A noble-metal-free system for photocatalytic hydrogen production from water

Article abstract of DOI:10.1002/chem.201302091

A series of heteroleptic copper(I) complexes with bidentate PP and NN chelate ligands was prepared and successfully applied as photosensitizers in the light-driven production of hydrogen, by using [Fe₃(CO)₁₂] as a water-reduction catalyst (WRC). These systems efficiently reduces protons from water/THF/triethylamine mixtures, in which the amine serves as a sacrificial electron donor (SR). Turnover numbers (for H) up to 1330 were obtained with these fully noble-metal-free systems. The new complexes were electrochemically and photophysically characterized. They exhibited a correlation between the lifetimes of the MLCT excited state and their efficiency as photosensitizers in proton-reduction systems. Within these experiments, considerably long excited-state lifetimes of up to 54μs were observed. Quenching studies with the SR, in the presence and absence of the WRC, showed that intramolecular deactivation was more efficient in the former case, thus suggesting the predominance of an oxidative quenching pathway.

Full text of DOI:10.1002/chem.201302091

DOI: 10.1002/chem.201302091
A Noble-Metal-Free System for Photocatalytic Hydrogen Production from
Water
Esteban Mejꢀa,[a] Shu-Ping Luo,[b] Michael Karnahl,[a] Aleksej Friedrich,[d]
Stefanie Tschierlei,[d] Annette-Enrica Surkus,[a] Henrik Junge,[a] Serafino Gladiali,[c]
Stefan Lochbrunner,[d] and Matthias Beller*[a]
Abstract: A series of heteroleptic cop-
over numbers (for H) up to 1330 were
obtained with these fully noble-metal-
free systems. The new complexes were
electrochemically and photophysically
characterized. They exhibited a correla-
tion between the lifetimes of the
MLCT excited state and their efficiency
as photosensitizers in proton-reduction
systems. Within these experiments, con-
siderably long excited-state lifetimes of
up to 54 ms were observed. Quenching
studies with the SR, in the presence
and absence of the WRC, showed that
intramolecular deactivation was more
efficient in the former case, thus sug-
gesting the predominance of an oxida-
tive quenching pathway.
_
per(I) complexes with bidentate PP
_
and NN chelate ligands was prepared
and successfully applied as photosensi-
tizers in the light-driven production of
hydrogen, by using [Fe3(CO)12] as
a
water-reduction catalyst (WRC).
These systems efficiently reduces pro-
tons from water/THF/triethylamine
mixtures, in which the amine serves as
a sacrificial electron donor (SR). Turn-
Keywords: copper · homogeneous
catalysis · hydrogen · photochemis-
try · sustainable chemistry
Introduction
during the daytime. To circumvent these drawbacks, efficient
ways of storing this energy in suitable secondary energy car-
riers are desirable.[2]
In the search for sustainable and abundant energy sources
as alternatives to the current fossil-fuel-based paradigm,
sunlight appears to be the most appealing candidate because
it bathes the Earth with an amount of energy that is about
10000 times larger than the amount that is required by
human technologies.[1] Although photovoltaic devices direct-
ly convert solar energy into electricity, the latter cannot be
efficiently stored as such and the systems can only operate
Starting in the 1930s, when it was suggested that hydrogen
derived from water splitting could be used as fuel for trans-
portation, the idea of changing the energy scheme of oil-de-
pendence into a “hydrogen economy”[3] is coming to maturi-
ty. Although the infrastructure for a hydrogen-based energy
system is still under development,[4] it is foreseeable that, for
hydrogen to succeed as an energy commodity, it should be
produced in a sustainable, economical, and environmentally
friendly manner, for which artificial photosynthesis is a feasi-
ble way to go.[5] The concept of artificial photosynthesis en-
circles those processes, biomimetic or not, in which solar
energy is harvested and stored in the form of chemical
bonds of highly energetic molecules, known as solar fuels.[6]
Currently, most homogeneous proton-reduction reactions
are carried out by using three-component systems that con-
sist of a light-harvesting chromophore or photosensitizer
(PS), a water-reduction catalyst (WRC), and an electron
donor or sacrificial reductant (SR).[7] Arguably, the photo-
sensitizer can be considered as the bottleneck of the whole
system. Since the beginning of this research field in the
1970s,[8] ruthenium complexes have played a key role,[9] al-
though a plethora of compounds based on other noble
metals, including iridium,[10] platinum,[11] and rhenium,[12]
have also been successfully employed. In contrast, efficient
noble-metal-free photosensitizers are still rare. Biomimetic
iron-,[13] zinc-,[14] and magnesium-based[15] systems have been
thoroughly investigated, albeit the reported activities are
still low.
[a] Dr. E. Mejꢀa,+ Dr. M. Karnahl, Dr. A.-E. Surkus, Dr. H. Junge,
Prof. Dr. M. Beller
Leibniz-Institut fꢁr Katalyse an der Universitꢂt Rostock e.V.
