Bio-inspired, side-on attachment of a ruthenium photosensitizer to an iron hydrogenase active site model

Article abstract of DOI:10.1039/b606659c

The first ruthenium-diiron complex [(-pdt)Fe₂(CO) ₅{PPh₂(C₆H₄CCbpy)}Ru(bpy) ₂]²⁺ 1 (pdt = propyldithiolate, bpy = 2,2′- bipyridine) is described in which the photoactive rutheniu

Full text of DOI:10.1039/b606659c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Bio-inspired, side-on attachment of a ruthenium photosensitizer to an iron
hydrogenase active site model†
Jesper Ekstro¨m,a Maria Abrahamsson,b Carol Olson,b Jonas Bergquist,c Filiz B. Kaynak,d Lars Eriksson,d
b
a
b
a,b
Licheng Sun,a,e Hans-Christian Becker, Bjo¨rn Akermark, Leif Hammarstro¨m* and Sascha Ott*
˚
Received 11th May 2006, Accepted 18th July 2006
First published as an Advance Article on the web 8th August 2006
DOI: 10.1039/b606659c
The first ruthenium–diiron complex [(l-pdt)Fe2(CO)5{PPh2(C6H4CCbpy)}Ru(bpy)2]2+ 1 (pdt =
propyldithiolate, bpy = 2,2ꢀ-bipyridine) is described in which the photoactive ruthenium
trisbipyridyl unit is linked to a model of the iron hydrogenase active site by a ligand directly attached to
one of the iron centers. Electrochemical and photophysical studies show that the light-induced MLCT
excited state of the title complex is localized towards the potential diiron acceptor unit. However, the
relatively mild potential required for the reduction of the acetylenic bipyridine together with the easily
oxidized diiron portion leads to a reductive quenching of the excited state, instead. This process results
in a transiently oxidized diiron unit which may explain the surprisingly high light sensitivity of
complex 1.
called proximal iron center FeP (Chart 1(a)).3,4 Considering the
importance of the enzyme in regard to future energy technologies
and the hydrogen society, it is not surprising that the bioinorganic
community has developed a great interest to model the active site
of these very useful enzymes.5,6 The objectives of this research
range from the pure modeling challenges7 to detailed mechanistic
and functional studies on the active site models.8–11
We have recently become interested in the use of Fe H2ase active
site mimics as potential catalysts for the photochemical production
of hydrogen and have thus introduced them as electron acceptors
in multi unit assemblies.12 In the first studies, the dithiolate
ligand served as the anchoring point for the attachment of the
photosensitizer, and the first ruthenium diiron dyads of this type
could be synthesized.13,14 However, the dithiolate ligand in the
enzyme is believed to contain a secondary amine, the purpose
of which is to shuttle protons to and from the distal iron center
FeD.15–17 This important functional role is impeded in the synthetic
diiron models as soon as they are functionalized at this position.
Nature has chosen to leave the dithiolate “naked” and attached the
electron transfer chain comprised by the [Fe4S4] cluster via a sulfur
ligand from a nearby cysteine residue to the H-cluster instead.3,4
Only a few examples of Fe H2ase active site mimics where
redox-active metal units are attached to one of the iron centers
have been reported,6,7 however, no photo-active unit was
Introduction
Hydrogen metabolism in nature is regulated by enzymes called
hydrogenases.1 Among these, iron hydrogenases (Fe H2ase) cat-
alyze the oxidation of hydrogen as well as the reduction of protons
with high turnover numbers and at mild potential.2 The catalytic
site of the Fe H2ase is comprised of two iron nuclei which are
in bonding distance and are tethered by a bridging, non-proteic
dithiolate. Biologically unusual CO and CN− ligands present the
low valent iron centers with remarkable electronic properties.
Electrons are shuttled to and from the active site via a [Fe4S4]
cluster which is directly coordinated through a cysteine to the so-
aDepartment of Organic Chemistry, Arrhenius Laboratory, Stockholm
University, 10691, Stockholm, Sweden
bDepartment of Photochemistry and Molecular Science, Uppsala University,
Box 523, 75120, Uppsala, Sweden
cDepartment of Physical and Analytical Chemistry, BMC, Uppsala Univer-
sity, Box 599, 75124, Uppsala, Sweden
dDepartment of Structural Chemistry, Arrhenius Laboratory, Stockholm
University, 10691, Stockholm, Sweden
eKTH Chemistry, Organic Chemistry, Royal Institute of Technology, 10044,
Stockholm, Sweden
† Electronic supplementary information (ESI) available: Spectral changes
of the UV/Vis spectrum of complex 1 upon exposure to room light. See
DOI: 10.1039/b606659c
Chart 1
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 4599–4606 | 4599
©

View Article Online
Table 1 Infrared data of [(l-pdt)Fe2(CO)6] and complexes 1–5 in the
ever included or investigated. Inspired by this motif, we have
synthesized dyad 1 in which the photosensitizer is linked to a model
of the Fe H2ase active site by a phosphine ligand in a similar side-
on fashion (Chart 1(b)). The diiron complexes were investigated
by crystallographic, spectroscopic and electrochemical techniques
and the photophysical properties of the ruthenium diiron dyad 1
were studied by steady-state and time-resolved spectroscopy.
carbonyl stretching frequency region (measured in THF solution at 25 ◦C)
Compounda
m(CO)/cm−1
[(l-pdt)Fe2(CO)6]
2072, 2032, 1990
2044, 1985, 1937
2000, 1955, 1938
2043, 1980, 1934
2043, 1980, 1934
2044, 1983,
[(l-pdt)Fe2(CO)5{PPh2(BrC6H4)}] 2
[(l-pdt)Fe2(CO)4{PPh2(BrC6H4)}2] 3
[(l-pdt)Fe2(CO)5{PPh2[C6H4CCSi(CH3)3]}] 4
[(l-pdt)Fe2(CO)5{PPh2(C6H4CCH)}] 5
[(l-pdt)Fe2(CO)5{PPh2(C6H4CCbpy)}Ru(bpy)2]2+ 1b
Results and discussion
1931c
a pdt = −SCH2CH2CH2S−. b As PF6 salt. c Measured in CH3CN.
