Fischer 2015
Fischer 2015
Fischer 2015
This study provides new insights into light-driven hydrogen mixture. Application of other iron(0) carbonyl compounds as
generation using an iridium photosensitizer (IrPS) and simple WRC precursors also results in the exclusive formation of
iron carbonyls as water reduction catalysts (WRCs). Stopped- [HFe(CO)4]¢ . Kinetic experiments show that the stability of the
flow rapid-scan FTIR and operando continuous-flow FTIR spec- system is primarily influenced by the applied amount of WRC
troscopy as well as time-dependent density functional theory precursor, whereas the reaction rate is mainly determined by
(TD-DFT) has been applied to study the reaction. The conver- the concentration of the IrPS. At least two loss channels could
sion of the WRC precursor [Fe3(CO)12] into the radicals be identified: light-induced CO dissociation from the WRC and
[Fe3(CO)11]C¢ and [Fe2(CO)8]C¢ as well as [Fe(CO)5] in the absence decomposition of the IrPS at high IrPS/WRC ratios, accompa-
of light in a solvent mixture of tetrahydrofuran, triethylamine, nied by a ligand transfer from the iridium towards the iron
and water has been studied quantitatively. During light-in- center of the WRC. To reveal the nature of the catalytically
duced hydrogen production in the presence of the IrPS, the tri- active complex, binding energies and charge-transfer probabili-
meric [HFe3(CO)11]¢ and the monomeric [HFe(CO)4]¢ anion ties of all coordination geometries of various IrPS···WRC com-
could be identified as major WRC species. The equilibrium be- plexes have been calculated. These computations indicate an
tween both species can be shifted completely towards increased probability of charge transfer for dimeric and trimer-
[HFe(CO)4]¢ by increasing the water content of the solvent ic iron carbonyl species.
Introduction
Efficient conversion and storage of solar energy in a convenient tion system consists of a sacrificial reductant (SR) as the elec-
chemical form such as hydrogen is one of the most prominent tron donor, a photosensitizer (PS) as the light-absorbing and
challenges of modern “green energy” research. From this per- primary charge-separating unit, as well as a water reduction
spective, photocatalytic water splitting represents a very prom- catalyst (WRC) as an agent for reducing aqueous protons and
ising reaction to produce hydrogen with an estimated theoreti- assembling them into molecular hydrogen.[3] For industrial ap-
cal efficiency of more than 15 %, generating a product that can plications, such a system must be active and stable in the
be used in fuel cells without emission of greenhouse gases.[1] long-term and be based on low-cost ingredients.[4] To satisfy
The water-splitting process is usually divided into the oxidation the latter requirement, noble-metal-free complexes of Co, Ni,
and reduction half reactions. By adding sacrificial oxidants or and Fe have been suggested as active WRCs.[3, 5] However, a so-
reductants, these half-reactions can be investigated separate- phisticated and time-consuming synthesis of these complexes
ly.[2] Accordingly, a homogeneous photocatalytic water reduc- is often necessary.
Beller and co-workers have reported the first photocatalytic
water reduction system utilizing simple iron carbonyls as the
[a] S. Fischer, E. Barsch, Prof. Dr. R. Ludwig
WRC.[6] The iridium complex [Ir(bpy)(ppy)2]PF6 (IrPS; bpy = 2,2’-
Abteilung fìr Physikalische Chemie
Institut fìr Chemie bipyridine; ppy = 2-phenylpyridine) is used as a PS and triethyl-
Universitt Rostock amine (TEA) as the SR in a THF/TEA/H2O solvent mixture
Dr.-Lorenz-Weg 1, 18059 Rostock (Germany) (Scheme 1). If [Fe3(CO)12] is applied as the WRC, a turnover
Fax: (+ 49) 381-498-6524
number (TON) of 200 hydrogen molecules per Fe is achieved
E-mail: ralf.ludwig@uni-rostock.de
after 3 h of light irradiation. This system has been further
[b] Dr. O. S. Bokareva, Dr. S. I. Bokarev, Prof. Dr. O. Kìhn
Institut fìr Physik, Universitt Rostock modified by replacing the IrPS by a heteroleptic copper com-
Albert-Einstein-Str. 23-24, 18059 Rostock (Germany) plex with bathocuproin and xantphos ligands, resulting in
E-mail: oliver.kuehn@uni-rostock.de a photocatalytic water reduction system exclusively consisting
[c] E. Barsch, Prof. Dr. R. Ludwig of earth-abundant elements.[7]
Leibniz-Institut fìr Katalyse an der Universitt Rostock e.V.
