue was subjected to TLC on silica gel. Elution with
n-hexane/CH2Cl2 (7:3 v/v) developed two bands. The faster mov-
ing band was unreacted 1, the second band afforded [Fe3(CO)7)(μ3-
PPh)2(μ-dppf)] (5) (12 mg, 12%) as red crystals after recrystalliza-
tion from hexane/CH2Cl2 at 4 °C. Characterization data for 5: Anal.
Calcd for C53H38Fe3O7P4: C, 56.15; H, 3.38. Found: C, 56.25; H,
(CDCl3): δ 318 (s).
2.3. Synthesis of [Fe3(CO)7(μ3-PPh)2(κ2-dppb)] (2)
A
solution of
1
(40 mg, 0.063 mmol) and 1,2-
bis(diphenylphosphino)benzene (dppb, 28 mg, 0.063 mmol) in
THF (20 mL) was refluxed for 12 h. The solvent was removed
under reduced pressure and the residue was separated by TLC on
silica gel. Elution with n-hexane/CH2Cl2 (7:3 v/v) developed two
bands. The faster moving band was unreacted 1 and the second
band afforded red microcrystalline [Fe3(CO)7(μ3-PPh)2(κ2-dppb)]
(2) (15 mg, 23%). Characterization data for 2 (2’): Anal. Calc. for
3.26. IR (ν(CO), CH2Cl2): 2027 vs, 1990 vs, 1962 m, 1944 w cm−1
.
.
1H NMR (CDCl3): 1H NMR (CDCl3): δ 7.80 - 7.40 (m, 20H), 7.14
-7.02 (m, 10H), 4.85 (s, 1H), 4.20 (d, J = 45 Hz, 2H), 3.94 (d,
J = 45 Hz, 2H), 3.59 (d, J = 30 Hz, 2H), 3.25 (s, 1H), 31P{1H}
NMR (CDCl3): 260.2 (t, J = 30, 62 Hz), 66.1 (t, J = 32, 60 Hz)
ESI-MS: m/z 1134.14; found; m/z 1134.28 calc. for [M+].
C49H34Fe3O6P4 0.5 CH2Cl2: C, 56.48; H, 3.35. Found: C, 56.55;
H, 3.25. IR (ν(CO), CH2Cl2): 2040 vs, 1996 vs, 1977 m, 1960 m,
1938 br, 1913 w cm−1
.
1H NMR (CDCl3): δ 7.75 (t, J = 5, 15 Hz,
1H), 7.62 – 7.21 (m, 30H), 7.00 (t, J = 5, 15 Hz, 1H), 6.82 (t, J = 10,
15 Hz, 1H), 6.59 (t, J = 10, 15 Hz, 1H). 31P{1H} NMR (CDCl3): δ
401.0 (ddd, J = 20, 123, 141 Hz), 241.3 (ddd, J = 8, 22, 123 Hz),
77.4 (d,d J = 8, 20 Hz), 65.1 (dd, J = 22, 141 Hz). ESI-MS: m/z
1042.22 found; m/z 1042.25 calc. [M+] for 2’ (1010/1009.93543
for 2). Recrystallization from hexane/CH2Cl2 at 4 °C afforded a
few red crystals that were shown to be [Fe3(CO)7(μ3-PPh)(μ3-
κ-O,2,3-κ2P-(OPPh)(κ2-dppb)] (2’) by X-ray crystallography (vide
infra).
2.7. Protonation experiments
Solutions (ca. 1 mL) of the complexes were prepared by dissolv-
ing each of complexes 1-5 in CH2Cl2 to give solutions with the fol-
lowing concentrations: 3.1 mM (1), 1.9 mM (2), 2.0 mM (3 and 4),
1.8 mM (5) To each solution, two equivalents of p-TsOH was added.
The acid-containing solution was instantly transferred to an IR cell
and monitored over time. The FTIR spectra were unchanged in the
presence of acid.
2.4. Synthesis of [Fe3(CO)7 (μ3-PPh)2(κ2-dppv)] (3)
A
solution of
1
(50 mg, 0.078 mmol) and cis-1,2-
2.8. X-ray structure determinations
bis(diphenylphosphino)ethene (dppv, 31.5 mg, 0.078 mmol) in
THF (20 mL) was refluxed for 20 h. The solvent was removed
under reduced pressure and the resultant residue was subjected
to TLC on silica gel. Elution with n-hexane/CH2Cl2 (7:3 v/v) de-
veloped one band, which gave [Fe3(CO)7(μ3-PPh)2(κ2-dppv)] (3)
(60 mg, 79%) as deep red crystals after recrystallization from
hexane/CH2Cl2 at 4 °C. Characterization data for 3: Anal. Calcd
for C45H32Fe3O7P4: C, 55.37; H, 3.30. Found: C, 55.51; H, 3.37.
IR (ν(CO), CH2Cl2): 2039 vs, 1995 vs, 1975 m, 1959 m, 1938 br,
Crystals of 1, 2’, 3 and 5 suitable for X-ray structure analysis
were grown by slow diffusion of hexane into a saturated CH2Cl2
solution at 4 °C. Crystals were immersed in cryo-oil, mounted in
a Nylon loop, and measured at a temperature of 120-170 K. The
X-ray diffraction data were collected on a Bruker Kappa Apex II
Duo or an Agilent Supernova diffractometer using Mo Kα radiation
˚
(λ = 0.71073 A). The APEX2 [47] or CrysAlisPro [48] program pack-
ages were used for cell refinements and data reductions. Struc-
tures were solved by direct methods or by charge flipping using
the SHELXS-97 [49] or SUPERFLIP [50] programs. A multi-scan, nu-
merical, or Gaussian absorption correction (SADABS [51] or CrysAl-
isPro [48]) was applied to all data. The solvent of crystallization
(CH2Cl2) in 2’ was slightly disordered. No stable disorder model
could be obtained without strict geometric restraints. Therefore, no
disorder model was used, but the chloride atoms were restrained
so that their Uij components approximate isotropic behavior. Struc-
tural refinements were carried out using SHELXL-97 or SHELXL-
2013 [49]. The hydrogen atoms were positioned geometrically and
1
1917 w cm−1
.
H NMR (CDCl3): δ 7.91-7.88 (m, 2H, CH CH), 7.71
=
– 6.83 (m, 30H). 31P{1H} NMR (CDCl3): δ 402.9 (ddd, J = 18, 111,
145 Hz), 238.4 (ddd, J = 8, 18, 111 Hz), 87.7 (dd J = 8, 18 Hz), 80.4
(dd, J = 18, 145 Hz). ESI-MS: m/z 976.28 found; m/z 976.16 calc.
for [M+].
2.5. Synthesis of [Fe3(CO)7)(μ3-PPh2)(κ2-dppe)] (4)
A THF solution (20 mL) of 1 (30 mg, 0.0471 mmol) and 1,2-
bis(diphenylphosphino)ethane (dppe, 18.7 mg, 0.0471 mmol) was
refluxed for 12 h. The solvent was removed under reduced pres-
sure and the residue was subjected to thin layer chromatography
on silica gel. Elution with n-hexane/CH2Cl2 (7:3 v/v) developed
two bands. The faster moving band was unreacted 1, the second
band afforded [Fe3(CO)7(μ3-PPh)2(κ2-dppe)] (4) (12 mg, 26%) as
red crystals after recrystallization from hexane/CH2Cl2 at 4°C. Char-
acterization data for 4: Anal. Calcd for C45H34Fe3O7P4: C, 55.25; H,
3.50. Found: C, 55.55; H, 3.65. IR (ν(CO), CH2Cl2): 2038 vs, 1995
˚
constrained to ride on their parent atoms, with C-H = 0.95–0.99 A
and Uiso = 1.2 Ueq(parent atom). Crystallographic details are sum-
marized in Table S1 (Supplementary Material).
