Models of the iron-only hydrogenase enzyme: Structure, electrochemistry and catalytic activity of Fe2(CO)3(μ-dithiolate)(μ,κ1,κ2-triphos)

Article abstract of DOI:10.1039/c9dt00700h

A series of diiron bis(2-diphenylphosphinoethyl)phenylphosphine (triphos) complexes Fe₂(CO)₃(μ-dithiolate)(μ,κ¹,κ²-triphos) (1-4) [dithiolate = 1 pdt; 2 edt; 3 adt (R = Bz), 4 (SMe)₂] have been prepar

Full text of DOI:10.1039/c9dt00700h

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: G. Hogarth, D. G.
Unwin, S. Ghosh, F. Ridley, M. G. Richmond and K. B. Holt, Dalton Trans., 2019, DOI:
10.1039/C9DT00700H.
This is an Accepted Manuscript, which has been through the
Volume 45 Number
1 7 January 2016 Pages 1–398
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
Accepted Manuscripts are published online shortly after
An international journal of inorganic chemistry
www.rsc.org/dalton
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-9226
PAPER
Joseph T. Hupp, Omar K. Farha et al.
Efficient extraction of sulfate from water using
framework
a Zr-metal–organic
rsc.li/dalton

Page 1 of 41
Dalton Transactions
Models of the iron-only hydrogenase enzyme: Structure, electrochemiVsietwrAyrticle Online
DOI: 10.1039/C9DT00700H
and catalytic activity of Fe2(CO)3(-dithiolate)(-triphos)
David Unwina, Shishir Ghosha,b, Faith Ridleya, Michael G. Richmondc, Katherine Holt*a
and Graeme Hogarth*b
a Department of Chemistry, University College London, 20 Gordon Street, London, WC1H 0AJ,
U.K
bDepartment of Chemistry, King’s College London, Britannia House, 7 Trinity Street, London,
SE1 1DB, U.K
cDepartment of Chemistry, University of North Texas, 1155 Union Circle, Box 305070,
Denton, TX 76203, USA
Abstract
A series of diiron bis(2-diphenylphosphinoethyl)phenylphosphine (triphos) complexes
Fe2(CO)3(-dithiolate)(-triphos) (1-4) [dithiolate = 1 pdt; 2 edt; 3 adt (R = Bz), 4 (SMe)2]
have been prepared and investigated as biomimics of the diiron site of [FeFe]-hydrogenases. The
triphos ligand bridges the diiron vector whilst also chelating to one iron and exist as a mixture of
basal-basal-apical (bba) and basal-basal-basal (bbb) isomers which differ in the mode of
chelation. In solution the bba and bbb forms do not interconvert on the NMR time scale, but the
bba isomers are fluxional, and at low temperature four forms of 1bba are seen as the
conformations for the trigonal twist of the Fe(CO)2P centre and pdt ring-flipping are frozen.
Crystallographic studies have established bba (pdt) and bbb (adt) ground state conformations
and in both there is a significant deviation away from the expected eclipsed conformation (Lap-
Fe-Fe-Lap torsion angle 0o) by 49.4 and 24.9o respectively, suggesting that introduction of
triphos leads to significant strain and DFT calculations have been used to understand the relative
energies of isomers. The electron rich nature of the diiron centre in 1-4 would suggest rapid
protonation, but while bridging hydride complexes such as [Fe2(CO)3(-pdt)(-triphos)(-
H)][BF4] (1H+) can be formed the process is slow. This behavior is likely a result of the high
energy barrier in forming the initial (not observed) terminal hydride which requires a significant
conformational change in triphos coordination. CV studies show that all starting compounds
oxidize at low potentials and the addition of [Cp2Fe][PF6] to 1 affords [Fe2(CO)3(-pdt)(-
1

Dalton Transactions
Page 2 of 41
triphos)][PF6] (1+) which has been characterised by IR spectroscopy. DFT studiDeOsI:s1u0.g10g3Ve9i/esCwt9ADartTic0le0O70nl0inHe
ground state for 1+ with a partially rotated Fe(CO)2P moiety that yields a weak semi-bridging
carbonyl with the adjacent Fe(CO)P2 group. No reduction peaks are seen for 1-4 within the
solvent window but 1H+ undergoes reduction at -1.7 V. All complexes act as proton-reduction
catalysts in the presence of HBF4∙Et2O. For 1, three separate processes are observed and their
dependence on acid concentration has been probed, and a mechanistic scheme is proposed based
on formation via a CECE process of 1(-H)H which can either slowly release H2 or be further
reduced. Relative contributions of the three processes to the total current were found to be
highly dependent upon the background electrolyte, being attributed to their relative abilities to
facilitate proton transfer processes. While 2 and 4 show similar proton reduction behaviour,
the adt complex 3 is quite different being attributed to facile protonation of nitrogen which is
followed by addition of a second proton at the diiron centre.
Keywords: iron-only hydrogenase, dithiolate, diiron, triphos, fluxionality, hydride
Introduction
Hydrogenases efficiently catalyse both the reduction of protons and oxidation of hydrogen, and
while there are three distinct types, by far the most widely studied are the so-called [FeFe]-
hydrogenases [1]. The active site of these (Chart 1a) feature a diiron centre spanned by a
bridging dithiolate group further ligated by three carbonyls and two cyanides, and being linked
to a [4Fe-4S] cluster via one of the thiolate groups of the latter. Over the past 20 years there has
been considerable effort expended into the development of [FeFe]-hydrogenase biomimics [2],
however, somewhat surprisingly given the tricarbonyl nature of the enzymes active site,
complexes based on an Fe2(CO)3L3(-dithiolate) framework have been relatively under-
explored [3-16], especially in comparison with those of the type Fe2(CO)4L2(-dithiolate) [17-
36]. Rauchfuss and co-workers have studied complexes of the type Fe2(CO)3(1-PR3)(2-
i
dppv)(-dithiolate) (R = Me, Pr, dppv = cis-1,2-bis(diphenylphosphino)ethene)) [6] and
Schollhammer and co-workers have prepared and investigated related phosphite, Fe2(CO)3{1-
P(OMe)3}(2-dppe)(-pdt) [15], and 1,10-phenanthroline, Fe2(CO)3(1-PPh3)(2-1,10-phen)(-
pdt) [16], complexes (Chart 1b-d). In each of these, a chelating ligand is ligated to one iron atom
and the monodentate to the second.
2

Page 3 of 41
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT00700H
X
BnN
Cys
S
X
S
S
(MeO)3P
OC
S
Me3P
OC
S
CO
CO
S
H2O
[4Fe4S]
(a)
S
Fe
Fe
Fe
Fe
PPh2
Fe
Fe
PPh2
OC
NC
CN
CO
OC
OC
Ph2P
Ph2P
C
O
(c)
(b)
X
Fe
BnN
Fe
S
S
CO
N
Ph3P
OC
Ph2
P
S
Fe
Fe
Et2P
OC
S
Fe
CO
OC
N
P
OC
Ph2
(d)
(e)
Chart. (a) Line drawing of active site of an [FeFe]-hydrogenase, (b-e) diiron tricarbonyl models
previously studied
As a result of the increased electron-density at the diiron centre following multiple carbonyl
substitutions, these complexes oxidise at very mild potentials making [Fe2(CO)3L3(-
dithiolate)]+ readily accessible and generally protonate rapidly with even weak acids. Further,
since both CO and substituted ligands are able to move between apical and basal sites, their
Fe2(CO)3L3 cores are highly flexible which can lead to significant structural rearrangements
upon protonation and oxidation. For example, the addition of H+ to Fe2(CO)3(1-PPh3)(2-1,10-
phen)(-pdt) results in rotation of the Fe(CO)2(PPh3) moiety such that the phosphine moves
from an apical to a basal site [16]. Likewise, oxidation of Fe2(CO)3(1-PMe3)(2-dppv)(-edt)
affords [Fe2(CO)3(1-PMe3)(2-dppv)(-edt)]+ in which the Fe(CO)(dppv) centre rotates [5].
While such complexes are good models of the active enzymatic site in [FeFe] hydrogenases,
similar low energy structural rearrangements might not be expected to be so facile when the
diiron centre is embedded into a protein matrix. Here it is more likely that steric factors and
secondary bonding interactions will significantly reduce the internal mobility of the diiron site.
With this in mind and in an attempt to limit, but not completely curtail, the structural flexibility
of the diiron centre, we investigated the coordination chemistry of the tridentate phosphine,
bis(2-diphenylphosphinoethyl)phenylphosphine (triphos), which can potentially both span the
iron-iron bond while also chelate to one iron atom, providing a more rigid Fe2(CO)3L3 core. In
preliminary work we detailed the synthesis and molecular structure of Fe2(CO)3(-pdt)(-
triphos) (1) in which the triphos ligand spanned the diiron centre [37] and showed that it could
3

Dalton Transactions
Page 4 of 41
View Article Online
be protonated by strong acids to afford a hydride-bridged cation. Herein we present a
DOI: 10.1039/C9DT00700H
continuation of these studies and report the synthesis of four diiron triphos complexes which
differ in the nature of the dithiolate-bridge. We have also carried out detailed electrochemical
and electrocatalytic studies of these complexes as proton-reduction catalysts, and the data are
supported by DFT calculations, which facilitate the development of a detailed picture of their
relationship to the active site of [FeFe]-hydrogenases.
Results and discussion
Synthesis and Characterization of Fe2(CO)3(-SXS)(-triphos) (1-4) - Heating toluene
solutions of Fe2(CO)6(-SXS) and triphos for ca. 16 h resulted in isolation of the trisubstituted
complexes Fe2(CO)3(-SXS)(-triphos) (1-4) after work-up (Scheme 1). For three-atom
thiolate-bridged 1 and 3 yields were high (> 90%) and work-up relatively simple. The non-
linked complex, Fe2(CO)6(-SMe)2, also gave a clean reaction but isolated yields were
significantly lower due to the higher solubility of 4 which makes isolation more difficult. The
reaction of Fe2(CO)6(edt) with triphos was quite different; a complex mixture of products
resulted and 2 could only be isolated in low yields through fortuitous crystallization from
CH2Cl2-MeOH. From IR and 31P{1H} NMR spectra of the reaction mixtures, it was clear that a
range of disubstituted products with one uncoordinated phosphorus centre were also products of
this reaction. A number of attempts were made to increase the yields of 2 by; prolonged heating,
the addition of the oxidative-decarbonylation reagent Me3NO and carrying out the reaction in
MeCN; but all either failed to give higher yields or indeed give any of the desired product.
Similarly, reactions of Fe2(CO)6(-SAr)2 (Ar = Ph, p-tol, C6F5) with triphos failed to yield the
targeted tricarbonyl products; the reasons for which are not known but this may be due (in part)
to fragmentation into mononuclear species.
X
X
Ph2
S
S
S
S
1
2
3
4
X = (CH2)3
X = (CH2)2
X = CH2N(Bz)CH2
S2X = (SMe)2
triphos, 110oC
- 3 CO
OC
OC
CO
CO
OC
OC
P
Fe
Fe
CO
Fe
Fe
CO
OC
Ph2P
P
Ph
Scheme 1
4

