Effect of the S-to-S bridge on the redox properties and H2 activation performance of diiron complexes related to the [FeFe]-hydrogenase active site

Article abstract of DOI:10.1039/c6dt02953a

Three biomimetic models of the [FeFe]-hydrogenase active site, namely diiron dithiolates of [(μ-edt){Fe(CO)₃}{Fe(CO)(κ²-PNP)}] (1, edt = ethane-1,2-dithiolate, PNP = Ph₂PCH₂N(nPr)CH₂PPh₂), [(μ-bdtMe){Fe(CO)₃}{Fe(CO)(κ²-PNP)}] (2, bdtMe = 4-methylbenzene-1,2-dithiolate), and [(μ-adtBn){Fe(CO)₃}{Fe(CO)(κ²-PNP)}] (3, adtBn = N-benzyl-2-azapropane-1,3-dithiolate), were prepared and structurally characterized. These complexes feature the same PNP ligand but different S-to-S bridges. Influence of the S-to-S bridge on the electrochemical properties and chemical oxidation reactivity of 1-3 was studied by cyclic voltammetry and by in situ IR spectroscopy. The results reveal that the S-to-S bridge has a considerable effect on the oxidation reactivity of 1-3 and on the stability of in situ generated single-electron oxidized complexes, [1]⁺, [2]⁺, and [3]⁺. The performances of [1]⁺ and [2]⁺ for H₂ activation were explored in the presence of a mild chemical oxidant, while rapid decomposition of [3]⁺ thwarted the further study of this complex. Gratifyingly, 1 was found to be catalytically active, although in a low turnover number, for H₂ oxidation in the presence of excess mild oxidant and a proton trapper under 1 atm H₂ at room temperature.

Full text of DOI:10.1039/c6dt02953a

Dalton
Transactions
View Article Online
View Journal
PAPER
Effect of the S-to-S bridge on the
redox properties and H2 activation
Cite this: DOI: 10.1039/c6dt02953a
performance of diiron complexes related
to the [FeFe]-hydrogenase active site†
Minglun Cheng,a Mei Wang,*a Dehua Zhenga and Licheng Suna,b
Three biomimetic models of the [FeFe]-hydrogenase active site, namely diiron dithiolates of
[(μ-edt){Fe(CO)3}{Fe(CO)(κ2-PNP)}] (1, edt
= ethane-1,2-dithiolate, PNP = Ph2PCH2N(nPr)CH2PPh2),
[(μ-bdtMe){Fe(CO)3}{Fe(CO)(κ2-PNP)}] (2, bdtMe = 4-methylbenzene-1,2-dithiolate), and [(μ-adtBn){Fe(CO)3}
{Fe(CO)(κ2-PNP)}] (3, adtBn = N-benzyl-2-azapropane-1,3-dithiolate), were prepared and structurally
characterized. These complexes feature the same PNP ligand but different S-to-S bridges. Influence
of the S-to-S bridge on the electrochemical properties and chemical oxidation reactivity of 1–3 was
studied by cyclic voltammetry and by in situ IR spectroscopy. The results reveal that the S-to-S bridge has
a considerable effect on the oxidation reactivity of 1–3 and on the stability of in situ generated single-
electron oxidized complexes, [1]+, [2]+, and [3]+. The performances of [1]+ and [2]+ for H2 activation were
explored in the presence of a mild chemical oxidant, while rapid decomposition of [3]+ thwarted the
further study of this complex. Gratifyingly, 1 was found to be catalytically active, although in a low turnover
number, for H2 oxidation in the presence of excess mild oxidant and a proton trapper under 1 atm H2 at
room temperature.
Received 26th July 2016,
Accepted 4th October 2016
DOI: 10.1039/c6dt02953a
www.rsc.org/dalton
Introduction
[FeFe]-hydrogenase ([FeFe]-H2ase) enzymes are a kind of
metalloenzymes which reversibly catalyze the redox of H2/H+ at
remarkable reaction rates.1–4 At the end of the last century, the
structure of [FeFe]-H2ases was elucidated by high-resolution
X-ray crystallography.5,6 Subsequent theoretical and spectro-
scopic studies suggest that the [FeFe]-H2ases exist in different
redox states, namely the super reduced state (Hsred),7 the
reduced state (Hred), the oxidized active state (Hox), the CO-
8
9,10
air
CO
Fig. 1 Hred and Hox states of the [FeFe]-H2ase active site.
inhibited state (H ), and the oxidized inactive state (H ).
ox
ox
Among them, Hred and Hox states (Fig. 1) have drawn extensive
attention as the former is believed to be an active state for H2
formation and the latter to be an active state for H2 uptake.
Over the past decade, studies on the mimicking of [FeFe]-
H2ases have been carried out mostly on the Hred models and
their electrochemical and photochemical proton reduction
performances,11–17 while less attention has been paid to
mimic the Hox state and the H2 oxidation reaction.18–23 To
date, only a few mixed-valence Hox models were found to be
capable of activating H2 under mild conditions. The pioneer-
ing work by Rauchfuss and co-workers revealed that the
mixed-valence complexes, [(μ-adtBn){Fe(CO)2(PMe3)}{Fe(CO)
aState Key Laboratory of Fine Chemicals, DUT-KTH Joint Education and Research
Center on Molecular Devices, Dalian University of Technology, Dalian 116024,
China. E-mail: symbueno@dlut.edu.cn
bDepartment of Chemistry, KTH Royal Institute of Technology, Stockholm, 10044,
Sweden
†Electronic supplementary information (ESI) available: Selected bond lengths
(Å) and angles (°) for 1–3 and 2′ (Table S1), crystal data and structural refine-
ments details for 1–3 and 2′ (Table S2), molecular structure of 2′ (Fig. S1), IR and
(κ2-dppv)}]+ (adtBn
dppv = cis-1,2-bis-(diphenylphosphino)ethene) and [(μ-adtBn)
= N-benzyl-2-azapropane-1,3-dithiolate,
31P NMR spectra (Fig. S2–S7), and additional CV of
1 (Fig. S8). CCDC
1495096–1495099. For ESI and crystallographic data in CIF or other electronic
{Fe(CO)3}{Fe(CO)(κ2-dppn)}]+ (dppn
=
1,8-bis(diphenylphos-
format see DOI: 10.1039/C6DT02953A
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 2 Structures of the [FeFe]-H2ase mimics 1–4.