2.9. Electrochemistry
Cyclic voltammetry experiments were carried out at room tem-
perature in deoxygenated CH2Cl2 or acetonitrile solutions with 0.1
M [NBu4][PF6] as the supporting electrolyte. The working electrode
was a 3 mm diameter glassy carbon electrode that was polished
with 0.3 μm alumina slurry, then washed with the solvent, prior
to each scan. The counter electrode was a platinum wire and a sil-
ver wire was employed as a quasi-reference electrode. All poten-
tials were referenced to the ferrocenium/ferrocene (Fc+/Fc) redox
vs, 1977 m, 1961 m, 1933 br cm−1 1H NMR (CDCl3): δ 1H NMR
.
(CDCl3): δ 7.94 – 6.72 (m, 30H), 2.91-2.68 (m, 2H), 2.52-2.02 (m,
2H). 31P{1H} NMR (CDCl3): δ 400.1 (ddd, J = 24, 117, 139 Hz), 240.1
(ddd, J = 10, 22, 117 Hz), 80.2 (dd J = 10, 22 Hz), 63.4 (dd, J = 24,
139 Hz). ESI-MS: m/z 978.23 found; m/z 978.18 calc. for [M+].
3

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
couple measured in situ at the end of each experiment. A Pine
WaveNow potentiostat was used for all electrochemical measure-
ments. Studies of electrocatalytic proton reduction were carried
out by adding equivalents of p-TsOH. Analyte concentrations and
scan rates are given in the caption to each relevant figure (vide in-
fra).
chemically and magnetically equivalent enantiomeric C2 structure).
Such libration/fluxionality has been studied previously for trinu-
clear carbonyl clusters of the iron triad [60], but was not studied
here.
3.2. Structural studies
2.10. Computational methodology
The relative orientation of the phosphinidene bridgehead
groups and the chelating diphosphine ligand could not be unam-
biguously determined on the basis of the IR and NMR spectro-
scopic studies. Thus, crystallographic studies were carried out on
clusters 1, 2’, 3 and 5. The crystal and structure solution data are
summarized in the supporting information (Table S1). Salient bond
length and angle data are presented in the Figure captions. The
crystal structure of 1 has been reported previously [46] and a view
of the molecular structure is shown in the supporting information
(Figure S1). The metric parameters for the triiron cluster are very
similar to those previously reported [46] and are not discussed fur-
ther here.
All calculations were performed with the hybrid meta
exchange-correlation functional M06 [52], as implemented by the
Gaussian 09 program package [53]. The iron atoms were described
by Stuttgart-Dresden effective core potentials (ECP) and an SDD
basis set [54], while a 6-31G(d’) basis set was employed for the
remaining atoms [55]. The ancillary phenyl groups were replaced
with methyl groups for computational expediency, and Grimme’s
dispersion correction was included in all optimizations [56].
The optimized structures represent a ground-state minimum
based on the Hessian matrix that displayed only positive eigen-
values. The natural charges (Q) and Wiberg bond indices were
computed using Weinhold’s natural bond orbital (NBO) program
(version 3.1) [57,58]. The geometry-optimized structures presented
here have been drawn with the JIMP2 molecular visualization and
manipulation program [59].
The molecular structure of the unusual complex 2’ consists of
an approximately equatorial triiron core with two Fe–Fe bonds
and one open edge (Figure. S2). The triiron unit is capped by the
mutual anti relationship of a triply bridging μ3-PPh ligand and
one phosphinidene-oxido (κ1-O,κ2P-(OPPh)) ligand. The origin of
the oxygen atom that has been inserted into a metal-phosphorus
bond to form 2’ is unclear; it may derive from (partially) oxi-
dized primary phosphine or the presence of adventitious water.
We have previously observed the insertion of an oxygen atom into
a metal-phosphorus bond in the case of a chelating diphosphine
derivative of [H4Ru4(CO)12], viz. 1,1-[H4Ru4(CO)10-(O-DUPHOS)]
3. Results and discussion
3.1. Synthesis and characterization
Heating of THF solutions of [Fe3(CO)9(μ3-PPh)2] (1) with the
diphosphines, dppb, dppv or dppe for 12-20 h gave the com-
plexes [Fe3(CO)7(μ3-PPh)2(κ2-dppb)] (2), [Fe3(CO)7(μ3-PPh)2(κ2-
dppv)] (3) and [Fe3(CO)7)(μ3-PPh)2(κ2-dppe)] (4) in moderate to
good yields (12–79%). Complexes 2, 3 and 4 were characterized
by elemental analysis, mass spectrometry, and by IR, 1H and
31P{1H} NMR spectroscopy, and, in the case of 3, X-ray crystal-
lography. Attempts to crystallize 2 gave the oxygenated deriva-
tive [Fe3(CO)7(μ3-PPh)(μ3-κ1-O,2,3-κ2P-(OPPh)(κ2-dppb)] (2’) (vide
infra). The complexes show ν(C-O) bands in the region 2038–
1913 cm−1, consistent with all terminal carbonyl groups. Shifts to
lower ν(C-O) frequencies by approximately 35–60 cm−1 were ob-
served upon substitution of carbonyl ligands in 1 by the diphos-
phines to afford 2, 3, and 4, consistent with the buildup of electron
density on the iron centres in the Fe3(CO)x core. The 31P{1H} NMR
spectra of clusters 2-4 exhibit two phosphinidene multiplets and
two phosphine multiplets at similar shifts for all three molecules,
consistent with their proposed asymmetric structures.
(DUPHOS
=
(-)-1,2-Bis[(2R,5R)-2,5-dimethylphospholano]benzene
[61]). Furthermore, Shi et al. observed the insertion of oxygen
into a phosphido-metal bond in the complex [Fe2(μ-PPh2)2(CO)6]
[31] and found that this could be suppressed by the rigorous ex-
clusion of dioxygen and water.
