Light-driven hydrogen generation: Efficient iron-based water reduction catalysts

Article abstract of DOI:10.1002/anie.200905115

(figure represented) Let your light shine: A novel state-of-theart system for light-driven proton reduction with iron catalysts was developed (see scheme; SR = sacrificial reagent). It consists of simple iron carbonyl complexes such as [Fe₃(CO)

Full text of DOI:10.1002/anie.200905115

Communications
DOI: 10.1002/anie.200905115
Hydrogen Generation
Light-Driven Hydrogen Generation: Efficient Iron-Based Water
Reduction Catalysts**
Felix Gꢀrtner, Basker Sundararaju, Annette-Enrica Surkus, Albert Boddien, Bjꢁrn Loges,
Henrik Junge, Pierre H. Dixneuf, and Matthias Beller*
Hydrogen is considered to be an attractive energy source,
since it can be used in a clean and highly efficient manner in
fuel cells. A current drawback for the wider use of hydrogen is
the dependence of hydrogen production on fossil resources by
reforming processes. Obviously, a more benign objective is
the conversion of the almost unlimitedly available energy of
sunlight into non-fossil-based energy carriers such as hydro-
gen using water as the hydrogen source.
redox cycles in photosystems II and I. In such a cascade an
iridium PS, for example, is excited by light irradiation and
subsequently reductively quenched by the SR to give the
reduced species Ir PSÀ. This compound immediately transfers
an electron onto the Fe WRC, which then reduces aqueous
protons to hydrogen. Thus both the WRC and the PS are
required for a catalytic cycle (Scheme 1).
Since the discovery of the Honda Fujishima effect in
1972,[1] photocatalytic water splitting into hydrogen and
oxygen has inspired researchers all over the world to develop
novel and economical photocatalysts for this reaction.[2] In the
past, mainly heterogeneous materials were developed for the
photocatalytic water splitting into H2 and O2, but only a few
homogeneous systems have been reported for this reaction to
date.[3] The lack of activity in homogenous systems is related
to the complexity of the multielectron processes during water
reduction (two electrons per molecule H2) and water oxida-
tion (four electrons per molecule O2). To improve the overall
process, the two half reactions water oxidation and water
reduction can be studied separately employing sacrificial
reagents (SR) that donate (water reduction)[4] or accept
(water oxidation) electrons in the catalytic system. For water
oxidation cerium(IV) reagents are often used, and remark-
able results were reported in the last few years.[5] In water
reduction systems amines such as triethylamine (TEA) or
triethanol amine (TEOA) are typically applied as electron
donors.
Scheme 1. Principle of the light-driven water reduction cascade with
iridium photosensitizer (Ir PS), sacrificial reagent (SR), and an iron
water reduction catalyst (WRC).
Beside the original ruthenium(II) bipyridine (bpy) photo-
sensitizers,[6] notable advancements for PS based on iridium[7]
and platinum[8] complexes were reported by the groups of
Bernhard and Eisenberg, respectively. Moreover, organic
dyes such as eosin[9] have been applied in water reduction
catalysis.
With respect to the water reduction catalysts (WRCs), to
date most work has focused on noble metals such as rhodium,
palladium, or platinum. Hence, in a state-of-the-art water
reduction from Bernard and co-workers employing an Ir PS
and a Rh WRC, an impressive turnover number (TON) of
5000 was reported.[7c] Notably, cobalt-based WRCs with
different oxime ligands resulted in high TONs up to 2100 in
a Pt PS/Co WRC system.[8a]
In the late 1970s, the first homogeneous multicomponent
systems for water reduction employing photosensitizers (PS)
and noble-metal water reduction catalysts (WRC) were
established.[6] The general principle of these water reduction
cascades was adapted from nature, where reduction equiv-
alents are generated by light in combination with coupled
To mimic nature more closely, supramolecular devices
have been assembled containing both the PS and the WRC
part in one molecule, connected by a linker.[10] However,
these so-called dyads are still less active than the multi-
component systems. Furthermore, nature can be a guideline
for further catalyst development: iron- and nickel-based
hydrogenases catalyze the reduction of protons with up to
9000 molecules of H2 per second and site.[11] Considering this
fact, surprisingly few examples employing Fe-based WRCs in
photocatalytic water reduction are known.[12] Sun, ꢀkermark,
and co-workers showed that light-driven water reduction is, in
principle, possible with iron complexes, although the reported
TON(Fe) of 4.3 for a Ru PS/Fe system is low.[12a] Parallel to
our work, a similar system has been further optimized.[12c]
[*] F. Gꢀrtner, Dr. A.-E. Surkus, A. Boddien, B. Loges, Dr. H. Junge,
Prof. Dr. M. Beller
Leibniz-Institut fꢁr Katalyse e.V. an der Universitꢀt Rostock
Albert-Einstein Strasse 29a, 18059 Rostock (Germany)
Fax: (+49)381-1281-5000
E-mail: Matthias.Beller@catalysis.de
B. Sundararaju, Prof. Dr. P. H. Dixneuf
Laboratoire Catalyse et Organomꢂtalliques
Institut Sciences Chimiques de Rennes
6226 CNRS-Universitꢂ de Rennes (France)
[**] This work has been supported by the State of Mecklenburg-
Vorpommern, the BMBF, and the DFG (Leibniz prize). F.G. thanks
the Fonds der Chemischen Industrie (FCI) for a Kekulꢂ grant.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200905115.
