(N-CnH2n-1)-1,3-Azapropanedithiolate (n = 5, 6, 7)-bridged diiron complexes as mimics for the active site of [FeFe]-hydrogenases: The influence of the bridge on the diiron complex

Article abstract of DOI:10.1039/b616236c

To further understand the effects of 1,3-azadithiolate (ADT) bridges on the structures and properties of diiron complexes, a novel family of (N-C _nH_2n-1)-1,3-azapropanedithiolate-bridged diiron complexes [Fe₂(CO)₆

Full text of DOI:10.1039/b616236c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
(N-CnH2nÀ1)-1,3-Azapropanedithiolate (n = 5, 6, 7)-bridged diiron
complexes as mimics for the active site of [FeFe]-hydrogenases: the
influence of the bridge on the diiron complexw
Youtao Si,ab Chengbing Ma,a Mingqiang Hu,ab Hui Chen,a Changneng Chen*a and
Qiutian Liua
Received (in Durham, UK) 7th November 2006, Accepted 28th March 2007
First published as an Advance Article on the web 24th April 2007
DOI: 10.1039/b616236c
To further understand the effects of 1,3-azadithiolate (ADT) bridges on the structures and
properties of diiron complexes, a novel family of (N-CnH2nÀ1)-1,3-azapropanedithiolate-bridged
diiron complexes [Fe2(CO)6(CH2S)2N–CnH2nÀ1] (n = 5, 6, 7) and their PMe3-disubstituted
complexes [Fe2(CO)4(PMe3)2(CH2S)2N–CnH2nÀ1] (n = 5, 6, 7) were synthesized as mimics of
[FeFe]-hydrogenases. Our studies suggest that the coordination configurations, CO stretching
1
frequencies, H and 31P NMR signals, redox potentials and electrocatalytic H2-production
processes of diiron complexes are closely pertinent to the electron-donating abilities and steric
chemistry of the ADT bridges.
hydrogenases, we have synthesized six (N-CnH2nÀ1)-1,3-aza-
propanedithiolate-bridged diiron compounds (1–6) that bear
typical structural similarities with the active site of [FeFe]-
hydrogenases. In complexes 4–6, PMe3 was selected to sub-
stitute for the CO on the iron atoms to mimic the CNÀ-
functionalized diiron complexes in nature. In this article, we
will describe the coordination configurations, spectroscopic
characterization, electrochemical properties and catalyzing
mechanism of H2 production processes promoted by com-
plexes 1–6.
Introduction
Hydrogenases are enzymes adept at catalyzing the oxidation
of molecular hydrogen or its production from protons and
electrons.1 According to the metals included in the enzymes,
organometallic hydrogenases are generally divided into three
major classes: [FeFe]-hydrogenases, [NiFe]-hydrogenases and
the newly known [Fe]-hydrogenases.2,3 [FeFe]-hydrogenases
are more O2 sensitive and committed to more efficient H2
production than [NiFe]-hydrogenases.4 It is the significant
ability of [FeFe]-hydrogenases that inspires many chemists
seeking active hydrogen production catalysts to synthesize
mimics of its active site.5–11 X-Ray crystallographic studies
on [FeFe]-hydrogenases isolated from Clostridium pasteuria-
num and Desulfovibrio desulfuricans have suggested that the
active site is composed of a ‘‘butterfly’’ diiron unit that is
linked to a typical Fe4S4 cuboidal subcluster by a cysteine-S
residue.12–14 The dithiolate bridge between the two iron atoms
of the 2Fe2S cluster has recently been suggested to be an
azadithiolate, SCH2NHCH2S (ADT),12,14 with the central N
atom playing an important role in the process of heterolytic
cleavage or formation of H2. Because of the key function of
the central N atom of ADT, some aromatic alkyl-functiona-
lized ADT complexes have been synthesized as models of the
active site.9,11,15 However, as far as we know, systematic
studies on the influence of different ADT bridges on diiron
complexes are absent, and little attention has been paid to
cycloalkyl-substituted ADT model compounds.
Results and discussion
Synthesis of complexes 1–6
Complexes 1–3 were prepared in several steps according to
literature procedures. Starting from Fe2S2(CO)6,16–18 we ob-
tained Fe2(SH)2(CO)6, which condensed with formaldehyde in
the presence of primary amines (namely cyclopentylamine,
cyclohexylamine and cycloheptylamine) to give the corre-
sponding diiron complexes in 70–80% yield.19,20 Disubstituted
complexes (4–6) were synthesized in moderate yields by treat-
ing complexes 1–3 with PMe3 in hexane, respectively.5,21 The
mixture was stirred for 24 h at room temperature under a
nitrogen atmosphere until the color of the solution turned
from red to purple. Single crystals suitable for X-ray analysis
were obtained by extracting and subsequently recrystallizing
from different freshly distilled solvents, namely pentane,
hexane or heptane, in light of the solubilities of the products.
