Diiron complexes with pendant phenol group(s) as mimics of the diiron subunit of [FeFe]-hydrogenase: Synthesis, characterisation, and electrochemical investigation

Article abstract of DOI:10.1002/ejic.201001092

Four diiron hexacarbonyl complexes, [Fe₂(μ-SCH ₂-o-C₆H₄OMe)₂(CO)₆] (4a), [Fe₂{μ-SCH₂-o,m-C₆H₃(OMe) ₂}₂-(CO)₆] (4b), [Fe₂{μ-SCH ₂-o,o′-C₆H₃(CO₂Me)(OMe)} ₂(CO)₆] (4c) and the demethylated form of complex 4a, [Fe₂(μ-SCH₂-o-C₆H₄OH) ₂(CO)₆] (5a), were synthesised and fully characterised. Complexes 4b and 4c were also structurally analysed. Electrochemical investigations revealed that the integrity of the bridging linkages of the examined diiron complexes significantly affect their reduction reversibility and catalysis through a coupled chemical reaction in a unique ECE mechanism, widely adopted by complexes with the core {Fe₂(CO)_4-6}. Demethylation of complexes 4a and 1Me, [Fe₂(μ-SCH ₂)₂CMe(CH₂-o-C₆H₄OMe)(CO) ₆], by BBr₃ led to complexes (5a and 1H, [Fe ₂(μ-SCH₂)₂CMe(CH₂-o-C ₆H₄OH)-(CO)₆]) with pendant phenol group(s), a weak acid. Deprotonation of the two complexes produced the pendant phenolate, which instantly intramolecularly substitutes the bound CO to yield species of the coordination form Fe^I-OR (R = phenolic moiety). Electrochemical investigation revealed that the pendant phenol groups in complexes 1H and 5a do not seem to improve their catalytic efficiency in proton reduction in the medium acetic acid/dichloromethane.

Full text of DOI:10.1002/ejic.201001092

FULL PAPER
DOI: 10.1002/ejic.201001092
Diiron Complexes with Pendant Phenol Group(s) as Mimics of the Diiron
Subunit of [FeFe]-Hydrogenase: Synthesis, Characterisation, and
Electrochemical Investigation
Ying Tang,[a] Zhenhong Wei,[a] Wei Zhong,[a] and Xiaoming Liu*[a,b]
Keywords: [FeFe]-hydrogenases / Enzyme models / Iron / Bridging ligands / Electrochemistry / Reduction
Four diiron hexacarbonyl complexes, [Fe2(μ-SCH2-o-
C6H4OMe)2(CO)6] (4a), [Fe2{μ-SCH2-o,m-C6H3(OMe)2}2-
(CO)6] (4b), [Fe2{μ-SCH2-o,oЈ-C6H3(CO2Me)(OMe)}2(CO)6]
(4c) and the demethylated form of complex 4a, [Fe2(μ-SCH2-
o-C6H4OH)2(CO)6] (5a), were synthesised and fully charac-
{Fe2(CO)4–6}. Demethylation of complexes 4a and 1Me,
[Fe2(μ-SCH2)2CMe(CH2-o-C6H4OMe)(CO)6], by BBr3 led to
complexes (5a and 1H, [Fe2(μ-SCH2)2CMe(CH2-o-C6H4OH)-
(CO)6]) with pendant phenol group(s), a weak acid. Depro-
tonation of the two complexes produced the pendant phenol-
terised. Complexes 4b and 4c were also structurally ana- ate, which instantly intramolecularly substitutes the bound
lysed. Electrochemical investigations revealed that the integ-
CO to yield species of the coordination form FeI–OR (R =
rity of the bridging linkages of the examined diiron com- phenolic moiety). Electrochemical investigation revealed that
plexes significantly affect their reduction reversibility and ca-
talysis through a coupled chemical reaction in a unique ECE
mechanism, widely adopted by complexes with the core
the pendant phenol groups in complexes 1H and 5a do not
seem to improve their catalytic efficiency in proton reduction
in the medium acetic acid/dichloromethane.
cyanide ligands. On the distal iron atom is located the bind-
ing site for a substrate, occupied by an exogenous ligand,
H2O or OH– in its resting state, Figure 1.
Introduction
Hydrogen is a potential energy vector, which is clean and
carbon-free since its “combustion” produces only energy
without by-products but only water. From water hydrogen
could be regenerated at the expense of solar energy under
the catalysis of hydrogenases as demonstrated by the prin-
ciple possessed by photosystem II (PSII). This would be an
ideal scenario, sustainable, renewable, and environmentally
benign, which is undoubtedly what mankind would like to
ideally achieve for our future. [FeFe]-hydrogenase is one of
the two metalloenzymes existing in nature, capable of re-
versibly and rapidly catalysing hydrogen evolution and oxi-
dation.[1] The crystal structures of the metalloenzymes iso-
lated from Desulfovibrio desulfuricans and Clostridium
pasteurianum, respectively, were revealed a decade ago.[2,3]
The active site, H-cluster, is composed of one {4Fe4S}-cub-
ane and one {2Fe2S}-subunit, Figure 1. The two iron atoms
are bridged by a dithiolate ligand, whose true nature is still
under debate,[2–5] and further coordinated by carbonyl and
Figure 1. The H-cluster of [FeFe]-hydrogenase.
Since the revelation of the detailed structural information
of the enzyme, mimicking of the diiron centre has attracted
tremendous attention because of the relevance of hydrogen
to energy, and a good collection of model complexes con-
taining either a {2Fe2S} or {2Fe3S} core have been re-
ported.[6–15] Some of those reported synthetic analogues du-
plicated delicately key structural features found in the H-
[a] Department of Chemistry / Institute for Advanced Study, cluster.[16–18] But attempts of preparing models to mimic
the bonding feature Fe–OH2(OH–) had not been success-
Nanchang University
Nanchang 330031, China
Fax: +86-791-3969254
ful[19–23] because such a bond is doomed to be labile as it is
E-mail: xiaoming.liu@ncu.edu.cn
[b] College of Biological, Chemical Sciences and Engineering,
Jiaxing University
necessary for this bond to be readily cleaved for substrate
binding in the enzymatic catalysis. Indeed, we recently re-
ported
a model complex, [Fe2(μ-SCH2)2CMe(CH2-o-
Jiaxing 314001, China
Fax: +86-573-83643937
C6H4OH)(CO)6] (1H), which bears a pendant phenol
group, Scheme 1.[24] Deprotonation of the pendant phenol
group led to an unstable diiron pentacarbonyl complex
E-mail: xiaoming.liu@mail.zjxu.edu.cn
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201001092.