The molecular structures of complexes 3 and 5 (Figs. 1 and
2, respectively) comprise an open triangular triiron core, i.e., two
–
Fe Fe bonds, spanned by two triply bridging (capping) phenyl
phosphinidene groups, one ’above’ and one ’below’ the triiron
plane. In complexes 2’ and 3, the bidentate diphosphine ligand
(dppb and dppv, respectively) chelates to a single iron atom, while
–
in 5, the dppf ligand spans an Fe Fe edge, and the structure con-
firms the expected C2 symmetry of the molecule. For complexes 2’
and 3, the chelating diphosphine ligand occupies two axial posi-
tions non-coplanar with the iron triangle; this also appears to be
the conformation in solution as suggested by the 31P NMR spec-
tra of the two complexes. For 2’ and 3, the metal-metal distances
that involve the central iron atom and that which is coordinated by
the diphosphine (Fe1–Fe2) are larger (for 2’; Fe1–Fe2 = 2.8445(6)
Treatment of 1 with dppf and two equivalents of trimethyl
amine-N-oxide in toluene afforded [Fe3(CO)7(μ3-PPh)2(μ-dppf)] (5)
as the major product. Complex 5 was fully characterized by a com-
bination of elemental analysis, mass spectrometry, and by IR, 1H
NMR, 31P{1H} NMR spectroscopy data together with a single crys-
tal X-ray diffraction study. The IR spectrum of 5 exhibits carbonyl
bands in the 2027–1944 cm−1 region, consistent with all carbonyl
ligands being terminally bonded. The 1H NMR spectrum of the
complex reveals peaks in the aromatic region for the phenyl pró-
tons, and the 31P{1H} NMR spectrum displays a triplet at δ 260.2
for two equivalent phosphinidene P-atoms and a triplet at δ 66.1
for two equivalent dppf phosphorus atoms, which is indicative of
a symmetric structure where the dppf ligand bridges the non-
bonded Fe-Fe “edge”, as depicted in Scheme 1 (vide supra). As the
dppf ligand is not expected to lie “in plane” with the triiron moi-
ety, a C2 symmetry is expected for the molecule (vide infra). There
is a possibility of libration (torsional rotation around the axis pass-
ing through the cyclopentadienyl ring centroids) and inversion of
the dppf ligand axis with respect to the triiron core (leading to a
˚
˚
A; 3: Fe1–Fe2 = 2.8525(3) A) than in the parent cluster 1 (Fe–
Feav = 2.7186(4), while the other iron-iron edge is shorter than
˚
than the parent cluster 1 (2’: Fe2–Fe3 = 2.6541(6) A; 3: Fe2–
˚
Fe3 = 2.6849(4) A). The reason for these changes in bond dis-
tances is most likely the electronic influence of the ligands. The
phosphine moieties are not as good π-acceptors as carbonyl lig-
ands. Therefore, buildup of electron density on the phosphine-
substituted iron (Fe1 for both 2’ and 3) means that there is less
need for a strong FeFe interaction with the central iron atom (Fe2
for both clusters); this is in turn compensated by a stronger in-
–
–
teraction between Fe2 and Fe3. The Fe Fe Fe bond angles are
87.384(18), 80.06(9), and 86.758(11) deg. for 2’, 3 and 5, respec-
tively, which are comparable to 81.087(13) deg. for the parent
cluster 1 and to 81.0(3) deg. for the structurally analogous clus-
ter [Fe3(CO)9(μ3-S)2] [62]. The diphosphine P1–Fe–P2 bite angles
are 87.15(3) and 85.450(16) deg. for 2’ and 3, respectively, which
are very similar to the corresponding angles in the structurally
4

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
2
˚
Fig. 1. (a) Molecular structure of [Fe3(CO)7(μ3-PPh)2(κ -dppv)] (3) with selected bond lengths (A) and angles (°); showing the atom numbering scheme. Thermal ellipsoids
are drawn at the 50% probability level and the hydrogen atoms are omitted for clarity. Key bond length and angle data: Fe(1)-P(3) 2.1960(4), Fe(1)-P(4) 2.2095(4), Fe(1)-P(1)
2.2466(4), Fe(1)-P(2) 2.2590(5), Fe(1)-Fe(2) 2.8525(3), Fe(2)-C(3) 1.7829(19), Fe(2)-C(2), 1.783(2) Fe(2)-C(4) 1.7880(19), Fe(2)-P(3) 2.2429(5), Fe(2)-P(4) 2.2543(5), Fe(2)-Fe(3)
2.6849(4), Fe(3)-C(7) 1.773(2), Fe(3)-C(6) 1.8026(18), Fe(3)-C(5) 1.8033(18), Fe(3)-P(4) 2.2145(5), Fe(3)-P(3) 2.2203(5), P(3)-Fe(1)-P(4) 70.960(16), P(1)-Fe(1)-P(2) 85.450(16),
P(4)-Fe(1)-Fe(2) 50.978(12), Fe(3)-Fe(2)-Fe(1) 80.068(9), Fe(1)-P(3)-Fe(3) 107.614(17), Fe(1)-P(3)-Fe(2) 79.965(16), Fe(3)-P(3)-Fe(2) 73.963(15), Fe(1)-P(4)-Fe(3) 107.342(18),
Fe(1)-P(4)-Fe(2) 79.429(16), Fe(3)-P(4)-Fe(2) 73.849(15). (b) A plot of the essential metal-ligand framework of 3; note that the orientation is different from that in (a).
˚
Fig. 2. (a) Molecular structure of [Fe3(CO)7)(μ3-PPh)2(μ-dppf)] (5) with selected bond lengths (A) and angles (°); and showing the atom numbering scheme. Thermal el-
lipsoids are drawn at the 50% probability level. The hydrogen atoms are omitted for clarity: Fe(1)-P(2) 2.2818(5), Fe(1)-P(1) 2.2789(5), Fe(2)-P(1) 2.2352(5), Fe(2)-P(2)
2.2358(5), Fe(1)-Fe(2) 2.6641(4), Fe(3)-P(1) 2.2435(5), Fe(3)-P(2) 2.2307(5), Fe(2)-Fe(3) 2.6704(4), Fe(3)-P(4) 2.2225(5), Fe(3)-P(3) 2.2203(5), P(1)-Fe(1)-P(2) 68.388(16), P(2)-
Fe(3)-P(1) 69.899(17), P(2)-Fe(1)-Fe(2) 53.069(13), Fe(3)-Fe(1)-Fe(2) 86.758(11), Fe(3)-P(2)-Fe(2) 110.23(2), Fe(1)-P(1)-Fe(2) 72.331(16), Fe(3)-P(1)-Fe(2) 109.78(2), Fe(1)-P(1)-
Fe(3) 72.377(15), Fe(1)-P(2)-Fe(3) 72.557(15), P(1)-Fe(1)-Fe(2) 53.075(13). (b) A plot of the essential metal-ligand framework of 5.
–
˚
analogous cluster [Fe3S2(CO)5(dppv)2] [44]. The Fe Pphosphine bond
taining the chelating disphosphine ligand 0.062 A longer than the
˚
distances are comparable to those observed for the same ligands
in related clusters, e.g., [Fe3S2(CO)5(dppv)2] [44] and [Fe3(CO)5(μ-
edt)2(κ2-dppv)] (edt = 1,2-ethanedithiol) [12a].
other Fe-Fe bond, whose distance is 2.667 A. The non-bonding dis-
˚
tance between the two wing-tip iron centers is 3.568 A, and this is
–
in good agreement with the corresponding non-bonding Fe Fe dis-
The bonding in cluster 4 was investigated by electronic struc-
ture calculations. Here we optimized the structure of [Fe3(CO)7(μ3-
PMe)2(κ2-dppe-Me4)], whose DFT structure (A) appears in the sup-
porting information (Figure S3). Species A is similar in nature to
tance found in clusters 1, 3, and 5. Table S2 in the supporting infor-
mation reports selected Natural charges (Q) [63], bond distances,
and Wiberg bond indices (WBIs) [64] in A. The WBIs for the two
–
Fe Fe bonds are 0.28 and 0.33, and the smaller WBI corresponds
–
the solid-state structure of cluster 3 with the Fe Fe bond con-
to the Fe—Fe bond containing the chelating diphosphine ligand. A
5

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
Table 1
Summary of electrochemical data from cyclic voltammetry of complexes 1–5.
Reduction processes E1/2/V vs Fc+/0
(ꢀEp/mV)
Oxidation processes E1/2/V vs Fc+/0
Complex
(ꢀEp/mV)
[Fe3(CO)9(μ3-PPh)2] (1)
–1.39 (195)
–1.66 (168)
–1.57 (395)
–1.57 (425)
–
[Fe3(CO)7(μ3-PPh)2(κ2-dppb)] (2)
[Fe3(CO)7(μ3-PPh)2((κ2-dppv)] (3)
[Fe3(CO)7)(μ3-PPh)2(κ2-dppe)] (4)
+0.11 (259)
+0.12 (409)
+0.08 (159)
+0.36 (169)a
+0.23 (130)
+0.54 (159)
+0.93 (130)
a
a
–1.55 (199)
–1.88 (160)
–1.89 (130)
[Fe3(CO)7)(μ3-PPh)2(μ-κ2-dppf)] (5)
Not a fully chemically reversible process (ireverse/iforward < 1.0).
a
WBI of 0.05 was computed for the wingtip irons, supporting the
absence of any significant bonding interaction between these iron
centers.
The composition of the HOMO and LUMO in species A was also
examined, and these orbitals are depicted in Figure S1. The HOMO
exhibits significant in-phase bonding between the iron centers in-
–
volved with the pair of Fe Fe bonds. One-electron oxidation of
the cluster (i.e. removal of a bonding electron from the HOMO) is
expected to weaken the bonding within of the Fe3(PMe)2 cluster
core thereby promoting possible rearrangement. The orbital inter-
action between the non-bonded wingtip iron centers is clearly an-
tibonding in nature and in keeping with the observed nido polyhe-
dron found in this genre of trimetallic clusters. The LUMO is com-
–
posed of antibonding Fe Fe bonds that are also antibonding with
respect to the capping phosphinidene ligands. Electron addition to
the LUMO is expected to promote the expansion, and possibly scis-
2
–
sion, of the Fe Fe(CO)(PP) bond (where PP = κ -dppe-Me4) within
the cluster. One-electron oxidation of the cluster (i.e. removal of
a bonding electron from the HOMO) is expected to weaken the
bonding within the Fe3(PMe)2 cluster core, thereby promoting pos-
sible rearrangement. The cyclic voltammetry of cluster 4 support
the proposed rearrangement of the cluster following both one-
electron reduction and one-electron oxidation (vide infra).