The present mechanistic study combines experimental and
Albert-Einstein-Str. 29a, 18059 Rostock (Germany)
theoretical methods such as IR spectroscopy and DFT calcula-
Supporting information and ORCID(s) from the author(s) for this article
are available on the WWW under http://dx.doi.org/10.1002/ tions to focus on the homogeneous photocatalytic system
cctc.201500872. based on IrPS and iron carbonyl WRC and to continue previous
ChemCatChem 2016, 8, 404 – 411 404 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Papers
ChemCatChem 2016, 8, 404 – 411 www.chemcatchem.org 405 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Papers
[HFe(CO)4]¢ under water reduction conditions (Figure SI6 in the
Supporting Information). The time needed for full conversion
ranges between 1.5 h for [Fe3(CO)12] and [HNEt3][HFe3(CO)11]
and 12 min for [Fe(CO)5]. In the case of [Fe(CO)5], transforma-
tion even partly occurs in the dark, which conforms to the
water gas shift reaction.[14]
Note that after the preformation reactions, [HFe(CO)4]¢ is the
only observable catalyst intermediate during water reduction.
To exclude these preformation reactions, [NEt4][HFe(CO)4] was
directly applied as the WRC precursor in the following kinetic
experiments.
ChemCatChem 2016, 8, 404 – 411 www.chemcatchem.org 406 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Papers
this work, it can be concluded that this process is the rate-lim-
iting step of the whole system. However, for IrPS amounts
higher than 0.08 mmol L¢1, we observed deviations from
pseudo-first-order kinetics (Figure SI12 and Table SI3 in the
Supporting Information). The reason for this may be the solu-
tion’s increased light absorption at a concentration range in
which the Lambert–Beer law is not valid anymore.
At low IrPS concentrations, from 0.01 to 0.16 mmol L¢1 (Fig-
ure 6 a, grey curves), an increasing hydrogen yield is observed
and [HFe(CO)4]¢ is the only IR-active species observed through-
out the whole reaction (Figure 6 c). At these conditions, the de-
composition of [HFe(CO)4]¢ is considered to be partly induced
by light irradiation, which causes CO dissociation from the
metal center[15] and leads to IR-inactive decomposition prod-
ucts.
To study the WRC behavior independently from the IrPS,
[NEt4][HFe(CO)4] was dissolved in the THF/TEA/H2O mixture. In
the dark, this solution is stable for at least 20 h. Irradiation of
this solution initiates the formation of [HFe3(CO)11]¢ from the
monomeric anion (Figure 6 b). Note that with [NEt4][HFe(CO)4]
as the WRC precursor, the trimeric species is only observable in
the absence of IrPS. The presence of even small amounts of
IrPS results in a fragmentation of the trimers , which again
leads to [HFe(CO)4]¢ as the only observable complex (Fig-
ure 6 c). The overall iron concentration curve in the absence of
IrPS (Figure 6 b, dashed violet) still differs from the [HFe(CO)4]¢
concentration curve in the case of low IrPS concentration (Fig-
ure 6 c, blue). This implies an additional loss channel within the
catalytic cycle.
At high IrPS concentrations, from 0.16 to 1.25 mmol L¢1 (Fig-
ure 6 a, black curves), the decomposition of [HFe(CO)4]¢ is even
more accelerated by increasing amounts of IrPS. When
[HFe(CO)4]¢ disappears, the hydrogen evolution rate decreases
dramatically. For the highest IrPS concentration, this happens
after 2 h (Figure 6 d). In that time, [HFe(CO)4]¢ is mainly con-
verted into [Fe(bpy)(CO)3] (1972.5 (s), 1890 (s) cm¢1; Figure 7,
green spectrum) and the unidentified species y (1965 (m),
1883 (s) cm¢1; Figure 7, orange spectrum). Both show a consid-
erably lower activity. In a consecutive reaction, the iron com-
plex z is formed (2033.5 (w), 1989.5 (s), 1892(w) cm¢1; Figure 7,
cyan spectrum), which does not show any catalytic activity.