Synthesis and IR spectroscopic characterization
Several synthetic and functional aspects have been considered
in the design of complex 1. The triphenylphosphine ligand was
identified as the most suitable ligand for the connection of the
ruthenium and the diiron portion as it can be functionalized
relatively easily and exhibits a higher stability towards oxidative
degradation than trialkylphosphines. An acetylene has been
chosen as the linking unit due to its appealing linear geometry
and rigidity. Finally, Ru(bpy)3 as photosensitizer displays a long
excited state lifetime combined with a high excited state energy.18
Two different strategies were pursued for the synthesis of the diiron
portion with the acetylenic triphenylphosphine ligand. In the first
approach, the introduction of the acetylene moiety was attempted
on the fully assembled [(l-pdt)Fe2(CO)5{PPh2(p-BrC6H4)}] where
one phenyl group on the phosphine is functionalized with a bromo
substituent. The second strategy relies on the synthesis of the
acetylenic ligand prior to its coordination to the diiron fragment.
(p-Bromophenyl)diphenylphosphine was synthesized from p-
bromophenyl lithium and chlorodiphenylphosphine according to
the literature procedure.19 Subsequent ligand exchange on the
parent [(l-pdt)Fe2(CO)6] with the functionalized phosphine using
trimethylamine N-oxide as decarbonylation agent in CH3CN
afforded the desired mono-substituted [(l-pdt)Fe2(CO)5{PPh2(p-
BrC6H4)}] 2 in 55% isolated yield (Scheme 1). The ligand substi-
tution can conveniently be followed by a change of the carbonyl
stretching frequencies in the respective IR spectra. Substitution
of one carbonyl ligand shifts the frequencies from 2072, 2032,
1990 cm−1 for the parent diiron hexacarbonyl starting material to
2044, 1985, 1937 cm−1 for 2 (Table 1). These values are in agreement
with those previously reported for related diiron complexes ligated
by one triphenylphosphine.20
The ligand exchange is sensitive to the reaction time and the
used amount of Me3NO as an increase of either one results in the
formation of disubstituted [(l-pdt)Fe2(CO)4{PPh2(p-BrC6H4)}2].
This reaction is characterized by a further shift of the IR carbonyl
frequencies towards lower energies (Table 1). Although well known
for trialkylphosphines and phenyldimethylphosphines,21 only one
example of double substitution with sterically more demanding
triarylphosphines has been reported, but the product was not
structurally characterized.22
Introduction of the acetylene moiety at the bromophenyl
substituent of 2 under classical Sonogashira cross coupling
conditions with [Pd(PPh3)2Cl2] and CuI using either conventional
or microwave heating was not successful. Similarly, an attempt
to couple the acetylene to the isolated bromophenylphosphine
ligand failed, reflecting the low reactivity of the electron-rich
bromoarene.23,24
To circumvent these problems, the acetylene was introduced at
the phosphine ligand prior to the complexation of the latter to the
diiron unit. Lithiation of 4-bromotrimethylsilylethynylbenzene25
with butyl lithium afforded the acetylenic phenyl lithium which
was subsequently reacted with chlorodiphenylphosphine to form
the protected acetylenic triphenylphosphine. Without further
purification, the acetylenic ligand was coordinated to the (l-
pdt)Fe2(CO)5 fragment, obtained from [(l-pdt)Fe2(CO)6] by treat-
ment of the latter with n-propylamine.24 After purification by
column chromatography and recrystallization, complex 4 could
be isolated in 47% overall yield (Scheme 2).
Protodesilylation of 4 proceeded in a solution of THF and
methanol in the presence of K2CO3 in good yield. With acetylene
Scheme 1 Reagents and conditions: (a) 2 eq. Me3NO, CH3CN, 2 h, 25 ◦C, 55%; (b) 10 eq. Me3NO, CH3CN, 2 h, 25 ◦C, 16%.
4600 | Dalton Trans., 2006, 4599–4606 This journal is The Royal Society of Chemistry 2006
©

View Article Online
Scheme 2 Reagents and conditions: (a) 1. BuLi, THF, −78 ◦C, 2 h; 2. PPh2Cl, THF, 2 h, −78 to 25 ◦C; (b) n-PrNH2, [(l-pdt)Fe2(CO)6], THF, heat, 22 h,
47% over two steps; (c) K2CO3, THF–CH3OH, 25 ◦C, 4 h, 81%; (d) [PdCl2(PPh3)2], CuI, Et3N, THF, CH3CN, 85 ◦C, 18 h, 59%.
˚
5 in hand, the attachment of the bipyridine ligand of the
photosensitizer was targeted. First attempts to react 5 with 5-
bromobipyridine26 under classical Sonogashira cross coupling
conditions failed and no defined product could be isolated. This
result is consistent with our previous findings that the iron
carbonyl core is rather unstable in the presence of polypyridine
ligands.24 We thus turned our focus on the reaction of 5 with
fully assembled [(5-Br-bpy)Ru(bpy)2]2+ and, most gratifyingly, we
found that 5 reacted smoothly with the latter complex to afford
the ruthenium–diiron dyad 1 in 59% isolated yield after column
chromatography and anion exchange with PF6. The ESI mass
spectrum of 1 features an isotopic pattern typical for ruthenium
complexes at m/z = 1357.11 as the base peak, corresponding to the
loss of one PF6 counter ion. Furthermore, NMR and elemental
analysis data are in agreement with the assigned structure. The
presence of the Fe2(CO)5PPh portion in 1 is expressed by the
characteristic IR absorptions in the carbonyl region, which are
very similar to those of precursors 4 and 5.