Fig. 3. CVs of [Fe3(CO)9(μ3-PPh)2] (1, 2 mM) in dichloromethane - 0.1M [NBu4][PF6]
at scan rates 10, 25, 50, 100, 250, 500 mV s−1 recorded at a glassy carbon work-
ing electrode. Potentials are relative to the ferrocene-ferrocenium couple. The inset
shows a plot of peak current vs square root of scan rate and the equal forward and
reverse currents indicate chemical reversibility for the reductions.
3.3. Electrochemical studies
The cyclic voltammetry of complexes 1–5 was investigated,
primarily to determine reduction potentials that might be re-
lated to electrocatalytic proton reduction activity (vide infra).
Dichloromethane was used as the primary solvent for these stud-
ies because dichloromethane is an innocent solvent in the context
of electrochemistry of metal carbonyl complexes. In the follow-
ing commentary, all potential data, including peak-to-peak sepa-
rations for couples, is taken from cyclic voltammograms at scan
rate 100 mV s−1 and is referenced against the Fc/Fc+ couple. The
potentials of the important redox processes observed in the cyclic
voltammograms are summarized in Table 1.
shows two quasi-reversible one-electron reduction processes at –
0.97 V (ꢀEp = 155 mV) and –1.51 V (ꢀEp = 145 mV), significantly
less negative than cluster 1 in CH2Cl2.
The electrochemical properties of the diphosphine-substituted
clusters 2 and 3 were investigated in CH2Cl2. Clusters 2 and
3 (Fig. 4) each exhibit one quasi-reversible oxidation, at +0.11
(ꢀEp = 259 mV) and +0.12 (ꢀEp = 409 mV), respectively, and
one quasi-reversible reduction at –1.57 (ꢀEp = 395 mV) and –
1.57 (ꢀEp = 425 mV), respectively. No additional features were ob-
served in the CVs from –2.5 to +1.0 V. Addition of the diphosphine
ligand in clusters 2 and 3 not only results in shifting of the reduc-
tion potentials to more negative values compared to 1, but now ox-
idative chemistry is seen within the solvent window. These obser-
vations are consistent with the increasing electron density within
the cluster cores of 2 and 3 compared to the parent cluster 1 (cf.
discussion above).
The electrochemistry of [Fe3(CO)9(μ3-PPh)2] (1) was investi-
gated in CH2Cl2, and also in THF. Very similar electrochemical be-
haviours for 1 were observed in both solvents. In CH2Cl2, cluster 1
exhibited two quasi-reversible, one-electron reduction processes at
–1.39 V (ꢀEp = 195 mV) and at –1.66 V (ꢀEp = 168 mV) (Fig. 3).
Plots of the peak currents against v1/2 are linear (see the insets
in Fig. 4) in accord with the Randles-Sevcik equation and indi-
cate that all redox processes are for species freely diffusing in so-
lution. This compares well with the behaviour observed by Ohst
and Kochi for the same complex [43]. Similarly, Perkinson et al.
[41] have reported that the clusters [Fe3(CO)9(μ3-P-p-C6H4X)(μ3-
Surprisingly, the electrochemistry of cluster 4 is more com-
plex than that of clusters 1–3. CVs of 4 to positive potentials
exhibit a quasi-reversible primary oxidation couple at +0.08 V
(ꢀEp = 155 mV) followed by a second oxidation couple at +0.36 V,
which diminishes in current relative to the primary oxidation pro-
cess with increasing scan rate (Fig. 5). To negative potentials, the
CVs of cluster 4 reveal a quasi-reversible primary reduction cou-
ple at –1.55 V followed by a second reduction couple at –1.88 V,
´
P-p-C6H4X) (X = X´ = NMe2, OCH3, Cl, CF3, CN)] exhibited two
quasi-reversible one-electron reduction processes. We have ob-
served the same behavior in cluster [Fe3(CO)9(μ3-Te)2] [65], which
6

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
Fig. 4. CVs of (a) [Fe3(CO)7(μ3-PPh)2(κ2-dppb)] (2, 2 mM)) with scan rates 25, 50, 100, 250, 500, 1000 mV s−1 and (b) [Fe3(CO)7{(μ3-PPh}2(κ2-dppv)] (3, 2 mM) with scan
rates 10, 25, 50, 100, 250, 500 mV s−1; other conditions as for Fig. 4. The insets show a plot of peak current vs square root of scan rate, and the unequal forward and reverse
currents and magnitudes of the plot slopes indicate less chemical reversibility for the reduction.
Fig. 6. CVs of [Fe3(CO)7{(μ3-PPh}2(μ-dppf)] (5) (2mM) with scan rate 10, 25, 50,
100, 250, 500, 1000, 2000 mV s−1; other conditions as for Fig. 4. The inset shows a
plot of peak current vs square root of scan rate and the equal forward and reverse
Fig. 5. CVs of [Fe3(CO)7{(μ3-PPh}2(κ2-dppe)] (4) (2mM) with scan rate 25, 50, 100,
250, 500, 1000, 2000 mV s−1; other conditions as for Fig. 4. The insets show a plot
currents indicate chemical reversibility for the reduction and oxidations.
of peak current vs square root of scan rate and the forward and reverse currents
indicate less chemical reversibility for the second reduction and the second oxida-
supported by the DFT computational results; the nature of the re-
tion.
duced forms of 4 is discussed below.
The cyclic voltammogram of the diphosphinoferrocene-
substituted triiron cluster
5
showed three consecutive quasi-
130 mV), +0.54
which loses current relative to the primary reduction current as
the scan rate increases. The current dependencies of the secondary
reduction and oxidation couples with scan rate indicate that these
processes arise from species produced by a chemical step follow-
ing the reduction and the oxidation of 4, respectively (at faster
scan rates there is less time for the chemical step to occur). Re-
arrangements of cluster 4 upon both reduction and oxidation are
reversible oxidation waves at +0.23 (ꢀEp
=
(ꢀEp = 159 mV) and +0.93 V (ꢀEp = 120 mV) and one quasi-
reversible reduction wave at –1.89 V (ꢀEp = 130 mV) (Fig. 6).
Substitution of the diphosphinoferrocene, dppf, ligand at the base
of the triiron cluster not only results in the first reduction poten-
tial shifting to more negative values compared to 1–4, but three
oxidation couples appear within the solvent window. The first and
7

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
tion waves increased in current and shifted to more negative po-
tential. Limiting catalytic currents were not reached with the max-
imum p-TsOH (45 equiv.) used.
Kochi and co-workers have reported that the protonation of the
dianion 12– occurs directly at the phosphinidene cap(s) and leads
to the formation of [Fe3(CO)9(μ3-PPh)(μ2-PPhH)]– at –78 °C [43].
Upon addition of up to 45 equivalents of acid, the first and sec-
ond reduction did not show a positive shift in potential. A shift
of these reductions in the positive direction would have been ex-
pected in the event that either of the species being reduced was
protonated prior to reduction. The observed electrochemical re-
sponse is suggestive of the first and second catalytic proton reduc-
tion steps involving formation of the monoanion 1– and the di-
anion 12–, respectively. Most probably subsequent protonation oc-
curs, initially leading to formation of [Fe3(CO)9(μ3-PPh)(μ2-PPhH)]•
or of [Fe3(CO)9(μ3-PPh)(μ2-PPhH)]–, respectively, and subsequent
spontaneous electrochemical reduction of the protonated species,
protonation again, and reductive elimination (of hydrogen) steps
would then lead to hydrogen evolution and regeneration of the cat-
alytically active 1– and/or 12– anions.