phino)naphthalene), reacted with 1 atm H2 at 20 °C in the 4-methylbenzene-1,2-dithiolate), and [(μ-adtBn){Fe(CO)3}
presence of FcBArF4 (Fc+ = [FeCp2]+, Cp = C5H5; BArF − = B(3,5- {Fe(CO)(κ2-PNP)}] (3). These complexes, together with the pre-
4
C6H3(CF3)2)4−) to generate intermediates containing both a viously reported analogous diiron complex [(μ-pdt){Fe(CO)3}
μ-hydride at the diiron center and a proton at the pendant N {Fe(CO)(κ2-PNP)}] (4, pdt = propane-1,3-dithiolate), feature the
atom.24,25 Subsequently,
a
diiron azadithiolate model, same pendant amine-containing diphosphine ligand, but have
[(μ-adtBn){Fe(CO)2(FcP*)}{Fe(CO)(κ2-dppv)}] (FcP*
=
Cp*Fe different S-to-S bridges (Fig. 2). The influence of the S-to-S
(C5Me4CH2PEt2), Cp* = C5Me5), was designed by the same bridge on the electrochemical properties and chemical oxi-
group with FcP* as a one-electron redox module to mimic the dation reactivity of 1–3 and on the stability of their corres-
redox function of an Fe4S4 cluster in the [FeFe]-H2ases, which ponding FeIIFeI mixed-valence complexes was studied by cyclic
gave the first example for the catalytic oxidation of 1 atm H2 by voltammetry and in situ IR spectroscopy. Among the Hox
a model of [FeFe]-H2ases in the presence of excess oxidant models of [1]+–[4]+, [1]+ and [4]+ were catalytically active for H2
(FcBArF4) and a proton trapper (P(o-tol)3) at room tempera- oxidation in the presence of FcBArF and P(o-tol)3 under mild
4
ture.26 Except for this special example, the previously reported conditions, whereas [2]+ was decomposed to a mononuclear
Hox models with a built-in amine in the center of the S-to-S iron complex [Fe(CO)(κ2-bdtMe)(κ2-PNP)] (2′) under the same
bridge activated H2 only stoichiometrically under 1 atm H2 in reaction conditions. Complex 2′ was readily oxidized to an FeIII
the presence of a mild oxidant and a proton trapper, because species (2″) in the presence of FcBArF4. Unfortunately, the poor
the stable μ-hydride intermediates formed in the reaction pro- stability of [3]+ thwarted the study on [3]+ for the H2 activation
cesses cannot be deprotonated by a weak base to recover the reaction.
starting neutral diiron complexes and therefore cannot finish
the catalytic cycle.
In our previous work, we found that the μ-hydride in the
diiron dithiolate complexes with a built-in amine in the
center of an unsymmetrically chelating diphosphine ligand,
Results and discussion
Preparation and characterization of 1–3
Complexes 1–3 were prepared from the reaction of [(μ-edt)
Fe2(CO)3], [(μ-bdtMe)Fe2(CO)6], and [(μ-adtBn)Fe2(CO)6] with
the PNP ligand, respectively, in refluxing toluene for 3 h.
Analytically pure complexes of 1–3 were obtained by chromato-
graphy and recrystallization of the crude products. The pro-
ducts were characterized by ESI-HRMS (electrospray ionization
Ph2PCH2N(nPr)CH2PPh2 (PNP), could be readily deprotonated
in the presence of a weak base.27,28 The DFT calculation
suggests that the deprotonation of the PNP-coordinating
diiron μ-hydride complex takes place via an indirect pathway,
in which the accessible pendant amine of the PNP ligand acts
as a proton shuttle so as to lower the energy barriers for the
deprotonation of diiron μ-hydride complexes.29 Moreover, we
found that complex [(μ-pdt){Fe(CO)3}{Fe(CO)(κ2-PNP)}] (pdt =
propane-1,3-dithiolate) catalytically activated H2 under 1 atm
H2 at 25 °C in the presence of excess Fc+ and P(o-tol)3.30 One
of the general questions arising from the study on this type of
Fe2S2 complexes concerns the influence of the S-to-S bridge on
the structure and reactivity of the underlying Fe2(CO)6−xLx
subunit. A better understanding of how the S-to-S bridge
affects the redox properties, stability and reactivity of the
underlying diiron centers will be helpful for the reasonable
design of new Hox models for H2 activation.
1
high resolution mass spectroscopy), IR, H and 31P{1H} NMR
spectroscopy. The HRMS data of the products are in good
agreement with the proposed compositions of 1–3. The 1H and
31P{1H} NMR spectra confirm that complexes 1–3 have similar
structures, in which one of the iron centers is chelated by the
PNP ligand. The IR data of the νCO absorptions of 1–3 are sum-
marized in Table 1, together with the νCO absorptions of the
Table 1 Comparison of νCO absorptions of the related FeIFeI and FeIIFeI
complexes
In this context, we tried to find the influence of the S-to-S
bridge on the redox properties and H2-uptake capability of the
Hox models bearing a PNP ligand. Here we report the prepa-
ration and molecular structures of three new Fe2S2 complexes,
νCO (cm−1),
νCO (cm−1),
Complex
n = 0 (FeIFeI)
n = +1 (FeIIFeI)
Ref.
[1]n
[2]n
[3]n
2028, 1956, 1900
2033, 1963, 1901
2026, 1955, 1901
2025, 1953, 1898
2088, 2031, 1901
2090, 2036, 1911
2086, 2027, 1908
2081, 2022, 1900
This work
This work
This work
27
[(μ-edt){Fe(CO)3}{Fe(CO)(κ2-PNP)}] (1, edt
=
ethane-1,2-
dithiolate), [(μ-bdtMe){Fe(CO)3}{Fe(CO)(κ2-PNP)}] (2, bdtMe = [4]n
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Fig. 3 Molecule structures of 1 (left), 2 (middle) and 3 (right) as ball and stick drawings. Hydrogen atoms are omitted for clarity.