Page 5 of 41
Dalton Transactions
Characterization of 1-4 as trisubstituted complexes was easily made on the basis oVfiewIARrticle Online
DOI: 10.1039/C9DT00700H
spectroscopy with each showing strong two carbonyl bands at ca. 1950 and 1890 cm-1. In
1
general, the H NMR spectra were complex and uninformative, although the inequivalence of
the two methyl groups in 4 was apparent. Most useful were 31P{1H} NMR spectra. For 1-3, two
isomers were observed at room temperature, each being characterized by two low-field
resonances between  80-90 and a higher field signal between  60-70. We assign these to basal-
basal-apical (bba) and basal-basal-basal (bbb) isomers respectively (Scheme 2), their ratios
varying with the bba isomer predominating for 1 and 3 (ratio ca. 4:1), while for 2 the bbb
isomer is the major species (ratio ca. 3:7). For Fe2(CO)3(-SMe)2(-triphos) (4), only a
single isomer was observed whose assignment was made relative to the chemical shifts of the
all-basal isomer 4bbb.
X
X
Ph2
P
S
S
S
S
OC
OC
CO
PPh2
Fe
Fe
Fe
Fe
OC
Ph2P
CO
OC
Ph2P
P
P
Ph
Ph
bba
basal-basal-apical (
)
bbb
basal-basal-basal (
)
Scheme 2
Confirmation of the co-existence of bba and bbb isomers derives from the molecular structures
of 1.CH2Cl2 (Figure 1) [37] and 3.H2O (Figure 2) which crystallise in their bba and bbb forms,
respectively. Important bond lengths and angles are summarised in Table 1 (and ESI). The
triphos ligand in both products binds the diiron unit in a  fashion, bridging the diiron
centre and chelating to one of the iron atoms. In 1bba, the two chelating phosphorus atoms bind
to the apical and one basal site, while in 3bbb the ligand chelates to both dibasal sites. The angle
formed by the chelate ligands at the iron mono-carbonyl centre of 88.37(3)o and 85.35(5)o in
1bba and 3bbb, respectively, and this points towards the high flexibility of this ligand. While
in most complexes of the type Fe2(CO)6-nLn(-dithiolate) the two iron centres are eclipsed
(torsion angle 0o) in both 1 and 3 there is a significant deviation from this with Pap-Fe-Fe-
COap torsional angles of 49.37o and 24.90o respectively, showing that coordination of the
triphos ligand requires some significant deviation from the prepared (eclipsed) ground state
structure. These structural features may account for the low yields of 3, the more rigid edt
5

Dalton Transactions
Page 6 of 41
View Article Online
ligand being less able to accommodate such changes. Coordination of triphos renDdOeIr: s10.t1h03e9/tCw9DoT00700H
sulfur atoms of the dithiolate-bridge inequivalent, thus completely breaking the symmetry of
the system and generating a highly asymmetric diiron centre. In both complexes, the central
atom(s) of the dithiolate-bridge are orientated towards the less sterically encumbered iron
dicarbonyl end of the molecule. For unlinked 4 only a single isomer was seen in the 31P{1H}
NMR spectrum, the chemical shifts of which suggest that it is the dibasal isomer. Here there
is less steric interaction between the substituents on the dithiolate groups and the triphos
ligand and this presumably accounts for the observation of a single isomer in solution. While
complexes of triphos have been known for over 50 years [38], as far as we are aware, there has
only been one previously crystallographically characterised example of this ligand acting in a
-fashion [39].
In Fe2(CO)4(-dithiolate)(2-diphosphine) complexes, apical-basal (ab) and dibasal (bb)
isomers generally interconvert only slowly in solution at room temperature on the NMR time
scale, although inequivalent phosphorus centres in the apical-basal isomers interconvert rapidly
31
at room temperature [18,29]. The room temperature P{1H} NMR spectrum of 1 in CD2Cl2
shows two sets of three resonances in an approximate 3:1 ratio assigned to isomers 1bba and
1bbb. The minor set, 1bbb, are all sharp and consist of two doublets at 89.7 (J 17.2 Hz) and
61.6 ppm (J 9.2 Hz) and a doublet of doublets at 86.3 ppm. The major set, 1bba, which are
observed at 83.3, 81.8 and 63.9 ppm are all broad at this temperature and the individual coupling
constants cannot be extracted. In each isomer, we assign the low-field resonances to the
phosphorus bound to the iron dicarbonyl unit, the intermediate resonance to the central
phosphorus atom of the triphos ligand and the high-field resonance to the chelate end of the
tridentate phosphine. Upon warming a d8-toluene solution of this mixture, there is no significant
change in the ratio of 1bba:1bbb and this suggests that the bba and bbb isomers do not
interconvert readily, consistent with slow interconversion of apical-basal (ab) and dibasal (bb)
coordination. Upon cooling a CD2Cl2 solution, resonances associated with 1bbb do not vary
significantly, however, those associated with 1bba broaden and collapse until at 183 K they are
replaced by twelve new, sharp resonances associated with four separate isomeric forms (labelled
I to IV see experimental section for details).
We propose that the four isomers of 1bba in the slow-exchange spectrum derive from
conformational changes through; (i)“flipping” of the Fe2S2C3 bridge across the Fe-Fe vector and
6

Page 7 of 41
Dalton Transactions
View Article Online
(ii) restricted motion of the phenyl rings and ethano-backbones associated with the triphos
DOI: 10.1039/C9DT00700H
ligand. Scheme 3 shows the different ring-flip conformers (i.e., “flippamers”) based on 1bba, of
which the structures of bba (L) and bba (R) [where the L (left) and R (right) designations refer
to the flip orientation of the pdt ligand] were computationally optimised in order to explore the
barrier for the flipping of the pdt ring in the structurally characterized form of this compound.
Figure 3 shows the optimised structure of 1bba, whose pdt ring adopts a distal orientation (left)
with respect to the chelated iron center [i.e., the Fe(2) center in the solid-state structure of 1bba
shown in Figure 1). The species bba (L) in Scheme 3 corresponds to the DFT-optimised
structure A depicted in Figure 3, while the pdt-flippamer is represented by bba (R) and whose
optimised structure is species B. The energy difference (G) between A and B favors the
former species but only by 0.4 kcal/mol, rendering the two isomers isoenergetic for practical
purposes. The computed barrier for pdt ring flipping in AB is 9.7 kcal/mol (G), and the
process proceeds through a flattening of the carbocyclic ring of the pdt ligand, as verified by the
optimised transition-state structure TSAB. Restricted motion of the phenyl rings and the ethano-
backbones associated with the triphos ligand are likely responsible for the other pair of
stereoisomers observed in the low-temperature NMR spectrum.
Ph2
P
Ph2
P
S
S
S
S
OC
OC
Fe
Fe
Fe
Fe
CO
CO
Ph2P
Ph2P
P
P
CO
CO
bba (R)
Ph
Ph
bba (L)
Scheme 3
In order to model the geometry of the bbb stereoisomer based on 1, we employed structure 3
(Figure 2) and replaced the aza moiety with a methylene group. The optimised structure (species
C) is fully consistent with this alternative stereoisomer. The pdt ring in C undergoes flipping to
afford species D through TSCD in a process similar to that found for the conversion of the
A→B.
While the interconversion of apical-basal sites of chelating diphosphine ligands (ab ⇌ ba) in
Fe2(CO)4(2-PP)(µ-SXS) derivatives is well-established and shown to proceed with low
activation barriers [18,29], the constraints imposed by the triphos ligand retard this isomerisation
7

Dalton Transactions
Page 8 of 41
View Article Online
path within the chelate portion of the ligand. DFT calculations confirmed that the (ab ⇌ ba)
DOI: 10.1039/C9DT00700H
isomerisation of the two phosphine ligands associated with the chelate ring in species A-D is
energetically prohibitive. The effect of tripodal rotation of the CO and PPh2 ligands about
each iron center was also computationally examined for species A and C, but no stable
structure could be found for the concerted rotation of the groups about the Fe(CO)P2 moiety
in either species. The Fe(CO)2P center is more flexible with respect to a tripodal rotation, and
we were able to optimize an alternative structure where the PPh2 moiety has exchanged sites
with the equatorial CO group. Here species A_Alt and C_Alt were successfully optimised
(Figure 5) and found to lie 12.8 and 13.0 kcal/mol higher in energy (G), respectively, than
their precursor. The unfavorable energy of A_Alt allows us to exclude it as a participant in
the above VT NMR discussion. Attempts to extract the transition-state structure for the two
tripodal rotations were unsuccessful.
Protonation studies - A key feature of iron-only hydrogenase models is their ability to bind a
proton(s) [40-48]. Protonation rates are controlled electronically, being correlated with the
energy of the metal-metal -bond (HOMO) which in turn is a function of the nature of the
electron-donating ligands attached to the diiron centre [48]. Consequently, Fe2(CO)3L3(-
dithiolate) complexes are expected to protonate rapidly. Addition of ca. 1-2 equivalents of
HBF4.Et2O to a CH2Cl2 solution of 1 led, however, only to a slow change (over ca. 2 h) in the IR
spectrum, but consistent with protonation at the diiron centre, initial carbonyl absorptions at
1947s and 1889vs cm-1 being replaced by new peaks at 2039vs, 1986s and 1964s cm-1 assigned
to [Fe2(CO)3(-pdt)(-triphos)(-H)][BF4] (1H+) (Scheme 4); the average blue shift of 88
cm-1 being consistent with protonation at the diiron centre. The salt was stable in air, and no
further change was noted even after sitting for 3 d. In the 31P{1H} NMR spectrum a single set of
three resonances was observed at 88.9, 85.2 and 46.4 ppm and upon cooling no significant
1
changes were observed. In the H NMR spectrum a doublet of triplets at  13.5 (J 29.2, 22.4
Hz) is assigned to the hydride resonance, both the position of couplings showing that it bridges
the diiron centres. The large high-field shift (ca. 17.5 ppm) of the resonance associated with the
chelate portion of the triphos ligand suggests that at least one of the PPh2 moieties has been
significantly perturbed, while in contrast, the other two phosphorus atoms change only slightly.
Unfortunately, we have been unable to crystallographically characterise 1H+ but on the basis of
spectroscopic measurements proposes that it adopts a bbb geometry (Scheme 4).
8