The ability of clusters 2 and 3 to act as proton reduction cata-
lysts was assayed next (Fig. 8). Both clusters show similar electro-
chemical responses to increasing amounts of acid (p-TsOH). As the
acid concentration increased, the reduction couple lost chemical
reversibility within the addition of one equivalent of acid and, with
increasing acid, gave rise to a catalytic wave for proton reduction
at ca. –1.84 V for 2 and at –1.75 V for 3. The catalytic wave reached
Fig. 7. CVs of complex [Fe3(CO)9(μ3-PPh)2] (1) (2 mM) in CH2Cl2-0.1 M [NBu4][PF6]
in the absence and presence of up to 45 molar equivalents of p-TsOH; scan rate
0.50 Vs–1, 3.0 mm diameter minidisk glassy carbon working electrode, potential
relative to the ferrocene/ferrocenium couple. See Supplementary Information for an
MPEG4 movie of the electrochemical response for 1 during proton reduction.
a limiting current ratio of ipcat/ip ≈ 10 at ca. 32 equivalents of p-
0
0
TsOH for 2 and of ipcat/ip ≈ 9 at ca. 13 equivalents of p-TsOH for
last oxidation couples can be attributed to oxidation of the triiron
centre as the middle couple at +0.54 V is almost certainly the
FeII/III couple of the ferrocene core of the coordinated dppf ligand.
The unreacted, free, dppf ligand shows an irreversible oxidation
at +0.20 V, which becomes reversible and shifts to more positive
potential at ca. +0.5–0.69 V when coordinated to a metal center
[66]. The chemical reversibility of the oxidation and reduction
couples was unchanged at all scan rates.
3. A second catalytic wave appears with increasing acid concen-
tration, at ca –2.1 V for 2 and at ca. –2.0 V for 3, which shows a
strong shift to negative potential with increasing acid concentra-
tion such that it moves out of the CV window at the highest acid
concentrations employed (see Fig. 8); an improved electrochemical
response is observed at these higher acid concentrations due to
better activation of the working electrode. The oxidation couples
of 2 and 3 sharpen, but are otherwise unperturbed, with increas-
ing amounts of acid. The sharpened response is again consistent
with better activation of the glassy carbon electrode with higher
concentrations of acid.
3.4. Electrocatalytic studies
A qualitative survey of the electrocatalytic activity of each com-
plex towards proton reduction was performed by cyclic voltam-
metry in the presence of p-TsOH. As the pKa of p-TsOH in
dichloromethane is unknown and only quasi-reversible electro-
chemical responses were observed for the clusters, quantitative
analyses of the proton reduction activity is not possible. In all
cases, blank runs of p-TsOH at the highest concentration used were
made (cf. Supplementary Material, Figures S4-S8). In all blank runs,
a small negative peak at ca –2.4 V was observed, and this reduc-
tive peak may be assumed to contribute to any reductive waves
at potentials lower –2.2 V, complicating any analysis of current re-
sponse at such negative potentials. A comparison between clusters
1–5 is indicative of their relative activity.
The electrochemical behaviour of 4 with increasing equivalents
of p-TsOH was also examined (Fig. 9). As increasing amounts of
p-TsOH were added, the cathodic currents of the first and sec-
ond reduction couples increased, consistent with proton reduction
0
catalysis. Current maxima were reached with ipcat/ip ≈ 3 at ca.
0
six equivalents of p-TsOH and ipcat/ip ≈ 20 at ca 32 equivalents
p-TsOH for the first and second catalytic waves (labelled ꢀ and
◦ in Fig. 9), respectively. Beyond 13 equivalents of added p-TsOH,
a new process (labelled ꢁ in Fig. 9) grows in and rapidly domi-
nates the catalytic wave(s) for proton reduction. The reduction po-
tential of the new (third) reduction process moves more negative
0
with increasing p-TsOH concentration and gives ipcat/ip ≈ 57 at
45 equivalents p-TsOH. This relative catalytic current is the high-
est observed for clusters 1–5 and, therefore, cluster 4 is the most
active of these catalysts in terms of rate.
Fig. 8 depicts the complicated response that was observed in
CVs of [Fe3(CO)9(μ3-PPh)2] (1) in the presence of p-TsOH. Beyond
three molar equivalents of acid added to 1, four successive waves
for catalytic reduction of protons are observed in the reductive
region. As increasing amounts of p-TsOH were added: (i) the re-
versibility of the first and second reductions was lost; (ii) the ca-
thodic currents of the first and second reduction couples (labelled
ꢀ and × in Fig. 7, respectively) increased concomitant with (iii)
a loss of reversibility and (iv) a shift to higher potential; (v) two
new peaks with high cathodic current (labelled ꢀ and ◦ in Fig. 7)
appear. Of note, the first catalytic reduction appears at –1.57 V,
which is a less negative potential than those observed for the
diphosphine-triiron cluster derivatives to be described next. With
increasing acid concentration, the second and third catalytic reduc-
The observation of increasing catalytic currents for proton re-
duction at the first and second reductions of 4 as increasing p-
TsOH is added indicates that both 4• and 4∗2– (the reduction
–
product of the species (4∗–) formed by the chemical step that
follows formation of 4•–; vide supra), are catalytically competent
species. However, the rates of catalysis by both 4• and 4∗2– satu-
–
rate at low equivalents of p-TsOH (by 6 and 32 equiv., respectively;
see above). Beyond 13 equiv. of p-TsOH, the newly appeared, third
reduction potential revealed at ca. –2.07 V relates to the process
that dominates the current for proton reduction. The requirement
for a large excess of acid and the prior formation of the dianion
8

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
Fig. 8. Cyclic voltammograms of complexes (a) [Fe3(CO)7(μ3-PPh)2(κ2-dppb)] (2, 2 mM) and (b) [Fe3(CO)7(μ3-PPh)2(κ2-dppv)] (3, 2 mM) in the absence and presence of
1–19 molar equivalents of p-TsOH (1 mM solution in CH2Cl2 supporting electrolyte [NBu4][PF6], scan rate 0.25 Vs–1, glassy carbon electrode, potential vs Fc+/Fc).
competent for H2-evolution (leading to the first and second cat-
alytic waves, respectively).
The nature of the one- and two-electron reduction products
originating from species A was explored in more detail by DFT
calculations. The optimized structures of the reduced species 2A−
and 3A2− were analyzed for their net spin densities [67]. Fig. 10
shows the spin density plots for 2A− and 3A2−. The addition
of one electron to A yields the radical anion 2A− whose struc-
–
ture reveals a significant expansion of the Fe Fe(CO)(PP) bond in
the starting cluster [Fe3(CO)7(μ3-PMe)2(κ2-dppeMe4)]. The initial
˚
˚
–
Fe −Fe(CO)(PP) bond distance of 2.729 A in A increases to 3.328 A in
2
–
A , resulting in a weakening of this Fe Fe bond. The WBI for the
–
Fe Fe(CO)(PP) bond decreases by 50% (WBI = 0.14) upon electron
–
accession. The other Fe Fe bond and the internuclear distance be-
tween the non-bonded, wingtip iron centers in 2A− are not signifi-
cantly affected based on WBIs of 0.31 and 0.05. These latter values
closely mirror the WBIs computed for neutral A. This regiospecific
elongation of the Fe Fe(CO)(PP) bond in 2A− helps eliminate steric
–
–
congestion between the metallic centers in the Fe Fe(CO)(PP) vec-
tor. The spin density plot of 2A− confirms that the majority of spin
is located on the iron atom of the Fe(CO)(PP) moiety, along with
a minor contribution from the CO ligand in the Fe(CO)(PP) moiety.
The uptake of a second electron furnishes the dianion that can ex-
ist as a diamagnetic singlet (A2−) or a triplet dianion (3A2−). Both
species were successfully optimized, and the triplet dianion was
computed to lie 14.1 kcal/mol (ꢀG) lower in energy than the cor-
responding singlet dianion. The large increase in the spin density
at the iron atom and CO ligand of the Fe(CO)(PP) moiety in 3A2−
indicates that the addition of the second electron occurs predomi-
nantly at these sites. The iron site in the Fe(CO)(PP) moiety of the
reduced species 2A− and 3A2− is electron rich and can serve as the
initial site for protonation in the presence of added acid, followed
by the ultimate release of H2.