1112
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1112–1120

Diiron Complexes with Phenol Groups as Mimics for [FeFe]-Hydrogenase
whose identity was established by FTIR, tandem mass spec- and complexes 4b, 4c and 5a were also crystallographically
trometry and DFT calculations. This was the first example analysed. Deprotonation of complex 5a produced a species
of a model possessing a FeI–OR (R = phenol moiety) bond. that may have the two phenolate groups coordinated to the
two iron atoms, [Fe2(μ-SCH2-o-C6H4O)2(CO)4]2– (6a), as
suggested by its infrared spectrum. These complexes in ad-
dition to two complexes (1Me and 1H) reported pre-
viously[24] were further electrochemically investigated to
probe how the bridging linkage affects the ECE mechanism
adopted widely by complexes with a core ({Fe2(CO)4–6}) in
their electron transfer and whether a pendant phenol, a
weak acid, would exert any influence on the catalytic re-
duction of a proton.
Scheme 1. Deprotonation of complex 1H and the formation of
complex 1H–.[24]
Results and Discussion
Herein, we report four diiron model complexes, [Fe2(μ-
SCH2-o-C6H4OMe)2(CO)6] (4a), [Fe2{μ-SCH2-o,m-C6H3-
Synthesis
(OMe)2}2(CO)6] (4b), [Fe2{μ-SCH2-o,oЈ-C6H3(CO2Me)-
(OMe)}2(CO)6] (4c) and [Fe2(μ-SCH2-o-C6H4OH)2(CO)6]
Complexes 4a–c were prepared from the reaction of
(5a), of which the last complex also possesses two pendant [Fe2(μ-S)2(CO)6], which was prepared following the litera-
phenol groups. All the complexes were fully characterised ture procedure,[25] with organic bromides (Schemes 2 and
Scheme 2. Synthesis of the bromides bearing methoxy groups.
Scheme 3. Synthesis of complexes 4a–c and 5a.
Eur. J. Inorg. Chem. 2011, 1112–1120
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1113

Y. Tang, Z. Wei, W. Zhong, X. Liu
FULL PAPER
3). As show in Scheme 2, the three aromatic bromides were Attempts to prepare complexes 5b and 5c were unsuccessful
prepared by bromination of derivatives of methylanisol. In because of the difficulty in purifying the demethylated
these complexes, the phenyl moiety is linked to the dimet- products. Complexes 4a and 5a show the same spectral en-
allic centre with a methylenethiolate. All the diiron com- velope with a 2 cm–1 “blue” shift after demethylation (Fig-
plexes possess six carbon monoxide ligands and show char- ure 2). This small shift in the infrared absorption bands re-
acteristic spectral absorption bands of diiron hexacarbonyl sults from the replacement of the methoxy group by a hy-
complexes, Table 1. Unlike complex 1H in which the bridg- droxy group in the conversion from complexes 4a to 5a.
ing linkage is a dithiolate,[24] the diiron centre in complexes But the replacement occurs on the phenyl ring and is far
4a–c is bridged by two identical monothiolates. Therefore, away from the diiron centre, and thus the resultant elec-
the organic moieties of the two thiolates can take three pos- tronic effect is rather limited.[24] As in complex 4a, its de-
sible conformations, syn, synЈ, and anti, as observed widely methylated product 5a possesses two isomers at a rather
in their analogues.[26–29] In the first conformation, both similar ratio to that in complex 4a.
phenyl rings point away from the diiron centre. The oppo-
site conformation is the synЈ form, but it is thermodynami-
Crystallographic Analysis
cally unfavoured and hardly observed. The most thermody-
namically stable form is the anti conformation. The distri-
bution of the two conformations depends on the nature of
Diffusion of hexanes into a dichloromethane solutions of
complexes 4b and 4c and storing of the mixtures at –20 °C
for several days afforded dark red crystals of complexes 4b
and 4c suitable for X-ray single crystal diffraction analysis.
Their selected bond lengths and angles are tabulated in
Table 2. Figure 3 shows the structural views of the two com-
plexes. Geometries of these complexes show no drastic dif-
ferences from currently existing diiron hexacarbonyl ana-
logues. Crystals for complex 5a were also crystallised in a
1
the organic moiety. The H NMR signals of the methylene
groups in the spectra of these complexes can be used to
determine the ratio of the two forms (anti/syn). For com-
plexes 4a, 4b and 4c, the ratios are approximately 8:1, 6:1
and 2:1, respectively. Apparently, the anti form in complex
4c is more stable relative to those in the other two com-
plexes. This can probably be attributed to the bulkiness of
the bridging linkage contributed by the acylmethyl ester as
shown in Scheme 2.
Table 2. Selected bond lengths [Å] and angles [°] of complexes 4b
and 4c.
Table 1. Infrared absorption bands [cm–1] of complexes 4a–c, 5a
and 6a.
Complex
4b
4c
Complex
υCO
Fe1–Fe2
Fe1–S1
Fe1–S2
Fe2–S1
Fe2–S2
Fe1–C1
Fe1–C2
Fe1–C3
2.5174(5)
2.2726(5)
2.2570(5)
2.2706(5)
2.2683(5)
1.7990(18)
1.7893(18)
1.7954(17)
1.815(2)
1.7883(18)
1.788(2)
99.22(8)
91.10(8)
99.11(8)
2.5224(14)
2.2624(17)
2.2550(18)
2.262(2)
2.2491(17)
1.815(6)
1.773(5)
1.792(5)
1.781(6)
1.793(5)
1.819(6)
90.8(2)
4a
2069.2, 2033.5, 1991.8
2069.8, 2033.9, 1993.3
2068.6, 2032.9, 1990.8
2070.9, 2035.9, 1994.6
2072.9, 2032.5, 1999.7, 1990.2
2073.4, 2032.9, 2000.2, 1990.7
2020.8, 1993.4, 1944.0, 1909.7
4b
4c
5a
1Me[a]
1H[a]
6a[b]
Fe2–C4
Fe2–C5
Fe2–C6
[a] Adopted from ref.[24] [b] In acetonitrile.