Albert-Einstein-Straße 29a, 18059 Rostock (Germany)
E-mail: matthias.beller@catalysis.de
Homepage: http://www.catalysis.de
[b] Dr. S.-P. Luo+
State Key Laboratory Breeding Base of
Green Chemistry-Synthesis Technology
Zhejiang University of Technology
310014 Hangzhou (P.R. China)
[c] Prof. Dr. S. Gladiali
Universitꢃ di Sassari, Dipartimento di Chimica e Farmacia
Via Vienna 2, 07100 Sassari (Italy)
[d] A. Friedrich, Dr. S. Tschierlei, Prof. Dr. S. Lochbrunner
Institute of Physics, University of Rostock
Universitꢂtsplatz 3, 18055 Rostock (Germany)
[+] These authors contributed equally to this work.
Supporting information for this article, including experimental infor-
mation on the synthesis and characterization of the copper com-
plexes, general conditions and detailed information on the photophys-
ical, electrochemical, and X-ray measurements and the photocatalytic
reaction conditions, is available on the WWW under http://dx.doi.org/
10.1002/chem.201302091.
15972
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15972 – 15978

FULL PAPER
The potential application of copper(I) complexes with
polypyridine ligands as photosensitizers has been long over-
looked, although their photo- and electrochemical proper-
ties are widely known, having found successful use in organ-
ic light-emitting diodes (OLEDs), light-emitting electro-
chemical cells (LECs), sensors, dye-sensitized solar cells
(DSSCs), etc.[16] In an early communication, the group of
Sauvage reported the use of homoleptic copper(I) com-
plexes with bipyridyl ligands for the photochemical reduc-
tion of water by using platinum (supported on titanium di-
oxide) as a WRC, with, in the best case, an overall turnover
number of 80 moles of H2 per mole of copper.[17]
Scheme 2. General synthetic procedure of copper complexes 1–12.
complexes were obtained as yellow-to-orange solids that
were stable to ambient conditions (air and moisture).
X-ray diffraction analysis of complexes 4 and 6 (Figure 1)
revealed a distorted tetrahedral geometry around the metal
center, as previously observed for this kind of com-
In a recent communication, we reported the successful
use of a series of heteroleptic copper(I) complexes with the
_
_
general formula [CuACTHNUTRGENN(GU NN)CAHTUNGTRNEN(UGN PP)]PF6 as photosensitizers in
photocatalytic proton-reduction systems with TONH up to
804 by using [Fe3(CO)12] as the water-reduction catalyst.[18]
Herein, we report an improved system in which turnover
numbers (TONH =nH/nCu, or the number of equivalents of
H+ that are reduced by one equivalent of copper) of up to
1330 were obtained. Selected complexes were characterized
by using photophysical and electrochemical methods and
some insights into the reaction mechanism were gained.
This system is, to the best of our knowledge, the most effi-
cient noble-metal-free homogeneous system for the photo-
catalytic reduction of protons reported to date.
Results and Discussion
Synthesis and structural characterization of the copper pho-
tosensitizers: All of the copper(I) complexes (Scheme 1)
were synthesized from [CuACHTNUGTRENUNG(MeCN)4]PF6 by using a straight-
forward one-pot tandem ligand-substitution method, accord-
ing to a previously reported procedure (Scheme 2).[18] The
Figure 1. Solid-state molecular structures (ORTEPs) of complexes 4·
ACHUTNERGN(UNG CH2Cl2)3.8 (A) and 6·ACHTUTGNERN(NUGN CH2Cl2)2.3 (B), as determined by X-ray crystallog-
raphy. Displacement ellipsoids are set at 50% probability. Hydrogen
atoms and solvent molecules are omitted for clarity.[19]
pounds.[16b] This distortion mainly arises from the considera-
_
ble differences in the bite angles of the bidentate PP and
_
NN ligands (for complex 4: N1-Cu-N2 80.04(7)8, P1-Cu-P2
118.70(2)8) and also from interACHTNUTRGENNUGliACHTUNGTERNNUGgand interactions. This dis-
tortion is clearly reflected in the difference between the
_
angles that are formed by the plane of the NN ligand with
each of the phosphorus atoms: In the isobutyl derivative (4),
these angles are 91.978 and À130.028, whereas, in the n-
hexyl analogue (6), these angles are 78.668 and À60.848. In
the former case (4), this huge difference renders one side of
Scheme 1. General structure of the copper(I) photosensitizers in this
work.