Page 9 of 41
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT00700H
S
Ph2
P
S
S
S
CO
OC
OC
OC
OC
Fe
Fe
HBF4.Et2O
CH2Cl2
Fe
Fe
PPh2
H
CO
Ph2P
P
Ph2P
P
Ph
Ph
1H+
1
Scheme 4
Addition of two equivalents of HBF4Et2O to CH2Cl2 solutions of 2 and 4 lead to similar
observations, consistent with the slow protonation of the metal-metal bond; however, in both
cases, the protonated product was considerably less stable than 1H+. For 2, new IR bands
developed at 2043 and 1999 cm-1 being attributed to 2H+. Attempts to isolate a pure product
were unsuccessful and all samples were accompanied by significant decomposition. Similarly
for 4, generation of new bands at 2031, 2003 and 1964 cm-1 was consistent with the
formation of 4H+, but a pure sample could not be isolated.
The protonation behaviour of 3 was expectedly different given the presence of the basic aza-
dithiolate moiety, which serves to direct protonation to the bridgehead nitrogen atom [49-55].
The neutral complex showed two carbonyl bands at 1951 and 1893 cm-1, and on the addition
of one equivalent of acid, these undergo a small shift and furnish three bands at 1970, 1922
and 1902 cm-1. This is consistent with previous reports that protonation at the nitrogen in the
bridge results in shifts of ca. 10-20 cm-1, in contrast to ca. 60-100 cm-1 shift for protonation at
the dimetal centre [48,49]. Addition of excess acid resulted in a further shift of the carbonyl
stretching bands to 2097, 2051, 2022 and 1990 cm-1 indicative of protonation at the iron-iron
+
bond. Thus it appears that [Fe2(CO)3(-adtH)(-triphos)][BF4] (3HN ) initially results and
that this is converted to doubly protonated [Fe2(CO)3(-adtH)(-triphos)(-H)][BF4]2
2+
(3H2 ) in the presence of excess acid (Scheme 5). Rauchfuss and co-workers have shown that
the related tri-substituted complex Fe2(CO)3(PMe3)(-adt)(2-dppv) also first forms an
ammonium cation, followed by the slow tautomerisation (t1/2 = 30 min in CH2Cl2 room
temperature) to the isomeric µ-hydride species [14]. Attempts to monitor the protonation of
1
both 1 and 3 by H NMR at low temperature were unsuccessful, signals simply broadening
upon addition of acid which may result from concomitant oxidation of the diiron centre (see
later).
9

Dalton Transactions
Page 10 of 41
H
H
Bz
Bz
Bz
View Article Online
DOI: 10.1039/C9DT00700H
N
N
N
S
S
S
S
S
S
Ph2
Ph2
P
Ph2
P
OC
OC
OC
P
xs HBF4 Et2O
CH2Cl2
HBF4 Et2O
CH2Cl2
Fe
Fe
Fe
Fe
Fe
Fe
OC
CO
OC
Ph2P
CO
CO
OC
H
P
P
P
Ph2P
Ph2P
Ph
Ph
Ph
2+
+
3H2
3HN
3
Scheme 5
The slow rate of protonation of 1 is unexpected. Pickett and co-workers [48] who have shown
that the process is second order with the rate (kH) showing a linear correlation with the energy
of the HOMO, as measured by E1/2ox. On the basis of the measured E1/2ox for 1 of -0.35 V (see
below), an extrapolated rate of ca. 4500 M-1s-1 is expected. Clearly protonation of 1 and the
other triphos complexes reported here must have a significant kinetic barrier, as rates (kH) are
nowhere near as high as predicted. Independent of the protonation kinetics, the final product
1H+ exhibits good thermodynamic stability. To better understand the slow protonation of 1
we considered the general protonation pathway as shown for disubstituted Fe2(CO)4(-
pdt)(PMe3)2, whose reactivity has been thoroughly studied [44]. Thus with [Et2OH]+ (the acid
used in this study), the initial formation of an intermediate in which this cation is bound to a
terminal carbonyl is proposed, followed by rearrangement to give a terminal hydride, a
transformation which also requires the development of a semi-bridging carbonyl. This species
is unstable and converts to more stable bridging hydride isomers, the thermodynamically
preferred species whose formation is accompanied by a significant rearrangement of the
positions of the PMe3 ligands (via Bailar twists) [56]. In mapping this general scheme onto
the protonation of 1, the formation of a carbonyl adduct should not be perturbed, but all other
transformations are likely to be significantly changed. Thus the formation of the semi-
bridging carbonyl is likely to be unfavourable at the sterically congested Fe2(triphos)
platform, while as shown above, movement of the triphos ligand between different
coordination sites has much a higher energy barrier(s) than the movement of monodentate or
bidentate phosphine ligands. We also note here that in the enzymes themselves it is terminal
hydrides that are responsible for hydrogen production; the thermodynamically stable bridging
hydride being off cycle [1], and its formation is prohibited by high activation barriers caused
by interactions of the H-cluster with surrounding amino acids [57]. Thus, the kinetically slow
protonation of 1H+ with a bridging hydride is not fully unexpected, and our triphos-tethered
compounds function as outliers as compared to other tricarbonyl complexes, Fe2(CO)3L3(-
10

Page 11 of 41
Dalton Transactions
View Article Online
dithiolate), whose protonation is both kinetically and thermodynamically favoured.
DOI: 10.1039/C9DT00700H
Unfortunately, we have been unable to identify any terminal hydride species in low-
temperature protonation studies. The slow protonation of 1 is expected to have a significant
effect on its behaviour as a proton reduction catalyst and, indeed, this has been
experimentally established (see later).
Electrochemical studies – Figure 6a (brown trace) shows the CV of 1 in CH2Cl2/0.1 M
TBAPF6 (TBA = NBu4). On sweeping first towards negative potentials, no reduction peak
was observed within the available potential window, consistent with high electron-density at
the diiron centre when ligated by three electron-donating phosphines. On scanning
anodically, 1 undergoes oxidation at E1/2 = -0.35 V (Table 1) in a process that was found to be
ox red
reversible (ip /ip ca. 1) over all scan rates studied (0.01 through 10 V s-1) (see ESI). The
stability of the oxidised [FeFe]+ species allowed chronocoulometry experiments to be carried
out, confirming unequivocally that the process was a one-electron transfer. Further
irreversible anodic peaks were noted above 0.3 V, attributed to the [FeFe]+/2+ process and
oxidation of resulting products.
Table 1: Oxidation potentials for the first reversible oxidation of complexes 1-4 and 1H+
Complex
1
2
3
4
1H+
E1/2 / V vs. Fc/Fc+
-0.35
-0.45
-0.35
-0.25
0.75
The edt complex 2 and unlinked 4 show similar electrochemical behaviour to 1, but the
oxidation potential is dependent on the identity of the bridging group, as summarised in Table
1. Complex 2 has E1/2 = -0.45 V for the [FeFe]0/+ couple, making it the most difficult of the
four complexes to oxidise, while 4 was easier to oxidise than 1, reversible oxidation taking
place at E1/2 = -0.25 V. A comparison of the relative oxidation peak currents for 1, 2 and 4
allows the oxidation to be assigned as a one-electron process, i.e. [FeFe]0/+ for all of the
complexes. The electrochemical reversibility of the [FeFe]0/+ process is unaffected by the
bridging group, indicating little structural rearrangement on the timescale of electron transfer
and hence a common mechanism for oxidation of these species. The difference in oxidation
potentials is likely attributed to different frontier MO energies resulting from differing
degrees of orbital overlap, dictated by the flexibility or constraint imposed by the bridging
group.
11

Dalton Transactions
Page 12 of 41
View Article Online
DOI: 10.1039/C9DT00700H
The adt-complex 3 (Figure 6b) has quite different oxidation behaviour. Oxidation occurs
reversibly at the same potential as for 1 (E1/2 = -0.35 V), but a second oxidation couple of
similar peak height is observed at E1/2 = -0.11 V, suggestive of a further reversible one-
electron oxidation. The scan rate dependence of the two couples (see ESI) shows the relative
peak heights and separations of the two processes do not change significantly with scan rate.
The behaviour observed may be attributed to the consecutive loss of two electrons from the
same orbital; thus the first couple is assigned to the [FeFe]0/+ process and the second to
[FeFe]+/2+. The ease of reversible oxidation of the [FeFe]+ cation for 3 as compared to 1 is
consistent with behaviour reported for Fe2(CO)3(-adt)(2-dppv)(PMe3) [9], which has been
attributed to stabilisation of the [FeFe]2+ dication by the nitrogen in the adt bridge. It was
proposed that an iron-nitrogen bond is formed, resulting in stabilisation of the dication in the
adt complex, while such an interaction between bridge and Fe centre is not possible for the
pdt complex. No attempt was made to isolate dications.
We have also recorded the CV of 1H+ in CH2Cl2/0.1 M TBAPF6 (Figure 6a – black trace).
Consistent with the removal of electron-density on the formation of the bridging hydride
across the iron-iron bond, the first oxidation of the complex is shifted ca. 1.1 V towards more
positive potentials (E1/2 = 0.74 V) as compared to 1. Unlike 1, the first reduction of 1H+ is
observed within the potential window of the solvent, appearing as a quasi-reversible couple at
ox red
E1/2 = -1.7 V (ip /ip = 0.65 at 0.1 V s-1). Both oxidation and reduction of 1H+ are one-
electron processes, with the response being stable under repeated cycling. The stability of the
reduced 35-electron species 1H on the electrochemical time scale is consistent with
voltammetry reported for some other bridging hydrides of phosphine substituted complexes
[40,58].
Oxidation of hydrogenase biomimics usually results in a concomitant shift of a carbonyl from
a terminal to semi-bridging site with the resultant twisting of the iron centres to a more
staggered geometry [5,6,59-61]. In order to further probe the structure and the stability of the
radical cation structures derived from 1, we have carried out DFT calculations. The
corresponding radical cations based on species A and C were successfully optimised, and the
resulting structures are shown in Figure 7. Spin contamination in the radical cations is
minimal, and Aox lies 6.0 kcal/mol lower in energy than Cox. The flexibility of the triphos
12