To further investigate the role of the ADT bridges in diiron
complexes, and to develop the biomimetic chemistry of [FeFe]-
Molecular structures of complexes 1–6
a Fujian Institute of Research on the Structure of Matter, Chinese
Academy of Sciences, Fuzhou 350002, China. E-mail:
ccn@fjirsm.ac.cn; Fax: +86 591/83792395; Tel: +86 591/83792395
b Department of Chemistry, Graduate University of Chinese Academy
of Sciences, Beijing 100049, China
The structures of the six compounds were unambiguously
determined by X-ray diffraction, and their molecular diagrams
are displayed in Fig. 1. Complexes 1–3 are (N-CnH2nÀ1)-1,
3-azapropanedithiolate (n = 5, 6, 7)-bridged all-carbonyl
diiron complexes, while 4–6 are their PMe3-disubstituted
derivatives. There are two molecules per asymmetric unit cell,
w Electronic supplementary information (ESI) available: Selected
bond lengths and angles. See DOI: 10.1039/b616236c
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1448 | New J. Chem., 2007, 31, 1448–1454

View Article Online
Table 1 Fe–Fe bond lengths of complexes 1–6, 8, 9, 10 and 11
Complex
Fe–Fe bond/A
1
2
3
4
5
6
8
2.5025(7)
2.5280(10)
2.5103(7)
2.5279(13)
2.6374(8)
2.6118(5)
2.5076
9
10
11
2.4924
2.554
2.5094
of the substitution process, namely, sterically less crowded.
This is the same reason for axial/basal phosphine orientation
in the least sterically encumbered diphosphine complex 4 and
basal/basal orientation in 5 and 6.
The Fe–Fe bonds in 4, 5 and 6 (2.5279(13), 2.6374(8) and
2.6118(5) A) are longer than those of their parent complexes 1,
2 and 3 (2.5025(7), 2.5280(10) and 2.5103(7) A), respectively,
which is mainly caused by the introduction of PMe3, which is a
stronger electron-donating ligand than CO. The Fe–Fe bond
lengths of 5 and 6 are also very close to those in the structures
of enzymes from Clostridium pasteurianum and Desulfovibrio
desulfuricans (ca. 2.6 A).12,13
The Fe–Fe bond of 2 is the longest in those of 1, 2 and 3,
and that of 5 is the longest in those of 4, 5 and 6. The reason
for this is that the cyclohexyl group in 2 and 5 is a stronger
electron donor than cyclopentyl or cycloheptyl. The cyclohex-
yl group makes the Fe atoms in 2 and 5 more electron rich and
thus elongates the Fe–Fe bonds. Similarly, the Fe–Fe bond in
1 (2.5025(7) A) is close in length to that of methoxyphenyl-
functionalized all-carbonyl complex [(m-SCH2)2N(C6H4OMe-
p)Fe2(CO)6] (8) (2.5076 A), while longer than that of Me-
functionalized all-carbonyl compound [(m-SCH2)2N(Me)Fe2-
(CO)6] (9) (2.4924 A).9 In addition, it has been reported that
the Fe–Fe bond of PPh3-substituted complex [(m-SCH2)2-
N(C6H4OMe-p)Fe2(CO)5PPh3] (10) (2.554 A) is longer than
that of its analogue [(m-SCH2)2N(C6H4Br-p)Fe2(CO)5PPh3]
(11) (2.5094 A) (Table 1).22 These results suggest that the
more strongly electron donating is the bridge, the longer the
Fe–Fe bond is.
Fig. 1 Crystal structures of 1–6 (4a and 4b, and 5a and 5b are two
pairs of optical isomers of 4 and 5, respectively).
representing the two different optical isomers in the crystal
structures of both 4 and 5 (Fig. 1: 4a and 4b for 4; 5a and 5b
for 5). In these structures, the 2Fe2S cores are all in a
‘‘butterfly’’ conformation, and each iron atom is coordinated
in a pseudo square-pyramidal geometry. Each structure has
two fused six-membered rings, one in the chair conformation
and the other in the boat conformation. For example, in
complex 1, the N(1)C(7)S(1)Fe(2)S(2)C(8) ring is in the chair
conformation while the N(1)C(7)S(1)Fe(1)S(2)C(8) ring in the
boat conformation.