Table 2 Selected bond lengths (A) for complexes 3 and 4 with estimated
standard deviations
3
4
Fe1–Fe2
Fe1–P1
Fe2–P2
Fe1–S1
Fe1–S2
Fe2–S1
Fe2–S2
Fe1–C1
Fe1–C4
2.5177(14)
2.256(2)
2.247(2)
2.277(2)
2.273(2)
2.275(2)
2.272(2)
1.768(10)
1.761(10)
Fe1–Fe2
Fe1–P1
2.521(3)
2.232(4)
Fe2–S1
Fe2–S2
Fe1–S1
Fe1–S2
Fe2–C1
Fe2–C2
Fe2–C3
Fe1–C4
Fe1–C5
S1–C11
S2–C13
C1–O1
C2–O2
C3–O3
C4–O4
C5–O5
C27–C28
2.240(4)
2.248(5)
2.255(4)
2.257(4)
1.805(17)
1.775(14)
1.776(16)
1.752(13)
1.780(11)
1.791(13)
1.889(14)
1.083(17)
1.110(16)
1.172(16)
1.139(15)
1.117(13)
1.172(19)
Fe2–C2
Fe2–C3
S1–C13
S2–C11
C1–O1
C2–O2
C3–O3
C4–O4
1.759(9)
1.708(9)
1.850(8)
1.809(8)
1.144(9)
1.164(8)
1.180(8)
1.138(9)
Molecular structures of 3 and 4
X-Ray diffraction analysis of complexes 3 and 4 shows the
usual square pyramidal geometry around the iron centers with
˚
Fe–Fe distances of 2.52 A (Table 2). As shown in Fig. 1,
the two triphenylphosphine ligands in 3 both reside in apical
positions similar to the situation in the dimethylphenylphosphine-
coordinated analogue.22 In contrast, trimethylphosphine ligands
prefer a mixed apical/basal coordination mode in the solid state.21
It thus seems that the bulkiness of the phenylphosphines directs
the ligands to the sterically least demanding apical positions.
Similarly, the acetylenic triphenylphosphine of complex 4
resides in an apical position in the solid state, (Fig. 2), consistent
with the crystal structure of other mono triphenylphosphine-
ligated diiron complexes.20,24 The acetylene with a typical bond
˚
length of 1.17 A points in the same direction as the dithiolate
ligand. The fact that the bulky triphenylphosphines exclusively
prefer the apical positions is encouraging for the use of these
diiron complexes as acceptors in multi unit assemblies. A defined
geometry and distance should give rise to simplified photophysical
properties such as monoexponential excited state decay functions.
Fig. 1 ORTEP (ellipsoids at 50% probability level) of 3. Hydrogen atoms
are drawn as small circles of arbitrary radius.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 4599–4606 | 4601
©

View Article Online
can be observed around 450 nm. The broad absorption in the
350–430 nm region arises from a ligand-centered (LC) transition
localized on the extended acetylenic bipyridine spacer.28 The sharp
peak around 290 nm is attributed to the p → p* transitions on the
auxiliary bipyridine ligands.18,29
Similar to the spectrum of reference complex 6, the spectrum
of 1 is dominated by strong p → p* transitions localized on the
bipyridine ligands at 287 nm.18,29 At least two poorly resolved
absorptions can be observed in the region between 320 and 370 nm.
By comparison with the spectra of complexes 5 and 6, we attribute
these transitions to the (l-pdt)Fe2(CO)5(PPh)3 portion of the
molecule as well as to the conjugated bpy–C≡C–PPh3 ligand.
However, the spectrum of 1 in this region is not a simple sum of the
spectra of 5 and 6. Moreover, the spectrum changes considerably
upon exposure of the sample to light and all spectral features in
this region lose intensity (see ESI†). Although it is not possible
to assign the degradation to either the conjugated linker or to the
diiron unit, it is clear that both processes could have an impact
on the photophysical behavior of the ruthenium chromophore.
The longest wavelength absorption band in the spectrum of 1
between 400–500 nm is largely unaffected by the photodegradation
and is attributed to the MLCT of the ruthenium chromophore,
overlapping the low-intensity absorption of the diiron unit. In
summary, all features of the UV/Vis spectrum of 1 have their
respective counterparts in the spectra of complexes 5 and 6,
although certain shifts and increased extinction coefficients can
be noted when a difference spectrum is constructed. Together with
the observed light-sensitivity of 1, it is clear from this data that the
electronic coupling between the ruthenium and diiron portions of
1 is not negligible.
Fig. 2 Molecular structure of 4 with displacement ellipsoids drawn at the
50% level. Hydrogen atoms are drawn as small circles of arbitrary radius.
Electronic absorption spectra
The electronic absorption spectra of dyad 1, of a representative di-
iron complex 5 and of a ruthenium reference complex [(bpy)2Ru(5-
bpy–C≡C–Ar)]2+ 627 (Ar = 2,5-bis(dodecyloxy)phenyl) with the
same conjugated phenylethynyl bridge in the 5-position of the
bipyridine as in 1, are shown in Fig. 3. Apart from a strong
absorption reaching into the far UV, the spectrum of complex
5 features a strong band with a maximum at 362 nm that is typical
for the (CO)5(PPh3) ligand set.24 A broad and featureless low-
intensity absorption in the visible region reaches up to 580 nm.