analogous complex 4.27,28 These diiron complexes each ing diiron complexes, the electrochemistry of 1–4 was studied
contain four carbonyl ligands and an unsymmetrically biden- by cyclic voltammetry in CH2Cl2. Complex 1 displayed a revers-
ox1
1=2
tate diphosphine ligand. In the IR spectra, the νCO absorption ible oxidation event at E
= −0.11 V (all potentials given in
of the lowest wavenumber around 1900 cm−1 is attributed to this paper are versus Fc+/0 potential.) with a peak potential
the vibration of the CO ligand in the Fe(CO)(PNP) unit. The Eopx1 = −0.05 V for the FeIFeI/FeIIFeI couple and an irreversible
other two νCO absorptions at higher wavenumbers are assigned oxidation peak at Eopx2 = +0.85 V for the FeIIFeI/FeIIFeII process
to the vibrations of CO ligands in the Fe(CO)3 unit. In general, (Fig. 4a), which are similar to the corresponding oxidation
the νCO bands of 1–4 display a tendency to blue shift as the events of 4 (Fig. 4d). It is noted that in addition to the
S-to-S bridge is changed from pdt to adtBn, edt, and then to reduction events for [1]2+/[1]+ at about +0.68 V and for [1]+/1 at
bdtMe, indicating that the change of the S-to-S bridge can −0.18 V, an extra small reduction peak appears at −0.51 V in
finely tune the electron back-bonding from the iron center to the second run of cyclic voltammetry (CV), which is not
the anti-bonding π*-orbital of the coordinated CO. Among observed in the first run scanning in the cathodic direction.
these diiron complexes, 2 exhibits the apparently higher wave- When the anodic scan returns before +0.65 V to avoid the
numbers for the νCO absorptions of the Fe(CO)3 unit due to second oxidation process, the reduction peak at −0.51 V does
the electron-withdrawing character of the aromatic bdt-bridge.
not appear in the subsequent cathodic scan. These obser-
The molecular structures of 1–3 were determined by single vations indicate that the reduction event at −0.51 V is the
crystal X-ray diffraction analyses, which are shown in Fig. 3 as reduction of the decomposed product of the second oxidation
ball and stick drawings. Selected bond lengths and angles for process and the structure of this product is currently not clear.
1–3 are listed in Table S1.† The Fe2S2 cores of these complexes Noticeably, such an extra weak-intensity reductive event also
have a butterfly framework and each iron atom is coordinated appears in the CV of 4 but not observed in the CVs of 2 and 3.
in a pseudo-square pyramidal geometry as in other biomimetic The potential of the first oxidation event (Eo1=x12) of 2 (Fig. 4b) is
diiron dithiolate complexes.31–33 In the solid state, the chelat- almost identical to the [1]+/1 couple, while the second irrevers-
ing PNP ligand lies in an apical/basal configuration at the Fe1 ible oxidation peak (Eopx2 = +0.60 V) of 2 is 270 mV less positive
atom of each complex, being analogous to the previously than the [1]2+/[1]+ couple. The CV of 3 shows a quasi-reversible
ox1
reported PNP-coordinating diiron dithiolate complex with a oxidation event at E = −0.20 V (Eopx1 = −0.08 V) and an irre-
1=2
pdt bridge.28 The Fe–Fe bond distances are in the order of versible peak at Eopx2 = +0.42 V for the further FeIIFeI to FeIIFeII
2 (2.5429(6) Å) < 1 (2.5618(7) Å) < 3 (2.5825(10) Å), indicating oxidation process (Fig. 4c).
that the stronger rigidity of the S-to-S bridge leads to a shorter
Table 2 summarizes the oxidation potentials of 1–4,30
Fe–Fe bond. It is worth noting that the Fe2–S bonds are together with those of [Fe2(μ-edt)(CO)3(κ2-dppv)(PMe3)] (5),34
0.013–0.015 Å longer than the Fe1–S bonds in the same [Fe2(μ-pdt)(CO)3(κ2-dppv)(PMe3)] (6), and [Fe2(μ-adtBn)(CO)3(κ2-
complex, and the Fe2–S bonds (2.2654–2.2875 Å) in 2 and 3 are dppv)(PMe3)] (7), to gain an insight into the influence of the
observably longer than those (2.2582–2.2639 Å) in 1. The S-to-S linker on the electrochemical redox properties of [FeFe]-
subtle difference in the Fe–S bond lengths may influence the H2ase models. The comparison of the oxidation potentials for
stability of the Fe2S2 core during the oxidation process, which the two series of diiron complexes (1–4 and 5–7) provides us
will be discussed in the following section.
some interesting hints. The oxidation potential related to the
FeIFeI/FeIIFeI couple is not very susceptible to the electronic
properties of the S-to-S bridge, especially when the bridges
Electrochemical properties of 1–4
To explore the influence of the S-to-S bridge in the second have the same number of atoms, while when the S-to-S bridge
coordination sphere on the redox properties of PNP-coordinat- is shortened from 3 to 2 atoms a 110–140 mV positive shift of
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 4 Cyclic voltammograms of (a) 1, (b) 2, (c) 3, and (d) 4 (1.0 mM) with full scan (black lines) and with anodic scan returned before the onset of
the second oxidation event (red lines) in 0.1 M nBu4NPF6/CH2Cl2 at a scan rate of 100 mV s−1
.
ΔEox (ΔEox = |Epox2 − Epox1|) is a significant electrochemical
datum that will directly influence the H2 activation perform-
ance of Fe2S2 complexes. Regardless of the number and the
type of phosphine ligands, complexes bearing an edt and a pdt
bridge display large ΔEox values (0.83–0.97 V), which prevent
the overlap of the two oxidation steps and guarantee the clean
formation of the one-electron oxidized intermediate under
controlled oxidation reaction. Furthermore, the chemical rever-
sibility of the first oxidation process evinced by the current
ratio, ipa/ipc (ipa = anodic peak current, ipc = cathodic peak
current), is about 0.96, 0.88, 0.78, and 1 for 1–4, respectively,
indicating that the S-to-S bridge does have a considerable
effect on the stability of these [FeFe]-H2ase models and their
oxidized derivatives.