Page 13 of 41
Dalton Transactions
View Article Online
ligand is assumed to be responsible for the computed energy difference, wiDthOI:t1h0.e103m9/Co9rDeT00700H
flexible bbb isomer able to undergo partial tripodal reordering of the ancillary ligands at the
Fe(CO)2P site to afford a semi-rotated structure in the gas phase. This partial tripodal rotation
of these ligands allows one of the CO groups to adopt a semi-bridging interaction (2.858 Å)
with the adjacent iron centre which helps to stabilize the cationic charge in the gas phase.
Chemical oxidation of 1 was carried out in situ upon addition of one equivalent of
[Cp2Fe][PF6]. The IR spectrum of the resulting complex 1+ showed a single band at 1947
cm-1, and we could find no evidence for a semi-bridging carbonyl as predicted by DFT.
Repeating the optimization of Cox in the presence of CH2Cl2 (implicit solvation) furnished a
similar structure (not shown) that exhibited a significantly weaker semi-bridging CO
interaction with the adjacent iron center (>3.00 Å) and confirmed the sensitive nature of the
secondary semi-bridging CO interaction with the iron center. While 1+ had some stability in
solution, all attempts to isolate a pure sample resulted in slow decomposition.
Electrocatalytic behaviour of 1 in the presence of HBF4∙EtO2 - Given that 1H+ undergoes
reversible oxidation at 0.75 V and reduction at -1.7 V, it is expected that addition of the
strong acid HBF4·Et2O to 1 should result in redox features at these potentials. Broadly this is
observed (Figure 8a), but protonation is not immediate, requiring the addition of 3–4 molar
equivalents of acid before the oxidation couple of the neutral complex at -0.35 V disappears
and is replaced with the reversible couple at 0.75 V assigned to oxidation of 1H+. The
relatively slow protonation of the highly basic 1 is consistent with earlier IR observations and
suggests a significant kinetic barrier to protonation. Addition of up to five equivalents of acid
results in increasing reduction currents at ca. -1.7 V, the potential at which 1H+ undergoes
reduction, suggesting proton reduction catalysis takes place at this potential (labelled A in
Figure 8a). This pathway to hydrogen evolution is proposed to proceed via a CECE
mechanism (A) as shown in Figure 8b. On addition of further acid equivalents, the catalytic
current at this potential does not increase further but reaches a limiting value (see Figure 8 c).
As the rate of this process rapidly becomes independent of proton concentration the rate-
limiting step is likely H2 elimination from 1(μH)H, as has been described for [Fe2(CO)4(2-
dppe)(-pdt)(-H)]+ [62]. H2 elimination first requires conversion of 1(μH)H to 1-H2 and
given the rigidity and basicity of 1 this step is expected to be slow.
13

Dalton Transactions
Page 14 of 41
A further catalytic process (labelled B) is observed at ca. -2.1 V. Given that DthOeI: 101.1(0μ3V9Hi/eCw)9HADrtTic0le0O70nl0inHe
species generated in process A is slow to eliminate H2 on the electrochemical timescale, it is
proposed that process B derives from the reduction of 1(μH)H at ca. -2.1 V and its
subsequent protonation, with hydrogen elimination taking place from resulting 1H(μH)H
(Figure 8b, B). Such a scenario has previously been described for catalysis involving
[Fe2(CO)4(2-dppe)(-pdt)(-H)]+, whereby the bridging hydride is proposed to behave as a
spectator ligand [61]. This mechanism has been demonstrated experimentally for mixed-
valence [Fe2(CO)2(2-dppv)2(-pdt)(-H)]+, where on addition of D+ only D2 was evolved,
with no mixed DH product observed [61]. In the present study, we know that 1H+ is
thermodynamically stable (as it can be readily isolated), and as shown in Figure 6a, the
complex 1H generated during reduction of 1H+ is stable towards electrochemical cycling.
Thus the possibility of a catalytic cycle where 1H acts as the catalyst, but the bridging
hydride plays no role in the catalytic cycle is feasible. The mechanistic details of the further
protonation of 1H are yet to be elucidated, with the metal centres, bridging sulphur sites or
the bridging hydride itself all being proposed in similar systems [61,62]. As shown in Figures
8a and 8c, process B reaches a limit after addition of about 5-6 acid equivalents, indicating
that hydrogen elimination becomes rate limiting in this process also.
After addition of 4 equivalents of acid, a new process (labelled C) begins to grow in at ca. -
1.6 V, some 100 mV positive of A (Figure 8a). This process has a non-linear relationship
with [H+] and it appears to be related to the limiting of processes A and B, as it emerges at
about the same acid concentration that they begin to reach their maximum rates (Figure 8c).
We attribute these currents to the reduction of intermediates of the catalytic A and B cycles,
which accumulate at the electrode interface during cycling. As H2 elimination steps for A and
B are rate-limiting, it follows that 1(μH)H and 1H(μH)H will build up in solution. Over the
time scale of electrochemical cycling, they slowly release H2, producing 1 and 1(μH)
respectively. Regeneration of 1 allows catalytic cycle A to proceed at -1.7 V. The fate of
1(μH) is more complex, as at potentials above -1.7 V there is a competition between its
oxidation to 1(μH)+ (the reduction of 1H+ was shown to be reversible in Figure 6 a) and
further protonation to give 1(μH)H+. Over time and in the presence of excess acid, increasing
concentrations of 1(μH)H+ are therefore likely to accumulate in solution, which we propose
give rise to the reduction currents at -1.6 V.
14

Page 15 of 41
Dalton Transactions
View Article Online
Effect of electrolyte composition on the electrocatalytic performance of 1 - The catalytic
DOI: 10.1039/C9DT00700H
performance of 1 and the relative contribution of peaks A, B and C to the total current was
found to depend strongly on background electrolyte. Figure 9 shows how the limiting
currents of processes A, B and C vary with concentration of HBF4 in the presence of 0.1 M
TBAPF6, TBABF4 and TBAClO4. As described above, in TBAPF6 process A reaches a
maximum rate after addition of only 4 equivalents of acid. In contrast, in TBAClO4 the
limiting current increases linearly with acid concentration up to the addition of 10
equivalents. In TBABF4 the behaviour is intermediate, with the rate of catalysis slowing after
addition of 5 equivalents, but without reaching a limiting value as observed in TBAPF6.
These observations suggest that the anion plays a significant role in key steps of catalytic
-
-
-
mechanism A, with the rate increasing in the order PF6 < BF4 < ClO4 .
Process B competes with A and hence its rate depends on the ease of H2 elimination from
1(μH)H in pathway A. This is reflected in the limiting currents for B on addition of 1-5
-
equivalents of acid, where the currents in the presence of PF6 are marginally higher than
-
-
those in BF4 which are higher than those in ClO4 (Figure 9). This is consistent with more
-
1(μH)H being available for reduction at -2.1 V in PF6 as H2 elimination is so slow under
-
these conditions. Conversely, in the presence of ClO4 where the larger currents for process A
suggest that H2 elimination from 1(μH)H is faster, currents due to B are initially the smallest
as there will be less 1(μH)H available at the electrode interface to partake in the reaction.
-
With the addition of further acid, process B begins to reach a constant rate in PF6 , suggesting
-
that H2 elimination (this time from 1H(μH)H) is rate-limiting. This is not the case for BF4
-
and ClO4 , where currents continue to increase with acid addition, indicating H2 elimination
steps are faster in the presence of these anions.
Process C has a similar, but more complex, acid dependence in all three electrolytes. In all
cases, currents at potential C only emerge after addition of 4-5 acid equivalents and thereafter
increase in a non-linear fashion. As described above, process C is attributed to the reduction
of 1(µH)H+, which accumulates at the electrode surface over time, due to slow H2
elimination from 1H(μH)H and subsequent protonation of 1(μH). The currents for this
-
process are greatest in the presence of ClO4 as H2 elimination is fastest under these
conditions, allowing higher concentrations of 1(µH)H+ to build up over time.
15

Dalton Transactions
Page 16 of 41
View Article Online
The role of anions in protonation and hydrogenation equilibria of metal hydrides has been
DOI: 10.1039/C9DT00700H
reported previously, for complexes such as [FeH(2-H2)(2-dppe)2]+ [63] and [W3(3-S)(-
S)3H3(2dmpe)3]+ [64] as well as Fe2(CO)2(2-dppv)2(-adt) [65]. In these studies it was
proposed that ion pairs form between the metal complex and the anion and this lowers
activation barriers for intramolecular proton transfer from one site to another. Our results
suggest that the anion may play a role in the hydrogen elimination step. As indicated in
Figure 8b, this requires the combination of two protons bound at different sites on the
molecule (e.g. to form 1-H2 from 1(μH)H or 1(μH)-H2 from 1H(μH)H) and we suggest that
the shuttling of H from one site on the molecule to another may be facilitated by ion pairing
-
-
to BF4 or ClO4 , hence catalysis is more efficient in the presence of these ions. Ion pairing
-
with PF6 may not be possible for steric or other reasons. Different solvents have also been
shown to influence protonation sites and equilibria [66], so it is also possible that the
TBAPF6, TBABF4 and TBAClO4 salts used in this study had varying water content, and the
presence of trace water may also influence site and rate of proton transfer. Due to the low
synthetic yield we were unable to test the electrocatalytic behaviour of 2, however 4
exhibited broadly similar trends to 1 (see ESI).
Electrocatalytic behaviour of 3 on the addition of strong acid – The behaviour of 3 on
addition of HBF4∙Et2O in the presence of TBAPF6 is very different to that observed for 1
(Figure 10a). Addition of one acid equivalent results in the immediate loss of the oxidation
peaks for 3 at -0.35 and -0.11 V and the appearance of a new oxidation peak at 0.15 V vs.
Fc/Fc+. The shift of oxidation potential by +0.5 V in the presence of acid is consistent with
protonation of the adt-bridge [49]. Thus, a shift of ca. +1.1 V is anticipated for protonation at
iron, but for 3 we observe no clear evidence of an oxidation peak at 0.75 V, suggesting that
protonation has taken place solely at the adt nitrogen. Addition of one equivalent of acid also
results in the emergence of a broad reduction peak at ca. -1.9 V (D), attributed to the
+
reduction of 3HN . This increases with the addition of up to 3-4 acid equivalents, indicating
H+ reduction catalysis at this potential. A possible CECE mechanism for the processes
+
contributing to D is shown in Figure 10c; after reduction of 3HN at -1.9 V, protonation of
3HN at the iron-iron bond is proposed, resulting in 3HNHFe+. As one would expect a species
+
protonated at iron to be reduced more readily than 3HN (as electron density has been
+
removed from the iron centre) 3HNHFe is likely to undergo rapid reduction at this potential
to give 3HNHFe. The final step of H2 elimination is rate-limiting and slow, as the catalytic
16