Chem. Eur. J. 2013, 19, 15972 – 15978
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15973

M. Beller et al.
the molecule more “open”, which means that the copper
atom is more exposed to a possible nucleophilic attack. This
result is an example of the so-called “rocking distortion” of
tetrahedral towards trigonal-pyramidal coordination geome-
tries,[16e] in which the P1 atom occupies the apical position
and the P2 atom occupies an equatorial position. This effect
is less pronounced in the latter case (6) and could (among
other factors) account for the differences in their photocata-
lytic performance, as discussed below.
Whereas the phenanthroline moiety remains structurally
unaltered upon coordination, the diphosphine scaffold suf-
fers considerable distortion. In free 4,5-bis(diphenylphosphi-
no)-9,9-dimethylxanthene (xantphos), the angle between the
planes that contain the two aromatic rings (those that are
connected by the oxygen atom) is only 18.28 (for the X-ray
structure of the ligand, see the Supporting Information, Fig-
ure SI1). Conversely, in complex 4, this interplanar angle is
36.98, whereas, in complex 6, it is 30.38. This result again re-
Figure 2. Normalized absorption (solid lines) and emission spectra (pho-
toluminescence intensity (PLI), dashed lines) for different copper(I) pho-
tosensitizers. Measurements were made in THF at room temperature.
flects the higher steric demand of the isobutyl substituents
on the NN ligand in comparison with the n-hexyl chains.
be concluded that the ground-state orbital energies are com-
parable in all the cases. Absorption and emission spectra of
the different copper photosensitizers were also measured in
MeCN (see the Supporting Information, Figure SI8), which
showed, in general, only a small solvent effect.[20]
_
In spite of these structural distortions, which could also
arise from crystal-packing effects, NMR analysis of all of the
complexes (1H, 13C, and 31P NMR) shows relatively simple
patterns, which is more in agreement with C2v symmetry.
These observations suggest a fluxional behavior in solution
(at room temperature), which makes, for instance, both
phosphorus atoms equivalent on the NMR timescale.
An interesting feature in some of the absorption spectra
is the growth of a small band in the proximity of 480 nm,
which probably arises from the formation of the homoleptic
_
complexes,[21] [Cu
ACHTNUTRGNEUGN(NN)2]PF6, owing to the ligand-redistribu-
tion equilibrium in solution, as observed previously for this
Photophysical studies: Electronic absorption and photolumi-
nescence spectra were measured for a representative selec-
tion of copper photosensitizers, namely complexes 1, 2, 3, 7,
8, and 9. The corresponding data are summarized in Table 1.
As observed for other CuI complexes of the general formula
kind of system.[16e] It is important to note that this band is
more prominent for complex 9, which has the most sterically
_
demanding NN ligand, thus enabling more-effective ligand
dissociation and redistribution.
Strong room-temperature photoluminescence is exhibited
by all of the complexes, which is caused by the deactivation
of the MLCT excited states (Figure 2 and Table 1). A com-
parison of the emission maxima (lmax) reveals the same sys-
_
_
[CuACHTUNGTRENNUNG(NN)ACHTUNGTRENNUNG
(PP)]+, the absorption bands that are attributed to
low-lying MLCT (metal-to-ligand-charge-transfer) transi-
tions occur at higher energies than those of the homoleptic
_
[Cu
N
tematic trend upon changing the alkyl substituents at the
_
390 nm) showed no significant shifts on changing either the
diphosphine scaffold or the substituents at the 2,9-positions
of the phenanthroline ligand (Figure 2), from which it can
2,9-positions of the NN ligands in both series of complexes
1–3 and 7–9. Increasing the bulkiness at the 2,9-positions of
_
the NN scaffold results in a blue-shift of the emission bands
(D1–3=23 nm and D7–9=27 nm), which means higher-
energy emissive states and a smaller energy difference be-
tween the absorption and emission maxima. This kind of
hypsochromic ligand effect, as previously observed in homo-
leptic[20,22] and heteroleptic[16a,e] cuprous phenanthroline de-
rivatives, reflects the effectiveness with which the Jahn–
Teller (flattening) distortion of the initially populated
Franck-Condon excited state is hindered. Because the
energy of the MLCT state depends on the dihedral angle
between the ligands and the potential-energy surface pres-
ents a local maximum at 908, conservation of the tetrahedral
geometry results in a larger energy gap between the ground
and excited states and, hence, higher emission energies.[22]
The steric effects of the ligands also have a profound in-
fluence on the excited-state dynamics of the corresponding
complexes, as reflected in the lifetime (t0) of the emissive
MLCT state and in the quantum yield (Fabs), thus exhibiting
Table 1. Electronic absorption and luminescence data for different [Cu-
_
_
A
ACHTUNGTRENNUNG(PP)]PF6 complexes in degassed THF at room temperature.
Absorption
Emission
t0
[ms][b]
[a]
Entry Complex lmax
e
lmax
Fabs
tcat.