Table 2 Redox potentials (V vs. Fc+/0) of 1–4 and some related diiron
complexes in nBu4NPF6/CH2Cl2
Complex
E
(V) Eopx1 (V) Epox2 (V) ΔEox a (V) Ref.
ox1
1=2
1 (edt, PNP)
−0.11 −0.05
−0.10 −0.04
−0.20 −0.09
−0.22 −0.12
−0.47 −0.41
−0.61 −0.54
+0.85 0.90
+0.60 0.64
+0.42 0.51
+0.80 0.93
+0.42 0.83
+0.43 0.97
This work
This work
This work
30
34
34
34
2 (bdtMe, PNP)
3 (adtBn, PNP)
4 (pdt, PNP)
5 (edt, dppv, PMe3)
6 (pdt, dppv, PMe3)
7 (adtBn, dppv, PMe3) −0.64 −0.58 −0.06 0.52
a ΔEox = |Epox2 − Epox1|.
the first oxidation potential is observed for both series. In con-
trast, the potential corresponding to the further oxidation of
FeIIFeI to FeIIFeII is highly dependent on the electronic pro-
Chemical oxidation of 1–3
perties of the S-to-S bridge but not very susceptible to the length Chemical oxidation reactions of 1–3 in CH2Cl2 were monitored
of the S-to-S bridge. It is worth noting that the second oxidation by in situ IR spectroscopy under N2 at room temperature using
potentials of the adt-bridging Fe2S2 complexes are considerably FcBArF as an oxidant. Upon addition of 1 equiv. of FcBArF to
4
4
less positive than those of their corresponding Fe2S2 complexes the solution of 1, the νCO absorptions of 1 at 2028, 1956, and
bearing a pdt bridge, e.g. Eopx2 = +0.42 V for 3 (adtBn) versus 1900 cm−1 decreased rapidly (Fig. 5a), accompanied by the
+0.80 V for 4 (pdt) and −0.06 V for 7 (adtBn) versus +0.43 V for appearance of a new νCO band at 2088 cm−1, while two other
6 (pdt). Such an apparent influence of the adt bridge on the new νCO bands at 2031 and 1901 cm−1 happened to overlap
FeIIFeI/FeIIFeII oxidation potential is most possibly due to the with the νCO absorptions of 1. Therefore the νCO band of 1 at
coordinative interaction of the iron center and the pendant 1956 cm−1 was used as an indicator of the reaction extent. The
amine of the adt bridge as verified in the case of 7.34
total disappearance of this band within 2 min suggests a rapid
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Fig. 5 Changes of νCO absorptions in the in situ IR spectra of 1 (a), 2 (b), and 3 (c) in CH2Cl2 with the addition of an equiv. of FcBArF4, and of the
in situ generated [1]+ (d), [2]+ (e), and [3]+ (f) in CH2Cl2, standing under N2 at room temperature over an extended time.
and complete conversion of 1. The obvious blue shifts of the bdtH2, CO, and diphosphine ligand in MeOH in the presence
first two νCO bands by 60 and 75 cm−1 indicate the formation of triethylamine, and used as catalysts for hydrogen evolution
of the one-electron oxidized product [1]+, and the broad low- reaction (HER).36,37 To attribute the two νCO absorptions in the
intensity band at 1901 cm−1 is typical of a semibridging CO final IR spectrum of Fig. 5e, we studied the oxidation reaction
ligand as reported for other FeIIFeI models.19–22 Similarly, the of 2′ by using an in situ IR spectroscopic technique. The νCO
first two νCO absorptions of 2 and 3 were blue-shifted rapidly absorption of 2′ at 1915 cm−1 completely disappeared within
by 57–73 cm−1 as 1 equiv. of FcBArF4 was added to the solution 3 min upon addition of 1 equiv. of FcBArF (Fig. S2†). In the
4
(Fig. 5b, c and Table 1), indicating the formation of [2]+ and meantime, a νCO band emerged at 1992 cm−1, suggesting that
[3]+, respectively.
the iron center of 2′ is oxidized from FeII to FeIII (2″). On the
Fig. 5d–f show that the in situ generated [1]+ in CH2Cl2 is basis of these observations, we assume that the initially gener-
more stable than [2]+ and [3]+, but it still slowly decomposed ated decomposed product of [2]+ is an FeII mononuclear
when standing in solution at room temperature under N2 for complex (2′), which is oxidized in situ to a FeIII complex by Fc+
several hours. The intensity of the νCO bands of [2]+ gradually in solution.
dropped and almost disappeared after the solution of [2]+
In comparison, the νCO absorptions of [3]+ totally vanished
stood for 2 h (Fig. 5e), accompanied by the appearance of two after the solution of [3]+ stood under N2 for about 10 min, and
νCO bands at 1915 and 1992 cm−1. As these two new νCO bands more than five low-intensity bands were observed in the region
did not emerge synchronously, they should be attributed to of 2105–1990 cm−1 (Fig. 5f). The quick decomposition of [3]+
two decomposed products. By removal of the solvent from the in solution does not allow for further study of [3]+ on H2 acti-
resulting solution and flash column chromatography, one of vation. We notice that the analogous complex, [Fe2(μ-adtBn)
the decomposed products of [2]+ was isolated as a black crys- (CO)4(κ2-dppv)]+, was also found to be unstable under identical
talline solid. The MS, IR, 1H and 31P{1H} NMR spectra revealed conditions and the reason for the instability of this mixed-
that the dinuclear complex [2]+ has split to a mononuclear valence complex was assumed to be disproportionation caused
complex by losing the Fe(CO)3 moiety. Similar disruption of by amine binding.25 Altogether, the stability of this series of
the Fe2S2 core was found during the ligand substitution reac- mixed-valence diiron complexes is in the order of [4]+ > [1]+ >
tion of all-CO diiron complexes bearing a bdt or its derivative [2]+ ≫ [3]+. The stability of the FeIIFeI intermediate is one of
bridge.35 The strong rigidity of the bdt bridge makes this type the key factors that determine the performance of Hox models
of diiron dithiolate complex less stable than those bearing flex- in the H2 activation due to the limit of the binding rate of H2
ible alkylene bridges. The crystallographic analysis further con- to the open coordination site at the rotated iron of the FeIIFeI
firmed that the isolated decomposed complex is a coordinately core.
unsaturated FeII complex, [Fe(CO)(κ2-bdtMe)(κ2-PNP)] (2′,
Activation of H2 by [1]+ in the presence of Fc+
Fig. S1†), which features a square pyramidal configuration
with the CO at the pyramid apex. Analogous complexes The reaction of [1]+ with 1 atm H2 in the presence of FcBArF
4
[Fe(CO)(κ2-bdt){κ2-(Ph2PCH2)2N(R)}] (R = CH2CH(OCH2CH3)2, in CH2Cl2 at 25 °C was studied by in situ IR spectroscopy. Two
CH2COONa) were prepared by the one-pot reaction of FeCl2, extra equiv. of FcBArF were added to the CH2Cl2 solution
4
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
containing the in situ generated [1]+, and H2 was bubbled into in situ IR spectra (Fig. S5†), indicating that part of 1 was
the solution for 10 min. As expected, the three νCO absorptions decomposed during the reaction. The 31P{1H} NMR spectrum
of [1]+ gradually decreased and eventually disappeared within of the resulting solution does not show discernable signals,
a period of 2 h (Fig. S3†). In contrast, it took 4 h for the com- possibly due to the disturbance of the decomposed paramag-
plete conversion of [4]+ under the same conditions,30 indicat- netic iron species existing in the solution.