By employing the same strategy as we described pre-
viously, we used BBr3 to demethylate complexes 4a–c, but
only complex 5a was obtained in a moderate yield.
C2–Fe1–C3
C2–Fe1–C1
C3–Fe1–C1
C2–Fe1–S2
C3–Fe1–S2
C1–Fe1–S2
C2–Fe1–S1
C3–Fe1–S1
C1–Fe1–S1
S2–Fe1–S1
S2–Fe1–Fe2
S1–Fe1–Fe2
C5–Fe2–C6
C5–Fe2–C4
C6–Fe2–C4
C5–Fe2–S2
C6–Fe2–S2
C4–Fe2–S2
C5–Fe2–S1
C6–Fe2–S1
C4–Fe2–S1
S2–Fe2–S1
S2–Fe2–Fe1
S1–Fe2–Fe1
97.6(2)
100.1(2)
94.42(16)
156.79(17)
101.57(16)
156.07(18)
85.62(16)
106.29(18)
80.50(6)
55.83(5)
56.12(5)
103.3(2)
92.1(2)
91.25(6)
102.75(6)
157.34(6)
157.97(6)
102.61(6)
88.34(6)
81.167(19)
56.414(16)
56.313(13)
90.68(9)
97.26(9)
99.77(10)
93.80(6)
158.20(7)
100.80(6)
157.81(6)
86.86(7)
104.90(6)
80.97(2)
98.6(3)
156.61(17)
98.63(17)
92.64(16)
87.26(16)
100.77(18)
160.2(2)
80.62(6)
56.05(5)
56.12(5)
55.987(11)
56.388(15)
Figure 2. Infrared spectra of complex 4a (solid line) and its demeth-
ylated product 5a (dashed line).
1114
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1112–1120

Diiron Complexes with Phenol Groups as Mimics for [FeFe]-Hydrogenase
similar procedure to give 5a·0.25CH2Cl2. But its poor qual- from hexacarbonyl to tetracarbonyl. By considering the
ity gave no satisfactory results in the structure refinements partial recovery of the parent complex and the spectral fea-
(Table S1 and Figure S1). Nevertheless, an approximate tures of the product upon deprotonation, it is assumed that
structure was generated, and the structural information re- the two pendant phenolates bind to the diiron centre, prob-
vealed that it is essentially analogous to the other two com- ably in a manner shown in Scheme 4. The structure pro-
plexes.
posed is based on the facts that the anti form is dominant
in complexes of this type and that it is not unprecedented
that one of the phenolate groups coordinates the iron atom
in a basal position rather than in an axial position.[26,33]
Plotting the absorption bands of a known diiron tetracar-
bonyl complex against the bands of the product produces a
linear correlation (Figure S2), which serves as further sup-
port for the proposed tetracarbonyl product. Despite our
attempts at using bulky organic cations, for example,
Ph4PBr or But4NBF4, isolating this product was not suc-
cessful because of its instability as we observed previously
for the monophenolate analogue.[24]
Figure 4. Infrared spectral variations for the interconversion from
complex 5a (solid line) to the product (presumably 6a) (dashed line)
and its partial recovery after acidification (dotted line).
Figure 3. Crystal structures of model complexes 4b (top) and 4c
(bottom) with ellipsoids drawn at a thermal probability of 50%
(hydrogen atoms were omitted for clarity reasons).
Coordination of the Phenol Groups in Complex 5a to the
Diiron Centre Upon Deprotonation
In our previous report, deprotonation of complex 1H led
to the formation of a monoanionic diiron pentacarbonyl
complex with a bond between FeI and the phenolate.[24]
Complex 5a was treated in the same manner in acetonitrile
Scheme 4. Deprotonation of the pendant phenol groups and their
and a group of neat absorption bands were observed, Fig- possible binding modes to the two iron atoms.
ure 4, which accompanied a distinct colour change upon
deprotonation from yellow–brown to dark red. Upon ad-
dition of an acid (HBF4·Et2O) under Ar, about 30% of the
Influence of Substituents of the Phenyl Rings in the
parent complex was recovered as estimated from the inten-
Complexes (4a–c, 5a, 1Me, and 1H) on Their
sities of the infrared absorption bands, Figure 4. Close in-
Electrochemical Behaviour
spection of the spectral profile of the product reveals a simi-
larity to those of diiron tetracarbonyl complexes.[23,30–32]
It is well understood that the reduction of complexes
Relative to the absorption bands of the parent complex 5a, possessing the core {Fe2(CO)n} (n = 4–6) is an ECE process
the absorption bands of the formed product are shifted to but often appears with one reduction peak or with the sec-
lower frequencies by about 85 cm–1 on average. This is com- ond reduction potential slightly more negative than the
parable to those analogous shifts found for the conversion first.[23,28,34–35] Indeed, the reduction of the complexes in-
Eur. J. Inorg. Chem. 2011, 1112–1120
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1115

Y. Tang, Z. Wei, W. Zhong, X. Liu
FULL PAPER
vestigated here followed the proposed ECE mechanism levels of HOMO and LUMO of a molecule, respectively,
when ferrocene was used as an external standard (data not these observations imply that the hydroxy groups decrease
shown). On the other hand, the redox behaviour of the the energy gap between the frontier orbitals more than that
complexes depends very much on the nature of the bridging observed for the non-hydroxy groups. It is certain that the
linkage of the diiron centre. The complexes examined in this hydroxy groups electronically influence the redox poten-
work fall into two categories, that is, complexes 1Me and tials, but exactly how the hydroxy groups exert the influence
1H is in one group, while complexes 4a–c and 5a are in the on the frontier orbitals is hard to explain without further
other. Furthermore, complexes 1H and 5a possess pendant theoretical exploration.
phenol groups, whereas in complexes 1Me and 4a–c, these
phenols are methylated. The variation of these bridging
linkages in both bulkiness and whether the two bridging
thiolates are from an integrated multidentate ligand or two
monothiolates makes these diiron complexes suitable for in-
vestigating the correlation between their electrochemical
properties and the bridging moieties. Particularly, the pen-
dant phenol group shows structural flexibility in its ability
to coordinate to the diiron centre upon deprotonation as
discussed above. Therefore, the pairs of complexes 1Me/1H
and 4a/5a can be appropriate systems to examine how the
pendant phenol group affects their electrochemical re-
duction and hence their catalysing proton reduction, i.e.,
whether this internal weak acid (pendant phenol) plays any
role in the catalysis of proton reduction. Such a correlation
would have particular interest with regard to guiding the
designing and assembling of novel artificial analogues for
the subunit of the [FeFe]-hydrogenase.