Page 17 of 41
Dalton Transactions
View Article Online
process becomes independent of acid concentration after addition of 3 acid eqDuOivI: a10l.e10n3t9s/.C9ADT00700H
further reduction process is noted at ca. -2.1 V, l due to the reduction of 3HNHFe, but this
appears not to be catalytic. Marked on Figure 10a is the position of process A observed for 1,
associated with species protonated at the Fe-Fe bond. For 3 no significant increase in current
is noted at these potentials, which suggests protonation at iron does not compete effectively
with protonation at nitrogen under these conditions.
In contrast, as shown in Figure 10b, the addition of excess acid to 3 in TBABF4 (green trace)
and TBAClO4 (orange trace) results in increased reduction currents at ca. -1.6 V (E and F) at
similar potentials to processes A and C observed for 1, suggesting the involvement of species
protonated at the iron-iron bond. This is consistent with IR observations of the shift of CO
bands to higher wavenumber on addition of excess acid to 3. The reduction potential for these
species is electrolyte-dependent and there is little reduction observed at this potential in
TBAPF6 (brown). A speculative mechanism for this catalytic process is shown in Figure 10c.
We propose that in excess acid 3HN+ undergoes further protonation to give 3HNHFe2+ and the
reduction of this species gives rise to currents observed at E and F in Figure 10b. Hydrogen-
elimination is again the rate limiting step in the process, resulting in slow regeneration of 3.
The similar dependency of the rate of this process on electrolyte anion identity is consistent
with the results observed for 1 and further suggests the anion plays a role in the H2
elimination step.
SUMMARY AND CONCLUSIONS
Introduction of the triphos ligand into hydrogenase biomimics allows access to complexes 1-
4 containing the biologically relevant Fe2(CO)3L3(-dithiolate) framework. In contrast to other
complexes of this type, which rely on separate addition of monodentate and bidentate ligands [3-
16], triphos complexes represent a new class of biomimics in which all three substituted ligands
are conjoined and this imposes some structural rigidity upon the molecule as well as sterically
inhibiting access to the central iron-iron bond. Structural and spectroscopic studies show that
a range of isomers are possible. An important finding is that although the diiron centre is
basic, protonation is slow; likely a result of the relative inaccessibility of a protonation site.
This is supported by computational work by De Gioia and co-workers who investigated
factors affecting the kinetics and thermodynamics of protonation of Fe2(CO)3(-
17

Dalton Transactions
Page 18 of 41
edt)(PMe3)(2-dppv) [8]; the presence of the bulky ligands significantly affectinDgOtI:h1e0.1r0a3V9tie/eCw9oADrftTic0le0O70nl0inHe
protonation. Although protonation of 1 affords the thermodynamically favoured bridging
hydride, 1H+, which likely proceeds via initial formation of a terminal hydride, De Gioia’s
work suggesting that the most stable terminal hydride is that bound to the least-substituted
iron centre [8]. In Fe2(CO)3(-edt)(PMe3)(2-dppv) formation of this terminal hydride led to
a trigonal rotation of this iron centre with formation of a bridging carbonyl ligand, a situation
also seen in the well-studied tetra-substituted model complexes Fe2(CO)2(2-dppv)2(-
dithiolate) [65-69] and Fe2(CO)2(PMe3)4(-dithiolate) [70]. We suggest that for 1-4 this is a
high energy process and thus renders formation of such a species unfavourable. With regards
to the hydrogenase enzyme, the active site is constrained within a protein structure, where
secondary interactions impose restrictions on the flexibility of the H-cluster and access to the
Fe-Fe bond. In particular, the formation of a bridging hydride species is prevented, as access
to the bond is blocked.
Given the slow rate of protonation of 1-4 it might be expected that this would be rate-limiting
in catalytic proton-reduction. However, it was found that for 1 and 3-4, H2 elimination is slow
and moreover their rates are highly dependent on electrolyte composition. For 1 and 4 the key
steps that are influenced by the identity of electrolyte anion seem to be combining of two
hydrides and their elimination as H2. These steps, involving proton-transfer from one site to
-
-
another, are facilitated by ions such as BF4 and ClO4 that can effectively form ion pairs with
-
the complex [15]. The larger, bulkier PF6 ion does not effectively aid this H transfer process,
and for the (-SMe)2 complex 4 showed evidence of inhibiting electrocatalysis altogether,
suggesting it may even block key sites. The slow rate of the H2 elimination may be attributed
partly to the thermodynamic stability of the intermediate bridging hydride species. This is
illustrated by the ease of isolation of 1(μH)+ and ability to reversibly cycle this molecule
through the 1-/0/1+ oxidation states; thus 1(μH) is very stable, and even on further
protonation and reduction the bridging hydride has a high tendency to remain intact. In
addition, the second proton may bind at a site removed from the bridging hydride, due to the
presence of the relatively bulky triphos ligand. This site could be a terminal hydride on one of
the iron centres or the sulphur of the bridging group. The rigidity of the complex imposes a
large barrier to the intramolecular proton transfer and its combination with the bridging
hydride; hence this step is very likely to be slow. For the adt complex we observe that the
triphos ligand imposes similar constraints on protonation at the Fe-Fe bond, with protonation
18

Page 19 of 41
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT00700H
-
-
-
at adt preferred. Again, the presence of BF4 or ClO4 (but not PF6 ) apparently facilitates H2
elimination from 3HNHFe, emphasising the role of effective intramolecular proton transfer in
the maintenance of a catalytic cycle.
EXPERIMENTAL SECTION
General Procedures and Starting Materials. Unless otherwise noted, all the reactions were
carried out under a nitrogen atmosphere using standard Schlenk techniques. Reagent-grade
solvents were dried using appropriate drying agents and distilled prior to use by standard
methods. Infrared spectra were recorded on a Nicolet 6700 FT-IR spectrometer in a solution
cell fitted with calcium fluoride plates, subtraction of the solvent absorptions being achieved
by computation. NMR spectra were run on a Bruker AMX400 instrument. Mass spectra were
recorded on a Varian Mat 312 mass spectrometer. Elemental analyses were performed by
Microanalytical Laboratories, University College London. Collman’s reagent and PPh3 were
purchased from Strem Chemical Inc. and used without further purification.
Preparation of Fe2(CO)3(-pdt)(-triphos) (1). A toluene solution (50 mL) of triphos
(0.267 g, 0.05 mmol) and Fe2(CO)6(-pdt) (0.221 g, 0.067 mmol) [71] was heated at reflux
for 16 h. Upon cooling to room temperature the solvent was removed in vacuo and the
residue washed with hexane to remove unreacted starting materials. Chromatography on
alumina eluting with light petroleum-diethyl ether (1:1) gave a faint yellow band which was
discarded. Eluting with dichloromethane gave a large red band which afforded a red-brown
solid identified as Fe2(CO)3(-pdt)(-triphos) (1) (0.387 g, 93 %). A second dark red band
eluted upon addition of a small amount of methanol but this product was not characterised.
Recrystallization of 1 from the slow mixing of methanol with a concentrated CH2Cl2 solution
1
afforded crystals suitable for crystallography. IR (CO)(CH2Cl2): 1947s, 1889vs cm-1; H
NMR (CD2Cl2, 298 K):  8.03 – 7.16 (m, 25H, Ph), 3.04 – 2.41 (m, 8H, PCH2), 2.36 (br, 2H,
SCH2), 2.29 (br, 2H, SCH2), 2.14 (br, 2H, CH2); 31P{1H} NMR (CD2Cl2, 298 K):  89.7 (d, J
17.2, 1bbb), 86.3 (dd, J 17.2, 9.2, 1bbb), 83.3 (br, 1bba), 81.8 (brd, J 23.4, 1bba), 63.9 (br,
1bba), 61.6 (d, J 9.2, 1bbb) ratio 1abb:1bbb ca. 7:3; (183 K)  91.74 (d, J 29.1, 1bbaI),
88.34 (d, J 28.1, 1bbaII), 87.78 (d, J 16.3, 1bbb), 86.11 (dd, J 16.3, 9.2, 1bbb), 83.52 (dd,
27.7, 9.4, 1bbaI), 80.70 (dd, m, 1bbaII+III), 77.48 (d, J 25.0, 1bbaIII), 76.90 (dd, J 25.3, 6.5,
1bbaIV), 74.42 (d, J 6.6, 1bbaIV), 71.99 (d, J 6.0, 1bbaIII), 70.64 (d, J 25.0, 1bbaIV), 64.88 (d,
19