ing that [1]+ is more active than [4]+ for H2 activation, but it is
The reaction of [2]+ with 1 atm H2 in the presence of
less stable than [4]+. After the reaction of [1]+ with H2/Fc+ for 2 equiv. of FcBArF was also studied. The in situ IR spectra
4
2 h, the resulting solution displayed four νCO bands at 2107, (Fig. S6†) show that [2]+ decomposes to the mononuclear
2051, 2003, and 1972 cm−1 in the IR spectrum (Fig. 6a). A com- complex regardless of whether or not the reaction was carried
parison of these νCO absorptions with those observed for the out under H2 or N2, and the decomposed FeII complex was oxi-
cationic products formed by the direct protonation of 1 in the dized to the FeIII complex (2″) due to the existence of an excess
presence of TfOH (Fig. 6b and S4†) reveals that the resulting of Fc+ in the solution. [2]+ cannot activate H2 in the presence
solution contains a mixture of [1(Hμ)(HN)]2+, [1(Hμ)]+, and of Fc+ under mild conditions.
[1(HN)]+, as well as the decomposed species of [1]+. Upon
Moreover, the catalytic activation of H2 by 1 was studied.
addition of an excess of P(o-tol)3 to the resulting solution, the First, FcPF6 was used for the oxidation of 1 to [1]+, and the fol-
neutral complex 1 was recovered partially as evidenced by lowing H2 oxidation reaction was carried out in the presence of
20 equiv. of FcBArF and 20 equiv. of P(o-tol)3 under 1 atm H2
4
at 25 °C. The supplemental FcBArF and P(o-tol)3 are required
4
to act as an electron acceptor and a proton trapper, respect-
ively, in the activation of H2 by Hox models. After 8 h of reac-
tion, the conversion of P(o-tol)3 to [HP(o-tol)3]+ was detected by
31P{1H} NMR spectroscopy (Fig. S7†), which gave a turnover of
2.3( 0.1) moles of H2 consumed per mol of 1. We found that
the bulky counter ion BArF − can stabilize the FeIIFeI inter-
4
mediate [1]+. If FcBArF4 was used for the oxidation of 1, the fol-
lowing H2 oxidation reaction by FcBArF is apparently slower
4
than using FcPF6 for the oxidation of 1 in the first step. It
−
might be possible that the F atoms of PF6 could help to
shuttle protons and thereby to facilitate the heterolytic clea-
vage of H2.38,39 In a reverse case, if FcPF6 was used for both
the oxidation of 1 and the following H2 oxidation reaction,
more decomposed species of 1 was detected by in situ IR spec-
troscopy during the H2 oxidation process. To the best of our
knowledge, among the [FeFe]-H2ase models so far reported,
Fig. 6 Selected region of the in situ IR spectra (a) for the νCO absorp-
only two diiron complexes have been reported to be catalyti-
tions of [1]+ and the resulting solution from the reaction of [1]+ and
cally active for the H2 oxidation reaction under mild con-
ditions. By comparison, the turnover numbers for H2 acti-
vation are 2 in 5 h for [(μ-adtBn){Fe(CO)2(FcP*)}{Fe(CO)(κ2-
FcBArF in CH2Cl2 under 1 atm H2 at 25 °C for 2 h. (b) The νCO absorp-
4
tions of the cationic products formed by direct protonation of 1 in the
presence of 1.5 equiv. of TfOH.
Scheme 1 Proposed pathway for H2 activation by [1]+ and Fc+ in the presence of P(o-tol)3.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
dppv)}] and 6.2 in 10 h for 4 under similar conditions.26,30 The to the mononuclear complex 2′ to release the ring strain
control experiments showed that no signal was detected at δ caused by the rigid bdtMe bridge in the diiron complex. In
−13.57 for [HP(o-tol)3]+ in the 31P{1H} NMR spectra when the contrast to 1 and 2, complex 3 with an adtBn bridge has a
reaction was carried out in the absence of 1 or Fc+ under other- small ΔEox value; the further oxidation of the in situ generated
wise identical conditions.
[3]+ leads to its rapid decomposition. Introduction of an
To have some insight into the mechanism of the H2 acti- internal amine base into the S-to-S bridge of the Hox model
vation reaction, we carried out the following control reactions. bearing a PNP ligand does not result in a concerted effect of
First, neither [1]+ nor Fc+ reacted with 1 atm H2 under mild these two pendant bases in H2 activation, but considerably
conditions. Second, [1]+ could not be oxidized by FcBArF as speeds up the decomposition of the Hox model. The results of
4
the second oxidation potential of 1 is more positive than the the present work clearly show that changing the S-to-S bridge
reduction potential of Fc+ (Fig. S8†). In view of these obser- can effectively tune the oxidation potentials and the stability of
vations, the heterolytic cleavage of the H–H bond should be a Hox models, and consequently bring about a great effect on
process involving [1]+, Fc+, and H2. The details of this synergis- the performance of the mixed-valence diiron complex for H2
tic process are currently not clear. Just as [4]+ and other activation.
reported [Fe2{(μ-S)2R}(CO)4(P^P)]+ models,20,21,30 [1]+ with an
open apical site at the iron center of the rotated Fe(CO)(PNP)
unit could allow a coordinative interaction of this iron center
Experimental section
with a molecule of H2. The H2 activation is possibly through
Materials and instruments
the similar pathway proposed for other Hox models as shown
in Scheme 1.29,30,40 Coordination of H2 to the open apical site
All reactions and operations related to organometallic com-
plexes were carried out under dry, oxygen-free N2 with stan-
dard Schlenk techniques. The solvents were dried and dis-
tilled prior to use according to the standard methods. The
reagents FcPF6 and P(o-tol)3 were purchased from Aldrich,
and LiB(H)Et3 from Aladdin. The other commercially avail-
able chemicals such as ethane-1,2-dithiol, toluene-3,4-
dithiol, benzylamine, n-propylamine, paraformaldehyde,
Fe(CO)5, and Cp2Fe were purchased from local companies.