Fig. 9. CVs of complex [Fe3(CO)7(μ3-PPh)2(κ2-dppe)] (4) (2 mM) in the absence
and presence of 1-45 molar equivalents of p-TsOH in CH2Cl2, supporting electrolyte
[NBu4][PF6], scan rate 0.25 Vs–1, glassy carbon electrode, potential vs Fc+/Fc).
suggests that the active species is derived from reduction of a pro-
tonation product such as (4 + mH+ + ne)(n-m)– where (n ≥ 3 and
m < n). The available data do not determine between protonation
in the active species at a capping or bridging phosphinidene phos-
phorus or iron atom. Infrared spectra of 4 in dichloromethane so-
lution are unperturbed by addition of acid; i.e., the neutral cluster
4 is not protonated by p-TsOH. It is not possible to characterize the
CVs for cluster 5 with increasing acid (p-TsOH) are presented
in Figure S9. As increasing acid was added, the reduction at
–1.89 V became more irreversible and grew in current reaching a
products of protonation of 4• or 4∗2–, because these species are
–
9

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
Fig. 10. Spin density plots for radical anion 2A− (left) and the triplet dianion 3A2− (right).
maximum icat/i0 ≈ 8 at ca 32 equiv. of acid. Interestingly, the sec-
ond and, particularly, the third oxidation couples of 5 were also
perturbed by the acid, with the anodic currents increasing and the
chemical reversibility decreasing. The result may indicate that the
cluster decomposes upon oxidation in the presence of acid.
tial leads, in general, to more electron rich species that are more
rapidly protonated, and thus faster proton reduction catalysis. The
compromise is that more negative potential corresponds to higher
energy consumption. There are interesting details in these results,
such as the differences and similarities in the behaviours of clus-
ters 2–5, which warrant further investigation.
4. Summary and conclusions
Appendix A. Supplementary data
The bicapped triiron cluster [Fe3(CO)9(μ3-PPh)2] (1) and new
electron rich derivatives of 1 formed by substitution with chelat-
ing diphosphines have been prepared. All clusters are capped by
an anti-arrangement of two triply bridging μ3-PPh phosphinidene
ligands. The diphosphines coordinate in a chelating (dppb, 2; dppv,
3; dppe; 4) or a bridging (dppf, 5) mode, as established by X-ray
crystallography for 3 and 5. The dppb ligand also coordinates in a
chelating mode in the crystal structure of cluster 2’, a derivative
of 2 containing an unusual phosphinidene-oxido ligand formed by
the insertion of an oxygen atom into a metal-phosphorus bond;
the source of the oxygen atom and the mechanism of formation of
2’ are unclear. All complexes have been examined as proton reduc-
tion catalysts in the presence of p-TsOH in CH2Cl2. Cluster 1 un-
dergoes successive, quasi-reversible, 1-electron reductions to pro-
Supplementary data to this article, including an MPEG4 movie
of the electrocatalytic response during proton reduction by
[Fe3(CO)9(μ3-PPh)2] (1), can be found online at https://doi.org/10.
1016/j.jorganchem.2019.XXXX. CCDC entries no 1896754, 1896755
and 1896756 contain the supplementary crystallographic data for
2’, 3 and 5, respectively. Copies of this information may be ob-
tained free of charge from The Director, CCDC, 12 Union Road,
Cambridge, CB2 1EZ, UK (fax: +44-1223-336033; e-mail: de-
posit@ccdc.cam.ac.uk or http://www.ccdc.cam.ac.uk). Atomic coor-
dinates and energies for all DFT-optimized structures are available
upon request (MGR).
Declaration of Competing Interest
duce the anion radical 1• and the triplet dianion 12– at –1.39 V
–
and –1.66 V, respectively. Both species appear competent for catal-
ysis of proton reduction, but the catalytic current at the primary
reduction at –1.39 V is low, indicative of low turnover at this po-
tential. The diphosphine-substituted clusters 2–5 exhibit different
electrochemical and electrocatalytic behaviour. The primary reduc-
tion processes appear at more negative potential than for 1, con-
sistent with the build-up of electron density in the Fe3 cluster
core upon diphosphine substitution. Clusters 2 and 3 show a single
quasi-reversible reduction couple at ca. –1.7 V. In the presence of
p-TsOH, strong catalytic waves for proton reduction are observed
at the potential of this primary reduction. The stronger catalytic
waves upon reduction of 2 and 3 compared to 1 may be correlated
with the build-up of cluster core electron density, and thus more
rapid protonation leading to hydrogen evolution. Cluster 4 shows
three reductive processes; in the presence of p-TsOH, only a weak
catalytic wave at the first reduction potential at –1.6 V was ob-
served and a much larger catalytic current appears at –2.0 V where
the reduction of a successor species produced at the first reduc-
tion occurs. A third, stronger, catalytic wave appears at higher acid
concentration but at very negative voltage where direct reduction
of p-TsOH may also occur. A close scrutiny of the electrocatalysis
experiments for 2 and 3 (Fig. 8) suggests that their electrocatalytic
behaviour is similar to that of 4, but less clear and pronounced.
Cluster 5 shows only a single reduction process at –1.95 V, which
gave only a weak catalytic current in the presence of acid. The re-
sults fit with the notion that reduction at more negative poten-
The authors declare no competing interests.
Acknowledgements
We thank the European Commission for the award of an Eras-
mus Mundus pre-doctoral fellowship to AR and MGR thanks the
Robert A. Welch Foundation (Grant B-1093) for funding. The DFT
calculations were performed at UNT through CASCaM, which is an
NSF-supported facility (CHE-1531468). We also wish to thank Dr.
David A. Hrovat who prepared the spin density plots reported here.
Supplementary materials
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.jorganchem.2021.
121816.
References
[1] C. Tard, C.J. Pickett, Chem. Rev. 109 (2009) 2245–2274.
[2] J.A. Cracknell, K.A. Vincent, F.A. Armstrong, Chem. Rev. 108 (2008) 2439–2461.
[3] P.M. Vignais, B. Billoud, Chem. Rev. 107 (2007) 4206–4272.
[4] M. Frey, ChemBioChem 3 (2002) 153–160.
[5] J.W. Peters, Curr. Opin. Struct. Biol. 9 (1999) 670–676.
[6] Y. Nicolet, C. Cavazza, J.C. Fontecilla-Camps, J. Inorg. Biochem. 91 (2002) 1–8.
[7] J.W. Peters, W.N. Lanzilotta, B.J. Lemon, L.C. Seefeldt, Science 282 (1998)
1853–1858.
10

A. Rahaman, G.C. Lisensky, M. Haukka et al.
Journal of Organometallic Chemistry 943 (2021) 121816
[8] (a) I.P. Georgakaki, L.M. Thomson, E.J. Lyon, M.B. Hall, M.Y. Darensbourg, Coord.
Chem. Rev. 238-239 (2003) 255–266; (b) D.J. Evans, C.J. Pickett, Chem. Soc.
Rev. 32 (2003) 268–275; (c) T.B. Rauchfuss, Inorg. Chem. 43 (2004) 14–26;
[27] P. Das, J.-F. Capon, F. Gloaguen, F.Y. Pétillon, P. Schollhammer, J. Talarmin,
K.W. Muir, Inorg. Chem. 43 (2004) 8203–8205.
[28] M.H. Cheah, S.J. Borg, S.P. Best, Inorg. Chem. 46 (2007) 1741–1750.
[29] M.H. Cheah, S.J. Borg, M.I. Bondin, S.P. Best, Inorg. Chem. 43 (2004) 5635–5644.
[30] R.E. Ginsburg, R.K. Rothrock, R.G. Finke, J.P. Collman, L.F. Dahl, J. Am. Chem.
Soc. 101 (22) (1979) 6550–6562.