The cyclopentyl moiety in the crystal structure of 1 resides in
a pseudo-equatorial position relative to the metallohetero-
cycle, while the cycloalkyl groups in the crystal structures of
2 and 3 are in axial positions. In the PMe3-disubstituted
complex 4, one PMe3 ligand is coordinated to an equatorial
site on an iron atom, which is similar to the conformation of
the benzyl-functionalized PMe3-disubstituted complex
[(C6H5CH2)N(SCH2)2Fe2(CO)4(PMe3)2] (7),10,11 while com-
plexes 5 and 6 feature a basal/basal configuration (Fig. 1). In
addition, it is notable that the introduction of phosphine
ligands may change the initial conformation of the substituted
ADT bridges in their crystal structures. In complex 4, the
cyclopentyl moiety is not in a pseudo-equatorial but an axial
position relative to the metalloheterocycle. The switch between
equatorial and axial positions of the cyclopentyl ring upon
phosphine substitution (1 to 4) could be explained by fluxion-
ality of the ADT bridge in solution and the steric requirement
Spectroscopic characterization of complexes 1–6
1
Complexes 1–6 were characterized by elemental analysis, H
and 31P NMR, and mass spectroscopy.
The mass spectra in API-ESI positive mode show the parent
ion peaks at m/z 455.0 [M + H]+ for 1, 470.0 [M + H]+ for 2,
484.0 [M + H]+ for 3, 552.0 [M + H]+ for 4, 565.9
[M + H]+ for 5 and 580.0 [M + H]+ for 6.
Owing to the electronic effects of the different cycloalkyl
groups, the 1H NMR signals of 2 are downfield-shifted
compared to those of 1 and 3. 1–3 each display a singlet for
their CH2S groups (d = 3.301 for 1, 3.385 for 2 and 3.226 for
3). Influenced by the ADT N atoms, the nearest H (9-H) atoms
of the cycloalkyl groups in 1–3 present themselves as singlets,
respectively (d = 3.009 for 1, 2.470 for 2 and 2.522 for 3),
while the signals for the other H atoms on the cycloalkyl
groups shift upfield (d = 1.722–1.260 for 1, 1.730–1.013 for 2
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1448–1454 | 1449

View Article Online
Table 2 Comparison of n(CO) bands in complexes 1–6, 7, 12 and 13
Complex
n(CO)/cmÀ1
Notes
1
2
3
4
5
6
12
13
7
2069(s), 2025(s), 1989(s)
2071(s), 2023(s), 1979(s)
2072(s), 2033(s), 1993(s)
1976(s), 1940(s), 1905(s)
1966(m), 1940(s), 1901(s)
2035(m), 1938(s), 1887(s)
2074(m), 2036(s), 1995(s)
1979(s), 1942(s), 1898(s)
1980, 1943, 1892
This work
This work
This work
This work
This work
This work
Refs. 6–8
Refs. 5–8
Refs. 10 and 11
Fig. 2 Cyclic voltammograms of 1, 2 and 3.
and 1.708–1.295 for 3). The introduction of PMe3 makes the
signals of the CH2S groups shift upfield and has little influence
on the signals of the H atoms of the cycloalkyl groups.
Interestingly, only a singlet in the 1H NMR spectra for the
PMe3 group is observed for 4, 5 and 6, respectively, implying
fast fluxionality of the PMe3 ligand in solution.
The 31P NMR spectra of 4–6 each show a singlet for their
PMe3 groups (d = 29.935 for 4, 24.108 for 5 and 24.085 for 6),
again indicating the fluxionality the PMe3 group in solution.
Because of a different conformation, the 31P NMR signal of 4
is downfield-shifted compared to those of 5 and 6.
sible oxidation peaks, respectively. The first reduction peaks at
À1.663 V for 1, À1.656 V for 2, À1.685 V for 3, À1.997 V for
4, À1.993 V for 5 and À1.999 V for 6 are ascribed to the first
one-electron reduction process from FeIFeI to FeIFe0, by
comparison with electrochemical studies of other ADT-
bridged diiron complexes.9–11,21 Similarly, the second reduc-
tion peaks at À2.217 V for 1, À2.240 V for 2, À2.207 V for 3,
À2.316 V for 4, À2.293 V for 5 and À2.351 V for 6, are
assigned to the one-electron reduction process from FeIFe0 to
Fe0Fe0. The first oxidation peaks of complexes 1–6, 0.547,
0.588, 0.575, À0.212, À0.205 and À0.221 V, respectively, are
ascribed to the one-electron oxidation process from FeIFeI to
FeIFeII. The second oxidation peaks for complexes 1–6, 0.879,
0.921, 0.903, À0.087, À0.0770 and À0.0777 V, respectively,
are assigned to the second electron-transfer process from
FeIFeII to FeIIFeII. Their electrochemical data, together with
those of complexes [(m-SCH2)2N(C6H4OMe-p)Fe2(CO)6] (8)
and [(m-SCH2)2N(C6H4OMe-p)Fe2(CO)5PPh2H] (14), are
given in Table 3.