Because some of the complexes are not sufficiently solu-
ble in acetonitrile, all the complexes, 4a–c, 5a, 1Me and 1H
were electrochemically examined in dichloromethane, and
their redox potentials are tabulated in Table 3. Their redox
processes are shown in Figures 5 and S3. In general, the
redox potentials are related to the capability of electron do-
nation of the bridging linkages. It is interesting to note the
difference between the reduction peak potential and the
oxidation peak potential of the investigated complexes,
redEc – oxEa, as shown in Table 3. The potential differences
are significantly affected by the replacement of a methoxy
group by a hydroxy group and much less by the variation
in the bridging linkages, although fluctuations in both re-
duction and oxidation potentials with the linkages are obvi-
ous. For complexes 4a–c and 1Me, the potential difference
is about 2.3 V, whereas for complexes 5a and 1H, which
Figure 5. Reduction behaviour of complex 4a–c, 5a (top) and 1Me
and 1H (bottom) in dichloromethane (scan rate = 0.1 Vs–1, 298 K).
were derived from demethylation of their parent complexes,
the potential difference is about 2.1 V. Since reduction and
The bridging linkages of the complexes not only exert an
oxidation potentials are directly correlated to the energy electronic influence on the redox potentials, but also affect
the reversibility of the reduction processes. As shown in
Figure 5, under the same conditions, complexes 1Me and
1H exhibit decent reversibility, whereas for the others, no
reversibility was observed. As mentioned earlier, the re-
Table 3. Redox potentials [V] of complexes 4a–c, 5a, 1Me and 1H
in dichloromethane.
[a]
Complex Reduction, Ec/Ea Oxidation, Ea
redEc – oxEa
duction of the diiron complexes follows a unique ECE
mechanism. The results observed in this work further shed
some light on the nature of the coupled chemical reaction.
It seems that one of the two bridging thiolate groups might
entirely de-coordinate in this chemical reaction. If this is
the case, the fact that the reversibility varies with the bridg-
ing linkage can be explained. In complexes 1Me and 1H,
the dithiolate that bridges the two iron atoms is an inte-
grated bidentate ligand, which allows the de-coordinated
4a
–1.656[b]
0.629
0.731
–2.29
–2.35
–2.34
–2.14
–2.33
–2.15
4b
–1.618[b]
4c
5a
1Me
1H
–1.740[b]
0.596
–1.501[b]
0.635, 0.879[c]
0.857
–1.550/–1.477
–1.508/–1.386
0.767
[a] Irreversible oxidation. [b] Irreversible reduction. [c] From a
daughter product and is not used in the calculation of the potential
difference.
1116
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1112–1120

Diiron Complexes with Phenol Groups as Mimics for [FeFe]-Hydrogenase
thiolate to dangle around and re-coordinate to the metal
centre when necessary. However, in the other complexes,
such a de-coordination would lead to a permanent depar-
ture of the de-coordinated thiolate group.
By considering the different reduction behaviours be-
tween the two categories of diiron complexes (Figure 5), it
is not surprising that they respond rather differently upon
the addition of an acid (acetic acid). Figure 6 shows the
catalytic reduction of a proton by complex 4a. As suggested
by their cyclic voltammograms (Figures S4 and S5), the ca-
talysis of the other two analogues (4b and 4c) is essentially
similar to that of complex 4a. For complex 1Me, catalysis
occurs at a potential more negative by ca. 300 mV relative
to the reduction potential of the first process, which is
hardly affected by the addition of the acid (Figure 7). This
is rather different from the catalysis by complexes 4a–c
(Figures 6, S4 and S5). The experimental observations sup-
port the suggestion that the monoanion generated from the
first one-electron reduction is not responsible for the cataly-
sis.[23] The striking difference in catalytic potentials for the
two categories of complexes is thought to be associated
with the difference in the chemical reaction coupled with
the electron-transfer process as discussed above.
Figure 7. Cyclic voltammograms of complex 1Me (C
=
3.24 mmolL–1) without and with the addition of the acid in dichlo-
romethane (scan rate = 0.1 Vs–1, 298 K).
Conclusions
In this work, four diiron hexacarbonyl complexes, 4a–c
and 5a, in addition to two complexes 1Me and 1H reported
previously,[24] were investigated electrochemically without
and with the presence of acetic acid in dichloromethane.
These complexes are believed to adopt the unique ECE
mechanism for electron transfer, in which the potential for
the second reduction could be either more positive or mar-
ginally more negative than the first one. Our results also
show that the bridging linkages of the diiron complexes sig-
nificantly affect the chemical reaction coupled to the first
electron-transfer process. When the two thiolate groups are
from two monothiolates, the coupled chemical reaction may
be associated with the partial deconstruction of the diiron
moiety, for example, the permanent cleavage of one of the
two thiolates. Such a destruction is much less significant for
complexes 1Me and 1H, as suggested by their quasirevers-
ible reductions; the bidentate ligand certainly made the dif-
ference. Demethylation of complexes 4a and 1Me led to
complexes (5a and 1H) with pendant phenol group(s).
However, this pendant weak acid does not seem to exert
much influence on the catalysis of proton reduction.
Figure 6. Cyclic voltammograms of complex 4a (C
=
3.30 mmolL–1) without and with the addition of acetic acid in di-
cholormethane (scan rate = 0.1 Vs–1, 298 K).