Dalton Transactions
Page 20 of 41
View Article Online
DOI: 10.1039/C9DT00700H
J
9.2, 1bbb), 61.44 (d,
J
9.8, 1bbaII), 57.66 (d,
J
9.8, 1bbaI), ratio
1bbb:1bbaI:1bbaII:1bbaIII:1bbaIV Ratio of isomers I:II:III:IVca 1:1:0.7:0.3:0.5; Anal. calc.
for Fe2P3S2O3C40H39: C, 57.42, H, 4.67; Found C, 57.54, H, 3.86.
Preparation of Fe2(CO)3(-edt)(-triphos) (2). A toluene solution (50 mL) of triphos
(0.077 g, 0.014 mmol) and Fe2(CO)6(-edt) (0.0.60 g, 0.016 mmol) [70] was heated at reflux
for 16 h. Upon cooling to room temperature, the solvent was removed in vacuo and the
residue washed with hexane to remove unreacted starting materials. A 31P{1H} NMR
spectrum of the crude reaction mixture revealed a very complex mixture of products.
Chromatography on alumina afforded a large number of bands which could not be fully
separated. The crude reaction mixture (from a second reaction) was then again washed with
hexanes and dried. The whole mixture was then dissolved in a minimum volume of
dichloromethane and this was layered with approximately twice the volume of methanol.
After mixing a brown solid was produced and this contained a small number of well-formed
large crystalline blocks of Fe2(CO)3(-edt)(-triphos) (2). These were used for
crystallographic and analytical characterization. The brown solid was found to be impure 2
and a second batch of pure 2 was produced upon repeating the crystallization procedure. The
overall yield of 2 is difficult to estimate but ca. 15 mg (13 %) of pure material was isolated.
1
IR (CO)(CH2Cl2): 1951s, 1902vs cm-1; H NMR (CD2Cl2, 298 K):  7.87 – 7.08 (m, 25H,
Ph), 3.43 – 1.27 (m, 12H, PCH2 + SCH2), 31P{1H} NMR (CD2Cl2, 298 K):  93.9 (d, J 13.9,
2bbb), 87.3 (d, J 26.5, 2abb), 86.9 (dd, J 13.9, 8.9, 2bbb), 84.7 (d, J 26.5, 2abb), 68.6 (s,
2abb), 65.8 (d, J 8.9, 2bbb) ratio 2abb:2bbb ca 3:7; Anal. calc. for Fe2P3S2O3C39H37: C,
56.93, H, 4.50; Found C, 57.06, H, 4.66.
Preparation of Fe2(CO)3(-adt)(-triphos) (3). A toluene solution (50 mL) of triphos
(0.237 g, 0.044 mmol) and Fe2(CO)6(-adt) (0.200 g, 0.049 mmol) [72] was heated at reflux
for 19 h. Upon cooling to room temperature the solvent was removed in vacuo and the
residue washed with hexane to remove unreacted starting materials. The dry solid was
dissolved in a minimum volume of dichloromethane and an excess of hexane was added
resulting in precipitation of a red solid identified as Fe2(CO)3(-adt)(-triphos) (3) (0.379
g, 94 %). Crystallization of 3 from the slow mixing of methanol with a concentrated CH2Cl2
solution afforded crystals suitable for crystallography. IR (CO)(CH2Cl2): 1952s, 1895vs cm-1;
1H NMR (CD2Cl2, 308 K):  8.25 – 7.15 (m, 30H, Ph), 4.40 – 1.90 (brm, 14H, CH2); 31P{1H}
20

Page 21 of 41
Dalton Transactions
View Article Online
NMR (CD2Cl2, 308 K):  89.4 (br, 3bbb), 84.9 (br, 3bbb), 81.4 (br, 3abb), 80.9 (br, 3abb),
DOI: 10.1039/C9DT00700H
65.3 (br, 3abb), 62.5 (br, 3bbb) ratio 3abb:3bbb ca 4:1; (223 K)  91.4 (br, 3abb), 88.57 (d,
J 13.0, 3bbb), 88.3 (s, 3abb), 84.9 (s, 3bbb), 82.9 (br, 3abb), 81.8 (br, 3abb), 78.1 (br,
3abb), 76.3 (s, 3abb), 74.1 (s, 3abb), 71.0 (s, 3abb), 64.63 (s, 3bbb), 61.5 (s, 3abb), 57.2
(br, 3abb); Anal. calc. for Fe2P3S2O3N1C46H44: C, 59.54, H, 4.75; Found C, 59.74, H, 4.88.
Preparation of Fe2(CO)3(-SMe)2(-triphos) (4). A toluene solution (50 mL) of
triphos (0.100 g, 0.019 mmol) and Fe2(CO)6(-SMe)2 (0.100 g, 0.027 mmol) [73] was heated
at reflux for 16 h. Upon cooling to room temperature, the solvent was removed in vacuo and
the residue washed with hexane to remove unreacted starting materials. The dry solid was
dissolved in a minimum volume of dichloromethane and an excess of hexane was added
resulting in precipitation of a red solid identified as Fe2(CO)3(-SMe)2(-triphos) (4)
(0.094 g, 61 %). IR (CO) (CH2Cl2): 1952s, 1895vs cm-1; 1H NMR (CDCl3, 253 K):  7.95 –
31
7.12 (m, 25H, Ph), 3.40 – 1.50 (m, 8H, PCH2), 1.25 (s, 3H, Me), 0.87 (s, 3H, Me); P{1H}
NMR (CDCl3, 253 K):  88.9 (d, J 19.4), 81.9 (dd, J 19.4, 9.7), 62.5 (d, J 9.7); Anal. calc. for
Fe2P3S2O3C39H39: C, 56.79, H, 4.73; Found C, 56.35, H, 4.86.
Preparation of [Fe2(CO)3(-pdt)(-triphos)(-H)][BF4] (1H+). A red CH2Cl2 solution
(ca. 5 mL) of 1 (20 mg) turned cloudy and pink upon addition of a slight excess of HBF4.Et2O.
IR studies showed the clean generation of new carbonyl bands consistent with the consumption
of 1. The mixture was stirred for 30 mins and the solvent was removed under reduced pressure
to give an oil which was washed with diethyl ether (2 x 1 ml) and dried to give [Fe2(CO)3(-
pdt)(-triphos)(-H)][BF4] (1H+) as a dry pink solid (12 mg, 55%). IR (CO)(CH2Cl2):
1
2039vs, 1986s, 1964s cm-1; H NMR (CD2Cl2):  7.80–7.32 (m, 25H, Ph), 3.55 – 1.62 (m,
31
14H, CH2), -13.5 (dt, 1H, J 29.2, 22.4 Hz); P{1H} NMR (CD2Cl2) 88.9 (s), 85.2(s), 46.4 (s)
ppm.
IR protonation studies on 1-4. CH2Cl2 solutions (ca. 1 mL) of 1-4 were prepared by
dissolving 2.5×10-3 mmol of the complex and two molar equivalent amount of HBF4∙Et2O
(0.680 µL) was added to each. The resultant acid-containing solution was immediately
transferred to an IR cell and monitored over time. For 1 no discernible change in the IR
spectrum was noted after 30 mins. After 2 h the formation of small amounts of 1H+ was
noted. For 2 a slow (ca. 1 h) lightening of the red solution lead to the generation of new IR
21

Dalton Transactions
Page 22 of 41
View Article Online
bands at 2043vs and 1999s cm-1 but over this time there was also substantial decomposition.
DOI: 10.1039/C9DT00700H
For 3 a rapid (ca. 1 min) lightening of the solution occurred, and new IR bands were
observed at 2042vs, 2021m and 1988m cm-1. This species also decomposed slowly and all
attempts to further characterize the product were unsuccessful. Addition of excess acid resulted
in the formation of new IR bands at 2097w, 2051vs, 2033s and 1990s cm-1. For 4 the dark red
solution lighted slowly and new IR bands at 2031s, 2003s and 1964vs cm-1 appeared alongside
those of 4. These (CO) bands slowly decayed with the concomitant decomposition of the
product over ca. 3 h.
Chemical oxidation of 1. [Cp2Fe][PF6] (0.828 mg, 2.5×10-3 mmol) was dissolved in a
minimum amount of CH2Cl2 and this solution was then added to a CH2Cl2 solution
containing 2.5×10-3 mmol of 1 in CH2Cl2. The mixture was immediately transferred to an IR
cell and monitored over time. IR absorptions for 1 were immediately replaced by a new
strong absorption at 1947 cm-1 which is attributed to the formation of [Fe2(CO)3(-
pdt)(-triphos)][PF6] (1+).
Electrochemistry. Electrochemistry was carried out in deoxygenated dichloromethane with
0.1 M TBAPF6 as supporting electrolyte. The working electrode was a 3 mm diameter glassy
carbon electrode which was polished with 0.3 μm alumina slurry prior to each scan. The
counter electrode was a Pt wire and the quasi-reference electrode was a silver wire. All CVs
were referenced to the Fc+/Fc redox couple. An Autolab potentiostat (EcoChemie,
Netherlands) was used for all electrochemical measurements. Catalysis studies were carried
out by adding equivalents of HBF4∙Et2O (Sigma-Aldrich).
X-ray crystallography. Single crystals were mounted on fibres and diffraction data collected
at 150 K on a Bruker SMART APEX diffractometer using Mo-Kα radiation (λ = 0.71073 Å).
Data collection, indexing and initial cell refinements were all done using SMART software
[74]. Data reduction was accomplished with SAINT [75] software and SADABS [76]
programs were used to apply empirical absorption corrections. The structures were solved by
direct methods or Patterson methods and refined by full matrix least-squares (SHELXTL,
V6.12) [77]. All non-hydrogen atoms were refined anisotropically and hydrogen atoms were
included using a riding model. Scattering factors were taken from International Tables for X-
ray Crystallography. Details of data collection and structure refinement are given in Table 2.
22

Page 23 of 41
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT00700H
Computational methodology. The DFT calculations reported here were performed with the
Gaussian 09 package of programs [78]. The calculations were carried out with the B3LYP
functional, which utilizes the Becke three-parameter exchange functional (B3) [79] combined
with the correlation functional of Lee, Yang, and Parr (LYP) [80]. The iron atoms were
described by Stuttgart-Dresden effective core potentials (ecp) and SDD basis set, while the 6-
31G(d’) basis set was employed for the remaining atoms. The geometry-optimised structures
reported here represent minima based on zero imaginary frequencies (positive eigenvalues) or
as transition states with one negative eigenvalue, as established by frequency calculations
using the analytical Hessian. Unscaled vibrational frequencies were used to make zero-point
and thermal corrections to the electronic energies. The geometry-optimised structures have
been drawn with the JIMP2 molecular visualization and manipulation program [81].
ASSOCIATED CONTENT
Text, tables and cif files giving details of the X-ray crystallographic structure determinations
of
1
and
3
are available free of charge at http://pubs.acs.org or
http://www.ccdc.cam.ac.uk/conts/retrieving.html as supplementary publications: CCDC Nos.
61463812 and 1897288 Union Road, Cambridge, CB2 1FZ, U.K.: fax: +44-1223-336033; e-
mail: deposit@ccdc.cam.ac.uk). Atomic coordinates and energies of all optimised stationary
points available. Some additional electrochemical information (Figures ES1-S5) and
experiments, crystallographic data and computational details are given in electronic
supplementary information.
AUTHOR INFORMATION
Corresponding Authors
Email: Graeme Hogarth (graeme.hogarth@kcl.ac.uk); Katherine Holt (k.b.holt@ucl.ac.uk)
Notes.
The authors declare no competing financial interests.
ACKNOWLEDGEMENTS
23