In the spectrum of the acetylenic ruthenium reference 6, the
characteristic metal-to-ligand charge transfer (MLCT) transition
Electrochemistry
Compared to [(l-pdt)Fe2(CO)6], the electrochemical reductions
and oxidations of all diiron complexes described herein are
shifted to more negative potentials, reflecting the electron donating
properties of the introduced phosphine ligands (Table 3). The
presence of two phosphines in complex 3 renders the irreversible
oxidation more facile by 300 mV compared to the mono-
substituted complexes. All mono-substituted complexes 1, 2, 4 and
5 exhibit comparable electrochemical behavior with an irreversible
oxidation around 0.38
0.02 V (all potentials herein are peak
potentials of differential pulse voltammetric experiments and are
given vs. Fc+/0). A second irreversible oxidation associated with the
diiron portion can be observed 300 mV after the first oxidation.
Since the first oxidation is irreversible, the second process may
be the oxidation of a species produced chemically after the
first oxidation event. In addition to these two processes, a third
reversible oxidation can be observed in the cyclic voltammogram
of dyad 1 at 0.91 V. Due to the reversibility of this process and its
characteristic potential,18 it can be assigned to the RuIII/II couple
in 1. It therefore seems that the ruthenium portion in complex 1
is intact, despite the preceding irreversible oxidations at the diiron
site.
On the reductive scan, the electronic effect of the phosphine
ligands results in cathodically shifted reduction potentials of all
complexes compared to [(l-pdt)Fe2(CO)6], although the effect
is somewhat less pronounced than in the oxidations. The most
electron rich complex 3 is reduced at most negative potential
Fig. 3 Electronic absorption spectra of compound 1 (—), 5 (---) and of
[(bpy)2Ru(5-bpy–C≡C–Ar)]2+ 627 (· · ·), recorded in CH3CN.
4602 | Dalton Trans., 2006, 4599–4606
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 3 Electrochemical data of [(l-pdt)Fe2(CO)6] and of complexes 1–5a
Compound
E
pa(RuIII/II)/V
E
pa/V
Epc (Fe2)/V
Epc (bpy)/V
[(l-pdt)Fe2(CO)6]
0.67
0.36, 0.64
0.05
0.36, 0.68
0.38, 0.68
0.40, 0.65
−1.60
−1.76
−1.92
−1.74
−1.76
−1.76
[(l-pdt)Fe2(CO)5{PPh2(BrC6H4)}] 2
[(l-pdt)Fe2(CO)4{PPh2(BrC6H4)}2] 3
[(l-pdt)Fe2(CO)5{PPh2C6H4CCSi(CH3)3}] 4
[(l-pdt)Fe2(CO)5{PPh2(C6H4CCH)}] 5
[(l-pdt)Fe2(CO)5{PPh2(C6H4CCbpy)}Ru(bpy)2]2+
1
0.91
−1.51, −2.01
a All potentials were recorded for deaerated, 1 mM solutions in CH3CN containing Bu4NPF6 (0.3 M) as supporting electrolyte and are given vs. Fc+/0
.
Anodic (Epa) and cathodic peak potentials (Epc) were recorded from differential pulse voltammetry (m = 0.1 V s−1). pdt = −SCH2CH2CH2S−.
(−1.92 V). The irreversible reduction associated with the diiron
portion in all mono-substituted complexes occurs at −1.76 V. In
dyad 1, this reduction is preceded by a reversible reduction at
−1.51 V. The reversibility of this wave, together with a comparison
deoxygenated acetonitrile at room temperature, and thus it seems
that the emission quenching in complex 1 is substantial. However,
the room temperature excited state lifetime of a deoxygenated
solution of complex 1 in acetonitrile, freshly prepared in the dark,
is 770 ns. Using the excited state lifetime of complex 6 (1200 ns) as a
reference,27 the quenching can be calculated to be only 35%. This
discrepancy in the observed excited state quenching of complex
1 is perhaps not surprising, considering the sensitivity to light
already detected during the acquisition of the UV/Vis data. Even
two consecutively recorded steady state emission spectra showed a
substantial increase in emission intensity. A similar result could be
obtained by exposing a solution of complex 1 to ambient light for
less than one minute. Nanosecond time-resolved emission exper-
iments, recording entire emission decay traces with a single laser
flash excitation, revealed an even more complicated behavior. The
first laser flash excitation gives an emission lifetime of 770 ns. With
continued flashing, the lifetime first decreases, but subsequently
increases. In parallel, a very short-lived emission (<1 ns) appears,
which could also be observed in time-correlated single photon
counting experiments with picosecond resolution. This behavior
can be explained by photodegradation processes on either the
acetylenic linker and/or the diiron unit that give rise to a changing
electronic coupling between the different parts of the compound.
Although the light-sensitive nature of 1 precludes the determi-
nation of quantitative photophysical data, it can with certainty
be concluded that the excited state is quenched to some degree
compared to that of the reference chromophore 6. In principle, this
quenching can be due to either energy or electron transfer. Since
the absorption and emission spectra do not overlap substantially,
Fo¨rster energy transfer is not likely to occur. Electron transfer can
potentially be of oxidative or reductive nature. The driving force
for electron transfer from the excited state of the ruthenium unit to
the diiron site can be calculated by the Rehm–Weller equation.33
Using the reduction potentials obtained from the electrochemical
studies, the excited state energy, and omitting the work term
arising from Coulombic interactions, the driving force for electron
transfer from the ruthenium excited state to the diiron portion
was calculated to be uphill by 0.66 V. However, a favorable driving
force of 100 mV can be calculated for the reductive quenching
of the excited state, resulting in a transiently oxidized diiron
portion. Since the oxidation of the diiron unit is completely
irreversible in the cyclic voltammetry, a similar decomposition
pathway may explain the observed light-sensitivity of complex 1.
In low temperature glasses, electron transfer is often hindered,
which can explain why no photo-degradation was observed in the
experiments conducted at 77 K.
of the potential of this process with that of complex 6 (Ered
=
−1.14 vs. SCE ≈ −1.52 vs. Fc+/0),27 leads to the assignment of this
reduction to be localized on the extended acetylenic bipyridine.