As indicated by the redox potentials in Table 3, demeth-
ylation causes noticeable variation in the energies of the
frontier orbitals for the pairs of complexes 4a and 5a, 1Me
and 1H. However, with regard to their catalysis, no essential
Experimental Section
difference is observed for these complexes (Figures 6, 7, S6 General Procedures: All reactions were carried out under an argon
atmosphere by using standard Schlenk technique. Solvents for reac-
tions and electrochemistry were freshly distilled by using appropri-
ate drying agents prior to use. Infrared spectroscopic data were
collected on a Scimitar 2000 (Varian) instrument in an appropriate
solvent by using a solution cell (spacer = 1 mm). NMR spectra
were recorded on an Avance DRX-400 or–600 (Bruker) instrument.
Electrochemistry was performed as described elsewhere.[23,28,33,36]
All potentials were quoted against the ferrocenium/ferrocene cou-
ple, FC+/FC. The microanalysis service was provided by the Centre
of Analysis and Testing at Nanchang University and Nanjing Uni-
versity.
and S7). The only noticeable difference observed is the neat
cyclic voltammograms observed for the complexes without
pendant phenol groups (4a–c and 1Me). This can probably
be attributed to the adsorbing preference on the surface of a
vitreous carbon electrode possessed by species with phenol
groups as the surface of a vitreous carbon electrode is rich
in functional groups, for instance, hydroxy and carboxylic
groups. However, the presence of a pendant phenol group
does not lead to a significant change in the catalytic effi-
ciency of proton reduction (Figures S8 and S9).
Eur. J. Inorg. Chem. 2011, 1112–1120
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1117

Y. Tang, Z. Wei, W. Zhong, X. Liu
FULL PAPER
Synthesis
Synthesis of 1,4-Dimethoxy-2-bromomethylbenzene (3b): Com-
pound 3b was prepared by following the same experimental pro-
cedure described for compound 3a by using compound 2b (6.1 g,
40 mmol), NBS (7 g, 40 mmol) and AIBN (160 mg, 1 mmol). The
compound was obtained as a needle-shaped crystalline solid (8.5 g,
92%). Mp: 71–73 °C. H NMR (CDCl3, 600 MHz): δ = 6.89 (d, J
= 2.4 Hz, 1 H, ArH), 6.79–6.83 (m, 2 H, ArH), 4.53 (s, 2 H, CH2),
3.84 (s, 3 H, OCH3), 3.76 (s, 3 H, OCH3) ppm.
Preparation of [Fe2(μ-S)2(CO)6]: After being flushed with Ar, a
three-necked round-bottomed flask (1 L) was charged with Fe-
(CO)5 (25 mL, 186 mmol), followed by dry MeOH (125 mL) and
finally freshly prepared aqueous KOH (75 mL, 75%). After stirring
for 30 min, the homogeneous solution was cooled to ice tempera-
ture in an ice bath, and then S8 (33 g, 1.03 mol) was slowly added.
After the addition and further stirring for 1 h, the reaction mixture
was treated with H2O (250 mL), hexanes (150 mL), and finally
NH4Cl (85 g, 1.6 mol). The ice bath was then removed, and the
reaction solution was stirred at room temperature for an additional
16 h. The reaction mixture was then extracted with petroleum ether
(4 ϫ 200 mL). The extracts were combined, and the solvents were
removed to give a crude product, which was purified by using flash
chromatography (eluent: petroleum ether) to produce a red solid
(9.7 g, 30%). Mp: 46–47 °C. IR [acetonitrile]: νCO = 2084.0, 2041.8,
2002.7 cm–1.
1
Synthesis of Methyl(3-methoxy-2-bromomethyl)benzoate (3c): The
same procedure was adopted to produce compound 3c as a white
solid (4.5 g, 96%) by using compound 2c (3.3 g, 18 mmol), NBS
(3.3 g, 18 mmol) and AIBN (60 mg, 0.4 mmol) in CCl4 (160 mL).
1
Mp: 112–114 °C. H NMR (CDCl3, 600 MHz): δ = 7.05 (m, 1 H,
ArH), 7.32 (t, J = 8.1 Hz, 1 H, ArH), 7.06 (d, J = 8.4 Hz, 1 H,
ArH), 5.0 5 (s, 2 H, CH2), 3.92 (s, 3 H, COOCH3), 3.91 (s, 3 H,
OCH3) ppm.
Synthesis of [Fe2(μ-SCH2-o-C6H4OMe)2(CO)6] (4a): A solution of
[Fe2(μ-S)2(CO)6] (1.2 g, 3.5 mmol) in dry THF (60 mL), precooled
to –78 °C with a dry-ice/acetone bath, was treated with LiBHEt3
(7 mL, 1 molL–1 in dry THF). After stirring the mixture for
30 min, compound 3a (1.4 g, 7.0 mmol) was added. The mixture
was stirred at –78 °C for 30 min and was then warmed to room
temperature for an additional 2 h. Removal of the solvent produced
a crude product, which was purified by using flash chromatography
(petroleum ether/ethyl acetate, 20:1) to give a red solid (1.6 g, 78%).
Mp: 141–142 °C. IR [dichloromethane]: νCO = 2069.2, 2033.5,
Synthesis of 2-Methylanisol, 2a: A stirred suspension of NaH (6.0 g,
150 mmol, 60% dispersion in mineral oil) in dry THF (120 mL)
was cooled to ice temperature in an ice bath. To the suspension
was slowly added a solution of o-cresol (1a, 10.8 g, 100 mmol) in
dry THF (20 mL). When there was no more hydrogen gas evolving,
CH3I (7.2 mL, 115 mmol) was added dropwise. The reaction mix-
ture was warmed to 85 °C and stirred overnight. Completion of the
reaction was confirmed by using TLC (petroleum ether/ethyl acet-
ate = 10:1). After removal of the solvent, water (50 mL) was added.
After extraction with ethylacetate (3ϫ 50 mL) and purification by
column chromatography (petroleum ether/ethyl acetate, 10:1), the
desired compound was obtained as a colourless liquid (11.8 g,
1
1991.8 cm–1. H NMR (CDCl3, 400 MHz): δ = 6.77–7.33 (m, 9 H,
ArH, anti/syn ratio 8:1), 3.94 (s, 3 H, anti-OCH3), 3.84 (s, 0.75 H,
syn-OCH3), 3.71 (s, 3 H, anti-OCH3), 3.67 (s, 2 H, anti-CH2), 3.61
(s, 0.5 H, syn-CH2), 3.25 (s, 2 H, anti-CH2) ppm. C22H18Fe2O8S2
(586.19): calcd. C 45.08, H 3.10; found C 44.94, H 3.29.