The accelerated decomposition of [HFe(CO)4]¢ at these high
IrPS concentrations leads to a declining overall hydrogen yield
Figure 6. Hydrogen evolution curves (a) and concentration (c Õ (number of
(Figure 6 a, black curves) and can be attributed to the destruc-
Fe per molecule)/absorption curves (b, c, d) of experiments with varying ini- tion of the IrPS and a transfer of the bipyridine ligand from the
tial concentrations of IrPS. Iron carbonyl species: [HFe(CO)4]¢ (blue), IrPS to the iron center of the WRC. This is supported by
[Fe(CO)3(bpy)] (green), y (orange), z (cyan), [HFe3(CO)11]¢ (red); a study by Bernhard and co-workers.[16] They observed a loss of
3·[HFe3(CO)11]¢ + [HFe(CO)4]¢ (dashed violet). Conditions: 0.0 (b), 1.6 (c), or
25.0 mmol (d) of IrPS, 8.0 mmol of [NEt4][HFe(CO)4], 20 mL of THF/TEA/H2O
the bpy ligands from the IrPS to form [Ir(ppy)2] + . Furthermore,
4:1:1, visible light (1.5 W), 25 8C. the formation and assignment of [Fe(bpy)(CO)3] is supported
by spectroscopic data[17] (Table SI5 in the Supporting Informa-
first-order behavior with respect to the IrPS (Figure SI11 in the tion) and DFT frequency calculations (Figure 7, calculated spec-
Supporting Information). In this range, the experimental results trum, dashed green).
are congruent with the findings of Neubauer et al.[8] They ex-
clusively investigated the IrPS subcycle by theoretical methods
The catalytically active complex
and photoluminescence spectroscopy at a IrPS concentration
of 0.013 mmol L¢1 and found the quenching of the excited IrPS [HFe(CO)4]¢ cannot be the catalytically active complex as its re-
by TEA to be the rate-limiting step. Considering the results of duction potential (¢1.98 V[18]) is too low for it to be reduced
ChemCatChem 2016, 8, 404 – 411 www.chemcatchem.org 407 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Papers
ure SI14 in the Supporting Information) and possibly remain
present in amounts below the detection limit.
This first electron transfer onto the WRC is a crucial reaction
step for the catalytically active species and was further investi-
gated by computational studies of IrPS···WRC complexes with
[HFe(CO)4]¢ , [HFe2(CO)8]¢ , and [HFe3(CO)11]¢ as the WRC. Bind-
ing energies and probabilities of charge transfer (CT) have
been calculated for all configurations of the IrPS···WRC com-
plexes with a Ir···WRC-center distance of 10 æ (Figure 8). The
binding maps of IrPS with the three iron carbonyl catalysts
have maxima or minima at the ligand positions. For the mono-
mer and dimer, the points indicating the catalyst position near
the ligand correspond to attractive regions with binding ener-
gies up to ¢0.08 eV for the monomer and ¢0.12 eV for the
dimer; the cavities between ligands provide unbound regions.