[ns][c]
[nm]
[mÀ1 cmÀ1
]
[nm]
1
2
3
4
5
6
1
2
3
7
8
9
389
389
387
386
389
386
5701.9
5838.5
4848.9
5432.7
4810.2
4215.6
569
565
546
545
535
518
0.08
0.27
0.75
0.05
0.34
0.52
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
[a] The quantum yield was measured at an optical density of about 0.1
(error: about 10%). [b] The lifetimes were measured in dilute solution
with an optical density of about 0.1, 0.02 mm (388 nm). [c] The lifetime
was measured under comparable catalytic conditions: 0.35 mm copper PS,
0.5 mm [Fe3(CO)12] (reduction catalyst), in a mixture of THF/triethyla-
mine/water (4:1:1).
15974
www.chemeurj.org
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15972 – 15978

Noble-Metal-Free Photocatalytic Hydrogen Production from Water
FULL PAPER
unprecedentedly high values of up to 54.1 ms and 74.5%, re-
spectively (for complex 3; Table 1, entry 3). These high
values are more astonishing considering the fact that the
measurements were performed in THF, which, as a Lewis
base, can typically act as an additional ligand.[23]
These results reflect the effective shielding of the metal
core that is imparted by the bulky ligands. For both proper-
ties, t0 and Fabs, the same trend is observed within series of
complexes 1–3 and 7–9 (that is: 1<2<3 and 7<8<9) and,
not surprisingly, it is derived from the ability with which the
flattening distortion of the cupric excited state is inhibited
(perfectly correlated with the bulkiness of the substituents:
Me<nBu<sBu). Under catalytic conditions (with a higher
concentration of PS and in the presence of TEA and water),
much shorter lifetimes (tcat, Table 1) were observed (see
below).
Figure 3. Cyclic voltammograms of complexes 1 (red), 3 (green), and 7
(blue) in MeCN solution. All potentials are referenced to the Fc/Fc+
couple. Inset: Oxidative differential pulse voltammograms (DPVs). Scan
, supporting electrolyte: 0.1m [Bu4N]PF6, Ag/AgCl/
LiClsat. reference electrode, glassy carbon working electrode, platinum
Upon excitation, the formally 3d9 metal center tends to
adopt a pseudo-square-planar geometry with two open coor-
dination positions in which one or two solvent molecules
can coordinate to form an exciplex intermediate.[23] This spe-
cies has a smaller energy gap to the ground state and, conse-
quently, the non-radiative decay pathway is favored; thus,
short lifetimes of the MLCT state are observed.[24] Effective
inhibition of this process, as in our case, leads to long-lived
excited states (with high-energy levels), along with high
quantum yields. The experimentally measured lifetimes (t0)
and quantum yields (Fabs, Table 1) were used to calculate
the radiative (kr) and non-radiative rate constants (knr) of
the MLCT excited states (see the Supporting Information,
Table SI1).[25] The obtained kr values are typical of hetero-
leptic copper complexes with phenanthroline and diphos-
phine ligands.[21] Conversely, the obtained knr values are up
to three orders of magnitude smaller than those for homo-
leptic phenanthroline complexes,[26] which indicates that the
non-radiative decay is much slower and accounts for the rel-
ative long lifetimes of the excited states, thus fulfilling one
of the prerequisites for a complex to be a successful photo-
sensitizer for solar-energy-conversion systems.
rate: 100 mVsÀ1
counter electrode, temperature: 258C.
Table 2. Electrochemical data of selected complexes in MeCN.[a]
Entry
Complex
Eox [V]
Ered [V]
DEoxÀred [V]
1
2
3
1
3
7
+0.93
À2.05
À2.12
À2.03
2.98
+1.02
3.14
+0.90 (+1.08)[b]
2.93 (3.11)[b]
[a] The measurements were carried out in dry MeCN under an argon at-
mosphere with 0.1m [Bu4N]PF6 as a supporting electrolyte. Peak poten-
tials are versus Fc/Fc+ (for potentials versus NHE and Ag/AgCl see the
Supporting Information, Table SI2)
phine ligand is oxidized, presumably at the sulfur atom. Be-
sides this peak, the substitution pattern in complex 3 causes
an oxidation potential that is 90 mV higher than that of
complex 1.