All reagents were used as received. Compounds PNP,41
of the rotated Fe unit is accompanied by or immediately fol-
lowed by the oxidation of the intermediate by Fc+; the H–H
bond of the coordinating H2 molecule is then heterolytically
cleaved with the assistance of the internal amine to form an
intermediate [1(HNHμ)]2+, which readily converts to [1(Hμ)]+
and [1(HN)]+ via partial deprotonation by P(o-tol)3 and finally
back to 1 by further deprotonation to fulfill a complete cata-
lytic cycle.
[(μ-edt)Fe2(CO)6],42 [(μ-bdtMe)Fe2(CO)6],43 [(μ-adtBn)Fe2(CO)6],44
and FcBArF
procedures.
were synthesized according to the literature
45,46
4
Conclusions
Infrared spectra were recorded with a JASCO FT/IR 430
spectrophotometer. In situ IR spectra were recorded on a
Mettler-Toledo-IR™ 15 System equipped with an MCT detector
and a Dsub AgX SiComp™ in situ probe. 1H and 31P NMR
spectra were recorded with a Bruker Avance III 500 spectro-
meter. Mass spectra were recorded on a LTQ Orbitrap XL
instrument.
In summary, three new [FeFe]-H2ase models (1–3) with the
same N-containing diphosphine ligand but with different S-to-S
bridges were prepared, and their molecular structures were
determined by single-crystal X-ray diffraction. Electrochemical
studies show that the electronic properties and rigidity of S-to-S
bridges in these complexes have a limited effect on the oxi-
dation potential for the FeIFeI/FeIIFeI couple but have a con-
siderable influence on the potential for the FeIIFeI/FeIIFeII oxi-
dation, while the potential for the FeIIFeI/FeIIFeII oxidation is
Synthesis of 1–3
less susceptible to the length of the S-to-S linker compared to A toluene solution (20 mL) containing [(μ-edt)Fe2(CO)6] (0.35 g,
the first oxidation potential. As a consequence, the S-to-S 1.0 mmol) and ligand PNP (0.46 g, 1.0 mmol) was refluxed for
bridge apparently influences the ΔEox (|Epox2 − Epox1|) value. To 3 h and the color of the solution turned from red to dark
design effective Hox models for H2 activation, this value should green. After the reaction, the solvent was removed under
be large enough to ensure that the second oxidation process reduced pressure. The residue was washed several times with
does not overlap with the first one. When the further oxidation hexane/CH2Cl2 (5 : 1, v/v) and then purified by chromatography
reaction of the mixed-valence complex is suppressed, the on a silica gel column with hexane/CH2Cl2 (3 : 1, v/v) as the
coordination of the H2 molecule to the Fe center and the oxi- eluent. A red powder was obtained from the collected orange
dation of the Hox model could occur in a synergetic manner. band after removal of the solvent. Yield of 1: 0.58 g (75%).
Accordingly, complex 1 with a large ΔEox value can be cleanly Single crystals suitable for X-ray analysis were obtained by
oxidized by Fc+ to give a metastable mixed-valence complex recrystallization of the product from hexane/CH2Cl2 solution
[1]+, which displayed catalytic activity for H2 oxidation in the under N2 at room temperature. 1H NMR (500 MHz, CDCl3):
presence of an excess oxidant and a proton trapper. Although δ 7.63–7.23 (m, 20H, 4Ph), 3.71, 3.17 (2br s, 4H, PCH2NCH2P),
the two single-electron oxidation events of 2 are well separated, 2.44 (br s, 2H, NCH2C), 1.82, 1.67 (2br s, 4H, SCH2CH2S), 1.26
[2]+ cannot activate H2 in the presence of Fc+; it decomposed (m, 2H, NCH2CH2), 0.58 (br s, 3H, CH2CH3). 31P{1H} NMR
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
(204 MHz, CDCl3): δ 53.58 (s). IR (CH2Cl2): νCO 2028, 1956, Catalytic oxidation of H2 by [1]+ in the presence of FcBArF
1900 cm−1. ESI-HRMS: m/z = 772.0260 (Calcd for [M + H]+: and P(o-tol)3
m/z = 772.0260).
4
The oxidant FcPF6 (3.3 mg, 0.01 mmol) was added to a CH2Cl2
solution (10 mL) of 1 (7.8 mg, 0.01 mmol) under N2. The
mixture was stirred for 10 min at 25 °C. Next, FcBArF4 (209 mg,
0.2 mmol) and P(o-tol)3 (61 mg, 0.2 mmol) were added to the
solution and H2 gas was bubbled through the solution for
10 min. The solution was stirred for 8 h at 25 °C under an
atmosphere of H2 (1.0 atm). The resulting solution was
detected by quantitative 31P{1H} NMR spectroscopy and a turn-
over of 2.3( 0.1) moles of H2 per mole of 1 was calculated on
the basis of the conversion of P(o-tol)3 to [HP(o-tol)3]+.
Quantitative 31P{1H} NMR spectra of the resulting solution
containing P(o-tol)3 and [HP(o-tol)3]+ were recorded using the
inverse gate proton decoupling technique for the suppression
of the nuclear Overhauser effect on a Bruker Avance III 500
with the following instrument settings: temperature = 298 K,
spectrometer frequency for 31P = 203.720 MHz, sweep width =
81 521.742 Hz, 32 768 data points, 90° excitation pulse.
The TON of H2 uptake was calculated by the following
equation:
Complexes 2 and 3 were prepared as dark green crystalline
solids with an essentially identical procedure to that used for
the preparation of 1. Yield of 2: 0.51 g (61%). 1H NMR
(500 MHz, CDCl3): δ 7.52–5.97 (m, 23H, 4Ph and C6H3), 3.85,
3.32 (2br s, 4H, PCH2NCH2P), 2.54 (t, 2H, NCH2C, J = 7.3 Hz),
1.81 (s, 3H, C6H3CH3), 1.36 (m, 2H NCH2CH2), 0.63 (t, 3H,
CH2CH3, J = 7.3 Hz). 31P{1H} NMR (204 MHz, CDCl3): δ 53.88.
IR (CH2Cl2): νCO 2033, 1963, 1901 cm−1. ESI-HRMS: m/z =
834.0387 (Calcd for [M + H]+: m/z = 834.0416).