In contrast, for the trimer, the binding energies in the cavities
Figure 7. Extracted experimental (solid) and calculated (dashed) pure com- reach values up to ¢0.17 eV, whereas the other areas are repel-
ponent spectra and calculated structures of species observed during photo- ling as a result of the larger size of the iron complex. The prob-
catalytic water reduction when applying [NEt4][HFe(CO)4] as the WRC. Spe-
cies: [HFe(CO)4]¢ (blue, top structure), [Fe(bpy)(CO)3] (green, bottom struc-
ability of CT from the IrPS to the WRC has been evaluated on
ture), y (orange), z (cyan). the basis of TD-DFT calculations of the lowest doublet excited
states. The spin-density of the IrPS···WRC complex in the
ground doublet state is always located on the p1*(bpy) orbi-
by the IrPS (reduction potential ¢1.38 V[19]). Instead, it can be tals, pointing to no CT without irradiation.
considered to be the resting state of the catalytically active If we compare the probability of electron transfer from IrPS
complex. to the Fe catalyst, the WRCs can be organized by their increas-
Irradiating a [HFe(CO)4]¢ solution without IrPS results in the ing photoactivity: [HFe(CO)4]¢ (0 %) < [HFe2(CO)8]¢ (5.3 %) <
formation of dimers and trimers. Experiments show that these [HFe3(CO)11]¢ (18 %), where the percentage of points where the
species are highly reactive towards the reduced IrPS. As ob- first excited state is of long-range CT nature is given in paren-
served by IR spectroscopy, they are rapidly reduced and con- thesis. In addition, if we consider the trimer placed on
verted into [HFe(CO)4]¢ under water reduction conditions (Fig- a sphere of 11 æ (Figure SI16 in the Supporting Information),
Figure 8. Calculated binding energies and charge-transfer configurations (white hollow circles) of IrPS···WRC complexes (a, b, and c); the WRC is located on
a sphere around the IrPS at the distance R = 10 æ, as shown in (d) for the example of WRC = [HFe(CO)4]¢ . The projections of atoms of the bpy and ppy ligands
onto the sphere are also given. (e) and (f) Examples of transition density differences for IrPS-localized and intermolecular transitions are shown. Blue and red
colors indicate regions of electron loss and gain, respectively.
ChemCatChem 2016, 8, 404 – 411 www.chemcatchem.org 408 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Papers
the repelling areas near the ligand positions disappear and the
binding map is similar to those of the monomer and dimer at
a distance of R = 10 æ. However, increasing the distance be-
tween IrPS and [HFe3(CO)11]¢ leads to a decreasing number of
points where CT is the first excited state, from 18 to 5.3 %,
which is again similar to the dimer case. Note that despite the
absence of CT-favorable points in the case of the monomer,
the reaction could nevertheless take place because higher ex-
cited states can be involved or the reaction can be assisted by
direct solvent complexation as proposed in previous work.[6b]
For the monomer placed on a sphere of 7 æ, the percentage of
favorable combinations (15 %) is comparable with that of the
trimer. However, a situation in which the WRC comes very
close to IrPS, “pushing aside” the solvent molecules, can hardly
be achieved.
These results confirm that the active complex can only be
a dimeric or trimeric iron carbonyl species. DFT calculations
predict negative values for the changes in the Gibbs free
energy (DG) for all steps in a water reduction catalytic subcycle
(Figure 9) starting with [HFe3(CO)11]¢ . This shows that the tri-
ChemCatChem 2016, 8, 404 – 411 www.chemcatchem.org 409 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Papers
creasing initial IrPS concentration, the hydrogen yield passes stirrer at constant and reproducible speed and was continuously
through a maximum at 0.16 mmol L¢1 (TON[Fe] = 186) and de- circulated by a microannular gear pump (HNP) through the ther-
creases at very high concentrations of IrPS. The accelerated de- mostated measuring cell of the IR spectrometer. The cell was
equipped with CaF2 windows and has an optical path length of
activation of the catalytic system at these high IrPS concentra-
100 mm. The FTIR measurements were carried out with a Bruker
tions is due to the bpy ligand transfer from the IrPS onto
Tensor 27 spectrometer with a mercury–cadmium–telluride (MCT)
[HFe(CO)4]¢ , resulting in the much less active [Fe(bpy)(CO)3]. detector. Tracking of gas evolution was realized by an automatic
TD-DFT calculations show that charge transfer from the IrPS gas burette (MESSEN NORD). Its temperature was kept constant at