Dalton Transactions
Page 24 of 41
View Article Online
We thank University College London for the provision of a studentship (DU), the EPSRC for
DOI: 10.1039/C9DT00700H
an Advanced Fellowship (KBH Grant No EP/D070538/1), the Commonwealth Scholarship
Commission (SG) and Professors Phillip Schollhamer and Jean Talarmin for hosting DU in
Brest enabling preliminary electrochemical measurements. GH thanks The Royal Society of
Chemistry for an International Authors Grant which allowed this work to be developed
during his visit to the University of North Texas and King’s College London for funding.
MGR acknowledges financial support from the Robert A. Welch Foundation (Grant B-1093-
MGR). Computational resources through the High-Performance Computing Services and
CASCaM at the University of North Texas are acknowledged, and we thank Prof. Michael B.
Hall (Texas A&M University) for providing us a copy of his JIMP2 program, which was
used to prepare the geometry-optimised structures reported here.
REFERENCES
1 For reviews, see for example: (a) D. Schilter, J.M. Camara, M.T. Huynh, S. Hammes-
Schiffer and T.B. Rauchfuss, Chem. Rev., 2016, 116, 8693-8749. (b) W. Lubitz, H. Ogata, O.
Rüdiger and E. Reijerse, Chem. Rev., 2014, 114, 4081–4148. (c) T. R. Simmons, G. Berggren, M.
Bacchi, M. Fontecave and V. Artero, Coord. Chem. Rev., 2014, 270–271, 127–150.
2 For reviews, see for example: (a) I.P. Georgakaki, L.M. Thomson, E.J. Lyon, M.B. Hall and
M.Y. Darensbourg, Coord. Chem. Rev., 2003, 238-239, 255-266. (b) D.J. Evans, C.J. Pickett,
Chem. Soc. Rev., 2003, 32, 268-275. (c) L. Sun, B. Åkermark and S. Ott, Coord. Chem. Rev.,
2005, 249, 1653-6663. (d) X. Liu, S.K. Ibrahim, C. Tard and C.J. Pickett, Coord. Chem. Rev.,
2005, 249, 1641-1652. (e) J.-F. Capon, F. Gloaguen, F.Y. Pétillon, P. Schollhammer and J.
Talarmin, Coord. Chem. Rev., 2005, 249, 1664-1676. (f) C. Tard and C.J. Pickett, Chem. Rev.,
2009, 109, 2245-2274. (g) J.-F. Capon, F. Gloaguen, F.Y. Pétillon, P. Schollhammer and J.
Talarmin, Coord. Chem. Rev., 2009, 253, 1476-1494. (h) N. Wang, M. Wang and L. Chen, L.
Sun, Dalton Trans., 2013, 42, 12059-12071. (i) S. Tschierlei, S. Ott and R. Lomoth, Energy
Env. Sci., 2011, 4, 2340-2352.
3 J.I. van der Vlugt, T.B. Rauchfuss and S.R. Wilson, Chem. Eur. J., 2006, 12, 90-98.
4 D. Chouffai, G. Zampella, J.-F. Capon, L. De Gioia, A. Le Goff, F.Y. Pétillon, P.
Schollhammer and J. Talarmin, Organometallics, 2012, 31, 1082-1091.
5 A.K. Justice, T.B. Rauchfuss and S.R. Wilson, Angew. Chem., Int. Ed., 2007, 46, 6152-6154.
6 A.K. Justice, L. De Gioia, M.J. Nilges, T.B. Rauchfuss, S.R. Wilson and G. Zampella, Inorg.
Chem., 2008, 47, 7405-7414.
24

Page 25 of 41
Dalton Transactions
View Article Online
7 M.T. Olsen, B.E. Barton and T.B. Rauchfuss, Inorg. Chem., 2009, 48, 7507-7509.DOI: 10.1039/C9DT00700H
8 G. Zampella, P. Fantucci and L. De Gioia, J. Am. Chem. Soc., 2009, 131, 10909-10917.
9 M.T. Olsen, T.B. Rauchfuss and S.R. Wilson, J. Am. Chem. Soc., 2010, 132, 17733-17740.
10 J.M. Camara and T.B. Rauchfuss, J. Am. Chem. Soc., 2011, 133, 8098-8101.
11 A. Silakov, M.T. Olsen, S. Sproules, E.J. Reijerse, T.B. Rauchfuss and W. Lubitz, Inorg.
Chem., 2012, 51, 8617-8628.
12 J.M. Camara and T.B. Rauchfuss, Nat. Chem., 2012, 4, 26-30.
13 J.C. Lansing, J.M. Camara, D.E. Gray and T.B. Rauchfuss, Organometallics, 2014, 33,
5897-5906.
14 B.E. Barton, M.T. Olsen and T.B. Rauchfuss, J. Am. Chem. Soc., 2008, 130, 16834-16835.
15 P.-Y. Orain, J.-F. Capon, F. Gloaguen, F.Y. Pétillon, P. Schollhammer, J. Talarmin, G.
Zampella, L. De Gioia and T. Roisnel, Inorg. Chem., 2010, 49, 5003-5008.
16 D. Chouffai, G. Zampella, J.-F. Capon, L. De Gioia, F. Gloaguen, F.Y. Pétillon, P.
Schollhammer and J. Talarmin, Inorg. Chem., 2011, 50, 12575-12585.
17 J.-F. Capon, F. Gloaguen, F. Y. Pétillon, P. Schollhammer and J. Talarmin, Eur. J. Inorg.
Chem., 2008, 4671-4681.
18 S. Ezzaher, J.-F. Capon, F. Gloaguen, F. Y. Pétillon, P. Schollhammer and J. Talarmin,
Inorg. Chem., 2007, 46, 3426-3428.
19 S. Ezzaher, J.-F. Capon, F. Gloaguen, F. Y. Pétillon, P. Schollhammer and J. Talarmin,
Inorg. Chem., 2007, 46, 9863-9872.
20 S. Ezzaher, J.-F. Capon, F. Gloaguen, N. Kervarec, F. Y. Pétillon, R. Pichon, P.
Schollhammer and J. Talarmin, C. R. Chim., 2008, 11, 906-914.
21 S. Ezzaher, J. -F. Capon, F. Gloaguen, F. Y. Pétillon, P. Schollhammer, J. Talarmin and N.
Kervarec, Inorg. Chem., 2009, 48, 48, 2-4.
22 S. Lounissi, J.-F. Capon, F. Gloaguen, F. Matoussi, F. Y. Pétillon, P. Schollhammer and J.
Talarmin, Chem. Commun., 2011, 47, 878-880.
23 D. Chouffai, G. Zampella, J.-F. Capon, L. De Gioia, A. Le Goff, F. Y. Pétillon, P.
Schollhammer and J. Talarmin, Organometallics, 2012, 31, 1082-1091.
24 S. Lounissi, G. Zampella, J.-F. Capon, L. De Gioia, F. Matoussi, S. Mahfoudhi, F.Y. Pétillon,
P. Schollhammer and J. Talarmin, Chem.-Eur. J., 2012, 18, 11123-11138.
25 C. Greco, P. Fantucci, L. De Gioia, R. Suarez-Bertoa, M. Bruschi, J. Talarmin and P.
Schollhammer, Dalton Trans., 2010, 39, 7320-7329.
26 F. I. Adam, G. Hogarth, I. Richards and B. E. Sanchez, Dalton Trans., 2007 2495-2498.
27 F. I. Adam, G. Hogarth and I. Richards, J. Organomet. Chem., 2007, 692, 3957-3968.
25