We have studied the CO stretching frequency in the region
1880–2075 cmÀ1 for complexes 1–6. The IR n(CO) data of 1–6,
together with those of [(m-PDT)Fe2(CO)6] (PDT = 1,3-pro-
panedithiolate) (12), [(m-PDT)Fe2(CO)4 (PMe3)2] (13) and
[Fe2(CO)4(PMe3)2(CH2S)2N–CH2–C6H5] (7), are listed in
Table 2 for comparison.5–8 Table 2 demonstrates that the
introduction of two phosphine ligands lowers the CO stretch-
ing frequency, which indicates an enhancement of the electron
density at the two iron atoms and a weakening of the CO triple
bonds. The IR behavior described above resembles that dis-
played by previously reported PDT-bridged analogs.20
The average n(CO) value of PDT-bridged all-carbonyl
complex 12 is higher than those of 1, 2 and 3 because
cycloalkyl-substituted ADT bridges are better electron donors
than PDT bridges.
Compared with the methoxyphenyl-functionalized all-car-
bonyl complex 8, complexes 1–3 are reduced with more
difficultly, indicating that the different electronic effects of
the cycloalkyl and C6H4OMe-p groups result in different
electron accumulations on the iron cores in each individual
complex.
The CV data of complexes 4–6 are shifted in a cathodic
direction compared to their parent complexes 1–3, respec-
tively. It is the stronger donor character of PMe3 that renders
the reduction of iron atoms more difficult. 4–6 are also reduced
with more difficultly than 14, resulting from both different
phosphine ligands and different ADT bridges.
The average n(CO) values of 1 and 3 are higher than that of
2. The cyclohexyl group enhances electron accumulation on
the Fe atoms of 2, leading to stronger back-bonding from the
Fe atoms to CO and a weakening of the CO triple bonds.
In the subset of complexes 4, 5, 6, 7 and 13, each one has a
wave number at or near 1940 cmÀ1, suggesting PMe3 has a
characteristic influence on the CO stretching frequency. This
means that the CO stretching frequency is influenced by both
the ADT bridges and the ligand on the iron atoms, in
particular the latter when it is a strong electron-donor.
The reduction behaviours of complexes 2, 3, 5 and 6 were
further studied by cyclic voltammetric techniques in the pre-
sence of HOAc. As shown in Fig. 4, after HOAc was added,
Cyclic voltammograms of complexes 1–6
The cyclic voltammograms (CV) of complexes 1–6 were
studied to evaluate the effects of different ADT bridges and
PMe3 on the redox properties of the iron cores of the model
complexes. All the CV measurements were carried out in
CH3CN (with 0.1 M n-Bu4NPF6 as the electrolyte) and
scanned in the cathodic direction, as indicated in Fig. 2 and
Fig. 3. 1–3 display one quasi-reversible reduction, one irrever-
sible reduction and two irreversible oxidation peaks, while 4–6
are manifested by two irreversible reduction and two irrever-
Fig. 3 Cyclic voltammograms of 4, 5 and 6.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1450 | New J. Chem., 2007, 31, 1448–1454

View Article Online
Table 3 Electrochemical data of 1–6, 8 and 14 (vs. Fc/Fc+
)
Epa/V
FeIFeI/FeIIFeI
Complex
Epa/V
FeIIFeI/FeIIFeII
Epc/V
FeIFeI/Fe0FeI
Epc/V
Fe0FeI/Fe0Fe0
1
2
3
4
5
6
8
14
0.547
0.588
0.575
À0.212
À0.205
À0.221
0.48
0.879
0.921
0.903
À0.087
À0.0770
À0.0777
0.81
À1.663
À1.656
À1.685
À1.997
À1.993
À1.999
À1.61
À2.217
À2.240
À2.207
À2.316
À2.293
À2.351
À2.10
0.26
0.49
À1.78
À2.22
the first electrochemically irreversible one-electron reduction
peak of 5 (À1.993 V vs. Fc/Fc+) grew, but its position didn’t
shift. This indicated that H+ was not attached to the model
complex at that time, while the second reduction peak shifted
to a more positive potential by 0.060 V after addition of one
equiv. of HOAc to a 1 mM solution of 5, implying that H+
was introduced. Addition of a second equiv. of HOAc made
the second reduction peak shift towards a more positive
potential by another 0.077 V while the position of the first
reduction peak did not shift. The addition of further acid led
to the growth of the second reduction peak height, and caused
its position to shift slightly to a more negative potential,
indicating that excess H+ didn’t incorporate into the model
complex. Complexes 2, 3 and 6, as shown in Fig. 5, had similar
catalytic cycles to that of 5. The observations described above
are indicative of catalytic proton reduction. The catalytic
activity of 5 was also supported by the electrolysis of a 1 mM
solution of 5 (5 ml, 5 mM) with excess HOAc (20 mM) at
À2.3 V. The initial rate of electrolysis was approximately 1.8
times that without 5. 12 F molÀ1 passed within 20 min,
corresponding to 6 turnovers.