to the WRC is only possible for dimeric or trimeric iron species. 25 8C. A condenser was installed in between the reaction vessel
As calculated, [HFe3(CO)11]¢ is theoretically able to run a WRC and the gas burette and operated at 5 8C to retain volatile sol-
subcycle as the catalytically active complex, producing hydro- vent.[24]
gen or decomposing into the resting state [HFe(CO)4]¢ and A typical experiment was carried out as follows: the whole appara-
[Fe2(CO)8]C¢ after the first reduction step. The latter species can tus was evacuated and purged with argon five times to provide
also be reduced and converted subsequently to form oxygen-free conditions. The reaction vessel was filled with 20 mL
[HFe(CO)4]¢ as shown experimentally. Finally, the resulting rest- of solvent. After temperature adjustment, an IR spectrum was
ing state, [HFe(CO)4]¢ , can be considered as a WRC reservoir to taken and used as the background. The solid catalysts were added
continuously produce the catalytically active dimeric and tri- to the system. After reaching thermal equilibrium between the gas
and condensed phases (10 min), the reaction was started by irradi-
meric species upon light irradiation, which is observed in ex-
ating the solution (1500 mW). An IR spectrum with 20 scans was
periments without IrPS. taken every 30 s. The monitored hydrogen evolution curve was
Concluding, the photocatalytic system with simple iron car- corrected for the system’s blind value, which was obtained in an
bonyls as the WRC is very dynamic. For further improvement experiment without catalysts.
of the system, at least two directions should be explored:
The pure component spectra and the associated absorption and
design of light-resistant trimeric or dimeric WRC compounds concentration profiles were extracted with an algorithm based on
to increase the system’s stability (e.g., addition of CO-substitut- factor analysis.[25] In Figures 2, 4, and 6 b, the concentrations of the
ing phosphine ligands[6b]) and improvement of the per- complexes were multiplied by the number of their iron atoms to
formance of the IrPS to enhance the system’s activity. Further- facilitate identification of conversion processes. This is indicated
more, the ground, excited, and reduced states of the IrPS with the entry “c Õ (number of Fe per molecule)”. In Figure 6 c and
should be inert towards ligand dissociation (for example, see 6 d, the absorption curves indicate the maximum of each species’
reference [20]) to allow reaction conditions with an excess of absorption spectrum as a function of time.
IrPS.
Theoretical methods
Experimental Section Quantum chemical calculations have been performed with (TD)-
DFT by using the long-range corrected density functional approach
All experiments were carried out in an argon atmosphere and
(LC-BLYP) to correctly account for charge-transfer excitations.[26]
under exclusion of air. Solvents were purified and degassed by
This approach has previously been used to investigate the vertical
standard procedures prior to use. The catalyst precursors were pur-
excitation energies of the bare IrPS[27] and it was applied further to
chased from a commercial supplier (Sigma–Aldrich). [Ir(bpy)-
describe other steps of the photocatalytic cycle.[28] An important
(ppy)2]PF6,[21] [HNEt3][HFe3(CO)11],[22] [NEt4][HFe(CO)4],[13] and [NEt4]
aspect concerns the choice of the so-called range separation pa-
[(NO)Fe(CO)3][23] were synthesized according to literature proce-
rameter, w, which defines the switching between short- and long-
dures.
range parts in the exchange-correlation potential. Following the
The time-resolved infrared experiments for the dark phase were strategy put forward in Ref. [29], this parameter was tuned such as
carried out on a stopped-flow unit (TgK Scientific, UK) combined to guarantee the fulfilment of Koopmans’ theorem; the details can
with a VERTEX 80 (Bruker) with rapid-scan extension. The solutions be found in a separate publication.[30] The value w = 0.18 bohr¢1
and the IR cell were thermostated to 25 8C. The IR cell has CaF2 has been applied for all calculations.
windows. The optical path length was specified by the supplier as
Numerical calculations and frequency calculations were performed
100 mm. The spectra (2 scans per spectrum) were taken with a reso-
by using the Gaussian 09[31] program with the LANL2DZ ECP[32] and
lution of 2 cm¢1. At a mirror velocity of 320 kHz, we were able to
basis set for Ir and Fe and the 6-31G(d)[33] basis set for all other
record an IR spectrum every 116 ms.