Dalton Transactions
Page 26 of 41
View Article Online
DOI: 10.1039/C9DT00700H
28 F. I. Adam, G. Hogarth, S. E. Kabir and I. Richards, C. R. Chim., 2008, 11, 890-905.
29 (a) S. Ghosh, G. Hogarth, N. Hollingsworth, K. B. Holt, I. Richards, M. G. Richmond, B.
E. Sanchez and D. Unwin, Dalton Trans., 2013, 42, 6775-6792. (b) S. Ghosh, G. Hogarth, N.
Hollingsworth, K. B. Holt, S. E. Kabir and B. E. Sanchez, Chem. Commun., 2014, 50, 945-
947. (c) S. Ghosh, S. Rana, N. Hollingsworth, M. G. Richmond, S. E. Kabir and G. Hogarth,
Inorganics, 2018, 6, 122-135.
30 (a) S. Ghosh, A. Rahaman, K.B. Holt, E. Nordlander, M.G. Richmond, S.E. Kabir and G.
Hogarth, Polyhedron, 2016, 116, 127-135. (b) F. Ridley, S. Ghosh, G. Hogarth, N.
Hollingsworth, K. B. Holt and D. G. Unwin, J. Electroanal. Chem., 2013, 703, 14-22.
31 S. Ghosh, B.E. Sanchez, I. Richards, M.N. Haque, K.B. Holt, M.G. Richmond and G.
Hogarth, J. Organomet. Chem., 2016, 812, 247-258.
32 (a) S. Rana, S. Ghosh, M. K. Hossain, A. Rahaman, G. Hogarth and S. E. Kabir, Trans.
Met. Chem., 2016, 41, 933-942. (b) S. Ghosh, S. E. Kabir and G. Hogarth, Trans. Met.
Chem., 2017, 42, 597-603.
33 A. K Justice, G. Zampella, L. De Gioia, T. B. Rauchfuss, J. I. van der Vlugt and S. R.
Wilson, Inorg. Chem., 2007, 46, 1655-1664.
34 N. Wang, M. Wang, T. Liu, T. Zhang and M. Darensbourg, L. Sun, Inorg. Chem., 2008, 47,
6948-6955.
35 P.-H.Zhao, M.-Y. Hu, J.-R. Li, Z.-Y. Ma, Y.-Z. Wang, J. He, Y.-L. Li and X.-F. Liu,
Organometallics, 2019, 38, 385-394.
36 P.-H. Zhao, Z.-Y. Ma, M.-Y. Hu, J. He, Y.-Z. Wang, X.-B. Jing, H.-Y. Chen, Z. Wang
and Y.-L. Li, Organometallics, 2018, 37, 1280-1290.
37 G. Hogarth and I. Richards, Inorg. Chem. Commun., 2007, 10, 66-70.
38 W. Hewertson and H.R. Watson, J. Chem. Soc., 1962, 1490-1494.
39 D.L. Dubois, A. Miedaner and R.C. Haltiwanger, J. Am. Chem. Soc., 1991, 113, 8753-8764.
40 A. Jablonskyté, J.A. Wright, S.A. Fairhurst, J.N.T. Peck, S.K. Ibrahim, V.S. Oganesyan
and C.J. Pickett, J. Am. Chem. Soc., 2011, 133, 18606-18609.
41 A. Jablonskyté, J.A. Wright and C.J. Pickett, Eur. J. Inorg. Chem., 2011, 1033-1037.
42 A. Jablonskyté, J.A. Wright and C.J. Pickett, Dalton Trans., 2010, 39, 3026-3034.
43 J.A. Wright and C.J. Pickett, Chem. Commun., 2009, 5719-5721.
44 C. Liu, J.N.T. Peck, J.A. Wright, C.J. Pickett and M.B. Hall, Eur. J. Inorg. Chem., 2011,
1080-1093.
45 J.A. Wright, L. Webster, A. Jablonskyté, P.M. Woi, S.K. Ibrahim and C.J. Pickett, Faraday
Discuss., 2011, 148, 359-371.
26

Page 27 of 41
Dalton Transactions
View Article Online
46 M.G.I. Galinato, C.M. Whaley, D. Roberts, P. Wang and N. Lehnert, Eur. J. Inorg. Chem.,
DOI: 10.1039/C9DT00700H
2011, 1147-1154.
47 S.L. Matthews and D.M. Heinekey, Inorg. Chem., 2010, 49, 9746-9748.
48 A. Jablonskyté, L.R. Webster, T.R. Simmons, J.A. Wright and C.J. Pickett, J. Am. Chem.
Soc., 2014, 136, 13038-13044.
49 L. Schwartz, G. Eilers, L. Eriksson, A. Gogoll, R. Lomoth and S. Ott, Chem. Commun.,
2006, 520–522.
50 G. Eilers, L. Schwartz, M. Stein, G. Zampella, L. De Gioia, S. Ott and R. Lomoth, Chem.
Eur. J., 2007, 13, 7075–7084.
51 S. Ott, M. Kritikos, B. Åkermark, L. Sun and R. Lomoth, Angew. Chem., Int. Ed., 2004,
43, 1006–1009.
52 F. Wang, M. Wang, X. Liu, K. Jin, W. Dong, G. Li, B. Åkermark and L. Sun, Chem.
Commun., 2005, 3221–3223.
53 J.-F. Capon, S. Ezzaher, F. Gloaguen, F. Y. Pétillon, P. Schollhammer and J. Talarmin,
Chem.–Eur. J., 2008, 14, 1954–1964.
54 H. Li and T. B. Rauchfuss, J. Am. Chem. Soc., 2002, 124, 726–727.
55 S. Jiang, J. Liu, Y. Shi, Z. Wang, B. Åkermark and L. Sun, Dalton Trans., 2007, 896–902.
56 G. Zampella, P. Fantucci and L. De Gioia, Chem. Commun., 2010, 8824-8826.
57 A.R. Finkelmann, M.T. Stiebritz and M. Reiher, Chem. Sci., 2014, 5, 215-221.
58 (a) W. Wang, M.J. Nilges, T.B. Rauchfuss and M. Stein, J. Am. Chem. Soc., 2013, 135,
3633-3639. (b) Y.-C. Liu, K.-T. Chu, Y.-L. Huang, C.-H. Hsu, G.-H. Lee, M.-C. Tseng and
M.-H. Chiang, ACS Catalysis, 2016, 6, 2559-2576.
59 T. Liu and M.Y. Darensbourg, J. Am. Chem. Soc., 2007, 129, 7008-7009.
60 C.M. Thomas, T. Liu, M.B. Hall and M.Y. Darensbourg, Inorg. Chem., 2008, 47, 7009-7024.
61 M.R. Carlson, D.L. Gray, C.P. Richers, W. Wang, P.-H. Zhao, T.B. Rauchfuss, V.
Pelmenschikov, C.C. Pham, L.B. Gee, H. Wang and S.P. Cramer, Inorg. Chem., 2018, 57, 1988-
2001.
61 S. Ezzaher, J.-F. Capon, N. Dumontet, F. Gloaguen, F.Y. Petillon and P. Schollhammer, J.
Talarmin, J. Electroanal. Chem., 2009, 626, 161-170.
63 M.G. Basallote, M. Besora, C.E. Castillo, M.J. Fernández-Trujillo, A. Lledós, F. Maseras
and M.A. Máñez, J. Am. Chem. Soc., 2007, 129, 6608-6618.
64 A.G. Algarra, M.G. Basallote, M.J. Fernández-Trujillo, R. Llusar, V.S. Safont and C.
Vicent, Inorg. Chem., 2006, 45, 5774-5784.
27

Dalton Transactions
Page 28 of 41
65 M.E. Carroll, B.E. Barton, T.B. Rauchfuss and P.J. Carroll, J. Am. Chem. Soc., 2012,V1ie3w4Ar,ticle Online
DOI: 10.1039/C9DT00700H
18843-18852.
66 B.E. Barton and T.B. Rauchfuss, Inorg. Chem., 2008, 47, 2261-2263.
67 A.K. Justice, M.J. Nilges, T.B. Rauchfuss, S.R. Wilson, L. De Gioia and G. Zampella, J. Am.
Chem. Soc., 2008, 130, 5293-5301.
68 W. Wang, T.B. Rauchfuss, L. Zhu and G. Zampella, J. Am. Chem. Soc., 2014, 136, 5773-
5782.
69 M.T. Huynh, W. Wang, T.B. Rauchfuss and S. Hammes-Schiffer, Inorg. Chem., 2014, 53,
10301-10311.
70 R. Zaffaroni, T.B. Rauchfuss, D.L. Gray, L. De Gioia and G. Zampella, J. Am. Chem. Soc.,
2012, 134, 19260-19269.
71 A. Winter, L. Zsolnai and G. Huttner, Zeit. Naturforsch. Teil B: Anorg. Chem. Org. Chem.,
1982, 37, 1430-1436.
72 J.D. Lawrence, H. Li and T.B. Rauchfuss, Chem. Commun., 2001, 1482-1483.
73 R.B. King, J. Am. Chem. Soc., 1962, 84, 2460.
74 SMART Version 5.628, Bruker AXS, Inc., Analytical X-ray Systems, 5465 East Cheryl
Parkway, Madison WI 53711-5373, 2003.
75 SAINT Version 6.36A, Bruker AXS, Inc., Analytical X-ray Systems, 5465 East Cheryl
Parkway, Madison WI 53711-5373, 2002.
76 G. M. Sheldrick, SADABS Version 2.10, University of Göttingen, 2003.
77 G. M. Sheldrick, Acta Cryst., 2008, A64, 112-122.
78 M.J. Frisch, et al., Gaussian 09, Revision E.01, Gaussian, Inc., Wallingford, CT, USA,
2009.
79 A.D. Becke, J. Chem. Phys., 1993, 98, 5648-5652.
80 C. Lee and W. Yang, R.G. Parr, Phys. Rev. B, 1993, 37, 785-789.
81 (a) JIMP2, version 0.091, a free program for the visualization and manipulation of
molecules: M.B. Hall and R.F. Fenske, Inorg. Chem., 1972, 11, 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.
28

Page 29 of 41
Dalton Transactions
Table 2 Crystallographic data and structure refinement for 1-2
View Article Online
Compound
empirical formula
formula weight
temp (K)
1.CH2Cl2
C41H41O3Fe2S2P3Cl2
921.37
2.0.25H2O
DOI: 10.1039/C9DT00700H
C46H44.5O3.25Fe2S2P3N1
932.06
150(2)
150(2)
wavelength (Å)
crystal system
space group
a (Å)
0.71073
0.71073
triclinic
triclinic
C2/c
P1
47.771(9)
11.747(2)
15.183(3)
90
9.4439(8)
9.6258(9)
22.437(2)
99.783(1)
95.331(2)
94.278(2)
1992.8(3)
2
b (Å)
c (Å)
α (°)
β (°)
97.984(4)
90
γ (°)
Volume (Å3)
Z
8438(3)
8
Dcalc (Mg m−3)
μ (Mo Kα) (mm−1)
F(000)
1.467
1.5369
0.944
1.1278
3860
948
crystal color
crystal size (mm)
θ range (°)
limiting indices
red
orange
0.14 × 0.13 × 0.09
2.45−28.34
−61 ≤ h ≥ 63
−15 ≤ k ≥ 15
–20 ≤ l ≥ 20
direct methods
356729
0.16 × 0.15 × 0.04
1.85−28.29
−12 ≤ h ≥ 12
−12 ≤ k ≥ 12
−29 ≤ l ≥ 28
direct methods
17384
structure solution
reflections collected
independent reflections (Rint)
max. and min. transmission
data/restraints/parameters
goodness of fit on F2
10136 (0.0550)
0.9199 and 0.8792
10136 / 0 / 518
1.049
9087 (0.0228)
0.9563 and 0.8403
9087 / 0 / 478
1.046
final R indices [I > 2σ(I)]
R1 = 0.0500
wR2 = 0.1107
R1 = 0.0736
wR2 = 0.1203
0.707 and −0.497
R1 = 0.0405
wR2 = 0.0950
R1 = 0.0510
wR2 = 0.0998
1.659 and −1.013
R indices (all data)
largest diff. peak and hole (e Å−3)
29