Because differences in the electronic inductive effects be-
tween the alkyl groups cannot be invoked, the shift in the
oxidation potentials should be attributed to the larger steric
effect of the sec-butyl group (in 3) compared to the methyl
group (in 1), thus more-effectively preventing the structure-
flattening that is associated with the oxidation of CuI to CuII
and, hence, stabilizing the cuprous state.[16b,21,27] Conversely,
the effect of the 2,9-substitution pattern cannot be observed
in the absorption spectra (Figure 2), because the spectra are
not influenced by structural changes that occur after the
MLCT process.[21]
Electrochemical studies: For complexes 1, 3, and 7, cyclic
and differential pulse voltammograms were recorded in
MeCN solution (Figure 3 and Table 2). Upon reduction, all
of the complexes display comparable behavior, with a rever-
sible single-electron-reduction couple. This peak is associat-
ed with a reduction of the phenanthroline ligand and, thus,
_
_
CÀ
with the formation of [CuII
known from previous studies of related [Cu
A
N
It is important to mention that, as previously reported for
_
A
[CuACTHUNGETRNNUG
(NN)2]+ complexes,[21] the oxidation process is not totally
plexes.[21] Comparing the reduction potentials of complexes
1 (À2.05 V) and 3 (À2.12 V versus Fc/Fc+), a shift of about
70 mV is observed, thus indicating an influence of the substi-
tution pattern at the 2- and 9-positions of the phenanthro-
line scaffold. In contrast, the influence of different phos-
phine ligands is negligible, as shown by complexes 1 and 7,
which only display a small difference of 30 mV between
their metal-based oxidation potentials (0.93 and 0.90 V
versus Fc/Fc+, respectively). Moreover, for complex 7,
a second oxidation peak can be observed, in which the phos-
reversible (Figure 3), because DEoxÀred also incorporates
energy changes that are associated with structural modifica-
tions (Jahn–Teller distortion and coordination number). For
the same reason, a direct correlation between the emission
energies and the reduction potentials (expected for d6 metal
complexes) is not observed.[28]
Nevertheless, the reduction and oxidation potentials of
the excited states can be calculated from their corrected
emission energies; that is, instead of the emission maxima,
the values that are obtained from the intersection of the
Chem. Eur. J. 2013, 19, 15972 – 15978
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15975

M. Beller et al.
normalized absorption and emission spectra of a photosensi-
tizer are used for the calculations.[25]
For example, upon light absorption, complex 1 displays
Eox and Ered potentials (vs. NHE) of +1.28 and À1.10 V re-
spectively, which indicates that it is able to undergo electron
transfer with both the WRC (oxidative quenching) and the
SR (reductive quenching). For details, see the Supporting
Information, Figure SI10.
Photocatalytic hydrogen production: In a recent communi-
cation, we reported the first effective application of hetero-
_
_
leptic [CuACHTUNGTRENNUNG(NN)ACHTUNGTRENNUNG(PP)]PF6 complexes as photosensitizers in
the photocatalytic water-reduction reaction by using the
readily available complex [Fe3(CO)12] as a water-reduction
catalyst and triethylamine as a sacrificial reductant.[18] At
Figure 4. Typical hydrogen-evolution curves for the photocatalytic reduc-
tion of water by using a copper(I) PS in the presence of [Fe3(CO)12] as
a WRC and TEA as a SR. For the reaction conditions, see Table 3.
that time, we realized that complexes that contained either
_
xantphos or thixantphos PP ligands (Scheme 1), along with
_
the bathocuproine NN moiety, displayed the highest activi-
contain xantphos derivatives (7–12). This effect can be at-
tributed to a faster electron-transfer rate from the excited
PS to the WRC (see below). On the other hand, the system-
atic change in the steric features of the phenanthroline li-
gands affects the thermodynamic properties of the excited
complex (see above), thereby hampering the non-radiative
decay by increasing the energy gap between the Cu d orbi-
tals and the ligand p* orbitals in the emissive states.[26]
In both series of complexes, that is, 1–6 and 7–12, the ob-
tained TONHs (Table 3) were found to be dependent on the
bulkiness of the alkyl residues at the 2,9-positions. This de-
pendence correlates nicely with the same effect that was ob-
served on the lifetimes of the excited complexes (t0,
Table 1). Thus, it appears clear that long-lived MLCT states
are beneficial for the activity of the catalysts. However, al-
though a rough correlation is observed (Figure 5), a direct
dependence could not be concluded.
ties, with TONHs of up to 886 and 874 (nH/nCu) for com-
plexes 1 and 7, respectively, under optimized conditions.
With that precedent in hand, the synthesis of several new
tailored copper complexes was undertaken, by systematical-
ly changing the steric bulkiness of the substituents proximal
to the nitrogen atoms (at the 2- and 9-positions) of the phe-
nanthroline ligand (complexes 1–12, Scheme 1). Once in
hand, these complexes were tested in the water-reduction
system. Most of them showed excellent activities with
TONHs of up to 1330 at room temperature with [Fe3(CO)12]
as a WRC and triethylamine (TEA) as a SR (Table 3). Typi-
cal hydrogen-evolution curves for selected photosensitizers
are shown in Figure 4.