˚
(d) L. Sun, B. Akermark, S. Ott, Coord. Chem. Rev. 249 (2005) 1653–1663;
(e) X. Liu, S.K. Ibrahim, C. Tard, C.J. Pickett, Coord. Chem. Rev 249 (2005)
1641–1652; (f) Y. Nicolet, C. Piras, P. Legrand, C.E. Hatchikian, J.C. Fontecilla–
Camps, Structure 7 (1999) 13–23.
[31] Y.-C. Shi, W. Yang, Y. Shi, D.-C. Cheng, J. Coord. Chem. 67 (2014) 2330–2343.
[32] C.G. Suriñach, M. Bhadbhade, S.B. Colbran, Organometallics 31 (2012)
3480–3491.
[9] (a) A. Rahaman, G.C. Lisensky, D.A. Tocher, M.G. Richmond, G. Hogarth,
E. Nordlander, J. Organomet. Chem. 867 (2018) 381–390; (b) S. Ghosh, A. Ra-
haman, K.B. Holt, E. Nordlander, M.G. Richmond, S.E. Kabir, G. Hogarth, Poly-
hedron 116 (2016) 127–135.
[33] A. Rahaman, C.G. Suriñach, A. Ficks, G.E. Ball, M. Bhadbhade, M. Haukka,
L. Higham, E. Nordlander, S.B. Colbran, Dalton Trans 46 (2017) 3207–3222.
[34] V.D. Patel, A.A. Cherkas, D. Nucciarone, N.J. Taylor, A.J. Carty, Organometallics
4 (1985) 1792–1800.
[10] (a) M. Bruschi, C. Greco, M. Kaukonen, P. Fantucci, U. Ryde, L.De. Gioia, Angew.
Chem. 121 (2009) 3555–3558; (b) C. Greco, P. Fantucci, L.De. Gioia, R.S. Bertoa,
M. Bruschi, J. Talarmin, P. Schollhammer, Dalton Trans. 39 (2010) 7320–7329;
(c) C. Greco, G. Zampella, L. Bertini, M. Bruschi, P. Fantucci, L. De Gioia, Inorg.
Chem. 46 (2007) 108–116.
[35] M. Shieh, C.-H. Ho, W.-S. Sheu, B.-G. Chen, Y.-Y. Chu, C.-Y. Miu, H.-L. Liu,
C.-C. Shen, J. Am. Chem. Soc. 130 (2008) 14114–14116.
[36] C.C. Borg-Breen, M.T. Bautista, C.K. Schauer, P.S. White, J. Am. Chem. Soc 122
(2000) 3952–3962.
[37] G.D. Williams, G.L. Geoffroy, R.R. Whittle, A.L. Rheingold, J. Am. Chem. Soc. 107
(1985) 729–731.
[38] M.T. Bautista, P.S. White, C.K. Schauer, J. Am. Chem. Soc. 113 (23) (1991)
8963–8965.
[39] (a) H.H. Ohst, J.K. Kochi, J. Chem. Soc., Chem. Commun. (1986) 121–123; (b)
H.H. Ohst, J.K. Kochi, J. Am. Chem. Soc. 108 (1986) 2908–2916.
[40] M. Shieh, C.-Y. Miu, K.-C. Huang, C.-F. Lee, B.-G. Chen, Inorg. Chem. 50 (16)
(2011) 7735–7748.
[11] C. Tard, X. Liu, D.L. Hughes, C.J. Pickett, Chem. Commun. (2005) 133–135.
[12] (a) A. Rahaman, S. Ghosh, D.G. Unwin, K.B.Holt S.B.-Modi, S.E. Kabir, E. Nord-
lander, M.G. Richmond, G. Hogarth, Organometallics 33 (2014) 1356–1366; (b)
S. Ghosh, G. Hogarth, K.B. Holt, S.E. Kabir, A. Rahaman, D.G. Unwin, Chem.
Commun. 47 (2011) 11222–11224; (c) A. Rahaman, S. Ghosh, S.B. Modi, A.F. Ab-
del-Magied, S.E. Kabir, M. Haukka, M.G. Richmond, G.C. Lisensky, E. Nordlan-
der, G. Hogarth, J. Organomet. Chem. 880 (2019) 213–222.
[13] (a) J.-F. Capon, F. Gloaguen, F.Y Pétillon, P. Schollhammer, J. Talarmin, Coord.
Chem. Rev. 253 (2009) 1476–1494; (b) G.A.N. Felton, C.A. Mebi, B.J. Petro,
A.K. Vannucci, D.H. Evans, R.S. Glass, D.L. Lichtenberger, J. Organomet. Chem.
694 (2009) 2681–2699.
[41] J.L. Perkinson, M.H. Baik, G.E. Trullinger, C.K. Schauer, P.S. White, Inorg. Chim.
Acta. 294 (1999) 140–152.
[14] (a) M.K. Harb, T. Niksch, J. Windhager, H. Görls, R. Holze, L.T. Lockett, N. Oku-
mura, D.H. Evans, R.S. Glass, D.L. Lichtenberger, M. El-Khateeb, W. Weigand,
Organometallics 28 (2009) 1039–1048; (b) M.K. Harb, U.-P. Apfel, J. Kübel,
H. Görls, G.A.N. Felton, T. Sakamoto, D.H. Evans, R.S. Glass, D.L. Lichtenberger,
M. El-Khateeb, W. Weigand, Organometallics 28 (2009) 6666–6675.
[15] L.-C. Song, B. Gai, Z.-H. Feng, Z.-Q. Du, Z.-J. Xie, X.-J. Sun, H.-B. Song,
Organometallics 32 (2013) 3673–3684.
[42] N.A. Pushkarevsky, D.A. Bashirov, T.G. Terent’eva, A.V. Virovets, E.V. Peresyp-
kina, H. Krautscheid, S.N. Konchenko, Russ. J. Coord. Chem. 32 (2006) 416–426.
[43] H.H. Ohst, J.K. Kochi, Inorg. Chem. 25 (1986) 2066–2074.
˚
˚
[44] W. Gao, J. Sun, M. Li, T. Akermark, K. Romare, L. Sun, B. Akermark, Eur. J. Inorg.
Chem (2011) 1100–1105.
[45] R.M. De Silva, M.J. Mays, G.A. Solan, J. Organomet. Chem. 664 (2002) 27–36.
[46] S.L. Cook, J. Evans, L.R. Gray, M. Webster, J. Organomet. Chem. 236 (1982)
367–374.
[47] Bruker AXSAPEX2-Software Suite for Crystallographic Programs, Bruker AXS,
Inc. Madison, WI, USA, 2009.
[16] (a) C.M. Thomas, O. Rüdiger, T. Liu, C.E. Carson, M.B. Hall, M.Y. Darensbourg,
Organometallics 26 (2007) 3976–3984; (b) L.-C. Song, Q.-L. Li, Z.-H. Feng,
X.-J. Sun, Z.-J. Xie, H.-B. Song, Dalton Trans. 42 (2013) 1612–1626.
[17] M.K. Harb, H. Görls, T. Sakamoto, G.A.N. Felton, D.H. Evans, R.S. Glass,
D.L. Lichtenberger, M. El- Khateeb, W. Weigand, Eur. J. Inorg. Chem. 25 (2010)
3976–3985.
[48] AgilentCrysAlisPro, Agilent Technologies inc, Yarnton, Oxfordshire, England,
2014.
[49] G.M. Sheldrick, Acta Cryst A64 (2008) 112–122.
[50] L. Palatinus, G. Chapuis, J. Appl. Cryst. 40 (2007) 786–790.
[51] G.M. Sheldrick, SADABS - Bruker Nonius scaling and absorption correction,
[18] (a) D. Seyferth, T.G. Wood, J.P. Fackler Jr, A.M. Mazany, Organometallics
3
(1984) 1121–1123; (b) D. Seyferth, T.G. Wood, R.S. Henderson, J. Organomet.
Chem. 336 (1987) 163–182.
Bruker AXS, Inc., Madison, Wisconsin, USA, 2012.