This transitional form is easily protonated, releasing H2 and
liberating the catalyst.
To the best of our knowledge, this is the first time the ECEC
mechanism has been proposed for a H2 production process
catalyzed by mimics of [FeFe]-hydrogenases, which was initi-
ally proposed only for [NiFe]-hydrogenases.23 This mechan-
ism is different from the EECC mechanism9 proposed for 8
and the CECE mechanism11 suggested for [(m-SCH2)2-
N(C6H4Br-p)Fe2(CO)6] (15). In our case, because the ADT
nitrogen atom was not protonated, the reduction potentials in
the catalytic process were more negative than those in the
similar process catalyzed by 15. However, it may be the richer
electron accumulation on the Fe atoms of our model com-
plexes, caused by the cycloalkyl groups, that makes one Fe
atom become protonated after the first electrochemical reduc-
tion, which is different from the EECC mechanism suggested
for 8. This indicates that the bridge influences the mechanism
of catalytic proton reduction.
Experimental section
Based on the previously reported electrochemical behaviors
of other diiron analogues9,10,19,21,23 and the electrochemical
data of our model complexes, we think an electro-
chemical–chemical–electrochemical–chemical (ECEC) mecha-
nism is rational for the catalytic cycles of complexes 2, 3, 5 and
6 (Scheme 1). For example, 5 is firstly electrochemically
reduced to 5À at its initial À1.993 V. Then, the reduced iron
atom is protonated to form 5H, which is followed by the
second iron atom electrochemical reduction leading to 5HÀ.
General procedures
All reactions were carried out under dry, oxygen-free nitrogen
using standard Schlenk techniques. Solvents were dried and
distilled prior to use according to standard methods. Com-
mercially available chemicals such as paraformaldehyde,
Fe(CO)5, PMe3 (1 M in THF), LiBEt3H, F3CCOOH and
CnH2nÀ1NH2 (n = 5, 6, 7) were reagent grade and used as
received. The starting complex [Fe2S2(CO)6] was prepared
according to the literature.16–18 Elemental analysis was carried
out on a Vario EL III Elemental Analyser. IR spectra were
taken on a Magna-75 FT-IR spectrophotometer using KBr
pellets in the range of 4000–400 cmÀ1. H NMR spectra were
1
collected on a Varian Unity 500NMR spectrometer. Mass
spectra were recorded on an DECAX-3000 LCQ Deca XP
instrument.
Synthesis of [Fe2(CO)6(CH2S)2N–C5H9] (1). Fe2S2(CO)6
(1 mM, 0.344 g) was dissolved in dry THF (40 ml) under a
nitrogen atmosphere and then cooled to À78 1C with acetone
and liquid nitrogen. After the solution was stirred for 30 min,
LiBEt3H (2 mM) was added dropwise very slowly. At the
midpoint of the addition, the color of the reaction mixture turned
from red to dark green; for the rest of addition it remained green.
After another 30 min, F3CCOOH (2 mM, 0.149 ml) was added.
The new mixture was stirred for an additional hour. The cool
solution was added to a mixture of paraformaldehyde (40 mM,
Fig. 4 Cyclic voltammogram of 5 (1.0 mM) with HOAc.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1448–1454 | 1451

View Article Online
Fig. 5 Cyclic voltammograms of (a) 2, (b) 3 and (c) 6 (1.0 mM) with HOAc.
1.2 g) and C5H9NH2 (1 mM, 1.98 ml) in THF that had been
stirred for 10 h and cooled to 0 1C. The combined mixture was
stirred for 24 h and the majority of the solvent was evaporated
under vacuum.19 The remaining solution was filtered through
silica gel. A red fraction was collected by elution with hexane.
Recrystallization of the crude product from freshly distilled
pentane in a fridge at À20 1C for several days gave compound
1 (0.342 g, 75%) as dark red crystals (found: C, 34.33; H, 2.89;
N, 3.10. Calc. for C13H13Fe2S2O6N: C, 34.31; H, 2.88; N,
3.08%); dH (500 MHz; CDCl3; Me4Si): 3.301 (4H, s, 2CH2S),
3.009 (1H, s, 9-H), 1.722–1.260 (8H, m, 4CH2); m/z 455.0
(49%, M + H+).