1
97%). H NMR (CDCl3, 400 MHz): δ = 7.16 (m, 2 H, ArH), 6.84
(m, 2 H, ArH), 3.82 (s, 3 H, OCH3), 2.22 (s, 3 H, CH3) ppm.
Synthesis of [Fe2{μ-SCH2C6H3(OMe)2}2(CO)6] (4b): Complex 4b
was prepared from [Fe2(μ-S)2(CO)6] (1.05 g, 3 mmol), LiBHEt3
(6 mL, 1 m in dry THF) and compound 2b (1.5 g, 6 mmol) follow-
ing the same procedure used to prepare complex 4a. Complex 4b
was obtained as a red solid (1.2 g, 63%). Mp: 136–137 °C. IR
[dichloromethane]: νCO = 2069.8, 2033.9, 1993.3 cm–1. 1H NMR
(CDCl3, 600 MHz): δ = 6.56–6.92 (m, 6 H, ArH, anti/syn 6:1), 3.89
(s, 3 H, anti-OCH3), 3.81 (s, 3 H, anti-OCH3), 3.81 (s, 1 H, syn-
OCH3), 3.75 (s, 1 H, syn-OCH3), 3.70 (s, 3 H, anti-OCH3), 3.65 (s,
3 H, anti-OCH3), 3.63 (s, 2 H, anti-CH2), 3.58 (s, 0.63 H, syn-CH2),
3.23 (s, 2 H, anti-CH2) ppm. 13C NMR (CDCl3, 600 MHz): δ =
209.98, 208.86, 153.56, 153.46, 153.28, 151.61, 151.45, 151.28,
128.61, 127.98, 127.68, 116.45, 116.10, 115.77, 114.17, 113.96,
113.75, 111.78, 111.60, 111.54, 55.80.55.78, 55.70, 55.63, 55.56,
55.43, 38.28, 36.83, 31.60, 23.85 ppm. C24H22Fe2O10S2 (646.26):
calcd. C 44.60, H 3.43; found C 44.46, H 3.30.
Synthesis of 1,4-Dimethoxy-2-methylbenzene, 2b: Compound 2b
was synthesised in the same procedure as described above for com-
pound 2a by using 2-methylhydroquinone (6.3 g, 50 mmol), NaH
(6 g, 150 mmol, 60% dispersion in mineral oil) and CH3I (10 mL,
150 mmol). It was isolated as a colourless liquid (6.8 g, 88%). H
NMR (CDCl3, 400 MHz): δ = 6.59–6.69 (m, 3 H, ArH), 3.69 (s, 3
H, OCH3), 3.66 (s, 3 H, OCH3), 2.18 (s, 3 H, CH3) ppm.
1
Synthesis of Methyl(3-methoxy-2-methyl)benzoate (2c): 3-Hydroxy-
2-methylbenzoic acid (1c, 3.2 g, 20 mmol) was dissolved in dry
MeOH (20 mL, 500 mmol) and mixed with concentrated H2SO4
(2 mL, 35.5 mmol, 95%). The solution was heated at reflux over-
night. Removal of the solvents gave a residue to which water
(30 mL) was then added. Extraction with ethyl acetate (3 ϫ 30 mL)
and routine work-up gave compound 2c as a colourless liquid
1
(3.3 g, 95%) without further purification for the next reaction. H
NMR (CDCl3, 400 MHz): δ = 7.30 (d, J = 7.76 Hz, 1 H, ArH),
7.07 (t, J = 7.98 Hz, 1 H, ArH), 6.86 (d, J = 8.12 Hz, 1 H, ArH),
4.53 (s, 2 H, CH2), 3.77 (s, 3 H, COOCH3), 3.71 (s, 3 H, OCH3),
2.33 (s, 3 H, CH3) ppm.
Synthesis of [Fe2{μ-SCH2C6H3(COOMe)OMe}2(CO)6] (4c): Com-
plex 4c was synthesised as described for 4a from [Fe2(μ-S)2(CO)6]
(1.2 g, 3.6 mmol), LiBHEt3 (7 mL, 1 m in dry THF) and compound
3c (1.8 g, 7 mmol) as a red solid (1.6 g, 65%). Mp: 174–176 °C. IR
[dichloromethane]: νCO = 2068.6, 2032.9, 1990.8 cm–1. 1H NMR
Synthesis of 2-Bromomethylanisol (3a): A solution of compound
2a (7.3 g, 60 mmol), NBS (10.7 g, 60 mmol) and AIBN (160 mg, (CDCl3, 600 MHz): δ = 7.56 (d, J = 7.2 Hz, 1 H, anti-ArH), 7.46
1 mmol) in CCl4 (250 mL) was heated at reflux for about 1 h until (d, J = 7.2 Hz, 1 H, syn-ArH), 7.37 (d, J = 7.2 Hz, 1 H, anti-ArH),
a white solid was floating on the surface. The mixture was filtered,
and the filtrate was then concentrated under reduced pressure to
produce a residue. Recrystallisation of the residue in hexanes af-
7.33 (t, J = 7.8 Hz, 1 H, anti-ArH), 7.25 (t, J = 9.6 Hz, 1 H, syn-
ArH), 7.21 (t, J = 7.8 Hz, 1 H, anti-ArH), 7.10 (d, J = 8.4 Hz, 1
H, anti-ArH), 7.00 (d, J = 7.8 Hz, 1 H, syn-ArH), 7.90 (d, J =
forded a white solid (11.5 g, 93%). Mp: 47–50 °C. 1H NMR 8.4 Hz, 1 H, anti-ArH), 4.07–4.64 (m, 24 H, PhCH2, PhOCH3,
(CDCl3, 600 MHz): δ = 7.30 (m, 2 H, ArH), 6.93 (m, 1 H, ArH), PhCOOCH3 anti/syn 2:1) ppm. 13C NMR (CDCl3, 600 MHz): δ =
6.88 (m, 1 H, ArH), 4.57 (s, 2 H, 2 ϫ CHH), 3.88 (s, 3 H, OCH3)
ppm. 13C NMR (CDCl3, 600 MHz): δ = 157.45, 130.89, 130.21, 131.33, 130.57, 130.24, 129.74, 129.17, 128.52, 128.42, 128.35,
126.07, 120.67, 110.95, 55.57, 29.06 ppm. 128.11, 122.89, 122.85, 122.59, 114.01, 113.83, 55.60, 55.51, 55.16,
210.30, 209.11, 167.52, 167.43, 167.21, 157.79, 157.55, 157.45,
1118
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1112–1120

Diiron Complexes with Phenol Groups as Mimics for [FeFe]-Hydrogenase
52.29, 52.18, 52.12, 34.47, 32.35, 21.42 ppm. C26H22Fe2O12S2
(702.28): calcd. C 44.47, H 3.16; found C 44.62, H 3.53.
graphic details of these complexes are summarised in Table 4.