9962
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 9962 –9965

Angewandte
Chemie
Recently, an elegant supramolecular device from zinc por-
phyrins and iron hydrogenase model complexes was used for
proton reduction.[12b] But again, only low catalyst activity was
observed (TON(Fe) ꢀ 4).
important to note that the reaction can be driven by only
visible light (Table 1, entries 8 and 9). The use of 385 or
420 nm UV cut-off filters[14] has two main effects. On the one
hand, the reaction rate decreases. On the other hand, the
lifetime of the catalytic system is prolonged, resulting in
approximately the same Fe TON of 200 after 15 h (Table 1,
entries 7 and 8). The extension of the stability of the system
might be due to less CO dissociation from iron carbonyl,
which is known to take place easily in the presence of UV
light.
Herein, we report an efficient system for homogeneous
reduction of aqueous protons to hydrogen consisting of
[Ir(bpy)(ppy)2]PF6 (ppy: phenyl pyridine) as the PS; iron(0)
carbonyl complexes as simple, cheap, readily available, and
abundant WRCs; and triethylamine as SR. Clearly, the
development of iron-based catalysts as a substitute for
noble metals is a major topic in catalysis.[13] Our initial
attempts for more benign WRCs focused on the use of various
Co0, CoI, CoII, FeI, FeII, and FeIII precursors. Typically, we
allowed the WRC precursor and [Ir(bpy)(ppy)2]PF6 to react
for 3 h in 10 mL of a THF/TEA/H2O (8:2:2) solution under
xenon light irradiation.[14a] All tests resulted only in low
hydrogen production, and no stable catalytic systems were
found (Table 1, entries 1–3). However, in the presence of
To investigate the scope of the new system, a fixed amount
of [Fe3(CO)12] (10.0 mmol Fe) and varied amounts of Ir PS
were used in a set of experiments (Figure 1). With 0.06 equiv-
Table 1: Investigations of various metal precursors for light-driven
hydrogen production from water with [Ir(bpy)(ppy)2]PF6 PS.
Entry[b]
Metal
precursor
VmLÀ1
3 h (15 h)
TON[a] WRC
3 h (15 h)
TON[a] Ir PS
3 h (15 h)
1[c]
2[c]
3[c]
4
Fe powder
[{CpFe(CO)2}2]
[Co2(CO)8]
[Fe(CO)5]
<1
3
3
26
–
13
12
114
–
33
33
282
5
6
7
[Fe2(CO)9]
[(cot)Fe(CO)3]
[Fe3(CO)12]
[Fe3(CO)12]
[Fe3(CO)12]
[Fe3(CO)12]
none
32
30
141
347
132
325
30 (47[d])
17 (45)
11 (32)
<1
132 (207[d])
325 (510[d])
Figure 1. Ir PS TON (blue squares) and Fe WRC TON (orange
triangles) with a fixed amount of [Fe3(CO)12] and varied Ir PS content.
Conditions: 10.0 mmol [Fe], 0.06–2.2 equiv [Ir], 10 mL THF/H2O/TEA
(8:2:2), 300 W Xe lamp, no filter, 3 h, 258C.