[19] (a) D. Seyferth, T.G. Wood, Organometallics
6
(1987) 2563–2567; (b)
[52] Y. Zhao, D.G. Truhlar, Theor. Chem. Acc. 120 (2008) 215–241.
[53] M.J. Frisch, et al., Gaussian 09, Revision E.01, Gaussian, Inc., Wallingford, CT,
USA, 2009.
D. Seyferth, T.G. Wood, Organometallics (1988) 714–718; (c) R.D. Adam,
7
S. Wang, Inorg. Chem. 24 (1985) 4449–4450; (d) R.B. King, F.-J. Wu, E.M. Holt,
Inorg. Chem. 25 (1986) 1733–1734; (e) R.B. King, F.-J. Wu, E.M. Holt,
J. Am. Chem. Soc. 109 (1987) 7764–7775; (f) J.S. McKennis, E.P. Kyba,
[54] D. Andrae, U. Haeussermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim. Acta 77
(1990) 123–141.
Organometallics
S.-T. Liu, H.O.A. Palacios, J.S. McKennis, Organometallics
(h) E.P. Kyba, M.C. Kerby, S.P. Rines, Organometallics
2
(1983) 1249–1251; (g) E.P. Kyba, R.E. Davis, C.N. Clubb,
(1986) 869–877;
(1986) 1189–1194;
[55] (a) G.A. Petersson, A. Bennett, T.G. Tensfeldt, M.A. Al-Laham, W.A. Shirley,
J. Mantzaris, J. Chem. Phys. 89 (1988) 2193–2218; (b) G.A. Petersson,
M.A. Al-Laham, J. Chem. Phys. 94 (1991) 6081–6090.
5
5
(i) T.C. Flood, F.J. Disanti, K.D. Campbell, Inorg. Chem 17 (1978) 1643–1646;
(j) A.L. Rheingold, Acta Crystallogr. C41 (1985) 1043–1045; (k) P.M. Treichel,
W.M. Douglas, W.K. Dean, Inorg. Chem. 11 (1972) 1615–1618.
[56] S. Grimme, S. Ehrlich, L. Goerigk, J. Comp. Chem. 32 (2011) 1456–1465.
[57] A.E. Reed, L.A. Curtiss, F. Weinhold, Chem. Rev. 88 (1988) 899–926.
[58] K.B. Wiberg, Tetrahedron 24 (1968) 1083–1096.
[20] (a) D. Seyferth, R.S. Henderson, L.-C. Song, Organometallics 1 (1982) 125–133;
(b) D. Seyferth, G.B. Womack, R.S. Henderson, M. Cowie, B.W. Hames,
Organometallics 5 (1986) 1568–1575.
[59] (a) M.B. Hall, R.F. Fenske, JIMP2, version 0.091, a free program for the visu-
alization and manipulation of molecules:, Inorg. Chem. 11 (1972) 768–775;
(b)J. Manson, C.E. Webster, M.B. Hall, Texas A&M University, College Station,
TX, 2006, http://www.chem.tamu.edu/jimp2/index.html.
[60] N. Begum, U.K. Das, M. Hassan, G. Hogarth, S.E. Kabir, E. Nordlander,
D.A. Tocher, Organometallics 26 (25) (2007) 6462–6472.
[21] (a) J.D. Lawrence, H. Li, T.B. Rauchfuss, Chem. Commun. (2001) 1482–1483;
(b) L. Hongxiang, T.B. Rauchfuss, J. Am. Chem. Soc. 124 (2002) 726–727; (c)
J.D. Lawrence, L. Hongxiang, T.B. Rauchfuss, M. Bénard, M.-M. Rohmer, Angew.
˚
Chem. Int. Ed. 40 (2001) 1768–1771; (d) S. Ott, M. Kritikos, B. Akermark, L. Sun,
[61] V. Moberg, P. Homanen, S. Selva, R. Persson, M. Haukka, T.A. Pakkanen,
M. Monari, E. Nordlander, Dalton Trans (2006) 279–288.
Angew. Chem., Int. Ed. 42 (2003) 3285–3288; (e) S. Ott, M. Borgström, M. Kri-
˚
tikos, R. Lomoth, J. Bergquist, B. Akermark, L. Hammarström, L. Sun, Inorg.
[62] C.H. Wei, L.F. Dahl, Inorg. Chem. 4 (1965) 493–499.
Chem. 43 (2004) 4683–4692.
[63] J.P. Foster, F. Weinhold, J. Am. Chem. Soc. 102 (1980) 7211–7218.
[64] (a) K.B. Wiberg, Tetrahedron 24 (1968) 1083–1096; (b) I. Mayer, J. Comput.
Chem. 28 (2007) 204–221.
[22] (a) P.M. Treichel, W.K. Dean, W.M. Douglas, Inorg. Chem. 11 (1972) 1609–1615;
(b) R. Bartsch, S. Hietkamp, S. Morton, O. Stelzer, J. Organomet. Chem. 222
(1981) 263–273; (c) J.P. Collman, R.K. Rothrock, R.G. Finke, E.J. Moore, F. Rose–
Munch, Inorg. Chem. 21 (1982) 146–156; (d) P.E. Garrou, Chem. Rev. 81 (1981)
229–266.
[65] A. Rahaman, G.C. Lisensky, J. Browder-Long, D.A Hrovat, M.G. Richmond,
E. Nordlander, G. Hogarth, Dalton Trans 49 (21) (2020) 7133–7143.
[66] (a) S. Ghosh, G. Hogarth, N. Hollingsworth, K.B. Holt, S.E. Kabir, B.E. Sanchez,
Chem. Commun. 50 (2014) 945–947; (b) A.C. Ohs, A.L. Rheingold, M.J. Shaw,
C. Nataro, Organometallics 23 (2004) 4655–4660; (c) D.L. DuBois, C.W. Eigen-
[23] (a) W. Clegg, Inorg. Chem. 15 (1976) 1609–1613; (b) M.R. Adams, J. Gal-
lucci, A. Wojcicki, G.J. Long, Inorg. Chem. 31 (1992) 2–4; (c) R.E. Ginsburg,
R.K. Rothrock, R.G Finke, J.P. Collman, L.F. Dahl, J. Am. Chem. Soc. 101 (1979)
6550–6562.
brot, J.A. Miedaner, J.C. Smart, R.C. Haltiwanger, Organometallics
5 (1986)
1405–1411; (d) B.D. Swartz, C. Nataro, Organometallics 24 (2005) 2447–2451;
(e) C. Nataro, A.N. Campbell, M.A. Ferguson, C.D. Incarvito, A.L. Rheingold,
J. Organomet. Chem. 673 (2003) 47–55; (f) G. Pilloni, B. Longato, B. Corain,
J. Organomet. Chem 420 (1991) 57–65; (g) B. Corain, B. Longato, G. Favero,
D. Ajö, G. Pilloni, U. Russo, F.R. Kreissl, Inorg. Chim. Acta. 157 (1989) 259–266.
[67] In the odd electron species 2A− and 3A2− the integer superscripts 2 and 3
refer to the doublet and triplet spin states, respectively, while the remaining
superscripts define the overall charge on each species.
[24] (a) R.E. Dessy, R. Kornmann, C. Smith, R. Haytor, J. Am. Chem. Soc. 90
(1968) 2001–2004; (b) Y.F. Yu, J. Gallucci, A. Wojcicki, J. Am. Chem. Soc 105
(1983) 4826–4828; (c) M.-H. Baik, T. Ziegler, C.K. Schauer, J. Am. Chem. Soc.
122 (2000) 9143–9154; (d) S.G. Shyu, A. Wojcicki, Organometallics 4 (1985)
1457–1459; (e) A. Wojcicki, Inorg. Chim. Acta 100 (1985) 125–133.
[25] R. Zaffaroni, T.B. Rauchfuss, A. Fuller, L. De Gioia, G. Zampella, Organometallics
32 (2013) 232–238.
[26] J.G.M. Van der Linden, J. Heck, B. Walther, H.-C. Böttcher, Inorg. Chim. Acta 217
(1994) 29–32.
11