CCDC-795447 (4b), -795448 (4c) and -795449 (5a) contain the sup-
plementary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Synthesis of [Fe2(μ-SCH2-o-C6H4OH)2(CO)6] (5a): To a precooled
solution of [Fe2(μ-SCH2C6H4OCH3)2(CO)6] (680 mg, 1.16 mmol)
in dry dichloromethane (50 mL) in a dry-ice/acetone bath was
added a solution of BBr3 (1 molL–1, 5 mL) in dry dichloromethane
under argon. After the addition, the reaction mixture was stirred
at –78 °C for 2 h and then warmed to room temperature for an
additional 6 h. The reaction mixture was cooled in an ice bath, and
methanol (2 mL) was slowly added to quench the reaction. Re-
moval of the solvents left a residue, which was purified by using
flash chromatography in a gradient eluting manner (petroleum
ether/ethyl acetate, 10:1, 8:1 and 5:1) to give a dark red solid
(377 mg, 58.2%). Mp: 138 °C. IR [dichloromethane]: νCO = 2070.9,
2035.9, 1994.6 cm–1. 1H NMR (CDCl3, 600 MHz): δ = 6.75–7.27
(m, 9 H, ArH, anti/syn 8:1), 5.34 (s, 1 H, anti-PhOH), 5.25 (s, 1 H,
anti-PhOH), 5.08 (s, 0.25 H, syn-PhOH), 3.70 (s, 2 H, anti-CH2),
Supporting Information (see footnote on the first page of this arti-
cle): Crystallographic data, the structural view of complex 5a, fur-
ther spectroscopic data and cyclic voltammograms are included.
Acknowledgments
We thank the National Natural Science Foundation of China
(Grant No. 20871064), the Ministry of Science and Technology
(China) (973 program, Grant No: 2009CB220009), and the State
Key Laboratory of Coordination Chemistry (Nanjing University,
China) for supporting this work.
3.63 (s, 0.5 H, syn-CH2), 3.26 (s,
2 H, anti-CH2) ppm.
C20H14Fe2O8S2 (558.14): calcd. C 43.04, H 2.53; found C 42.65, H
2.71.
[1] D. J. Evans, C. J. Pickett, Chem. Soc. Rev. 2003, 32, 268–275.
[2] J. W. Peters, W. N. Lanzilotta, B. J. Lemon, L. C. Seefeldt, Sci-
ence 1998, 282, 1853–1858.
Deprotonation of Complex 5a: To NaH (60% in oil, 13.2 mg,
0.33 mmol) in a Schlenk tube precooled with an ice bath was added
a solution of complex 5a (40 mg, 0.079 mmol) in dry acetonitrile
(5 mL) under rapid stirring. The colour of the solution changed
quickly from yellow–brown to dark red. The reaction was complete
within 2 h, as confirmed by infrared spectroscopic monitoring. IR
[acetonitrile]: νCO = 2020.8, 1993.4, 1944.0, 1909.7 cm–1.
[3] Y. Nicolet, C. Piras, P. Legrand, C. E. Hatchikian, J. C. Fontec-
illa-Camps, Structure 1999, 7, 13–23.
[4] U. Ryde, C. Greco, L. De Gioia, J. Am. Chem. Soc. 2010, 132,
4512–4513.
[5] B. E. Barton, M. T. Olsen, T. B. Rauchfuss, J. Am. Chem. Soc.
2008, 130, 16834–16835.
X-ray Single-Crystal Diffraction: Standard procedures were used
for mounting the crystals on a Bruker 98 APEX2 area-detector
diffractometer at 296(2) K. The crystals were routinely wrapped
with paraffin oil before being mounted. Intensity data were col-
lected by using Mo-Ka radiation (λ = 0.71073 Å) under 296(2) K
with a φ- and ω-scan mode. The SAINT and SADABS programs
in the APPEX(II) software package were used for integration and
absorption correction. The structures were solved by using direct
methods with the SHELXS-97 program and refined on F2 with
XSHELL6.3.1; all non-hydrogen atoms were anisotropically mod-
elled. All hydrogen atoms were geometrically positioned. Crystallo-
[6] X. M. Liu, S. K. Ibrahim, C. Tard, C. J. Pickett, Coord. Chem.
Rev. 2005, 249, 1641–1652.
[7] F. F. Xu, C. Tard, X. F. Wang, S. K. Ibrahim, D. L. Hughes, W.
Zhong, X. R. Zeng, Q. Y. Luo, X. M. Liu, C. J. Pickett, Chem.
Commun. 2008, 606–608.
[8] L. C. Song, G. H. Zeng, S. X. Lou, H. N. Zan, J. B. Ming,
Q. M. Hu, Organometallics 2008, 27, 3714–3721.
[9] P. Li, M. Wang, L. Chen, J. H. Liu, Z. B. Zhaoa, L. C. Sun,
Dalton Trans. 2009, 1919–1926.
[10] B. E. Barton, M. T. Olsen, T. B. Rauchfuss, Curr. Opin. Bi-
otechnol. 2010, 21, 292–297.
[11] P. Volkers, T. B. Rauchfuss, J. Inorg. Biochem. 2007, 101, 1748–
1751.
[12] C. Tard, C. J. Pickett, Chem. Rev. 2009, 109, 2245–2274.
[13] M. L. Singleton, J. H. Reibenspies, M. Y. Darensbourg, J. Am.