8[e]
9[f]
10[g]
11[h]
12[i]
75 (198)
48 (141)
–
–
–
184 (488)
119 (347)
–
–
–
<1
<2
[Fe3(CO)12]
[a] TON=n(H)/n(WRC or PS). [b] Reaction conditions: 18.5 mmol Fe,
7.5 mmol [Ir(bpy)(ppy)2]PF6, 10 mL THF/H2O/TEA (8/2/2), 300 W Xe
lamp,[14a] 258C. [c] No more activity after 3 h. [d] No more activity after
6 h. [e] 385 nm cut-off filter was used. [f] 420 nm cut-off filter was used.
[g] No Ir PS added. [h] No Fe WRC added. [i] No TEA added.
alents Ir PS, the TON for Ir is 1863 after 3 h, and up to 3035
turnovers were observed after 6 h, which are the highest
TONs observed. Visible-light irradiation (420 nm cut-off
filter) resulted in a TON of 2500 (Ir PS) after 20 h. In an
additional set of experiments with fixed amounts of PS
(7.5 mmol) and varied amounts of [Fe3(CO)12] (Figure S2,
Supporting Information), a maximum TON of 322 for the Fe
WRC without filter and 400 with a 420 nm cut-off filter was
obtained with 0.62 equivalents of Fe. These TONs are the best
ever reported for an iron WRC system in light-driven water
reduction and can compete with the known cobaloxime
systems.[16]
To ensure that the source of liberated hydrogen is water,
experiments with D2O instead of water were carried out. By
applying a thermal conductivity detector for gas analysis in
GC, we can easily distinguish H2 and D2 by the opposite
algebraic signs of the integral when helium is used as carrier
gas.[17] Indeed, it was confirmed that water is the only source
of hydrogen in the reaction. Furthermore, the necessity of
light was demonstrated in an experiment in which any light
irradiation on the reaction vessel was excluded in constant
time intervals. In the dark gas evolution is interrupted, and it
resumes immediately after light is introduced again (Fig-
ure S3, Supporting Information).
simple iron(0) carbonyl complexes, the light-driven reduction
of aqueous protons took place without addition of any ligand
(Table 1, entries 4–9). Control experiments (Table 1,
entries 10–12) without the addition of either SR, PS, or
WRC clearly demonstrate that all three components are
essential in the catalytic system. No hydrogen evolution or
only traces were observed when either the sacrificial reagent
TEA, the Ir PS, or iron carbonyl was not present in the
mixture.
As shown in Table 1, approximately the same activity is
obtained for mononuclear [Fe(CO)5] as for the oligomeric
iron carbonyl complexes [Fe2(CO)9], [Fe3(CO)12], and [(cot)-
Fe(CO)3] (cot: cyclooctatetraene; Table 1, entries 4–7). The
evolved gas was collected in an automatic burette and
analyzed by GC analysis.[14b] Besides traces of THF, TEA,
and CO, only hydrogen was detected.
Next, we investigated the influence of UV irradiation on
hydrogen production in the presence of [Fe3(CO)12]. It is
Angew. Chem. Int. Ed. 2009, 48, 9962 –9965
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9963

Communications
The mechanism for the photocatalytic proton reduction
by iron carbonyl WRCs might be similar to the one proposed
Experimental Section
All catalytic experiments were carried out under an argon atmos-
phere with exclusion of air. THF, TEA, and doubly distilled water
were degassed with standard laboratory methods prior to use. The
catalyst precursors were purchased from commercial suppliers
(Aldrich). [Ir(bpy)(ppy)2]PF6 was synthesized according to a liter-
ature procedure.[7c] The amount of gas liberated was measured by an
automatic gas burette. Details on the equipment, which is also used
for hydrogenation reactions, have been published elsewhere.[20] In
addition, GC for analyzing gases was used (gas chromatograph HP
6890N, carboxen 1000, TCD, external calibration). The light source
was a 300 W Xe lamp.[14]
I
À
by Sun, ꢀkermark, and co-workers for thiolate-bridged Fe
FeI dimers.[12a] The electrochemical properties of [Fe(CO)5]
and [Fe3(CO)12] in THF in the presence and absence of water
were investigated previously. It was shown that [Fe3(CO)12]
degrades to mononuclear [Fe(CO)5] and other iron carbonyl
compounds in THF solution.[18] Monomeric [Fe(CO)5] shows
one reduction peak at À1.8 V versus Ag/AgCl, which is
consistent with our own measurements (for electrochemical
details, see the Supporting Information). In the same study it
was demonstrated that [HFe(CO)4]À could be formed by
electrochemical reduction of [Fe(CO)5] in a THF/water
mixture.[18]
Typical procedure for light-driven water reduction: A double-
walled thermostatically controlled reaction vessel was evacuated and
purged with argon five times to remove any other gas. The iridium
sensitizer (7.5 mmol) and [Fe3(CO)12] (18.5 mmol) were added as
powders in a Teflon crucible or as a solution from a freshly prepared
stock solution. A solution of THF/H2O/TEA (8:2:2, 10 mL) was
added, and room temperature was maintained. The reaction was
started after stirring at ambient temperature for 8 min by switching on
the light source. The variability of gas evolution is typically 1–10%.