Finally, a further reduction of 1 can be observed at −2.01 V and is
attributed to the reduction of an auxiliary bipyridine ligand. The
electrochemically most easily reduced ligand at a metal complex
is usually the same ligand that is involved in the photo-induced
MLCT excited state. Since, in the case of complex 1, this ligand
is the acetylenic bipyridine, it appears that the excess electron
would reside close to the acceptor in the lowest MLCT state. This
situation is preferable and may facilitate a subsequent charge shift
to the acceptor unit.
Steady-state and time-resolved spectroscopy
The highest energy peak in the vibrational progression of the
emission spectrum at 77 K is known to give a good estimate of
the excited state energy of a ruthenium polypyridyl complex.30–32
Complex 1 displays an emission maximum at 616 nm in buty-
ronitrile glass at 77 K which corresponds to an excited state
energy of 2.01 eV (Table 4). This value is similar to that found
for reference complex 6 (2.04 eV)27 but somewhat lower than
that of [Ru(bpy)3]2+ (2.12 eV).18 This difference in excited state
energy can be rationalized by the better electron accepting capacity
of the conjugated phenylethynyl bipyridine, compared to that of
the unsubstituted bipyridine ligand. The excited state lifetime of
complex 1 at 77 K is 2.5 ls, which is similar to that of other
ruthenium(II)–polypyridine complexes.18
Complex 1 shows room-temperature MLCT-type emission
centered at 670 nm, while, as expected, the diiron complex 5
shows no emission at ambient temperatures. For the reference
chromophore 6 the emission maximum lies at 666 nm.27 The
emission yield of 1 is ca. 6% of that of the reference complex 6 in
Table 4 Emission maxima and lifetimes, in deoxygenated acetonitrile
solutions, freshly prepared in the dark, at room temperature and in
butyronitrile glass at 77 K
Room temperature
77 K
Compound
kmax/nm
s/ns
kmax/nm
s/ls
Complex 1
670
666
770
1200
616
610
2.5
Not given
Complex 627
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 4599–4606 | 4603
©

View Article Online
Table 5 Selected crystal data for complexes 3 and 4
Conclusions
3
4
With the synthesis of complex 1, we have found a viable strategy
for the construction of the first model of the iron hydrogenase
active site where a photosensitizer is attached to the diiron site
directly via a ligand. From X-ray crystallography it appears
that the triphenylphosphine ligand prefers exclusively the apical
position, which results in a well-defined geometry of the RuFe2
system. Electrochemical studies show that the bridging acetylenic
bipyridine is the most easily reduced ligand. It can thus be
anticipated that the light-induced MLCT excited state of 1 is
localized towards the potential diiron electron acceptor which is a
favorable feature to be included in future assemblies of this type.
As a result of the easily reduced acetylenic ligand and the mild
oxidation potential of the diiron portion, the ruthenium excited
state in 1 can be quenched reductively by the diiron site. The
transiently oxidized diiron site is known to be unstable which
may explain the surprisingly high light-sensitivity of complex 1.
In future developments of the system, the redox properties of the
diiron portion have to be adjusted to more favorable values in order
to fulfill the electronic requirements for electron transfer from the
ruthenium chromophore to the diiron unit. Inclusion of an amine
in the dithiolate linker or further ligand substitutions on the diiron
core with electron donating ligands would enable protonation of
the diiron site and thus render a subsequent reduction more facile.
Such an electronic modulation would not only move the potential
required for the reduction of the diiron site within reach of the
ruthenium excited state, but would also diminish the unwanted
reductive quenching of the ruthenium excited state and decrease
the associated photodegradation of the system.
Empirical formula
T/K
C48H44Br2Fe2O4P2S2 C31H29Fe2O5PS2Si
293(2)
293(2)
716.42
Triclinic
Mr/g mol−1
Crystal system
Space group
1082.41
Monoclinic
P21/n
9.3219(9)
22.8707(16)
22.4947(17)
90
96.750(7)
90
4762.6(7)
4
¯
P1
˚
a/A
13.001(13)
8.539(5)
14.994(10)
90.09(5)
92.35(7)
90.09(6)
1663(2)
2
1.431
1.118
Prism
0.72, 0.99
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
3
˚
V/A
Z
Dc/g cm−3
1.510
2.484
Needle
0.41, 0.88
l(Mo-Ka)/mm−1
Crystal shape
T
N
N
min, Tmax
meas, Nuniq, Rint
obs, Npar, S (GoF)
32469, 14784, 0.1008 16952, 7707, 0.0930
3150, 518, 0.745
2716, 344, 0.935
0.1015, 0.2732
0.2356, 0.3295
−0.56, 0.87
R1, wR2 both with (I > 2r(I)) 0.0678, 0.1624
R1, wR2 (all data)
0.3178, 0.2224
−0.609, 1.180
−3
˚
Dqmin, Dqmax/e A
were refined with anisotropic displacement parameters and the
hydrogens, which were placed at geometrically calculated positions
and let to ride on the atoms they were bonded to, were given
isotropic displacement parameters calculated as nUeq. for the non-
hydrogen atoms with n = 1.5 for methyl hydrogens (–CH3) and
n = 1.2 for methylenic (–CH2–) and aromatic hydrogens. Selected
crystal data are given in Table 5 and selected bond lengths in
Table 2. Molecular structures of complexes 3 and 4 are shown in
Fig. 1 and 2.
Experimental
CCDC reference numbers 607112 and 607113.