A comparison of the gas-evolution curves in Figure 4
clearly reveals the effects of both the diphosphine and phe-
nanthroline ligands. On the one hand, changing the diphos-
phine scaffold influences the kinetics of the catalytic reac-
tion, which is more rapid with the complexes that contain
thixantphos derivatives (1–6) than with complexes which
To estimate the contribution of the WRC to the quench-
ing processes, we also measured the lifetime of complex
1 under catalytic conditions in THF with TEA and water
(4:1:1) but without the WRC. In this case, the obtained life-
time is 730 ns, which is much shorter than in the dilute THF
solution (t0, Table 1). This difference results from the in-
Table 3. Results of photocatalytic water-reduction reactions by using
copper(I) catalysts as the PS in the presence of [Fe3(CO)12] as the WRC
and TEA as the SR.[a]
[b]
Entry
Complex
T [h]
Volume H2 [mL]
TONH
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
9
24
40
60
45
24
65
8
6
6
10
6
6
37
41
57
9.6
54
17
37
46
54
44
49
43
862
954
1330
224
1270
400
873
1085
1251
1029
1147
1005
9
10
11
12
10
11
12
[a] Reaction conditions: copper PS (3.5 mmol), [Fe3(CO)12] (5 mmol),
10 mL THF/TEA/water (4:3:1), 258C, Xe-light irradiation (output:
1.5 W), without a light filter. Gas evolution was quantitatively measured
by using automatic gas burettes; gas analysis was performed by using
GC. The values are the averages of at least two experiments. The results
differ between 1 and 17%. [b] TONH =nH/nCu.
Figure 5. Plot of the turnover numbers (TONHs) for the photocatalytic re-
duction of water versus the lifetimes (in THF) of the corresponding pho-
tosensitizers. Reaction conditions are given in Table 3.
15976
www.chemeurj.org
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15972 – 15978

Noble-Metal-Free Photocatalytic Hydrogen Production from Water
FULL PAPER
creased concentration of the photosensitizer, which enhan-
ces the number of molecular collisions and the correspond-
ing quenching processes. In addition, exciplex quenching
may occur, owing to the presence of water and TEA.[23]
However, quenching by the WRC in the catalytic system
(180 ns) is four-times more effective than all of the other
decay channels put together (tcat., Table 1). Because good
photocatalytic performance can only be achieved if effective
electron transfer takes place, we suggest that electron-trans-
fer processes contribute to the quenching. In this case, it is
plausible to assume that oxidative quenching owing to the
WRC is the dominant electron-transfer pathway. With this
observation in mind and taking into account our previous
mechanistic investigations on an analogous system that used
Ir-based photosensitizers,[10b,29] we propose the reaction
mechanism as shown in Scheme 3.
the complexes was assessed. This new molecular system rep-
resents, to the best of our knowledge, the most efficient
noble-metal-free method for the homogeneous photocatalyt-
ic production of H2 from water.
Acknowledgements
Thanks to Dr. A. Spannenberg and Dr. H.-J. Drexler for performing the
crystallographic measurements and to Dr. N. Rockstroh for proofreading
of the manuscript. Financial support by the BMBF through the program
“Spitzenforschung und Innovation in den neuen Lꢂndern” is gratefully
acknowledged.
[1] D. Gust, T. A. Moore, A. L. Moore, Faraday Discuss. 2012, 155, 9–
26.
[2] N. Armaroli, V. Balzani, Angew. Chem. 2007, 119, 52–67; Angew.
Chem. Int. Ed. 2007, 46, 52–66.
[3] J. A. Turner, Science 2004, 305, 972–974.
[4] R. F. Service, Science 2004, 305, 958–961.
[5] A. Magnuson, M. Anderlund, O. Johansson, P. Lindblad, R.
Lomoth, T. Polivka, S. Ott, K. Stensjç, S. Styring, V. Sundstrçm, L.
Hammarstrçm, Acc. Chem. Res. 2009, 42, 1899–1909.
[6] a) J. J. Concepcion, R. L. House, J. M. Papanikolas, T. J. Meyer,
Proc. Natl. Acad. Sci. USA 2012, 109, 15560–15564; b) S. Styring,
Faraday Discuss. 2012, 155, 357–376; c) V. Balzani, A. Credi, M.
Venturi, ChemSusChem 2008, 1, 26–58; d) T. R. Cook, D. K. Dogu-
tan, S. Y. Reece, Y. Surendranath, T. S. Teets, D. G. Nocera, Chem.
Rev. 2010, 110, 6474–6502.
[7] a) A. J. Esswein, D. G. Nocera, Chem. Rev. 2007, 107, 4022–4047;
b) P. Du, R. Eisenberg, Energy Environ. Sci. 2012, 5, 6012–6021;
c) W. T. Eckenhoff, R. Eisenberg, Dalton Trans. 2012, 41, 13004–
13021.
[8] a) J. M. Lehn, J. P. Sauvage, Nouv. J. Chim. 1977, 1, 449–451; b) K.