This scheme can be adapted to the newly developed
catalytic system. In the catalytic cycle, electrons for the
reduction of the iron WRC are provided by the reduced
Ir PSÀ. This species is generated by photoexcitation of Ir PS
and subsequent reduction with TEA as SR (Scheme 1,
Scheme 2). Thus triethylamine is the final electron source. It
Received: September 12, 2009
Published online: November 20, 2009
Keywords: energy conversion · homogeneous catalysis · iron ·
.
photochemistry · water splitting
[1] A. Fujishima, K. Honda, Nature 1972, 238, 37 – 38.
[2] For recent reviews on heterogenous semiconductors catalysts,
see: a) A. Kudo, Y. Miseki, Chem. Soc. Rev. 2009, 38, 253 – 278;
b) F. E. Osterloh, Chem. Mater. 2008, 20, 35 – 54.
[3] a) H. Kunkely, A. Vogler, Angew. Chem. 2009, 121, 1713 – 1715;
Angew. Chem. Int. Ed. 2009, 48, 1685 – 1687; b) S. W. Kohl, L.
Weiner, L. Schwartsburd, L. Konstantinovski, L. J. W. Shimon,
Y. Ben-David, M. A. Iron, D. Milstein, Science 2009, 324, 74 – 77.
[4] Reviews on homogenous water reduction cascades: a) M. Wang,
Y. Na, M. Gorlov, L. Sun, Dalton Trans. 2009, 6458 – 6467;
b) L. L. Tinker, N. D. McDaniel, S. Bernhard, J. Mater. Chem.
2009, 19, 3328 – 3337.
Scheme 2. Proposed mechanism for the reduction of aqueous protons
by iron carbonyls and PS.[12a,19]
[5] X. Sala, I. Romero, M. Rodrꢁguez, L. Escriche, A. Llobet,
Angew. Chem. 2009, 121, 2882 – 2893; Angew. Chem. Int. Ed.
2009, 48, 2842 – 2852.
has been used before in combination with iridium sensitiz-
ers.[7] The oxidative degradation pathways of tertiary amines
have been studied previously, and different radical and
[6] a) M. Kirch, J.-M. Lehn, J.-P. Sauvage, Helv. Chim. Acta 1979, 62,
1345 – 1384; b) A. Moradpour, P. Keller, H. Kagan, Nouv. J.
Chim. 1978, 2, 547 – 549; c) K. Kalyanasundaram, J. Kiwi, M.
Grꢂtzel, Helv. Chim. Acta 1978, 61, 2720 – 2730; d) C. V.
Krishnan, N. Sutin, J. Am. Chem. Soc. 1981, 103, 2141 – 2142.
[7] a) J. I. Goldsmith, W. R. Hudson, M. S. Lowry, T. H. Anderson,
S. Bernhard, J. Am. Chem. Soc. 2005, 127, 7502 – 7510; b) L. L.
Tinker, N. D. McDaniel, P. N. Curtin, C. K. Smith, M. J. Ireland,
S. Bernhard, Chem. Eur. J. 2007, 13, 8726 – 8732; c) E. D. Cline,
S. E. Adamson, S. Bernhard, Inorg. Chem. 2008, 47, 10378 –
10388; d) L. L. Tinker, S. Bernhard, Inorg. Chem. 2009, 48,
10507 – 10511; e) P. N. Curtin, L. L. Tinker, C. M. Burgess, E. D.
Cline, S. Bernhard, Inorg. Chem. 2009, 48, 10498 – 10506.
[8] a) P. Du, J. Schneider, G. Luo, W. W. Brennessel, R. Eisenberg,
Inorg. Chem. 2009, 48, 4952 – 4962; b) P. Du, K. Knowles, R.
Eisenberg, J. Am. Chem. Soc. 2008, 130, 12576 – 12577; c) J.