atoms. For DFT/TD-DFT calculations, no symmetry restrictions were
The experimental setup (Scheme SI1) for time-resolved operando applied. The environment effects of THF solvent have been includ-
continuous-flow FTIR experiments under light irradiation consists ed through the Polarizible Continuum Model (IEFPCM).[34] Calculat-
of a reaction vessel connected to a FTIR-spectrometer and to an ed IR bands were plotted with a half width of 8 cm¢1 and harmonic
automatic gas burette. Hence, this setup provides the tracking of IR frequencies were corrected by a scaling factor f = 0.9795, which
gas evolution and the simultaneous acquisition of IR spectra and was chosen to give best agreement of calculated with experimen-
differs from the setup of the Beller group[6] only by the additional tal data.
IR unit.
As no stable complexes between IrPS and WRC are supposed to
A mercury vapor lamp (LUMATEC) was attached to the double- be formed in solution, the binding energies and energies of the 10
walled thermostatically controlled reaction vessel at a fixed posi- lowest excited states have been calculated for 152 configurations
tion by a fiber optic cable. This lamp emits visible light between for each WRC, similar to our previous study of interaction between
380 and 700 nm. The reaction solution was stirred by a magnetic IrPS and TEA.[8] The center of masses of the iron catalysts (Mc) were
ChemCatChem 2016, 8, 404 – 411 www.chemcatchem.org 410 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Papers
placed on spherical grids around the IrPS. The relative orientation [14] R. G. Pearson, H. Mauermann, J. Am. Chem. Soc. 1982, 104, 500 – 504.
of iron catalyst towards the Ir–Mc axis was kept fixed; no geometry [15] a) M. Poliakoff, J. J. Turner, J. Chem. Soc. Dalton Trans. 1974, 2276 – 2285;
relaxation was included. The distance between Ir and Mc was 10 æ b) R. L. Whetten, K. J. Fu, E. R. Grant, J. Chem. Phys. 1983, 79, 4899 –
4911; c) P. J. Krusic, D. J. Jones, D. C. Roe, Organometallics 1986, 5, 456 –
(Figure 9). For the monomeric WRC, this study was repeated for
460; d) L. BaÇares, T. Baumert, M. Bergt, B. Kiefer, G. Gerber, Chem. Phys.
a distance of 7 æ (Figure SI13 in the Supporting Information);
Lett. 1997, 267, 141 – 148; e) L. BaÇares, T. Baumert, M. Bergt, B. Kiefer,
owing to steric hindrance, it was not done for the dimer and G. Gerber, J. Chem. Phys. 1998, 108, 5799 – 5811; f) P. Portius, J. Yang, X.-
trimer forms. Z. Sun, D. C. Grills, P. Matousek, A. W. Parker, M. Towrie, M. W. George, J.
Am. Chem. Soc. 2004, 126, 10713 – 10720.
[16] L. L. Tinker, N. D. McDaniel, P. N. Curtin, C. K. Smith, M. J. Ireland, S. Bern-
Acknowledgements hard, Chem. Eur. J. 2007, 13, 8726 – 8732.
[17] a) M. DelaVarga, R. Costa, R. Reina, A. NfflÇez, M. Ýngel Maestro, J.
This work has been partly supported by the BMBF within the Maha, J. Organomet. Chem. 2003, 677, 101 – 117; b) F. Calderazzo, S. Fa-
laschi, F. Marchetti, G. Pampaloni, J. Organomet. Chem. 2002, 662, 137 –
project “Light2Hydrogen” (Spitzenforschung und Innovation in 143; c) H. K. van Dijk, D. J. Stufkens, A. Oskam, J. Organomet. Chem.
den Neuen Lndern), by the European Union (European Social 1988, 340, 227 – 238.
Funds, ESF) within the project “PS4H” and by the Ministry for [18] A. M. Bond, P. A. Dawson, B. M. Peake, B. H. Robinson, J. Simpson, Inorg.
Education, Science and Culture of Mecklenburg-Vorpommern. We Chem. 1977, 16, 2199 – 2206.