Kalyanasundaram, J. Kiwi, M. Grꢂtzel, Helv. Chim. Acta 1978, 61,
2720–2730; c) J. Kiwi, M. Grꢂtzel, J. Am. Chem. Soc. 1978, 100,
6314–6320; d) A. Moradpour, E. Amouyal, P. Keller, H. Kagan,
Nouv. J. Chim. 1978, 2, 547–549; e) M. Kirch, J. M. Lehn, J. P. Sauv-
age, Helv. Chim. Acta 1979, 62, 1345–1384.
[9] a) A. Abbotto, N. Manfredi, Dalton Trans. 2011, 40, 12421–12438;
b) G. C. Vougioukalakis, A. I. Philippopoulos, T. Stergiopoulos, P.
Falaras, Coord. Chem. Rev. 2011, 255, 2602–2621.
Scheme 3. Proposed reaction mechanism for the photocatalytic produc-
_
_
(NN)]+) as the photo-
tion of hydrogen by using complexes 1–12 ([CuACTHNUTRGENNGU(PP)ACHTUTGNRENNUGN
sensitizer (PS), [Fe3(CO)12] as a water-reduction catalyst (WRC), and
triethylamine (TEA) as a sacrificial reductant (SR).
[10] a) B. F. DiSalle, S. Bernhard, J. Am. Chem. Soc. 2011, 133, 11819–
11821; b) F. Gꢂrtner, A. Boddien, E. Barsch, K. Fumino, S. Losse,
H. Junge, D. Hollmann, A. Brꢁckner, R. Ludwig, M. Beller, Chem.
Eur. J. 2011, 17, 6425–6436; c) F. Gꢂrtner, D. Cozzula, S. Losse, A.
Boddien, G. Anilkumar, H. Junge, T. Schulz, N. Marquet, A. Span-
nenberg, S. Gladiali, M. Beller, Chem. Eur. J. 2011, 17, 6998–7006;
d) F. Gꢂrtner, S. Denurra, S. Losse, A. Neubauer, A. Boddien, A.
Gopinathan, A. Spannenberg, H. Junge, S. Lochbrunner, M. Blug, S.
Hoch, J. Busse, S. Gladiali, M. Beller, Chem. Eur. J. 2012, 18, 3220–
3225.
Conclusion
In summary, we have reported a noble-metal-free system
that is capable of reducing protons from water by using light
as the primary energy source, thereby resulting in the effi-
cient generation of molecular hydrogen at room tempera-
ture. This three-component system includes a copper-based
photosensitizer, an iron-based water-reduction catalyst, and
triethylamine as a sacrificial reductant. The new heteroleptic
copper complexes introduced herein were fully character-
ized by using spectroscopic, X-ray, photophysical, and elec-
trochemical techniques. Notably, they displayed strong lumi-
nescence with extremely long lifetimes of the excited states
and high stability towards atmospheric conditions. Thanks to
their modularity and ease of synthesis, the effect of the
steric features of the ligands was investigated and the corre-
lation with the optical properties and catalytic activities of
[11] a) P. W. Du, K. Knowles, R. Eisenberg, J. Am. Chem. Soc. 2008, 130,
12576–12577; b) R. Okazaki, S. Masaoka, K. Sakai, Dalton Trans.
2009, 6127–6133; c) W. G. Wang, F. Wang, H. Y. Wang, C. H. Tung,
L. Z. Wu, Dalton Trans. 2012, 41, 2420–2426.
[12] a) W. N. Jiang, J. H. Liu, C. Li, Inorg. Chem. Commun. 2012, 16, 81–
85; b) B. Probst, C. Kolano, P. Hamm, R. Alberto, Inorg. Chem.
2009, 48, 1836–1843; c) H. Takeda, K. Koike, T. Morimoto, H. Inu-
maru, O. Ishitani, Adv. Inorg. Chem. 2011, 63, 137–186; d) H. Y.
Wang, W. G. Wang, G. Si, F. Wang, C. H. Tung, L. Z. Wu, Langmuir
2010, 26, 9766–9771; e) P. Kurz, B. Probst, B. Spingler, R. Alberto,
Eur. J. Inorg. Chem. 2006, 2966–2974; f) B. Probst, M. Guttentag,
A. Rodenberg, P. Hamm, R. Alberto, Inorg. Chem. 2011, 50, 3404–
3412; g) B. Probst, A. Rodenberg, M. Guttentag, P. Hamm, R. Al-
berto, Inorg. Chem. 2010, 49, 6453–6460.
Chem. Eur. J. 2013, 19, 15972 – 15978
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15977

M. Beller et al.
[13] a) D. Streich, Y. Astuti, M. Orlandi, L. Schwartz, R. Lomoth, L.
Hammarstrçm, S. Ott, Chem. Eur. J. 2010, 16, 60–63; b) S. Tschier-
lei, S. Ott, R. Lomoth, Energy Environ. Sci. 2011, 4, 2340–2352.