NMR spectra were recorded on Varian spectrometers (300 or
400 MHz). The infrared spectra were recorded on Perkin-
Elmer spectrum 1. The elemental analysis were performed in
Analytische Laboratorien, Lindlar, Germany. Microwave heating
experiments was performed in a SmithCreator from Biotage
(Personal Chemistry AB) Uppsala, Sweden. Chromatographic
purification was carried out on Merck silica gel 60 (230–400
mesh). The mass-spectrometry experiments were done on a Bruker
Daltonics BioAPEX-94e superconducting 9.4 tesla FTICR mass
spectrometer (Bruker Daltonics, Billerica, MA, USA) (FTICR-
ESI MS). Electronic absorption spectra were measured on a
Hewlett-Packard 8453 instrument, in 1 × 1 cm quartz cuvettes.
Single-crystal X-ray diffraction patterns were recorded with
an Oxford Diffraction Excalibur diffractometer equipped with a
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b606659c
Electrochemical measurements were carried out in a three-
electrode cell connected to an Autolab potentiostat with a GPES
electrochemical interface (Eco Chemie). The working electrode
was a glassy carbon disc (diameter 2 mm, freshly polished). A
graphite rod served as a counter and a non-aqueous Ag/Ag+
served as a reference electrode (CH Instruments, 10 mM AgNO3
in acetonitrile, E1/2 = 80 mV vs. Fc+/0). All potentials are given
vs. the Fc+/0 couple. Solutions were prepared from dry acetonitrile
˚
(Merck, spectroscopic grade, dried with MS 3 A) and contained
ca. 1 mM of the analyte and 0.3 M tetrabutylammonium hexaflu-
orophosphate (Fluka, electrochemical grade) as supporting elec-
trolyte. Before all measurements, the stirred solutions were purged
with solvent saturated argon to remove residual oxygen and the
experiments were kept under an atmosphere of argon at all times.
Steady state emission measurements were performed on a
SPEX-Fluorolog II fluorimeter, and corrected for different detec-
tor sensitivity at different wavelengths. 1 × 1 cm quartz cuvettes
were used for room temperature measurements. Steady state
emission spectra at 77 K were measured in a liquid-nitrogen
filled cold finger Dewar. Low temperature measurements were
performed in butyronitrile, purified by distillation over P2O5. All
samples were prepared in darkness.
˚
sapphire-3 CCD on a Mo-radiation source (k = 0.71073 A) with
x-scans at different φ to fill Ewald sphere. The sample-detector
distance was 50 mm. Indexing, cell refinements and integration of
reflection intensities were performed with the Crysalis software.34
Numerical absorption correction was performed with the program
X-RED35 verifying the crystal shape with program X-shape.36
The structure was solved by direct methods using SHELXS9737
giving electron density maps where most of the non-hydrogen
atoms could be resolved. The rest of the non-hydrogen atoms were
located from difference electron density maps and the structure
model was refined with full matrix least square calculations on
F2 using the program SHELXL97–2.38 All non-hydrogen atoms
Time resolved emission measurements were performed with
a frequency tripled Q-switched Nd:YAG laser from Quantel
4604 | Dalton Trans., 2006, 4599–4606
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
producing <10 ns flashes. Excitation light at 460 nm was obtained
in an OPO. The emission was detected at right angle with a
monochromator and a P928-type PMT. The PMT output was
recorded on a Philips digital oscilloscope (2 Gsamples s−1) and
analysed with the Applied Photophysics LKS60 software or Igor
Pro 5.03. Excited state lifetimes at 77 K were measured in a
butyronitrile glass (distilled over P2O5) in a cold finger Dewar
filled with liquid nitrogen.
were grown by slow diffusion of pentane into a toluene solution of
the complex at 7 ◦C (Found: C, 51.91; H, 4.19. C31H29Fe2O5PS2Si
requires C, 51.97; H, 4.08%); mmax/cm−1 (THF) 2043, 1980, 1934;
dH (300 MHz; CDCl3) 7.69–7.58 (m, 6 H), 7.51–7.47 (m, 2 H),
7.45–7.42 (m, 6 H), 1.74 (m, 2 H), 1.50 (m, 1 H), 1.40 (m, 2 H),
1.08 (m, 1H), 0.25 (s, 9 H).
[(l-pdt)Fe2(CO)5{PPh2(C6H4CCH)}] (5)
THF was distilled over Na and CH2Cl2 over CaH2 prior
Iron complex 4 (1.6 g, 2.3 mmol) was dissolved in a mixture of
THF (70 ml) and methanol (30 ml). K2CO3 (7 g, 50 mmol) was
added and the solution was stirred at room temperature for 4 h.
The solvent was removed by rotary evaporation and the product
was purified by chromatography on silica with pentane–CH2Cl2 =
3 : 1 as eluent. The product was obtained as a red solid (1.2 g, 81%)
(Found: C, 52.32; H, 3.29. C28H21Fe2O5PS2 requires C, 52.20; H,
3.29); mmax/cm−1 (THF) 2044, 1985, 1936; dH (300 MHz; CDCl3)
7.71–7.61 (m, 6 H), 7.53 (m, 2 H), 7.47–7.42 (m, 6 H), 3.18 (s, 1 H),
1.74 (m, 2 H) 1.51 (m, 1 H), 1.43 (m, 2 H), 1.10 (m, 1 H).
˚
to use. CH3CN was dried using molecular sieve (4 A).
[(l-pdt)Fe2(CO)6],39 diphenyl(4-bromophenyl)phosphine19 and 5-
bromo-2,2ꢀ-bipyridine26 were synthesized according to literature
procedures.
[(l-pdt)Fe2(CO)5{PPh2(BrC6H4)}] (2)
[(l-pdt)Fe2(CO)6] (0.12 g, 0.31 mmol) was dissolved in CH3CN
(20 ml). Trimethylamine N-oxide dihydrate (0.070 g, 0.63 mmol)
was added and the solution was stirred for 20 min. Diphenyl(4-
bromophenyl)phosphine (0.21 g, 0.62 mmol) was added and
the solution was stirred for 2 h. The solvent was removed
under reduced pressure and the crude product was purified by
chromatography on silica with pentane–toluene = 5 : 1 as eluent.