Zhang, P. Du, J. Schneider, P. Jarosz, R. Eisenberg, J. Am. Chem.
Soc. 2007, 129, 7726 – 7727.
+
cationic species, including the radical N(Et)3 C, were proposed
as the primary reaction product.[4a,15] From the resulting
hydrido iron carbonyl species [HFe(CO)4]À hydrogen is
evolved by protonolysis (Scheme 2), which is also postulated
for biomimetic hydrogenase models in electrochemical
hydrogen production from water.[12,19]
In conclusion, we have developed a novel, convenient,
bimetallic system for water reduction cascades consisting of
simple, cheap, and readily available iron(0) carbonyls; TEA
as a sacrificial reagent; and an iridium photosensitizer. The
activity for the iron water reduction catalyst is the highest for
any iron water reduction system known to date and can
compete with the known cobaloxime water reduction cata-
lysts.
[9] T. Lazarides, T. McCormick, P. Du, G. Luo, B. Lindley, R.
Eisenberg, J. Am. Chem. Soc. 2009, 131, 9192 – 9194.
[10] a) S. Rau, B. Schꢂfer, D. Gleich, E. Anders, M. Rudolph, M.
Friedrich, H. Gꢃrls, W. Henry, J. G. Vos, Angew. Chem. 2006,
9964 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 9962 –9965

Angewandte
Chemie
118, 6361 – 6364; Angew. Chem. Int. Ed. 2006, 45, 6215 – 6218;
Angew. Chem. 2006, 118, 6361 – 6364; b) M. Elvington, J. Brown,
S. M. Arachchige, K. J. Brewer, J. Am. Chem. Soc. 2007, 129,
10644 – 10645; c) H. Ozawa, M. Haga, K. Sakai, J. Am. Chem.
Soc. 2006, 128, 4926 – 4927.
[14] a) For properties of the light source (Lot Oriel 300 W Xe-lamp,
LSB530) and UV cut-off filters (ITOS, GG420, or GG385), see
www.lot-oriel.com; b) for a typical gas evolution curve, see the
Supporting Information, Figure S1.
[15] a) S. D. Ross, Tetrahedron Lett. 1973, 14, 1237 – 1240; b) P. J.
DeLaive, T. K. Foreman, C. Giannotti, D. G. Whitten, J. Am.
Chem. Soc. 1980, 102, 5627 – 5631; c) S. G. Cohen, A. Parola,
G. H. Parsons, Chem. Rev. 1973, 73, 141 – 161.
[16] For a review on Fe-based hydrogen generation catalysts, see R.
Lomoth, S. Ott, Dalton Trans. 2009, 9952 – 9959.
[11] M. Frey, ChemBioChem 2002, 3, 153 – 160.
[12] a) Y. Na, M. Wang, J. Pan, P. Zhang, B. ꢀkermark, L. Sun, Inorg.
Chem. 2008, 47, 2805 – 2810; b) A. M. Kluwer, R. Kapre, F.
Hartl, M. Lutz, A. L. Spek, A. M. Brouwer, P. W. N. M. van
Leeuwen, J. N. H. Reek, Proc. Natl. Acad. Sci. 2009, 106, 10460 –
10465; c) see Ref. [64] in R. Lomoth, S. Ott, Dalton Trans. 2009,
9952 – 9959.
[17] Thermal conductivity/E-4 W/mK (Fa. Messer, Switzerland):
H2 (1861) > He(1500) > D2 (1310).
[18] N. El Murr, A. Chaloyard, Inorg. Chem. 1982, 21, 2206 – 2208.
[19] R. Mejia-Rodriguez, D. Chong, J. H. Reibenspies, M. P. Soriaga,
M. Y. Darensbourg, J. Am. Chem. Soc. 2004, 126, 12004 – 12014.
[20] 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, ChemSusChem 2008, 1, 751 – 758.
[13] a) S. Enthaler, K. Junge, M. Beller in Iron Catalysis in Organic
Chemistry (Eds.: B. Plietker), Wiley-VCH, Weinheim, 2008,
pp. 125 – 142; b) S. Enthaler, K. Junge, M. Beller, Angew. Chem.
2008, 120, 3363 – 3367; Angew. Chem. Int. Ed. 2008, 47, 3317 –
3321.
Angew. Chem. Int. Ed. 2009, 48, 9962 –9965
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9965