Chem. Soc. 2010, 132, 8870–8871.
[14] W. M. Gao, J. L. Sun, T. Akermark, M. R. Li, L. Eriksson,
L. C. Sun, B. Akermark, Chem. Eur. J. 2010, 16, 2537–2546.
[15] S. Ezzaher, A. Gogoll, C. Bruhn, S. Ott, Chem. Commun. 2010,
46, 5775–5777.
Table 4. Crystallographic data and processing parameters for com-
plexes 4b,c.
Complex
4b
4c
Empirical formula
Fw
T / K
Crystal system
Space group
a [Å]
b [Å]
c [Å]
C24H22Fe2O10S2 C26H22Fe2O12S2
646.26
296(2)
triclinic
702.28
296(2)
triclinic
[16] C. Tard, X. M. Liu, S. K. Ibrahim, M. Bruschi, L. De Gioia,
S. C. Davies, X. Yang, L. S. Wang, G. Sawers, C. J. Pickett, Na-
ture 2005, 433, 610–613.
¯
¯
P1
P1
8.9004(17)
9.5959(18)
17.285(4)
101.806(3)
95.464(3)
107.528(2)
1358.3(5)
2
11.018(8)
11.061(8)
13.150(9)
88.948(8)
79.530(9)
70.571(8)
1484.6(18)
2
[17] T. Liu, M. Y. Darensbourg, J. Am. Chem. Soc. 2007, 129, 7008–
7009.
[18] C. M. Thomas, T. Liu, M. B. Hall, M. Y. Darensbourg, Inorg.
Chem. 2008, 47, 7009–7024.
[19] P. I. Volkers, T. B. Rauchfuss, S. R. Wilson, Eur. J. Inorg. Chem.
2006, 4793–4799.
[20] J. F. Capon, S. Ezzaher, F. Gloaguen, F. Y. Petillon, P.
Schollhammer, J. Talarmin, T. J. Davin, J. E. McGrady, K. W.
Muir, New J. Chem. 2007, 31, 2052–2064.
[21] C. M. Thomas, O. Rudiger, T. Liu, C. E. Carson, M. B. Hall,
M. Y. Darensbourg, Organometallics 2007, 26, 3976–3984.
[22] A. K. Justice, T. B. Rauchfuss, S. R. Wilson, Angew. Chem. Int.
Ed. 2007, 46, 6152–6154.
a [°]
b [°]
c [°]
V / Å3
Z
qcalcd. / gcm–3
F(000)
1.580
1.571
660
716
Reflections collected (unique) 12866(6842)
12625(7290)
7290(0.0282)
1.073
0.0583
0.1350
Reflections [Rint
]
6842(0.0170)
1.017
0.0264
Goodness-of-fit on F2
[a]
Final R1 [I Ͼ2σ(I)]
[23] Y. W. Wang, Z. M. Li, X. H. Zeng, X. F. Wang, C. X. Zhan,
Y. Q. Liu, X. R. Zeng, Q. Y. Luo, X. M. Liu, New J. Chem.
2009, 33, 1780–1789.
[a]
Final wR2
0.0684
[a] R1 = ∑||Fo| – |Fc||/∑|Fo| and wR2 = [∑(|Fo – Fc |)2/∑(wFo2)2]1/2
.
2
2
Eur. J. Inorg. Chem. 2011, 1112–1120
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1119

Y. Tang, Z. Wei, W. Zhong, X. Liu
[24] W. Zhong, Y. Tang, G. Zampella, X. Wang, X. Yang, B. Hu, [31] X. Zhao, I. P. Georgakaki, M. L. Miller, R. Mejia-Rodriguez,
FULL PAPER
J. Wang, Z. Xiao, Z. Wei, H. Chen, L. De Gioia, X. Liu, Inorg.
Chem. Commun. 2010, 13, 1089–1092.
[25] J. L. Stanley, T. B. Rauchfuss, S. R. Wilson, Organometallics
2007, 26, 1907–1911.
[26] X. Zhao, Y. M. Hsiao, C. H. Lai, J. H. Reibenspies, M. Y. Dar-
ensbourg, Inorg. Chem. 2002, 41, 699–708.
[27] W. Zhong, G. Zampella, Z. M. Li, L. De Gioia, Y. Q. Liu,
X. R. Zeng, Q. Y. Luo, X. M. Liu, J. Organomet. Chem. 2008,
693, 3751–3759.
[28] X. H. Zeng, Z. M. Li, Z. Y. Xiao, Y. W. Wang, X. M. Liu, Elec-
trochem. Commun. 2010, 12, 342–345.
[29] X. Ru, X. H. Zeng, Z. M. Li, D. J. Evans, C. X. Zhan, Y. Tang,
L. J. Wang, X. M. Liu, J. Polym. Sci., Part A: Polym. Chem.
2010, 48, 2410–2417.
[30] A. Le Cloirec, S. P. Best, S. Borg, S. C. Davies, D. J. Evans,
D. L. Hughes, C. J. Pickett, Chem. Commun. 1999, 2285–2286.
C. Y. Chiang, M. Y. Darensbourg, Inorg. Chem. 2002, 41,
3917–3928.
[32] R. Mejia-Rodriguez, D. S. Chong, J. H. Reibenspies, M. P. Sor-
iaga, M. Y. Darensbourg, J. Am. Chem. Soc. 2004, 126, 12004–
12014.
[33] C. Zhan, X. Wang, Z. Wei, D. J. Evans, X. Ru, X. Zeng and
X. Liu, Dalton Trans. 2010, DOI: 10.1039/C1030DT00687D.
[34] G. A. N. Felton, C. A. Mebi, B. J. Petro, A. K. Vannucci, D. H.
Evans, R. S. Glass, D. L. Lichtenberger, J. Organomet. Chem.
2009, 694, 2681–2699.
[35] J. F. Capon, F. Gloaguen, F. Y. Petillon, P. Schollhammer, J.
Talarmin, Coord. Chem. Rev. 2009, 253, 1476–1494.
[36] Z. M. Li, X. H. Zeng, Z. G. Niu, X. M. Liu, Electrochim. Acta
2009, 54, 3638–3644.
Received: October 13, 2010
Published Online: January 11, 2011
1120
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1112–1120