Yield of 3: 0.44 g (51%). 1H NMR (500 MHz, CDCl3):
δ 7.56–6.59 (m, 25H, 5Ph), 3.77, 3.16 (2br s, 4H, PCH2NCH2P),
2.93 (s, 2H, NCH2Ph), 2.74 (d, 2H, SCH2N, J = 10.0 Hz), 2.44
(br s, 2H, NCH2C), 2.33 (d, 2H, NCH2S, J = 10.5 Hz), 1.26
(m, 2H, NCH2CH2), 0.57 (br s, 3H, CH2CH3). 31P{1H} NMR
(204 MHz, CDCl3): δ 53.42. IR (CH2Cl2): νCO 2026, 1955,
1901 cm−1. ESI-HRMS: m/z = 877.0823 (Calcd for [M + H]+:
m/z = 877.0838).
Decomposition of [2]+ to 2′
TON ¼ A1 ꢀ no=½2ðA1 þ A2Þ ꢀ ncatꢁ ðmol H2 per mol catalystÞ
The complex [2]+ was freshly formed from the reaction of 2
(0.50 g, 0.6 mmol) and FcPF6 (0.20 g, 0.6 mmol) in CH2Cl2
under Ar. The solution was stirred for 6 h at 25 °C. After the
reaction, the solvent was removed under reduced pressure
and the residue was purified by chromatography on a silica
gel column with hexane/CH2Cl2 (1 : 3, v/v) as the eluent.
A black powder was obtained from the collected black band
after removal of the solvent. Yield of 2′: 0.24 g (58%). Single
crystals suitable for X-ray analysis were obtained by recrystal-
lization of the product from hexane/CH2Cl2 solution under
N2 at room temperature. 1H NMR (500 MHz, CDCl3):
δ 7.93–6.91 (m, 23H, 4Ph and C6H3), 3.97 (t, 2H, PCH2N, J =
10.1 Hz), 3.27 (t, 2H, PCH2N, J = 10.4 Hz), 2.66 (t, 2H,
NCH2C, J = 10.0 Hz), 2.34 (s, 3H, C6H3CH3), 1.57 (m, 2H
NCH2CH2), 0.88 (m, 3H, CH2CH2CH3). 31P{1H} NMR
A1 = the integrated area for the 31P{1H} NMR signal of [HP(o-
tol)3]+, A2 = the integrated area for the 31P{1H} NMR signal of
P(o-tol)3, no = the total mol of P(o-tol)3 added to the solution,
and ncat = total mol of the catalyst in solution.
Electrochemical studies
Cyclic voltammograms were obtained in a three-electrode cell
under Ar using CHI 630D electrochemical work station. The
working electrode was a glassy carbon disc (diameter 3 mm)
polished with 3 and 1 μm diamond pastes and sonicated in
ion-free water for 15 min prior to use. The reference electrode
was a non-aqueous Ag+/Ag (0.01 M AgNO3 in CH3CN) electrode
and the counter electrode was platinum wire. Methylene chlor-
ide was freshly distilled over P2O5 before being used for
electrochemical studies. A solution of 0.1 M nBu4NPF6 (Fluka,
electrochemical grade) in CH2Cl2 was used as a supporting
electrolyte and was degassed by bubbling with dry argon for
10 min before measurement. The ferricinium/ferrocene (Fc+/
Fc) couple was used as an internal reference and all potentials
given in this work are referred to the Fc+/0 potential.
(204 MHz, CDCl3): δ 50.84 (s). IR (CH2Cl2): νCO 1915 cm−1
.
ESI-HRMS: m/z = 694.1181 (Calcd for [M + H]+: m/z =
694.1219). The same product 2′ was isolated from the result-
ing solution of the reaction of [2]+ and excess FcBArF under
4
1 atm H2 at 25 °C for 8 h.
Activation of H2 by [1]+ in the presence of FcBArF
4
Determination of the structures of 1–3
The oxidant FcPF6 (13.3 mg, 0.04 mmol) was added to a
CH2Cl2 solution (5 mL) of 1 (30.8 mg, 0.04 mmol) under Ar. The single crystal X-ray diffraction data were collected with a
The solution was stirred for 20 min at 25 °C and the reaction Bruker Smart Apex II CCD diffractometer with graphite-mono-
was monitored by in situ IR spectroscopy. After complex 1 was chromated Mo-Kα radiation (λ = 0.071073 Å) at 296 K using the
completely oxidized to [1]+ according to IR spectra, FcBArF
ω–2θ scan mode. Data processing was accomplished with the
4
(83.6 mg, 0.08 mmol) was added to the resulting solution and SAINT processing program.47 Intensity data were corrected for
H2 gas was bubbled through the solution for 10 min. The solu- absorption by using the SADABS program.48 All structures were
tion was stirred under an atmosphere of H2 (1.0 atm) at 25 °C. solved by direct methods and refined on F2 against full-matrix
After the νCO bands of [1]+ disappeared (about 2 h), P(o-tol)3 least-squares methods by using the SHELXTL 97 program
(24.3 mg, 0.08 mmol) was added to the resulting solution to package.49 All non-hydrogen atoms were refined anisotropically.
recover the starting complex 1.
Hydrogen atoms were located by geometrical calculation.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Details of crystallographic data and processing parameters are 19 T. Liu and M. Y. Darensbourg, J. Am. Chem. Soc., 2007, 129,
summarized in Table S2.†
7008–7009.
20 A. K. Justice, T. B. Rauchfuss and S. R. Wilson, Angew.
Chem., Int. Ed., 2007, 46, 6152–6154.
21 A. K. Justice, L. D. Gioia, M. J. Nilges, T. B. Rauchfuss,
S. R. Wilson and G. Zampella, Inorg. Chem., 2008, 47, 7405–
7414.
Acknowledgements
We are grateful to the National Natural Science Foundation of
China (Grant No. 21373040, 21673028, 21120102036 and 22 M. L. Singleton, N. Bhuvanesh, J. H. Reibenspies and
21361130020), the Basic Research Program of China (No.
2014CB239402), the Ph.D. Program Foundation of Ministry of
M. Y. Darensbourg, Angew. Chem., Int. Ed., 2008, 47, 9492–
9495.
Education of China (No. 20130041110024), the Swedish Energy 23 A. K. Justice, M. J. Nilges, T. B. Rauchfuss, S. R. Wilson,
Agency, the Swedish Research Council and the K & A
Wallenberg Foundation for the financial support of
this work.
L. De Gioia and G. Zampella, J. Am. Chem. Soc., 2008, 130,
5293–5301.
24 M. T. Olsen, B. E. Barton and T. B. Rauchfuss, Inorg. Chem.,
2009, 48, 7507–7509.