The product was obtained as a red solid (0.12 g, 55%) (Found: C,
44.76; H, 2.93. C26H20BrFe2O5PS2 requires: C, 44.76; H, 2.88%);
mmax/cm−1 (THF) 2044, 1985, 1937; dH (300 MHz; CDCl3) 7.69–
7.64 (m, 4 H), 7.57–7.55 (m, 4 H), 7.45–7.42 (m, 6 H), 1.76 (m, 2
H) 1.53 (m, 1 H) 1.45 (m, 2 H) 1.10 (m, 1 H).
[(l-pdt)Fe2(CO)5{PPh2[C6H4CC(2,2ꢀ-bipyridin-5-
yl)]}Ru(bpy)2](PF6)2 (1)
[(5-Br-bpy)Ru(bpy)2](PF6)2 (0.21 g, 0.22 mmol) was dissolved in
CH3CN (30 ml). A solution of iron complex 5 (0.24 mg, 0.37
mmol) in THF (10 ml) was added together with [Pd(PPh3)2Cl2]
(0.035 g, 0.05 mmol) and Et3N (10 mL). The solution was
degassed with argon and CuI (64 mg, 0.33 mmol) was added.
The reaction mixture was stirred at 85 ◦C for 18 h. After
cooling to room temperature, the solvent was removed under
reduced pressure and the product was purified by chromatography
on silica using CH3CN–H2O–KNO3(aq) = 90 : 5 : 1 as eluent.
The resulting red product was dissolved in a small amount of
CH3CN. A saturated aqueous solution of NH4PF6 was added and
the product was extracted with CH2Cl2, dried over anhydrous
Na2SO4, filtered and concentrated in vacuo. The product was
afforded as a red solid (195 mg, 59%) (Found C, 46.15; H, 3.03; N,
5.73. C58H43Fe2N6O5P3RuS2 requires C, 46.39; H, 2.89; N, 5.60%);
mmax/cm−1 (CH3CN) 2044, 1983, 1931; dH (400 MHz; CD3CN)
8.52–8.48 (m, 6 H), 8.16 (dd, 1 H), 8.10–8.03 (m, 5 H) 7.86 (s,
1 H), 7.82 (d, 1 H), 7.73–7.66 (m, 9 H), 7.63 (s, 1 H), 7.57–7.49
(m, 8 H), 7.44–7.38 (m, 5 H) 1.78 (m, 2 H), 1.50 (m, 3 H), 1.08
(m, 1 H); kmax/nm (e/dm3 mol−1 cm−1) (CH3CN) 287 (64500), 323
(39700), 340 (sh, 39200), 441 (11000); m/z (ESI) 1357.11 (100%)
[M − PF6]+. [M − PF6]+ requires 1356.83.
[(l-pdt))Fe2(CO)4{PPh2(BrC6H4)}2] (3)
[(l-pdt)Fe2(CO)6] (0.12 g, 0.31 mmol) was dissolved in CH3CN
(20 mL). Trimethylamine N-oxide dihydrate (0.35 g, 3.1 mmol)
was added and the solution was stirred for 20 min. Diphenyl(4-
bromophenyl)phosphine (0.42 g, 1.2 mmol) was added and the
solution was stirred for 2 h. The solvent was removed by reduced
pressure and the crude product was purified by chromatography
on silica with pentane–toluene = 1 : 5 as eluent. The product was
obtained as a red solid (0.050 g, 16%). Crystals for X-ray analysis
were grown by slow solvent evaporation from a THF solution of
the complex at 7 ◦C (Found: C, 50.84, H, 3.39. C43H34Br2Fe2O4P2S2
requires: C, 51.02; H, 3.39%); mmax/cm−1 (THF) 2000, 1955, 1938;
dH (300 MHz; CDCl3) 7.69 (m, 8 H) 7.63–7.52 (m, 8 H) 7.40 (m,
12 H) 0.80 (m, 4 H) 0.70 (m, 2 H).
[(l-pdt)Fe2(CO)5{PPh2[C6H4CCSi(CH3)3]}] (4)
Acknowledgements
To 1-bromo-4-trimethylsilylethynylbenzene (1.2 g, 4.8 mmol) in
dry THF (30 ml) was added n-BuLi (1.6 M, 3.0 mL, 4.8 mmol)
under argon at −78 ◦C. The solution was stirred for 2 h before
chlorodiphenylphosphine (0.9 mL, 5 mmol) was added. The
reaction mixture was stirred for an additional 2 h, the solvent
removed by evaporation under reduced pressure and the residue
dissolved in EtOAc and passed through a silica column. The
resulting phosphine (an oil) and [(l-pdt)Fe2(CO)6] (1.9 g, 4.9
mmol) were dissolved in dry THF (100 mL) and n-propylamine (10
mL) and refluxed for 22 h. The solvent was removed under reduced
pressure and the crude product was purified by chromatography
on silica with pentane–toluene = 4:1 as eluent. The product was
obtained as a red solid (1.6 g, 47%). Crystals for X-ray analysis
Financial support for this work was provided by the Swedish
Energy Agency, the Swedish Research Council and NEST-STRP:
SOLAR-H (EU contr. nr. 516510).
References
1 M. W. W. Adams, Biochim. Biophys. Acta, 1990, 1020, 115–145.
2 M. Frey, ChemBioChem, 2002, 3, 152–160.
3 J. W. Peters, W. N. Lanzilotta, B. J. Lemon and L. C. Seefeldt, Science,
1998, 282, 1853–1858.
4 Y. Nicolet, C. Piras, P. Legrand, E. C. Hatchikian and J. C. Fontecilla-
Camps, Structure, 1999, 7, 13–23.