Synthesis of [Fe2(CO)6(CH2S)2N–C7H13] (3). Compound 3
was prepared using a procedure similar to that used for
preparing 1, except that C7H13NH2 (1 mM, 3.00 ml) was used
as the primary amine. Recrystallization of the crude product
from freshly distilled heptane in a fridge at À20 1C for several
days gave compound 3 (0.34 g, 70%) as dark red crystals
(Found: C, 37.10; H, 3.44; N, 2.99. Calc. for C15H17Fe2S2O6N:
C, 37.29; H, 3.55; N, 2.90%); dH (500 MHz; CDCl3; Me4Si):
3.226 (4H, s, 2CH2S), 2.522 (1H, s, 9-H), 1.708–1.295 (12H, m,
6CH2); m/z 484.0 (100%, M + H+).
Synthesis of [Fe2(CO)4(PMe3)2(CH2S)2N–C5H9] (4). PMe3
(2 mM, 2 ml) was added to a solution of complex 1 (0.228 g,
0.5 mM) in 10 ml of freshly distilled hexane under a nitrogen
atmosphere. After 24 h of stirring, the color of the solution
changed from red to purple. The solvent and unreacted PMe3
were then removed under vacuum. The resulting dark brown
residue was extracted with toluene and recrystallized from
freshly distilled pentane to give 4 (0.166 g, 60%) as dark red
crystals (Found: C, 37.10; H, 5.58; N, 2.49. Calc. for
C17H31Fe2S2O4NP2: C, 37.04; H, 5.67; N, 2.54%); dH (500
MHz; CDCl3; Me4Si): 3.190 (4H, s, 2CH2S), 2.830 (1H, s,
13-H), 1.782–0.868 (8H, m, 4CH2), 1.259 (18H, s, 2PMe3); dP
(202 MHz; CDCl3; Me4Si): 29.935 (PMe3); m/z 552.0 (100%,
M + H+).
Synthesis of [Fe2(CO)6(CH2S)2N–C6H11] (2). Compound 2
was prepared using a procedure similar to that used for
preparing 1, except that C6H11NH2 (1 mM, 2.24 ml) was used
as the primary amine. Recrystallization of the crude product
from freshly distilled hexane gave compound 2 (0.37 g, 80%)
as red crystals (Found: C, 35.90; H, 3.19; N, 2.88. Calc. for
C14H15Fe2S2O6N: C, 35.85; H, 3.22; N, 2.99%); dH (500 MHz;
CDCl3; Me4Si): 3.385 (4H, s, 2CH2S), 2.470 (1H, s, 9-H),
1.730–1.013 (10H, m, 5CH2); m/z 470.0 (100%, M + H+).
Synthesis of [Fe2(CO)4(PMe3)2(CH2S)2N–C6H11] (5). Com-
pound 5 was prepared using a procedure similar to that used
for preparing 4, except that [Fe2(CO)6S2(CH2)2N–C6H11]
(0.235 g, 0.5 mM) was used as the starting all-carbonyl
complex. The resulting dark brown residue was extracted with
toluene, followed by a recrystallization from freshly distilled
hexane to give 5 (0.164 g, 58%) as dark red crystals (Found: C,
38.10; H, 5.78; N, 2.49. Calc. for C18H33Fe2S2O4NP2: C,
38.25; H, 5.88; N, 2.48%); dH (500 MHz; CDCl3; Me4Si):
3.245 (4H, s, 2CH2S), 2.370 (1H, s, 13-H), 1.700–1.011 (10H,
m, 5CH2), 1.482 (18H, s, 2PMe3); dP (202 MHz; CDCl3;
Me4Si): 24.108 (PMe3); m/z 565.9 (100%, M + H+).
Synthesis of [Fe2(CO)4(PMe3)2(CH2S)2N–C7H13] (6). Com-
pound 6 was prepared using a procedure similar to that used
for preparing 4, except that [Fe2(CO)6S2(CH2)2N–C7H13]
(0.241 g, 0.5 mM) was used as the starting all-carbonyl
complex. The resulting brown residue was extracted with
toluene, followed by a recrystallization from freshly distilled
Scheme 1 ECEC mechanism proposed for the H2 production process
catalyzed by our complexes (X = 2, 3, 5 or 6).