25 J. M. Camara and T. B. Rauchfuss, J. Am. Chem. Soc., 2011,
133, 8098–8101.
26 J. M. Camara and T. B. Rauchfuss, Nat. Chem., 2012, 4, 26–30.
Notes and references
1 M. W. W. Adams and E. I. Stiefel, Science, 1998, 282, 1842– 27 N. Wang, M. Wang, T. Zhang, P. Li, J. Liu and L. Sun,
1843.
Chem. Commun., 2008, 5800–5802.
2 R. Cammack, Nature, 1999, 397, 214–215.
3 Y. Nicolet, C. Cavazza and J. C. Fontecilla-Camps, J. Inorg.
28 N. Wang, M. Wang, L. Liu, K. Jin, L. Chen and L. Sun,
Inorg. Chem., 2009, 48, 11551–11558.
Biochem., 2002, 91, 1–8.
4 D. J. Evans and C. J. Pickett, Chem. Soc. Rev., 2003, 32, 268–
29 Y. Wang and M. S. G. Ahlquis, Dalton Trans., 2013, 42,
7816–7822.
275.
30 N. Wang, M. Wang, Y. Wang, D. Zheng, H. Han,
M. S. G. Ahlquist and L. Sun, J. Am. Chem. Soc., 2013, 135,
13688–13691.
5 Y. Nicolet, C. Piras and P. Legrand, Structure, 1999, 7, 13–
23.
6 J. W. Peters, N. L. William, J. L. Brian and C. S. Lance, 31 X. Zhao, I. P. Georgakaki, M. L. Miller, R. Mejia-Rodriguez,
Science, 1998, 282, 1853–1858.
7 A. Adamska, A. Silakov, C. Lambertz, O. Ruediger,
C.-Y. Chiang and M. Y. Darensbourg, Inorg. Chem., 2002,
41, 3917–3928.
T. Happe, E. Reijerse and W. Lubitz, Angew. Chem., Int. Ed., 32 F. Gloaguen, J. D. Lawrence, T. B. Rauchfuss, M. Benard
2012, 51, 11458–11462. and M.-M. Rohmer, Inorg. Chem., 2002, 41, 6573–6582.
8 Z. Chen, B. J. Lemon, S. Huang, D. J. Swartz, J. W. Peters 33 D. Morvan, J.-F. Capon, F. Gloaguen, A. L. Goff,
and K. A. Bagley, Biochemistry, 2002, 41, 2036–2043.
9 Z.-P. Liu and P. Hu, J. Am. Chem. Soc., 2002, 124, 5175–
5182.
10 J.-F. Capon, F. Gloaguen, F. Y. Petillon, P. Schollhammer
and J. Talarmin, Coord. Chem. Rev., 2009, 253, 1476–1494.
11 I. P. Georgakaki, L. M. Thomson, E. J. Lyon, M. B. Hall and
M. Marchivie, F. Michaud, P. Schollhammer, J. Talarmin
and J.-J. Yaouanc, Organometallics, 2007, 26, 2042–2052.
34 M. T. Olsen, T. B. Rauchfuss and S. R. Wilson, J. Am. Chem.
Soc., 2010, 132, 17733–17740.
35 L. Schwartz, P. S. Singh, L. Eriksson, R. Lomoth and S. Ott,
C. R. Chim., 2008, 11, 875–889.
M. Y. Darensbourg, Coord. Chem. Rev., 2003, 238–239, 255– 36 M. Beyler, S. Ezzaher, M. Karnahl, M.-P. Santoni,
266.
R. Lomoth and S. Ott, Chem. Commun., 2011, 47, 11662–
12 J.-F. Capon, F. Gloaguen, P. Schollhammer and J. Talarmin,
Coord. Chem. Rev., 2005, 249, 1664–1676.
13 X. Liu, S. K. Ibrahim, C. Tard and C. J. Pickett, Coord.
Chem. Rev., 2005, 249, 1641–1652.
11664.
37 J. M. Gardner, M. Beyler, M. Karnahl, S. Tschierlei, S. Ott
and L. Hammarström, J. Am. Chem. Soc., 2012, 134, 19322–
19325.
14 G. A. N. Felton, C. A. Mebi, B. J. Petro, A. K. Vannucci, 38 Z. M. Heiden and T. B. Rauchfuss, J. Am. Chem. Soc., 2009,
D. H. Evans, R. S. Glass and D. L. Lichtenberger,
131, 3593–3600.
J. Organomet. Chem., 2009, 694, 2681–2699.
39 A. Macchioni, Chem. Rev., 2005, 105, 2039–2074.
15 C. Tard and C. J. Pickett, Chem. Rev., 2009, 109, 2245– 40 H.-J. Fan and M. B. Hall, J. Am. Chem. Soc., 2001, 123,
2274. 3828–3829.
16 S. Tschierlei, S. Ott and R. Lomoth, Energy Environ. Sci., 41 A. L. Balch, M. M. Olmstead and S. P. Rowley, Inorg. Chim.
2011, 4, 2340–2352. Acta, 1990, 168, 255–264.
17 N. Wang, M. Wang, L. Chen and L. Sun, Dalton Trans., 42 F. Gloaguen, J. D. Lawrence, M. Schmidt, S. R. Wilson and
2013, 42, 12059–12071. T. B. Rauchfuss, J. Am. Chem. Soc., 2001, 123, 12518–12527.
18 M. Razavet, S. J. Borg, S. J. George, S. P. Best, S. A. Fairhurst 43 A. Winter, L. Zsolnai and G. Huttner, Z. Naturforsch., B:
and C. J. Pickett, Chem. Commun., 2002, 700–701.
Anorg. Chem. Org. Chem., 1982, 37, 1430–1436.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
44 J. D. Lawrence, H. Li and T. B. Rauchfuss, Chem. Commun., 47 G. M. Sheldrick, SHELXTL97 Program for the Refinement of
2001, 1482–1483. Crystal Structure, University of Göttingen, Germany, 1997.
45 M. Brookhart, B. Grant and A. F. Volpe Jr., Organometallics, 48 Software packages SMART and SAINT, Siemens Energy &
1992, 11, 3920–3922. Automation Inc., Madison, Wisconsin, 1996.
46 J. L. Bras, H. Jiao, W. E. Meyer, F. Hampel and J. A. Gladysz, 49 G. M. Sheldrick, SADABS, Absorption Correction Program,
J. Organomet. Chem., 2000, 616, 54–66.
University of Göttingen, Göttingen, Germany, 1996.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016