[14] a) P. Poddutoori, D. T. Co, A. P. S. Samuel, C. H. Kim, M. T. Vagnini,
M. R. Wasielewski, Energy Environ. Sci. 2011, 4, 2441–2450; b) H.
Yamaguchi, T. Onji, H. Ohara, N. Ikeda, A. Harada, Bull. Chem.
Soc. Jpn. 2009, 82, 1341–1346; c) Y. Zorlu, F. Dumoulin, M.
Durmus, V. Ahsen, Tetrahedron 2010, 66, 3248–3258.
Angew. Chem. 2013, 125, 437–441; Angew. Chem. Int. Ed. 2013, 52,
419–423.
[19] CCDC-909701 (4) and 909702 (6) contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[20] M. K. Eggleston, D. R. McMillin, K. S. Koenig, A. J. Pallenberg,
Inorg. Chem. 1997, 36, 172–176.
[15] a) Y. Amao, K. Aoki, J. Biobased Mater. Bioenergy 2008, 2, 51–56;
b) Y. Amao, T. Hirakawa, Int. J. Hydrogen Energy 2010, 35, 6624–
6628; c) Y. Amao, T. Hirakawa, N. Himeshima, Catal. Commun.
2008, 9, 131–134; d) Y. Amao, Y. Maki, Y. Fuchino, J. Phys. Chem.
C 2009, 113, 16811–16815; e) Y. Tomonou, Y. Amao, Biometals
2003, 16, 419–424.
[16] a) N. Armaroli, G. Accorsi, M. Holler, O. Moudam, J. F. Nierengart-
en, Z. Zhou, R. T. Wegh, R. Welter, Adv. Mater. 2006, 18, 1313–
1316; b) D. G. Cuttell, S. M. Kuang, P. E. Fanwick, D. R. McMillin,
R. A. Walton, J. Am. Chem. Soc. 2002, 124, 6–7; c) S. B. Harkins,
J. C. Peters, J. Am. Chem. Soc. 2005, 127, 2030–2031; d) C. S. Smith,
C. W. Branham, B. J. Marquardt, K. R. Mann, J. Am. Chem. Soc.
2010, 132, 14079–14085; e) S.-M. Kuang, D. G. Cuttell, D. R. McMil-
lin, P. E. Fanwick, R. A. Walton, Inorg. Chem. 2002, 41, 3313–3322;
f) N. Alonso-Vante, J.-F. Nierengarten, J.-P. Sauvage, J. Chem. Soc.
Dalton Trans. 1994, 1649–1654; g) T. Bessho, E. C. Constable, M.
Graetzel, A. Hernandez Redondo, C. E. Housecroft, W. Kylberg,
M. K. Nazeeruddin, M. Neuburger, S. Schaffner, Chem. Commun.
2008, 3717–3719.
[21] C. T. Cunningham, K. L. H. Cunningham, J. F. Michalec, D. R.
McMillin, Inorg. Chem. 1999, 38, 4388–4392.
[22] D. R. McMillin, J. R. Kirchhoff, K. V. Goodwin, Coord. Chem. Rev.
1985, 64, 83–92.
[23] N. A. Gothard, M. W. Mara, J. Huang, J. M. Szarko, B. Rolczynski,
J. V. Lockard, L. X. Chen, J. Phys. Chem. A 2012, 116, 1984–1992.
[24] J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd ed.,
Springer, New York, 2010.
[25] D. V. Scaltrito, D. W. Thompson, J. A. OꢅCallaghan, G. J. Meyer,
Coord. Chem. Rev. 2000, 208, 243–266.
[26] E. C. Riesgo, Y.-Z. Hu, F. Bouvier, R. P. Thummel, D. V. Scaltrito,
G. J. Meyer, Inorg. Chem. 2001, 40, 3413–3422.
[27] M. T. Miller, P. K. Gantzel, T. B. Karpishin, Inorg. Chem. 1999, 38,
3414–3422.
[28] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. von
Zelewsky, Coord. Chem. Rev. 1988, 84, 85–277.
[29] D. Hollmann, F. Gꢂrtner, R. Ludwig, E. Barsch, H. Junge, M. Blug,
S. Hoch, M. Beller, A. Brꢁckner, Angew. Chem. 2011, 123, 10429–
10433; Angew. Chem. Int. Ed. 2011, 50, 10246–10250.
[17] A. Edel, P. A. Marnot, J. P. Sauvage, Nouv. J. Chim. 1984, 8, 495–
498.
[18] S.-P. Luo, E. Mejꢀa, A. Friedrich, A. Pazidis, H. Junge, A.-E. Surkus,
R. Jackstell, S. Denurra, S. Gladiali, S. Lochbrunner, M. Beller,
Received: May 31, 2013
Published online: October 10, 2013
15978
www.chemeurj.org
ꢄ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15972 – 15978