[19] J. I. Goldsmith, W. R. Hudson, M. S. Lowry, T. H. Anderson, S. Bernhard, J.
thank Matthias Beller, Henrik Junge, Felix Grtner, Michael Kar- Am. Chem. Soc. 2005, 127, 7502 – 7510.
nahl, Nils Rockstroh, and Petra Bartels for providing chemicals [20] a) L. L. Tinker, S. Bernhard, Inorg. Chem. 2009, 48, 10507 – 10511; b) D. N.
and expertise, Fabio Ragaini for helpful hints, Peter Kumm and Chirdon, W. J. Transue, H. N. Kagalwala, A. Kaur, A. B. Maurer, T. Pintauer,
Martin Riedel for technical support, as well as Bastian Hoffmann S. Bernhard, Inorg. Chem. 2014, 53, 1487 – 1499.
[21] E. D. Cline, S. E. Adamson, S. Bernhard, Inorg. Chem. 2008, 47, 10378 –
for supporting the stopped-flow experiments. 10388.
[22] W. McFarlane, G. Wilkinson, W. Hìbel, in Inorganic Syntheses (Ed.: H. F.
Holtzlaw), Wiley, 2007, pp. 181 – 183.
Keywords: homogeneous catalysis · photocatalysis · proton
[23] B. Plietker, A. Dieskau, Eur. J. Org. Chem. 2009, 775 – 787.
reduction · time-resolved IR spectroscopy · water reduction [24] a) B. Loges, H. Junge, B. Spilker, C. Fischer, M. Beller, Chem. Ing. Tech.
2007, 79, 741 – 753; b) A. Boddien, B. Loges, H. Junge, M. Beller, Chem-
[1] a) S. Chu, H. Michel, H. Dotan, H. Mayes, S. Sheehan, Nature 2013, 502, SusChem 2008, 1, 751 – 758.
S60 – S61; b) F. E. Osterloh, J. Phys. Chem. Lett. 2014, 5, 2510 – 2511. [25] K. Neymeyr, M. Sawall, D. Hess, J. Chemom. 2010, 24, 67 – 74.
[2] R. Eisenberg, Science 2009, 324, 44 – 45. [26] a) M. Chiba, T. Tsuneda, K. Hirao, J. Chem. Phys. 2006, 124, 144106; b) H.
[3] Z. Han, R. Eisenberg, Acc. Chem. Res. 2014, 47, 2537 – 2544. Iikura, T. Tsuneda, T. Yanai, K. Hirao, J. Chem. Phys. 2001, 115, 3540 –
[4] J. Marshall, Nature 2014, 510, 22 – 24. 3544; c) Y. Tawada, T. Tsuneda, S. Yanagisawa, T. Yanai, K. Hirao, J. Chem.
[5] P. Du, R. Eisenberg, Energy Environ. Sci. 2012, 5, 6012 – 6021. Phys. 2004, 120, 8425 – 8433.
[6] a) F. Grtner, B. Sundararaju, A.-E. Surkus, A. Boddien, B. Loges, H. [27] S. I. Bokarev, O. S. Bokareva, O. Kìhn, J. Chem. Phys. 2012, 136, 214305.
Junge, P. H. Dixneuf, M. Beller, Angew. Chem. Int. Ed. 2009, 48, 9962 – [28] a) S. I. Bokarev, D. Hollmann, A. Pazidis, A. Neubauer, J. Radnik, O. Kuhn,
9965; Angew. Chem. 2009, 121, 10147 – 10150; b) F. Grtner, A. Boddien, S. Lochbrunner, H. Junge, M. Beller, A. Bruckner, Phys. Chem. Chem.
E. Barsch, K. Fumino, S. Losse, H. Junge, D. Hollmann, A. Brìckner, R. Phys. 2014, 16, 4789 – 4796; b) S. I. Bokarev, O. S. Bokareva, O. Kìhn,
Ludwig, M. Beller, Chem. Eur. J. 2011, 17, 6425 – 6436. Coord. Chem. Rev. 2015, 304-305, 133 – 145.