5 D. J. Evans and C. J. Pickett, Chem. Soc. Rev., 2003, 32, 268–275.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 4599–4606 | 4605
©

View Article Online
6 L.-C. Song, Acc. Chem. Res., 2005, 38, 21–28.
23 G. J. R. Singh, J. Org. Chem., 1989, 54, 4453–4457.
7 C. Tard, X. Liu, S. K. Ibrahim, M. Bruschi, L. D. Gioia, S. C. Davies,
X. Yang, L.-S. Wang, G. Sawers and C. J. Pickett, Nature, 2005, 433,
610–613.
24 (a) S. Ott, M. Borgstro¨m, M. Kritikos, R. Lomoth, J. Bergquist, B.
˚
Akermark, L. Hammarstro¨m and L. Sun, Inorg. Chem., 2004, 43,
4683–4692; (b) L. Schwartz, J. Ekstro¨m, R. Lomoth and S. Ott, Chem.
Commun., 2006, DOI: 10.1039/b608260b.
8 F. Gloaguen, J. D. Lawrence, T. B. Rauchfuss, M. Benard and M.-M.
Rohmer, Inorg. Chem., 2002, 41, 6573–6582.
25 R. P. Hsung, C. E. D. Chidsey and L. R. Sita, Organometallics, 1995,
14, 4808–4815.
˚
9 S. Ott, M. Kritikos, B. Akermark, L. Sun and R. Lomoth, Angew.
Chem., Int. Ed., 2004, 43, 1006–1009.
26 F. M. Romero and R. Ziessel, Tetrahedron Lett., 1995, 36, 6471–
10 J.-F. Capon, F. Gloaguen, P. Schollhammer and J. Talarmin, Coord.
Chem. Rev., 2005, 249, 1664–1676.
6474.
27 F. Chaignon, J. Torroba, E. Blart, M. Borgstro¨m, L. Hammarstro¨m
and F. Odobel, New J. Chem., 2005, 29, 1272–1284.
28 U. H. F. Bunz, Chem. Rev., 2000, 100, 1605–1644.
29 J.-P. Sauvage, J.-P. Collin, J.-C. Chambron, S. Guillerez, C. Coudret, V.
Balzani, F. Barigelletti, L. D. Cola and L. Flamigni, Chem. Rev., 1994,
94, 993–1019.
30 J. Treadway, B. Loeb, R. Lopez, P. A. Anderson, F. R. Keene and T. J.
Meyer, Inorg. Chem., 1996, 35, 2242–2246.
31 L. Hammarstro¨m, F. Barigelletti, L. Flamigni, M. T. Indelli, N.
Armaroli, G. Calogero, M. Guardigli, A. Sour, J.-P. Collin and J.-P.
Sauvage, J. Phys. Chem. A, 1997, 101, 9061–9069.
11 L. Schwartz, G. Eilers, A. Gogoll, L. Eriksson, R. Lomoth and S. Ott,
Chem. Commun., 2006, 520–522.
˚
12 L. Sun, B. Akermark and S. Ott, Coord. Chem. Rev., 2005, 249, 1653–
1663.
˚
13 S. Ott, M. Kritikos, B. Akermark and L. Sun, Angew. Chem., Int. Ed.,
2003, 42, 3285–3288.
14 H. Wolpher, M. Borgstro¨m, L. Hammarstro¨m, J. Bergquist, V. Sund-
˚
stro¨m, S. Styring, L. Sun and B. Akermark, Inorg. Chem. Commun.,
2003, 989–991.
15 H.-J. Fan and M. B. Hall, J. Am. Chem. Soc., 2001, 123, 3828–3829.
16 Y. Nicolet, A. L. D. Lacey, X. Vernede, V. M. Fernandez, E. C.
Hatchikian and J. C. Fontecilla-Camps, J. Am. Chem. Soc., 2001, 123,
1596–1602.
17 Z.-P. Liu and P. Hu, J. Am. Chem. Soc., 2002, 124, 5175–5182.
18 A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and A. v.
Zelewsky, Coord. Chem. Rev., 1988, 84, 85–277.
32 J. V. Caspar and T. J. Meyer, Inorg. Chem., 1983, 22, 2444–2453.
33 D. Rehm and A. Weller, Isr. J. Chem., 1970, 8, 259–271.
34 Xcalibur CCD system, CrysAlis Software system, Version 1.170, Oxford
Diffraction, Oxford Diffraction Ltd, 2003.
35 X-RED, Absorption Correction Program, version 1.09, Stoe & Cie
GmbH Darmstadt, Germany.
19 C. M. Thomas and J. C. Peters, Inorg. Chem., 2004, 8–10.
20 M. M. Hasan, M. B. Hursthouse, S. E. Kabir and K. M. A. Malik,
Polyhedron, 2001, 20, 97–101.
36 X-SHAPE, Crystal Optimization for Numerical Absorption Correction,
version 1.2, Stoe & Cie GmbH Darmstadt, Germany.
37 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467–473.
38 G. M. Sheldrick, Computer Program for the Refinement of Crystal
Structures, Go¨ttingen, Germany, 1997.
21 X. Zhao, I. P. Georgakaki, M. L. Miller, R. Mejia-Rodriguez, C.-Y.
Chiang and M. Y. Darensbourg, Inorg. Chem., 2002, 41, 3917–3928.
˚
22 P. Li, M. Wang, C. He, G. Li, X. Liu, C. Chen, B. Akermark and L.
39 D. Seyferth, R. S. Henderson and L. Song, Organometallics, 1982, 125–
133.
Sun, Eur. J. Inorg. Chem., 2005, 2506–2513.
4606 | Dalton Trans., 2006, 4599–4606
This journal is
The Royal Society of Chemistry 2006
©