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1452 | New J. Chem., 2007, 31, 1448–1454

View Article Online
Table 4 Crystallographic data summary for complexes 1–6a
1
2
3
4
5
6
Empirical formula
Formula weight
Space group
a/A
b/A
c/A
a/1
C
13H13Fe2NO6S2 C14H15Fe2NO6S2 C15H17Fe2NO6S2 C17H31Fe2NO4P2S2 C18H33Fe2NO4P2S2 C19H35Fe2NO4P2S2
455.06
P2(1)/n
7.7471(9)
11.7596(15)
19.604(2)
90
90.778(5)
90
1785.8(4)
4
1.693
469.09
P2(1)/c
9.6577(2)
13.08730(10)
14.9696(2)
90
92.7930(10)
90
1889.81(5)
4
1.649
483.12
P2(1)/n
10.604(3)
12.035(4)
15.428(4)
90
92.439(6)
90
1967.2(10)
4
1.631
551.19
ꢀ
P1
565.22
P2(1)/c
22.819(5)
10.912(2)
23.588(5)
90
118.581(2)
90
5158.0(19)
8
1.456
579.24
P2(1)/n
11.0584(6)
22.4454(10)
11.6423(6)
90
113.615(2)
90
2647.7(2)
4
1.453
12.4237(5)
14.1988(6)
14.6021(6)
98.3330(10)
91.9110(10)
94.4710(10)
2538.28(18)
8
b/1
g/1
V/A3
Z
r
calc/g cmÀ3
1.442
l(Mo-Ka)/A
0.71073
293(2)
1.885
0.0480
0.0928
0.71073
293(2)
1.784
0.0521
0.1259
0.71073
293(2)
1.717
0.0310
0.0894
0.71073
293(2)
1.454
0.0785
0.2815
0.71073
293(2)
1.433
0.0568
0.0973
0.71073
293(2)
1.398
0.0410
0.1335
b
T/K
m/mmÀ1
Rb
wRc
a
P
CCDC reference numbers 622703–622708. For crystallographic data in CIF or other electronic format see DOI: 10.1039/b616236c. R = ||Fo|
1
P
P
P
À |Fc||/ |Fo|. wR = [ w(|Fo| À |Fc|)2/ wFo ]2.
c
2
heptane to give 6 (0.144 g, 50%) as dark red crystals (Found:
C, 39.35; H, 6.10; N, 2.49. Calc. for C19H35Fe2S2O4NP2: C,
39.40; H, 6.09; N, 2.42%); dH (500 MHz; CDCl3; Me4Si):
3.129 (4H, s, 2CH2S), 2.426 (1H, s, 13-H), 1.700–0.866 (12H,
m, 6CH2), 1.488 (18H, s, 2PMe3); dP (202 MHz; CDCl3;
Me4Si) 24.085 (PMe3); m/z 580.0 (100%, M + H+).
Conclusions
Selecting an appropriate ADT bridge is important for obtain-
ing a good mimic of the active site of [FeFe]-hydrogenases,
which can reduce protons to H2 at a relatively high potential.
The electronic effect of the ADT is a key factor on the redox
properties of the model diiron complex because the bridge
influences the H2 production mechanism.
X-Ray structure determination of complexes 1–6. Single
crystals of complexes 2 and 4 were mounted on a Siemens
Smart CCD diffractometer with Mo-Ka radiation (l =
0.71073 A). Single crystals of complexes 1, 3, 5 and 6 were
performed on a Mercury-CCD diffractometer equipped with
graphite-monochromated Mo-Ka radiation (l = 0.71073 A).
All data were collected at 293(2) K using a o-2y scanning
mode. An empirical absorption correction was made of the
multi-scan type. The structure was solved by direct methods
and refined by full-matrix least-squares techniques using the
SHELXL-97 program.24 Anisotropic displacement parameters
were refined for all non-hydrogen atoms. The hydrogen atoms
were added in a riding model and not refined. Crystallographic
data of 1–6 are outlined in Table 4.
Acknowledgements
We are grateful to the National Key Foundation of China (no.
20633020), the Science & Technology Innovation Foundation
for the Young Scholar of Fujian Province (no. 2005J059), and
the National Nature Science Foundation of China (no.
20471061) for their financial support of this work.
References
1 (a) M. Y. Darensbourg, E. J. Lyon and J. J. Smee, Coord. Chem.
Rev., 2000, 206–207, 533–561; (b) M. Y. Darensbourg, E. J. Lyon,
X. Zhao and I. P. Georgakaki, Proc. Natl. Acad. Sci. U. S. A.,
2003, 100, 3683–3688; (c) D. J. Evans and C. J. Pickett, Chem. Soc.
Rev., 2003, 32, 268–275; (d) X. Liu, S. K. Ibrahim, C. Tard and C.
J. Pickett, Coord. Chem. Rev., 2005, 249, 1641–1652.
2 (a) M. Frey, ChemBioChem, 2002, 3, 153–160; (b) S. P. J. Albracht,
Biochem. Acta, 1994, 1188, 167–204; (c) A. Volbeda, M. H.
Charon, C. Piras, E. C. Hatchikian, M. Frey and J. C. Fontecil-
la-Camps, Nature, 1995, 373, 580–587; (d) R. Cammack, Nature,
1999, 397, 214–215; (e) M. W. W. Adams and E. I. Stiefel, Science,
1998, 282, 1842–1843; (f) J. Alper, Science, 2003, 299, 1686–1687.
3 M. Korbas, S. Vogt, W. Meyer-Klaucke, E. Bill, E. J. Lyon, R. K.
Thauer and S. Shima, J. Biol. Chem., 2006, 281, 30804–30813.
4 M. W. W. Adams, Biochim. Biophys. Acta, 1990, 1020, 115–145.
5 L. Schwartz, G. Eilers, L. Eriksson, A. Gogoll, R. Lomoth and S.
Ott, Chem. Commun., 2006, 520–522.
6 X. Zhao, I. P. Georgakaki, M. L. Miller, J. C. Yarbrough and M.
Y. Darensbourg, J. Am. Chem. Soc., 2001, 123, 9710–9711.
7 E. J. Lyon, I. P. Georgakaki, J. H. Reibenspies and M. Y.
Darensbourg, Angew. Chem., Int. Ed., 1999, 38, 3178–3180.
8 D. Chong, I. P. Georgakaki, R. Mejia-Rodriguez, J. Sanabria-
Chinchilla, M. P. Soriaga and M. Y. Darensbourg, Dalton Trans.,
2003, 4158–4163.
Electrochemistry
Acetonitrile (Aldrich Chemicals, spectroscopy grade) was the
solvent used for the electrochemistry, with a solution of 0.1 M
n-Bu4NPF6 in MeCN being used as the electrolyte. The
electrolyte solution was de-gassed by bubbling argon through
it for 10 min before measurements were carried out. All
electrochemistry results were obtained at a scan rate of
100 mV sÀ1 by using a CHI660A potentiostat and a three-
electrode cell under argon. The working electrode was a glassy
carbon disc (diameter 3 mm) polished with 1 mm diamond
pastes and sonicated in ion-free water for 10 min. The
reference electrode was a Ag/AgCl electrode (3 M KCl in
H2O) and the auxiliary electrode was a platinum wire. All
potential data are quoted against the Fc/Fc+ potential.
9 L.-C. Song, J.-H. Ge, X.-G. Zhang, Y. Liu and Q.-M. Hu, Eur. J.
Inorg. Chem., 2006, 3204–3210.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1448–1454 | 1453

View Article Online
10 S. Ott, M. Kritikos, B. Akermark, L.-C. Sun and R. Lomoth,
Angew. Chem., Int. Ed., 2004, 43, 1006.
18 D. Seyferth and R. S. Henderson, J. Organomet. Chem., 1981, 218,
C34–C36.
11 S. Ott, M. Kritikos, B. Akermark and L.-C. Sun, Angew. Chem.,
2003, 115, 3407–3410.
19 H.-X. Li and T. B. Rauchfuss, J. Am. Chem. Soc., 2002, 124,
726–727.
12 Y. Nicolet, C. Piras, P. Legrand, E. C. Hatchikian and J. C.
Fontecilla-Camps, Structure, 1999, 7, 13–23.
13 J. W. Peters, W. N. Lanzilotta, B. J. Lemon and L. C. Seefeldt,
Science, 1998, 282, 1853–1858.
20 P. Li, M. Wang, C.-J. He, G.-H. Li, X.-Y. Liu, C.-N. Chen, B.
Akermark and L.-C. Sun, Eur. J. Inorg. Chem., 2005, 2506–2513.
21 W. Dong, M. Wang, X.-Y. Liu, K. Jin, G.-H. Li, F.-J. Wang and
L.-C. Sun, Chem. Commun., 2006, 305–307.
14 Y. Nicolet, B. J. Lemon, J. C. Fontecilla-Camps and J. W. Peters,
Trends Biochem. Sci., 2000, 25, 138–143.
22 S. Ott, M. Borgstrom, M. Kritikos, R. Lomoth, J. Bergquist, B.
¨
Akermark, L. Hammarstrom and L. Sun, Inorg. Chem., 2004, 43,
¨
15 H. Fan and M. B. Hall, J. Am. Chem. Soc., 2001, 123, 3828–3829.
16 D. Seyferth, R. S. Henderson and L.-C. Song, Organometallics,
1982, 1, 125–133.
17 J. D. Lawrence, H. Li and T. B. Rauchfuss, Chem. Commun., 2001,
1482–1483.
4683–4692.
23 A. Perra, E. S. Davies, J. R. Hyde, Q. Wang, J. McMaster and M.
Schroder, Chem. Commun., 2006, 1103–1105.
¨
24 G. M. Sheldrick, SHELXL-97, Program for the refinement of
crystal structures, University of Gottingen, Germany, 1997.
¨
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1454 | New J. Chem., 2007, 31, 1448–1454