[7] a) S.-P. Luo, E. Meja, A. Friedrich, A. Pazidis, H. Junge, A.-E. Surkus, R. [29] T. Stein, L. Kronik, R. Baer, J. Am. Chem. Soc. 2009, 131, 2818 – 2820.
Jackstell, S. Denurra, S. Gladiali, S. Lochbrunner, M. Beller, Angew. Chem. [30] O. S. Bokareva, G. Grell, S. I. Bokarev, O. Kìhn, J. Chem. Theory Comput.
Int. Ed. 2013, 52, 419 – 423; Angew. Chem. 2013, 125, 437 – 441; b) E. 2015, 11, 1700 – 1709.
Meja, S.-P. Luo, M. Karnahl, A. Friedrich, S. Tschierlei, A.-E. Surkus, H. [31] Gaussian 09, revision D. 01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Junge, S. Gladiali, S. Lochbrunner, M. Beller, Chem. Eur. J. 2013, 19, Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
15972 – 15978; c) S. Fischer, D. Hollmann, S. Tschierlei, M. Karnahl, N. nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
Rockstroh, E. Barsch, P. Schwarzbach, S.-P. Luo, H. Junge, M. Beller, S. A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
Lochbrunner, R. Ludwig, A. Brìckner, ACS Catal. 2014, 4, 1845 – 1849. K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
[8] A. Neubauer, G. Grell, A. Friedrich, S. I. Bokarev, P. Schwarzbach, F. Grt- Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro,
ner, A.-E. Surkus, H. Junge, M. Beller, O. Kìhn, S. Lochbrunner, J. Phys. M. J. Bearpark, J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, R.
Chem. Lett. 2014, 5, 1355 – 1360. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S.
[9] D. Hollmann, F. Grtner, R. Ludwig, E. Barsch, H. Junge, M. Blug, S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox,
Hoch, M. Beller, A. Brìckner, Angew. Chem. Int. Ed. 2011, 50, 10246 – J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat-
10250; Angew. Chem. 2011, 123, 10429 – 10433. mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
[10] a) Y. A. Belousov, T. A. Kolosova, V. N. Babin, Polyhedron 1989, 8, 603 – Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dan-
611; b) Y. A. Belousov, T. A. Belousova, Polyhedron 1999, 18, 2605 – 2608; nenberg, S. Dapprich, A. D. Daniels, ©. Farkas, J. B. Foresman, J. V. Ortiz,
c) Y. A. Belousov, E. F. Brin, Polyhedron 2001, 20, 2765 – 2769; d) A. B. J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford, CT, USA, 2013.
Yurii, Russ. Chem. Rev. 2007, 76, 41 – 58; e) P. J. Krusic, J. San Filippo, B. [32] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270 – 283.
Hutchinson, R. L. Hance, L. M. Daniels, J. Am. Chem. Soc. 1981, 103, [33] a) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972, 56, 2257 –
2129 – 2131; f) P. J. Krusic, J. Am. Chem. Soc. 1981, 103, 2131 – 2133; g) F. 2261; b) P. C. Hariharan, J. A. Pople, Theor. Chim. Acta 1973, 28, 213 –
Ragaini, D. L. Ramage, S. Song Jeong, G. L. Geoffroy, A. L. Rheingold, J. 222.
Am. Chem. Soc. 1993, 115, 12183 – 12184; h) F. Ragaini, Organometallics [34] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999 – 3094.
1996, 15, 3572 – 3578.
[11] K. Farmery, M. Kilner, R. Greatrex, N. N. Greenwood, J. Chem. Soc. A
1969, 2339 – 2345.
[12] M. Y. Darensbourg, D. J. Darensbourg, H. L. C. Barros, Inorg. Chem. 1978,
17, 297 – 301. Received: August 5, 2015
[13] I. Y. Guzman-Jimenez, J. W. van Hal, K. H. Whitmire, Organometallics Revised: September 11, 2015
2003, 22, 1914 – 1922. Published online on November 3, 2015
ChemCatChem 2016, 8, 404 – 411 www.chemcatchem.org 411 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim