Water oxidation with molecularly defined iridium complexes: Insights into homogeneous versus heterogeneous catalysis

Article abstract of DOI:10.1002/chem.201201472

Molecularly defined Ir complexes and different samples of supported IrO₂ nanoparticles have been tested and compared in the catalytic water oxidation with cerium ammonium nitrate (CAN) as the oxidant. By comparing the activity of nano-scaled supported IrO₂ particles to the one of organometallic complexes it is shown that the overall activity of the homogeneous Ir precursors is defined by both the formation of the homogeneous active species and its conversion to Ir^IV-oxo nanoparticles. In the first phase of the reaction the activity is dominated by the homogeneous active species. With increasing reaction time, the influence of nano-sized Ir-oxo particles becomes more evident. Notably, the different conversion rates of the homogeneous precursor into the active species as well as the conversion into Ir-oxo nanoparticles and the different particle sizes have a significant influence on the overall activity. In addition to the homogeneous systems, IrO₂@MCM-41 has also been synthesized, which contains stabilized nanoparticles of between 1 and 3 nm in size. This latter system shows a similar activity to IrCl₃·xH₂O and complexes 4 and 5. Mechanistic insights were obtained by in situ X-ray absorption spectroscopy and scanning transmission electron microscopy. Ir walks a fine line: Molecularly defined Ir complexes and IrO₂ nanoparticles have been applied in the water oxidation reaction with cerium ammonium nitrate (CAN) as oxidant and compared (see scheme). The conversion of the first to the latter has been investigated by means of XANES, EXAFS, and STEM. Copyright

Full text of DOI:10.1002/chem.201201472

FULL PAPER
DOI: 10.1002/chem.201201472
Water Oxidation with Molecularly Defined Iridium Complexes: Insights into
Homogeneous Versus Heterogeneous Catalysis
Henrik Junge,[a] Nicolas Marquet,[a] Anja Kammer,[a] Stefania Denurra,[a, b]
Matthias Bauer,*[c] Sebastian Wohlrab,[a] Felix Gꢀrtner,[a] Marga-Martina Pohl,[a]
Anke Spannenberg,[a] Serafino Gladiali,[a, b] and Matthias Beller*[a]
Abstract: Molecularly defined Ir com-
plexes and different samples of sup-
ported IrO2 nanoparticles have been
tested and compared in the catalytic
water oxidation with cerium ammoni-
um nitrate (CAN) as the oxidant. By
comparing the activity of nano-scaled
supported IrO2 particles to the one of
organometallic complexes it is shown
that the overall activity of the homoge-
neous Ir precursors is defined by both
the formation of the homogeneous
active species and its conversion to
IrIV-oxo nanoparticles. In the first
phase of the reaction the activity is do-
minated by the homogeneous active
species. With increasing reaction time,
the influence of nano-sized Ir-oxo par-
ticles becomes more evident. Notably,
the different conversion rates of the
homogeneous precursor into the active
species as well as the conversion into
Ir-oxo nanoparticles and the different
particle sizes have a significant influ-
ence on the overall activity. In addition
to
the
homogeneous
systems,
IrO2@MCM-41 has also been synthe-
sized, which contains stabilized nano-
particles of between 1 and 3 nm in size.
This latter system shows a similar activ-
ity to IrCl3·xH2O and complexes 4 and
5. Mechanistic insights were obtained
by in situ X-ray absorption spectrosco-
py and scanning transmission electron
microscopy.
Keywords: heterogeneous catalysis ·
homogeneous catalysis · iridium ·
nanoparticles · oxygen · water oxi-
dation
Introduction
is more difficult, because it is endothermic (E0 =1.23 eV at
pH 0.0) and four electrons have to be transferred to an elec-
tron acceptor per molecule of generated oxygen [Eq. (1)].[4]
Photocatalytic water-splitting to oxygen and hydrogen with
the help of sunlight is of significant interest for the develop-
ment of more benign energy technologies.[1] When applying
organometallic catalysts for this process most efforts have
focused on hydrogen generation, because it is expected to
play an important role as secondary energy carrier for
energy supply and storage.[2] Whereas the reduction of pro-
tons to hydrogen, a two electron process, has been investi-
gated with considerable success,[3] water oxidation to oxygen
Unfortunately, so far only few semiconducting materials
are able to perform overall water splitting.[5] Hence, to de-
velop more efficient catalysts the overall process is typically
separated into its two half reactions, water oxidation and
water reduction, which are studied autonomously. In this re-
spect, previous efforts to oxidize water with molecularly de-
fined catalyst precursors mainly focused on ruthenium-
based systems. Pioneering work has been performed by
Meyer and co-workers,[6] followed by other groups.[4]
[a] Dr. H. Junge, Dr. N. Marquet, A. Kammer, S. Denurra,
Dr. S. Wohlrab, F. Gꢀrtner, Dr. M.-M. Pohl, Dr. A. Spannenberg,
Prof. Dr. S. Gladiali, Prof. Dr. M. Beller
Recently, iridium complexes have been identified to be
highly active water oxidation catalysts.[7] It is important to
note that the majority of the catalyst systems applied so far
for oxygen generation makes use of strong oxidants such as
Leibniz-Institut fꢁr Katalyse e.V.
Albert-Einstein-Straße 29a, 18059 Rostock (Germany)
Fax : (+49)381-1281-5000
E-mail: matthias.beller@catalysis.de
[b] S. Denurra, Prof. Dr. S. Gladiali
Universitꢂ di Sassari
ꢀ
cerium ammonium nitrate (CAN), iodate (IO3 ), peroxodi-
ACTHNUTRGNEUNG
sulfate (S2O82ꢀ), Co3+ ammonia compounds or [Ru(bipy)3]3+
Dipartimento di Chimica e Farmacia,
Via Vienna 2, 07100 Sassari (Italy)
(Scheme 1).[8–14] Nevertheless, recently ruthenium complexes
allowed also for photocatalytically driven reactions.[15–17]
Apart from organometallic complexes also heterogeneous
materials like RuO2 and IrO2 are known to oxidize water in
the presence of sacrificial oxidants.[18] Therefore, it is inter-
esting to get more insights into the real catalyst species. To
distinguish between an active homogeneous or heterogene-
ous catalyst, different methods such as kinetic isotope effect
[c] Prof. Dr. M. Bauer
Technische Universitꢀt Kaiserslautern
Fachbereich Chemie,
Erwin-Schrçdinger-Straße 54
67663 Kaiserslautern (Germany)
E-mail: bauer@chemie.uni-kl.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201472.
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Scheme 1. Water oxidation in a two component system with CeIV as sacri-
ficial oxidant. WOC=water oxidation catalyst.
investigations, UV/Vis, TEM, powder X-ray diffraction,
XPS, analytical thermogravimetry/differential thermal analy-
sis (TG/DTA) measurements, or an electrochemical quartz
crystal nanobalance (EQCN) have been used.[7c,8a,19–21] For
example, it has been observed by Crabtree that kinetic iso-
tope effects[8a] differ when soluble iridium catalysts or IrOx
nanoparticles are used. In addition, the deposition of meas-
urable amounts of the soluble catalysts on electrodes on
short timescales (minutes) was reported.[20]
In our previous work, we demonstrated that simple
IrCl3·xH2O and IrACHTUNGTRENNUNG(acac)3 (iridiumacetylacetonate) constitute
highly efficient catalysts for water oxidation. By using
IrCl3·xH2O, we were able to detect Ir-containing nanoparti-
cles in the reaction mixture with sizes of <2 nm in a ceria
matrix by using high-angle annular dark field scanning trans-
mission electron microscopy (HAADF-STEM) combined
with EDX.[22] Similar results were reported by Grotjahn for
molecularly defined Ir complexes.[19] In addition, the degra-
dation of the iridium compounds has been investigated in
different organic solvents by NMR measurements.[14,19]
These observations prompted us to focus further on whether
the metal precursors lead to molecularly defined species or
nm-sized particles as active catalysts for water oxidation. In
this paper, we present efficient water oxidation catalysis
with both Ir complexes as well as nano-scaled IrO2 particles.
To characterize the active catalysts in situ, X-ray absorption
near edge structure (XANES) and extended X-ray absorp-
tion fine structure (EXAFS) measurements, as well as scan-
ning transmission electron microscopy (STEM) investiga-
tions were performed.
Figure 1. Synthesis of chloro-bridged dimers [{(C^N)2IrACTHUNTRGNE(UGN m-Cl)}2] 1–5.
(i.e., 0.5 mmol of the dimeric complex) was used as the cata-
lyst in water (10 mL) containing CeIV (CAN) (1700 mmol) as
the sacrificial oxidant. The evolved gas was quantitatively
measured by using an automatic gas burette[25] and analyzed
by GC. In fact, all of the complexes showed significant activ-
ities and the conversion was in between 92 and 98% with
respect to the oxidant CAN and provided approximately
10 mL of oxygen. The experiments are summarized in
Table 1. For comparison, results for Ir
have also been included.
ACHTUNGTRENUN(GN acac)3 and IrCl3·xH2O
Table 1. Oxygen evolution rates for the molecularly defined water oxida-
tion catalyst precursors using CeIV as oxidant.
Entry[a] Catalyst
precursor
[Ir]
TOF
TOFmax Ref.
[b,c]
[b,d]
amount [mmol]
A
]
[hꢀ1
ACHTUNGTERNNUNG ]
Results and Discussion
1
2
3
4
5
6
7
8
9
Ir
(acac)3
1.08
755
161
570
103
79
170
188
313
276
1177
303
966
364
317
463
498
735
604
[22]
[22]
IrCl3·xH2O 1.04
At the start of this work, the Ir-phenylazole complexes 1–5
(Figure 1) were tested as pre-catalysts in the water oxidation
half reaction to generate molecular oxygen. These com-
plexes are available based on our experience in the synthesis
of new Ir-based photosensitizers for the photocatalytic hy-
drogen generation from water.[23] The chloro-bridged dimers
1
2
3
4
5
6
7
1.06
1.04
1.01
1.10
1.08
1.04
1.04
this paper
this paper
this paper
this paper
this paper
this paper
this paper
[{(C^N)2IrACHTUNGTRENNUNG(m-Cl)}2] can be synthesized through various pro-
[a] General conditions: H2O (10 mL), CAN (1700 mmol), ca. 1 mmol cat.,
258C, dark. [b] TOF values are given in mol(O2)·mol(Ir)ꢀ1 hꢀ1. [c] TOF
was calculated as an average TOF between 0% and 85–89% of CAN
conversion for entries 1 and 2 and between 0 and 73–82% of CAN con-
version for entries 3 to 9. After that, the kinetics drop due to the deple-
tion of CeIV concentration. [d] TOFmax has been defined as the highest
observed TOF during the reaction by graphical display of TOF versus
time (see, for example, the Supporting Information, SI6-3 to SI6-6). The
reproducibility is in the range of 1–15%.
tocols,[24] however we have chosen the previously described
procedure, shown in Figure 1 (top), in which 1–5 are accessi-
ble by direct cyclometalation of IrCl3·xH2O with the respec-
tive HC^N ligands.[23] This is a convenient approach to
obtain the desired product without the necessity of purifica-
tion steps. The corresponding 2-phenylazole ligands are
commercially available. In a typical experiment, Ir (1 mmol)
&
2
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Homogeneous Versus Heterogeneous Catalysis
Among all of the chloro-
FULL PAPER
bridged dimers [{(C^N)2IrACTHNUTRGENUG(N m-
Cl)}2] 1–5, complex 1 showed
the highest activity (Table 1,
entry 3), which is only slightly
less than the activity of Ir-
A
(Table 1,
entry 1).
Whereas when using 4 and 5
(Table 1, entries 6 and 7), ap-
proximately one third of the
overall activity and half of the
TOFmax were achieved. Note-
worthy, the Ir precursors 2 and
3
gave lower TOF values
(Table 1, entries 4 and 5). In ad-
dition, from the most promising
ligand 2,5-diphenyloxazole (2,5-
Figure 2. Synthesis of monomeric water oxidation catalysts 6 and 7.
dpo), the complexes [Cp*Ir
ACHTUNGTRENNUNG
dpo)Cl] (6) and [Ir(2,5-dpo) CHTUGNTRENNUNG
G
were synthesized. The syntheses have been performed ac-
cording to the reported procedures of the analogue phenyl-
pyridine (ppy) derivatives starting either from the dimeric
iridium Cp*-chloro complex and the 2,5-dpo ligand or from
1, respectively (Figure 2).[8b,26,27] In the case of [Cp*Ir
ACTHNUTRGNE(NUG 2,5-
dpo)Cl] (6), crystals suitable for X-ray crystal structure anal-
ysis were obtained from a CH2Cl2/hexane mixture.[28] De-
tailed information for the structure is provided in the Sup-
porting Information (Figure S2-1). The analytical data of the
new compounds 5,[29] 6, and 7 (1H and 13C NMR, IR spectro-
scopy and MS spectrometry are listed in the Experimental
Section). Both of the Ir complexes 6 and 7 were more active
than 2–5, but less active than 1 (Table 1, entries 8 and 9).
Next, we were interested to determine if the iridium
oxide nanoparticles were formed during the water oxidation
applying the dimeric Ir species. Therefore, typical reaction
mixtures from the catalytic experiments containing 1 and 4,
Figure 3. XANES spectra of the pre-catalysts (gray dotted dashed line)
IrO2, IrCl3·xH2O, IrACTHNUTRGNEUNG(acac)2, 1, 4 and the corresponding spectra recorded
IrCl3·xH2O, and Ir
C
in course of water oxidation reactions (black line).
IrO2 were investigated in more detail by X-ray absorption
spectroscopy (XAS) measurements. Whereas the XANES
part of XAS is mainly sensitive to the oxidation state of an
X-ray-absorbing atom by the shape and position of the
white line (first resonance after the edge jump), the EXAFS
(extended X-ray absorption fine structure) part yields type,
number, and distance of coordinating atoms to the X-ray ab-
sorber.[30–33]
Figure 3 shows the XANES spectra at the Ir L3-edge of
the pre-catalysts IrO2, IrCl3·xH2O, IrACHTNUTRGNE(NUG acac)3, 1 and 4 in com-
parison to the spectra under water oxidation conditions. The
corresponding EXAFS spectra are given in Figure 4. The
structural parameters obtained by fitting these spectra with
theoretical models are summarized in Table 2. In general,
there are two indicators of the oxidation state in the
XANES spectra. The first is the intensity of the so-called
white-line (sharp resonance after the edge rise), which gains
intensity with increasing oxidation state[30] due to the in-
creasing number of final states of the p!d transition. The
second is the weaker signal after the white line, which is
Figure 4. Fourier-transformed EXAFS spectra of the pre-catalysts (gray
dotted dashed line) IrO2, IrCl3·xH2O, Ir
ACHTUNGERTN(NUNG acac)2, 1, 4 and the correspond-
ing spectra recorded in course of water oxidation reactions (black line).
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

M. Bauer, M. Beller et al.
Table 2. Results of fitting the experimental EXAFS data with theoretical models. The average values from the
LC-XANES fit for oxidation state determination are also given.
ysis. For solid IrO2 the crystal
structure data[33] could be well
reproduced within the error
Sample
Abs-Bs
N(Bs)
R
[ꢄ]
(Abs-Bs)
s
E
LC-XANES fit
A
]
IrO2/Precatalyst[b]
bar, with
a sixfold nearest
ꢀ
IrO2
Ir O
ꢀ
Ir Ir
6.9ꢂ0.7
3.1ꢂ0.3
2.5ꢂ0.3
6.1ꢂ0.6
1.4ꢂ0.1
3.0ꢂ0.3
3.0ꢂ0.3
1.1ꢂ0.1
3.4ꢂ0.3
2.3ꢂ0.2
6[a]
1.99ꢂ0.02
3.15ꢂ0.04
3.54ꢂ0.04
1.99ꢂ0.02
3.14ꢂ0.04
2.18ꢂ0.02
2.36ꢂ0.02
3.50ꢂ0.03
2.16ꢂ0.02
2.40ꢂ0.02
2.01ꢂ0.02
2.93ꢂ0.03
3.19ꢂ0.03
2.04ꢂ0.02
3.01ꢂ0.03
3.36ꢂ0.03
2.00ꢂ0.02
2.36ꢂ0.02
2.39ꢂ0.02
2.80ꢂ0.03
2.02ꢂ0.02
2.48ꢂ0.02
3.24ꢂ0.03
2.01ꢂ0.02
2.35ꢂ0.02
2.37ꢂ0.02
2.80ꢂ0.03
2.04ꢂ0.02
2.56ꢂ0.02
3.28ꢂ0.03
0.008ꢂ0.001
0.013ꢂ0.001
0.005ꢂ0.001
0.003ꢂ0.001
0.002ꢂ0.001
0.009ꢂ0.001
0.005ꢂ0.001
0.009ꢂ0.001
0.002ꢂ0.001
0.004ꢂ0.001
0.006ꢂ0.001
0.013ꢂ0.001
0.002ꢂ0.001
0.011ꢂ0.001
0.015ꢂ0.002
0.012ꢂ0.001
0.039ꢂ0.004
0.112ꢂ0.011
0.039ꢂ0.004
0.081ꢂ0.016
0.084ꢂ0.005
0.032ꢂ0.003
0.032ꢂ0.006
0.005ꢂ0.001
0.025ꢂ0.003
0.004ꢂ0.001
0.081ꢂ0.016
0.018ꢂ0.005
0.002ꢂ0.003
0.025ꢂ0.006
ꢀ
neighbor Ir O contribution and
an Ir–Ir shell with two atoms at
3.16 ꢄ. The more distanced Ir–
Ir shell could only be fitted
with 2.5 Ir, which is due to par-
ticle size effects even in the ref-
erence. In contrast, under water
oxidation conditions, a sixfold
oxygen coordination is still
found; however, the first Ir–Ir
coordination is reduced by a
factor of two, and the second
Ir–Ir contribution is not detect-
ed anymore. This is in agree-
ment with a reduced particle
size of IrO2 under water oxida-
tion conditions.
ꢀ
Ir Ir
ꢀ
IrO2 WOC
Ir O
ꢀ
Ir Ir
ꢀ
IrCl3·xH2O
Ir O
ꢀ
Ir Cl
ꢀ
Ir Ir
ꢀ
IrCl3·xH2O WOC
Ir O
ꢁ28.5/71.5%
ꢁ41/59%
ꢀ
Ir Cl
ꢀ
Ir
Ir
1
N
Ir O
Ir-C
6[a]
3[a]
ꢀ
Ir C
ꢀ
Ir O
6.1ꢂ0.6
4.1ꢂ0.4
1.3ꢂ0.1
4[a]
ꢀ
Ir C
ꢀ
Ir Ir
[c]
[c]
ꢀ
Ir C/N
2[a]
ꢀ
Ir Cl
4[a]
ꢀ
Ir C
ꢀ
Ir C
6.1ꢂ1.2
5.6ꢂ0.6
0.8ꢂ0.1
0.5ꢂ0.2
4[a]
ꢀ
1 WOC
Ir O/C
ꢁ52/48%
ꢀ
Ir Cl
The white line intensity of
IrCl3·xH2O during the course of
the reaction is significantly en-
hanced in comparison to the
solid pre-catalyst IrCl3·xH2O.
Examination of the second res-
onance at ꢁ11225 eV reveals
ꢀ
Ir Ir
[c]
ꢀ
4
Ir C/N
2[a]
ꢀ
Ir Cl
4[a]
ꢀ
Ir C
ꢀ
Ir C
6.1ꢂ1.2
6.7ꢂ0.6
0.4ꢂ0.1
1.6ꢂ0.2
[c]
ꢀ
4 WOC
Ir O/C
ꢁ82/18%
ꢀ
Ir Cl
ꢀ
Ir Ir
still
a
contribution
of
[a] Values were fixed to the expected value of the molecular/crystal structure. [b] Mean value of the obtained
ranges given in the text are listed. [c] Carbon/nitrogen/oxygen cannot be distinguished as neighbors by
EXAFS.
IrCl3·xH2O. Thus, the white line
increase can be due to oxida-
tion state changes and particle
size effects, as mentioned
rather intense for Ir0,[30] of lower intensity for IrIII, and final-
ly disappears for IrIV.[31] As can be seen in all the XANES
spectra (Figure 3), the white line increases, whereas the
second resonance, observed in all IrIII pre-catalysts, is re-
duced in intensity during water oxidation reaction. A signifi-
cant contribution of an IrIV species is therefore suspected in
all spectra recorded under water oxidation conditions.
above. Fitting the spectrum of IrCl3·xH2O under water oxi-
dation conditions by linear combination of IrCl3·xH2O and
IrO2 yields a composition of 68–75% IrCl3·xH2O and 25–
32% IrO2. The white line area and the second resonance
were fit independently, resulting in a range of composition
percentage. The EXAFS results given in Table 2 confirm
ꢀ
this quantification. The first Ir O contribution increases
To quantify the amount of IrIII and IrIV fractions in the
spectra, a linear combination (LC) fit approach was used by
fitting the WOC spectra with the pre-catalyst and IrO2 refer-
ence spectra. Because the XANES white line is both affect-
ed by the oxidation state and the particle size,[32] this LC fit
was not restricted to the area around the white line, as usu-
ally carried out. Instead, the second resonance was used as a
unique fingerprint of the oxidation state and was fitted inde-
pendently to circumvent particle size effects. The XANES
spectra of solid IrO2 and IrO2 under reaction conditions
showed only an intense white line at ꢁ11215 eV, and the
second signal, which is indicative of oxidation states lower
than IrIV, is missing. During water oxidation, the white line
intensity is slightly increased. Since there is only IrIV pres-
ent, this increase can be attributed to the formation of
smaller particles compared to the solid pre-catalyst. This ob-
servation is also reflected in the results of the EXAFS anal-
slightly compared to the solid pre-catalyst. In contrast, the
ꢀ
Ir Cl coordination number with 2.3 is reduced to around
75% during the course of the reaction. Whereas with solid
ꢀ
IrCl3·xH2O a third Ir Ir shell can be fitted, but for the reac-
tion mixture (due the low content of IrO2) this is not possi-
ble.
Results for Ir
ACHTUNGTRENNUNG
Compared to the pre-catalyst the white line of IrAHCTUNGTRENNUNG
the reaction mixture is increased, and the second resonance
showed a significant contribution of IrIII. The linear combi-
nation fit of the spectra yields 55–63% IrACTHNURTGNENG(U acac)3 and 45–
37% IrO2. Structural parameters achieved by fitting the ex-
perimental EXAFS spectra with theoretical models given in
Table 2 are in good agreement with these results: The first
ꢀ
Ir O shell still contains six oxygen backscatterers, but the
ꢀ
intensity of the first Ir C contribution characteristic of Ir-
(acac)3 is reduced to 4 atoms. This corresponds to 60% Ir-
&
4
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Homogeneous Versus Heterogeneous Catalysis
FULL PAPER
ACHTUNGTRENNUNG(acac)3. The presence of IrO2 is reflected in the third contri-
bution consisting of 1.3 Ir atoms at a slightly increased dis-
tance compared with solid IrO2.
By using complex 1, a slightly smaller amount (44–52%)
of IrIII is observed under water oxidation conditions in the
LC-XANES fit, which corresponds to 48–56% IrO2. These
results are also reflected in the EXAFS results of Table 2.
The isolated complex 1 was fitted with good quality by using
the molecular structure. Since carbon and nitrogen cannot
be distinguished as backscatterers, they are summarized in
one C/N shell. The characteristic structural features of 1 are
the two chlorine neighbors per Ir center that follow from
the dimeric structure shown in Figure 1. Moreover, a higher
carbon shell consisting of more distanced atoms of the li-
gands is observed. In the reaction mixture slightly less than
six atoms are found in the first coordination shell. Since the
ꢀ
Ir Cl contribution is still seen, the remaining 1 is supposed
to be present in the reaction mixture. Thus, also coordinat-
ing carbon can be present in the first shell. As carbon and
oxygen backscatterers cannot be distinguished, the nearest
neighbor shell is labeled O/C in Table 2. Only 0.8 Cl are
Figure 5. HAADF-STEM images of molecularly defined Ir complexes
ꢀ
after the water oxidation reaction a) Ir
inset e) Ir-containing particle.
ACHTNUGRTEN(NNUG acac)3, b) IrCl3·xH2O, c) 1, d) 4,
found in the Ir Cl shell, indicating the presence of 40% of
1, which is in good agreement with the XANES results. Sup-
ꢀ
posed that the third Ir Ir shells originated only from IrO2
ꢀ
species in the reaction mixture, the real Ir Ir coordination
sumption of CAN cannot be excluded. However, all samples
have been treated in the same manner and therefore the dif-
ferences should be significant. Starting from IrACHTUNGTRENNUNG(acac)3,
IrCl3·xH2O and 1 (Figure 5a-c and the Supporting Informa-
tion, SI4-1-SI4-3) Ir-containing nanoparticles of 0.5 nm to
number after de-averaging in the IrO2 particles should be
around 0.9. It is therefore slightly lower than in the other
WOC reaction mixtures studied so far.
Investigating complex 4, a further reduction of the IrIII
fraction under water oxidation conditions is observed. The
2 nm were detected. Whereas the IrACHTNUTRGNEUNG(acac)3 sample contains
LC fit results yield a composition of about 18% IrIII and
predominantly particles in the range of 1 nm, smaller parti-
cles have been formed when starting from 1 (<1 nm). For
IrCl3·xH2O, the size distribution is between approximately
0.5 and 2 nm.[22] These results are in agreement with the ob-
servations of Grotjahn, who detected Ir-containing nanopar-
ticles with sizes between 2 and 10 nm.[19] On the other hand,
starting from 4 (Figure 5d and the Supporting Information
SI4-4) no nanoparticles on the Ce matrix were detected,
since these have obviously agglomerated to give bigger par-
ticles in the 10 mm range (Figure 5e and the Supporting In-
formation SI4-4).
Finally, we were interested in the activity of IrO2 with re-
spect to different particle sizes. Applying commercially
available IrO2, which consists of agglomerates of particles
with a primary particle size of 1 to 3 nm (see the Supporting
Information, Figure SI5-2), much less activity has been ob-
served (Table 3, entry 1). In addition, three samples of iso-
lated and stabilized IrO2 particles on porous supports with
sizes between 8 and 1 nm were synthesized. The applied
nanoporous glasses (NPG)[34] and mesoporous silicas[35] are
attractive catalyst supports to disperse catalysts and to limit
particle growth in the order of their inherent pore size.[36]
Their pore size distribution can be adjusted during the syn-
thetic route. By applying these materials as supports for
nanoparticles, effects such as particle ripening and leaching
can be overcome. Additionally, highly and thermally stable
surface areas of mesoporous silicas are favorable for a varie-
IV
ꢀ
82% Ir . The EXAFS results, mainly the Ir Cl shell sup-
port this ratio, since from the Ir Cl coordination number of
0.4 a fraction of 20% remaining 4 is calculated. However,
the Ir Ir shell showed a rather high coordination number of
1.6, according to a real coordination number in the IrO2 par-
ticles of around 2. Therefore, larger particles of IrO2 may be
present in this reaction mixture. Having a closer look at the
Ir Ir coordination number in all WOC mixtures, a relative
estimation of the IrO2 particle sizes can be given: 4 (NIrꢀIr
ꢀ
ꢀ
ꢀ
=
2ꢂ0.4)>IrO2 (NIrꢀIr =1.4ꢂ0.3)ꢁIr(acac)3 (NIrꢀIr =1.3ꢂ
G
0.3)>1(NIrꢀIr=0.9ꢂ0.2). It has to be mentioned, that in case
of IrCl3·xH2O the amount of formed IrO2 is very small ac-
cording to the LC-XANES analysis. The coordination
ꢀ
number of the Ir Ir shell in such a case can be easily below
the detection limit of the EXAFS analysis.
All these findings are in general in good agreement with
the results of TEM measurements. The HAADF-STEM
images of IrACHTUNGTRENNUNG(acac)3, IrCl3·xH2O, 1 and 4 have been taken
after acting as precursors in the water oxidation reaction
(Figure 5). To improve the chance to detect Ir nanoparticles,
both, the concentration of Ir and the ratio of iridium to
cerium have been increased ([Ir] (15 mmol) and CAN
(3.4 mmol) in 50 mL water). After the reaction was finished,
the solution has been concentrated to 10 mL. Thus, an influ-
ence of the modified reaction conditions onto the samples
as well as a degradation of the Ir complexes after full con-
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

M. Bauer, M. Beller et al.
ty of oxide syntheses. Normally, the pore sizes of NPGꢅs
range from 10 to 1000 nm and pore volumes between 0.1
and 1.1 cm3gꢀ1 and inner surfaces up to 500 m2gꢀ1 are ach-
ieved. Much lower pore sizes and higher surface areas can
be achieved with mesoporous silicas. These materials also
possess surface areas up to more than 1000 m2 gꢀ1, whereas
the pore sizes can be controlled and adjusted in the diame-
ter range from 1.5 to more than 20 nm. For the synthesis of
the supported IrO2 samples, two NPGꢅs (NPG-63 and NPG-
27) and one mesoporous silica (MCM-41) have been chosen.
The supported IrO2 materials have been investigated in
more detail by means of XRD and TEM. The XRD patterns
of the samples IrO2@NPG-63, IrO2@NPG-27, IrO2@MCM-
41 and commercial IrO2 are shown in the Supporting Infor-
mation (Figure SI5-1) and compared with reported reflex
positions for tetragonal iridium(IV) oxide. The phase purity
and the nanocrystalline character of IrO2 incorporated in
NPG-63 and NPG-27 is clearly shown. Average crystallite
sizes of 7.7 nm for IrO2@NPG-63 and 7.3 nm for
IrO2@NPG-27 were estimated by using the Scherrer equa-
tion. The pore sizes of the glasses
Figure 6. HAADF-TEM image of IrO2@MCM-41; inset: high-resolution
BF image of crystalline IrO2 from IrO2@MCM-41.
applied (dpore =63/27 nm) did not
have a direct influence on the
size of the IrO2 crystallites
Table 3. Oxygen evolution rates for the IrO2 nanoparticles using CeIV as oxidant.
Entry[a]
Catalyst
precursor
[Ir]
TOF
TOFmax
Ref.
formed within. Therefore,
smaller pore size of the silica sup-
port, like MCM-41 with dpore
a
[b,c]
[b,d]
amount [mmol]
G
]
[hꢀ1
ACHTUNGTERNNUNG ]
1
2
3
4
IrO2 (Strem)
1.28
1.05
1.27
1.15
35
8.0
18
347
143
238
343
[22]
=
4.8% IrO2@NPG-63
5.5% IrO2@NPG-27
9.3% IrO2@MCM-41
this paper
this paper
this paper
3.8 nm, is required to influence
the particle growth. The XRD
pattern of IrO2@MCM-41 showed
no clear signals that could be as-
signed to the metal oxide.
158
[a] General conditions: H2O (10 mL), CAN (1700 mmol), IrO2 (1 mmol), 258C, dark. [b] TOF values are given
in mol(O2)·mol(Ir)ꢀ1 hꢀ1. [c] TOF was calculated between 0 and 85–89% of CAN conversion for entry 1, 0 and
75% of CAN conversion for entries 3 and 4. After that, the kinetics drop due to the depletion of CeIV concen-
tration. The experiment in entry 2 has been stopped after 18 h at 30% conversion. [d] TOFmax has been de-
fined as the highest observed TOF during the reaction by graphical display of TOF versus time (see for exam-
ple, the Supporting Information, SI6-1 and SI6-2). Reproducibility is in the range of 1–25%, except for
Further investigations by using
TEM indicated the presence of
highly dispersed and crystalline IrO2@NPG-63 (30%).
IrO2 species in the mesoporous
material;
the
sample
IrO2@MCM-41 showed a small apparent particle size be-
tween 1 and 3 nm and a crystalline structure (Figure 6).
The three synthesized IrO2 samples were tested in the
oxygen evolution reaction under the same conditions similar
to the molecularly defined Ir complexes (Table 3).
IrO2@NPG-63 and IrO2@NPG-27 with particle sizes of
7.7 nm and 7.3 nm, respectively, showed much less activity
(Table 3, entries 2 and 3) than the commercially available
IrO2, which itself consists of aggregates of IrO2 particles
with primary particle sizes from 1 to 3 nm. Hence, an influ-
ence of the particle size on the activity is apparent. Interest-
ingly, when IrO2@MCM-41 is used with isolated and well-
dispersed IrO2 particles possessing sizes between 1 and
3 nm, much better activities are observed. Comparing the
average TOF values with that of the homogeneous Ir-pre-
cursors, IrO2@MCM-41 is equal to IrCl3·xH2O and may also
compete with complexes 2–5, although it is less active than
tivity in this complex system is a result of both homogene-
ous and heterogeneous catalysis. Starting from stable coordi-
natively saturated Ir complexes, active homogeneous species
as well as active nano-sized iridium-oxo species are formed.
The latter nanoparticles can be furthermore converted into
bulk iridium oxide and other deactivated species (Figure 7).
After an induction period, which lasts less than 10 min for
1
and IrACTHNGUETRNNU(G acac)3 and approximately 20 or 40 min for
IrCl3·xH2O and 4, respectively (see the Supporting Informa-
tion, Figure SI6-3 to SI6-6), an increasing TOF is observed.
Comparing the activities in this period, starting from homo-
geneous and heterogeneous catalyst precursors, it becomes
obvious that the TOFmax values of up to 1200 hꢀ1 should
arise mainly from the active homogeneous Ir complexes, if
no strong deleterious effects of supports are present. There-
fore, the overall activity seems to be dominated by the
active homogeneous species, which are formed by a rate ac-
cording to k1 (Figure 7). However, the generation of IrIV-oxo
nanoparticles is expected to start already in this phase, prob-
ably after the generation of significant amounts of the active
1 and IrACHTUNGTRENNUNG(acac)3.
As a result of the various catalytic experiments and the
mechanistic investigations, we conclude that the catalyst ac-
&
6
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Homogeneous Versus Heterogeneous Catalysis
FULL PAPER
Long-term catalyst tests revealed that the active species
resulting from 1 can be used in longer term experiments of
up to 11 days (see the Supporting Information, Figure SI7-
1); they stay active after full conversion and addition of a
new charge of CAN (Figure SI7-2). Thus, relatively high
turnover numbers of 8139 and 7250, respectively, were ach-
ieved.
Finally, it is interesting to note that the Ir-oxo nanoparti-
cles show a somewhat different behaviour during water oxi-
dation. With IrO2 by itself (as well as on supports), no in-
duction period is observed, which is in agreement to a previ-
ous report.[19] The TOF is initially very high and then drops
and remains quite stable until the depletion of CAN be-
comes evident, which leads to a further decrease (see the
Supporting Information, Figure SI6-1 and SI6-2).
Figure 7. Hypothesis for the conversion of molecularly defined Ir com-
plexes into Ir-O-species.
homogeneous species (k2 in Figure 7). After 10–20 min for 1
and IrACHTUNGTRENNUNG(acac)3 and approximately 90 min for 4 and 75 min for
IrCl3·xH2O, respectively, the maximum TOF is reached. Its
time and value are determined by the ratio of k1 and k2. At
the peak TOF, the conversion of CAN has reached 30 to
50%. Hence, the depletion of CAN as the reason for the ac-
tivity decrease seems to be unlikely. This has been proven
by an additional experiment with a higher amount of CAN
and the same amount of 1, which was allowed to react for
seven days. In this experiment the highest TOF is also ob-
tained within the first 20 min, when only 2% of CAN has
been consumed (see the Supporting Information, Figure
SI6-7).
After having reached the maximum activity the TOF is
more and more defined by the IrIV-oxo nanoparticles. De-
pending on k2, there are significant differences in the extent
of the conversion to the IrIV-oxo nanoparticles due to the
different ligands in the Ir complexes. For example, after one
Conclusion
The activity of seven iridium complexes and different heter-
ogeneous iridium oxide materials in the cerium-driven oxi-
dation of water to oxygen has been compared. A combina-
tion of catalytic experiments, in situ X-ray absorption spec-
troscopy, XRD, and STEM measurements have shown that
both molecular iridium species and IrIV-oxo nanoparticles
are present in the water oxidation reactions. By applying or-
ganometallic precursors in the first phase of the reaction,
the most active species are generated from the homogene-
ous precursors. During the course of the reaction nano-sized
Ir-oxo species are formed that catalyze the water oxidation
with lower activities, as it has been shown for preformed
IrIV-oxo nanoparticles. However, this could at least in part
be also an effect of the applied supports. The obtained cata-
lyst activities of organometallic iridium complexes are a
result of different conversion rates of the pre-catalyst into
the active species as well as into Ir-oxo nanoparticles. De-
pending on the ligand, the particle size of the formed nano-
particles is different and has an important influence on the
activity of the catalyst system. Thus, the tetrakis-(2,5-diphe-
ꢀ
hour IrACHTUNGTRENNUNG(acac)3 and 1 were converted to Ir O species with up
to 40 and 50%, respectively, whereas starting from 4 up to
82% and starting from IrCl3·xH2O only 25% IrIV-oxo nano-
particles have been generated within the same time. In the
case of complexes with organic ligands, this conversion is ac-
companied by CO2 generation due to a degradation of the
ligands. In these cases the analysis of the gas phase after fin-
ishing the reactions resulted in significant contents of
carbon dioxide, between 0.17 and 0.30 vol%. For 6, an even
higher CO2 content has been observed (1.1 vol%), which is
very likely due to the sensitivity of the Cp* ligand with re-
spect to oxidative degradation; this has also been observed
for other complexes.[21] During this phase of the major con-
nyloxazole)-m-(dichloro)diiridiumACHTUNGRTENNGU(III) and IrCAHTUNGTRNE(NUGN acac)3 belong
to the most powerful systems studied here. Finally,
IrO2@MCM-41 with stabilized nanoparticles of comparable
size (1–3 nm) has been prepared, which shows comparable
activity to IrCl3·xH2O and complexes 4 and 5.
ꢀ
tribution of Ir O species, the particle size of the formed iri-
dium oxide nanoparticles is another important parameter in-
fluencing the activity. Whereas the highest activities for IrO2
particles have been observed in the range of 1–3 nm, parti-
cles of more than 7 nm are significantly less active. Notably,
the stabilization of such small nanoparticles on a porous
support, as in the case of IrO2@MCM-41, has a positive in-
fluence. At the end of the reaction the catalyst activity
drops due to the depletion of CAN. Furthermore, during all
phases, an agglomeration to bigger particles (k3 in Figure 7)
and a deactivation by the formation of inactive species (k4,
k5, and k6 in Figure 7) might occur.
Experimental Section
General
(dichloro)diiridiumAHCTUNTGRENNUNG
protocol
for
the
synthesis
of
tetrakis-(C^N)-m-
(III) 1–5: The cyclometalated iridium dimers were
prepared by a procedure adapted from Bernhard et al.[14] in moderate
yields. The ligands were purchased from commercial suppliers (Aldrich
and Alfa Aesar). Stoichiometric quantities of the appropriate cyclometa-
lating ligand (2.0 equiv, typically 1.0 mmol) were added to a mixture of
iridium chloride (1.0 equiv, typically 0.5 mmol) in an ethoxyethanol/water
mixture (3:1, 22 mL). The reaction mixture was heated under reflux
(1208C) with constant stirring for 24 h. The resulting precipitate was col-
lected by suction filtration and washed with diethyl ether, and dried to
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

M. Bauer, M. Beller et al.
yield the product [{(C^N)2Ir-m-Cl}2]. With the exception of compound 5
(which is a new compound), compounds 1–4 are known already and the
published analytical data correlate with our measured results.
chloride precursor were calculated according to the specific pore volumes
and overall densities of the porous materials to ensure a comparable cat-
alyst loading of all supports that was fixed at 9.3, 5.5 and 4.8 wt%, re-
spectively. After completed incipient wetness impregnation the materials
were dried at 608C. Afterwards the materials were calcined at 4008C
under air for 1 h using heating and cooling ramps of 58Cminꢀ1. The cal-
cined materials were washed three times with water and dried at 608C.
The results of XRD and TEM are provided in the Supporting Informa-
tion (SI5–1–3)
Synthesis of tetrakis-(4-methyl-2-phenylthiazole)-m-(dichloro)diiridium-
ACHTUNGTRENNUNG
(III) 5: T: 908C, t=30 h. Orange solid. Yield: 36%. 1H NMR (300 MHz,
CD2Cl2): d=7.47 (4H, d, J
ACHTUNGTRENNUNG
A
E
ACHTUNGTRENNUNG
7.3 Hz), 2.17 ppm (12H, s, CH3); 13C NMR (100 MHz, CD2Cl2): d=179.7,
156.4, 142.9, 142.3, 133.3, 128.9, 124.1, 121.9, 112.2, 18.7 ppm; ATR-IR:
n˜ =3099 (m), 3049 (w), 1579 (m), 1458 (m), 1282 (m), 1097 (m), 1024
(m), 759 (m), 731 (s), 720 cmꢀ1 (s); MS (ESI-TOF/HRMS): m/z: calcd for
C20H16IrN2S2: 541.038434; found: 541.03943 [C20H16IrN2S2]+.
General procedure for oxygen evolution experiments: The catalytic ex-
periments were carried out in an argon atmosphere and under the exclu-
sion of air. Solvents were purified and degassed with standard procedures
prior to use. Each experiment was performed in a glass reactor thermally
equilibrated at 258C through a double mantle with a temperature-con-
trolled circulating water bath. The reactor was always covered with alu-
minum foil during the reaction to exclude light effect.[22] The amount of
evolved gases was quantitatively measured via an automatic gas bu-
rette[25] and qualitatively by GC (gas chromatograph HP7890N, carboxen
1000, TCD, external calibration). In a typical experiment, CAN (ca.
930 mg, 1.7 mmol) was introduced in the reactor that was than evacuated
and filled with argon five times. Next, degassed and distilled water
(10 mL) was added and the stirred solution was thermally equilibrated at
258C for several minutes. Then the catalyst (ca. 1 mmol [Ir]) was intro-
duced in the reactor as a solid weighed in a Teflon crucible and gas evo-
lution was recorded until the end of the reaction. After each reaction, a
gas sample (ca. 5 mL) was taken and analyzed by gas chromatography
(GC). All measured volumes have been corrected by the blank volume
(0.7 mL). During the long term experiment the blank volume fluctuates
between 0 and 1 mL.
Synthesis of [Cp*Ir(2,5-diphenyloxazole)Cl] 6: Under an argon atmos-
phere, a slurry of [Cp*IrCl2]2 (370 mmol), 2,5-diphenyloxazole (767 mmol)
and NaOAc (1.295 mmol) in CH2Cl2 (10 mL) was stirred at room temper-
ature for 2 days and subsequently filtered over celite. The solvent was
evaporated and the yellow solid was dried under vacuum. For purifica-
tion, the solid was dissolved in CH2Cl2 (minimum amount) and filtered
over celite. The solvent was removed under reduced pressure and the
product dried under vacuum overnight to give a yellow solid. Yield:
74%. 1H NMR (300 MHz, CD2Cl2): d=7.81 (1H, d, J
7.76 (2H, d, J(H,H)=6.8 Hz), 7.68 (1H, d, J(H,H)=7.5 Hz), 7.53 (1H,
s), 7.49 (2H, t, J(H,H)=7.5 Hz), 7.40 (1H, t, J(H,H)=7.3 Hz), 7.22 (1H,
t, J(H,H)=7.4 Hz), 7.09 (1H, t, J(H,H)=7.4 Hz), 1.77 ppm (15H, s,
Cp); 13C NMR (75 MHz, CD2Cl2): d=174.9, 162.6, 152.0, 136.6, 131.7,
ACHTUNGTREN(NUNG H,H)=7.3 Hz),
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
H
130.8, 129.5, 129.4, 127.7, 124.4, 124.0, 122.5, 119.8, 88.3 (5C, CpAr), 9.4
(CpMe) ppm; ATR-IR: n˜ =3129 (w), 2914 (w), 1597 (m), 1481 (m), 1393
(m), 1180 (m), 1051 (m), 765 (s), 731 (s), 718 (s), 705 cmꢀ (m); MS (ESI-
TOF/HRMS): m/z: calcd for C25H25IrNO: 548.156556; found: 548.15735
[MꢀCl]+. The pure compound was obtained as orange crystals which
turned to yellow when scratched for collection. Crystals suitable for X-
ray diffraction were grown from a CH2Cl2/hexane mixture.
X-ray absorption spectroscopy: X-ray absorption measurements were
performed at the XAS beamline at the ꢄngstrçmquelle Karlsruhe
(ANKA). The synchrotron beam current was between 80–140 mA at 2.5
GeV storage ring energy. A SiACTHUNRTGNEUNG(111) double crystal monochromator was
Synthesis of [Ir(2,5-diphenyloxazole)2ACHTUNTRGNE(UNG H2O)2]CF3SO3 7: Under an argon
used for measurements at the Ir L3-edge (11.215 keV). The second mon-
ochromator crystal was tilted for optimal harmonic rejection. To perform
operando studies, that is, under the original conditions, the spectra were
recorded in fluorescence mode using a hyperpure Germanium detector
with a cell that could be securated and filled under inert atmosphere.
Measurements started 60 min after start of the reaction. In all cases
stable spectral conditions were found after the acquisition of the first
spectrum (30–60 min). Only in case of 4 evolving gas made the first spec-
tra useless and stable conditions were found after 150 min. Measurements
were carried out for 5–12 h, within that time window no changes in the
spectra were detected. More details can be found in the Supporting Infor-
mation.
atmosphere, a solution of AgOTf in MeOH (10 mL) was added to a
slurry of the dimer [(2,5-diphenyloxazole)2Ir-(m-Cl)]2 in CH2Cl2 (10 mL).
The reaction mixture was stirred overnight at room temperature. The sol-
ution was then filtered over celite and the filtrate was evaporated to
yield a dark-orange oil. The oil was diluted with CH2Cl2 and hexane was
added. Solvents were removed under reduced pressure and the product
was obtained as a yellow solid which was dried under vacuum. Yield:
91%; 1H NMR (300 MHz, [D6]DMSO): d=8.32 (2H, s); 8.05 (4H, d, J-
A
R
ACHTUNGTRNE(NUNG H,H)=
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
(75 MHz, [D6]DMSO): d=172.8, 151.7, 144.9, 131.9, 131.0, 129.9, 129.3,
128.6, 126.1, 124.9, 124.7, 124.6, 120.4 ppm; ATR-IR: n˜ =3049 (w), 1597
(m), 1491 (m), 1460 (m), 1208 (s), 1172 (s), 1041 (s), 1012 (s), 739 (m),
658 (m), 626 cmꢀ (s); MS (EI): m/z: calcd for C15H12IrNO2: 431.5; found:
432 [MꢀDPOꢀOTfꢀH2O+H]+; m/z: calcd for C30H20IrN2O2: 632.7;
found: 633 [MꢀOTfꢀH2O]+; m/z: calcd for C31H20N2O5SF3Ir: 781.8;
found: 782 [Mꢀ2H2O]+; MS (ESI-TOF/HRMS): m/z: calcd for
C30H20N2O2Ir: 633.115; found: 633.1141, [MꢀOTfꢀH2O]+, m/z: calcd for
CF3O3S: 148.95257; found: 148.95259, [M]+.
TEM measurements: For the TEM measurements, samples of IrACHTUNGTRENNUNG(acac)3,
IrCl3·xH2O, 1 and 4 were taken after acting as precursors in the water ox-
idation reaction, following the same protocol as described previously but
with a higher concentration of Ir. The Ir precursor (15 mmol [Ir]) and
CAN (3.4 mmol) were added to H2O (50 mL). After completion of the
reaction, the solution was concentrated to approximately 10 mL. The
samples were then collected by depositing a drop of the pre-treated solu-
tion on a carbon-supported copper grid mesh 300. The TEM measure-
ments were performed at 200 kV on a JEM-ARM200F (JEOL) that is
aberration-corrected by a CESCOR (CEOS) for the scanning transmis-
sion (STEM) applications. The microscope is equipped with a JED-2300
(JEOL) energy-dispersive x-ray-spectrometer (EDXS) for chemical anal-
ysis. High-angle annular dark field (HAADF) and EDXS imaging were
operated with spot size 5c and a 50 mm condenser aperture.
Synthesis of supported IrO2 on nanoporous silica or glasses: MCM-41
with a pore size of 3.8 nm was supplied by Sꢁd-Chemie (Germany).
Granulated nanoporous glasses specified by pore sizes of 27 nm (NPG-
27) and 63 nm (NPG-63) were purchased from VitraBio GmbH (Stei-
nach, Germany). The supports were characterized before usage with ni-
trogen adsorption using the five-point BET method and scanning and
transmission electron microscopy. The syntheses of IrO2 particles within
these glasses were achieved by a process including an impregnation and
calcination step. In the case of MCM-41, incipient wetness impregnation
was applied to impregnate 800 mg of MCM-41 with 150 mg (0.369 mmol)
H2IrCl6 dissolved in 1.5 mL water (IrO2@MCM-41). Nanoporous glasses
were impregnated in the same way by a solution of 106 mg (0.260 mmol)
of H2IrCl6 dissolved in water (1.0 mL) for 1.0 g of NPG-27 (IrO2@NPG-
27) or a solution of 91 mg (0.224 mmol) of H2IrCl6 dissolved in water
(1.3 mL) for 1.0 g NPG-63 (IrO2@NPG-63). The concentrations of the
Acknowledgements
Parts of this work have been supported by the BMBF within the project
“Light2Hydrogen” (Spitzenforschung und Innovation in den Neuen
Lꢀndern), by the Ministry for education, science and culture of Mecklen-
burg-Vorpommern and the European Union (Investing in our Future,
European Regional Development) within the project “Nano4Hydrogen”.
&
8
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Homogeneous Versus Heterogeneous Catalysis
FULL PAPER
We thank P. Bartels, Ch. Fischer and A. Koch for their excellent analyti-
cal and technical support. F.G. thanks the Fonds der Chemischen Indus-
Tong, S. Andersson, L. Sun, Chem. Eur. J. 2010, 16, 4659–4668;
e) L. Duan, Y. Xu, P. Zhang, M. Wang, L. Sun, Inorg. Chem. 2010,
49, 209–215.
trie (FCI) for
a Kekulꢆ grant. M.B. acknowledges the synchrotron
ANKA (Karlsruhe, Germany) for provision of beamtime and the Carl-
Zeiss foundation for financial support.
[16] M. D. Kꢀrkꢀs, E. V. Johnston, E. A. Karlsson, B.-L. Lee, T. ꢄker-
mark, M. Shariatgorji, L. Ilag, ꢈ. Hansson, J.-E. Bꢀckvall, B. ꢄker-
mark, Chem. Eur. J. 2011, 17, 7953–7959.
[17] S. Roeser, P. Farras, F. Bozoglian, M. Martinez-Belmonte, J. Benet-
Buchholz, A. Llobet, ChemSusChem 2011, 4, 153.
[1] a) H. Arakawa, M. Aresta, J. N. Armor, M. A. Barteau, E. J. Beck-
man, A. T. Bell, J. E. Bercaw, C. Creutz, E. Dinjus, D. A. Dixon, K.
Domen, D. L. DuBois, J. Eckert, E. Fujita, D. H. Gibson, W. A.
Goddard, D. W. Goodman, J. Keller, G. J. Kubas, H. H. Kung, J. E.
Lyons, L. E. Manzer, T. J. Marks, K. Morokuma, K. M. Nicholas, R.
Periana, L. Que, J. Rostrup-Nielson, W. M. H. Sachtler, L. D.
Schmidt, A. Sen, G. A. Somorjai, P. C. Stair, B. R. Stults, W. Tumas,
Chem. Rev. 2001, 101, 953–996; b) Q. Schiermeier, J. Tollefson, T.
Scully, A. Witze, O. Morton, Nature 2008, 454, 816–823; c) N. S.
Lewis, D. G. Nocera, Proc. Natl. Acad. Sci. USA 2006, 103, 15729–
15735.
[2] a) J. A. Turner, Science 2004, 305, 972; b) N. Armaroli, V. Balzani,
Angew. Chem. 2007, 119, 52–67; Angew. Chem. Int. Ed. 2007, 46,
52–66; c) L. Schlapbach, A. Zꢁttel, Nature 2001, 414, 353–358.
[3] For reviews on light-driven water reduction systems, see: 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; c) E. D. Cline, S. Bernhard, Chimia 2009, 63, 709–
713; d) T. S. Teets, D. G. Nocera, Chem. Commun. 2011, 47, 9268–
9274.
[18] See for example: a) A. Harriman, I. J. Pickering, J. M. Thomas, P. A.
Christensen, J. Chem. Soc. Faraday Trans. 1 1988, 84, 2795–2806;
b) R. Nakamura, H. Frei, J. Am. Chem. Soc. 2006, 128, 10668–
10669; c) T. Nakagawa, N. S. Bjorge, R. W. Murray, J. Am. Chem.
Soc. 2009, 131, 15578–15579; d) M. Hara, T. E. Mallouk, Chem.
Commun. 2000, 1903–1904; e) S. D. Tilley, M. Cornuz, K. Sivula, M.
Grꢀtzel, Angew. Chem. 2010, 122, 6549–6552; Angew. Chem. Int.
Ed. 2010, 49, 6405–6408; f) S. J. Kwon, F.-R. F. Fan, A. J. Bard, J.
Am. Chem. Soc. 2010, 132, 13165–13167; g) P. G. Hoertz, Y. I. Kim,
W. J. Youngblood, T. E. Mallouk, J. Phys. Chem. B 2007, 111, 6845–
6856; h) M. Yagi, E. Tomita, T. Kuwabara, J. Electroanal. Chem.
2005, 579, 83–88; i) G. S. Nahor, P. Hapiot, P. Neta, A. Harriman, J.
Phys. Chem. 1991, 95, 616–621.
[19] D. B. Grotjahn, D. B. Brown, J. K. Martin, D. C. Marelius, M.-C.
Abadjian, H. N. Tran, G. Kalyuzhny, K. S. Vecchio, Z. G. Specht,
S. A. Cortes-Llamas, V. Miranda-Soto, C. van Niekerk, C. E. Moore,
A. L. Rheingold, J. Am. Chem. Soc. 2011, 133, 19024–19027.
[20] N. D. Schley, J. D. Blakemore, N. K. Subbaiyan, C. D. Incarvito, F.
D’Souza, R. H. Crabtree, G. W. Brudvig, J. Am. Chem. Soc. 2011,
133, 10473–10481.
[4] For reviews on water oxidation catalysts: a) X. Sala, I. Romero, M.
Rodrꢇguez, L. Escriche, A. Llobet, Angew. Chem. 2009, 121, 2882–
2893; Angew. Chem. Int. Ed. 2009, 48, 2842–2852; b) H. Yamazaki,
A. Shouji, M. Kajita, M. Yagi, Coord. Chem. Rev. 2010, 254, 2483–
2491.
[21] D. Hong, M. Murakami, Y. Yamada, S. Fukuzumi, Energy Environ.
Sci. 2012, 5, 5708–5716.
[22] N. Marquet, F. Gꢀrtner, S. Losse, M.-M. Pohl, H. Junge, M. Beller,
ChemSusChem 2011, 4, 1598–1600.
[5] For reviews on heterogeneous water-splitting catalysts see: a) X.
Chen, S. Shen, L. Guo, S. S. Mao, Chem. Rev. 2010, 110, 6503–6570;
b) A. Kudo, Y. Miseki, Chem. Soc. Rev. 2009, 38, 253–278; c) F. E.
Osterloh, Chem. Mater. 2008, 20, 35–54; d) K. Maeda, K. Domen, J.
Phys. Chem. C 2007, 111, 7851–7861.
[23] F. Gꢀrtner, S. Denurra, S. Losse, A. Neubauer, A. Boddien, A. Go-
pinathan, A. Spannenberg, H. Junge, S. Lochbrunner, M. Blug, S.
Hoch, J. Busse, S. Gladiali, M. Beller, Chem. Eur. J. 2012, 18, 3220–
3225.
[24] M. S. Lowry, W. R. Hudson, R. A. Pascal, Jr., S. Bernhard, J. Am.
Chem. Soc. 2004, 126, 14129–14135.
[25] For details on the experimental set-up see: 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.
[26] C. Scheeren, F. Maasarani, A. Hijazi, J.-P. Djukic, M. Pfeffer, Orga-
nometallics 2007, 26, 3336–3345.
[27] B. Schmid, F. O. Garces, R. J. Watts, Inorg. Chem. 1994, 33, 9–14.
[28] CCDC-866799 contains the supplementary crystallographic 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.
[29] Although compound 5 has been already mentioned as an intermedi-
ate in our previous publication (see ref. [22]), no analytical data
have been provided so far.
[6] S. W. Gersten, G. J. Samuels, T. J. Meyer, J. Am. Chem. Soc. 1982,
104, 4029–4030.
[7] a) M. Yagi, A. Syouji, S. Yamada, M. Komi, H. Yamazaki, S. Tajima,
Photochem. Photobiol. Sci. 2009, 8, 139; b) N. D. McDaniel, F. J.
Coughlin, L. L. Tinker, S. Bernhard, J. Am. Chem. Soc. 2008, 130,
210–217; c) R. H. Crabtree, Chem. Rev. 2012, 112, 1536–1554.
[8] a) J. D. Blakemore, N. D. Schley, D. Balcells, J. F. Hull, G. W. Olack,
C. D. Incarvito, O. Eisenstein, G. W. Brudvig. R. H. Crabtree, J. Am.
Chem. Soc. 2010, 132, 16017–16029; b) J. F. Hull, D. Balcells, J. D.
Blakemore, C. D. Incarvito, O. Eisenstein, G. W. Brudvig, R. H.
Crabtree, J. Am. Chem. Soc. 2009, 131, 8730–8731.
[9] A. Savini, G. Bellachioma, G. Ciancaleoni, C. Zuccaccia, D. Zuccac-
cia, A. Macchioni, Chem. Commun. 2010, 46, 9218–9219.
[10] R. Lalrempuia, N. D. McDaniel, H. Mꢁller-Bunz, S. Bernhard, M.
Albrecht, Angew. Chem. 2010, 122, 9959–9962; Angew. Chem. Int.
Ed. 2010, 49, 9765–9768.
[11] W. I. Dzik, S. E. Calvo, J. N. H. Reek, M. Lutz, M. A. Ciriano, C.
Tejel, D. G. H. Hetterscheid, B. de Bruin, Organometallics 2011, 30,
372–374.
[12] D. G. H. Hetterscheid, J. N. H. Reek, Chem. Commun. 2011, 47,
2712–2714.
[13] T. P. Brewster, J. D. Blakemore, N. D. Schley, C. D. Incarvito, N.
Hazari, G. W. Brudvig, R. H. Crabtree, Organometallics 2011, 30,
965–973.
[14] A. Savini, P. Belanzoni, G. Bellachioma, C. Zuccaccia, D. Zuccaccia,
A, Macchioni, Green Chem. 2011, 13, 3360–3374.
[30] D. Lahiri, V. Subramanian, B. A. Bunker, P. V. Kamat, J. Chem.
Phys. 2006, 124, 204720-1-7.
[31] A. Minguzzi, C. Locatelli, G. Cappelletti, M. Scavini, A. Vertova,
P. F. Ghigna, S. Rondinini, J. Phys. Chem. A 2012, 116, 6497–6504.
[32] a) E. Bus, D. E. Ramaker, J. A. van Bokhoven, J. Am. Chem. Soc.
2007, 129, 8094–8102; b) Y. Akdogan, C. Vogt, M. Bauer, H. Ber-
tagnolli, L. Giurgiu, E. Roduner, Phys. Chem. Chem. Phys. 2008, 10,
2952–2963; c) N. Guo, B. R. Fingland, W. D. Williams, V. F. Kisper-
sky, J. Jelic, W. N. Delgass, F. H. Ribeiro, R. J. Meyer, J. T. Miller,
Phys. Chem. Chem. Phys. 2010, 12, 5678–5693.
[33] A. A. Bolzan, C. Fong, B. J. Kennedy, C. J. Howard, Acta Crystal-
logr. B 1997, 53, 373–380.
[34] a) J. Weitkamp, Handbook of Porous Solids, (Eds.: F. Janowski, D.
Enke; F. Schꢁth, K. S. W. Sing), Wiley-VCH, 3, 2002, 1432–1434;
b) S. Wohlrab, A. Janz, M.-M. Pohl, S. Kreft, D. Enke, A. Kçckritz,
A. Martin, B. Lꢁcke, Stud. Surf. Sci. Catal. 2010, 175, 315–319.
[15] a) L. Tong, L. Duan, Y. Xu, T. Privalov, L. Sun, Angew. Chem. 2011,
123, 465–469; Angew. Chem. Int. Ed. 2011, 50, 445–449; b) L.
Duan, Y. Xu, L. Tong, L. Sun, ChemSusChem 2011, 4, 238–244;
c) Y. Xu, A. Fischer, L. Duan, L. Tong, E. Gabrielsson, B. ꢄker-
mark, L. Sun, Angew. Chem. 2010, 122, 9118–9121; Angew. Chem.
Int. Ed. 2010, 49, 8934–8937; d) L. Duan, Y. Xu, M. Gorlov, L.
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

M. Bauer, M. Beller et al.
[35] a) C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli, J. S.
Beck, Nature 1992, 359, 710–712; b) J. S. Beck, J. C. Vartuli, W. J.
Roth, M. E. Leonowicz, C. T. Kresge, K. D. Schmitt, C. T. D. Chu,
D. H. Olson, E. W. Sheppard, S. B. McCullen, J. B. Higgins, J. L.
Schlenker, J. Am. Chem. Soc. 1992, 114, 10834–10843.
[36] a) F. Schꢁth, A. Wingen, J. Sauer, Microporous Mesoporous Mater.
2001, 44–45, 465–476; b) A. Taguchi, F. Schꢁth, Microporous Meso-
porous Mater. 2005, 77, 1–45.
Received: April 28, 2012
Published online: && &&, 0000
&
10
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Homogeneous Versus Heterogeneous Catalysis
FULL PAPER
Ir walks a fine line: Molecularly
defined Ir complexes and IrO2 nano-
particles have been applied in the
water oxidation reaction with cerium
ammonium nitrate (CAN) as oxidant
and compared (see scheme). The con-
version of the first to the latter has
been investigated by means of
Iridium
H. Junge, N. Marquet, A. Kammer,
S. Denurra, M. Bauer,* S. Wohlrab,
F. Gꢀrtner, M.-M. Pohl,
A. Spannenberg, S. Gladiali,
M. Beller* ..................... &&&& — &&&&
Water Oxidation with Molecularly
Defined Iridium Complexes: Insights
into Homogeneous Versus Heteroge-
neous Catalysis
XANES, EXAFS, and STEM.
Water oxidation…
…is a significant and challenging
goal. Succesfully applied molecu-
larly defined Ir complexes are
thought to be converted into Ir-
oxo nanoparticles that may also
act as catalysts. Thus, the iridium
performs “tightrope-walking”
between homogeneous and hetero-
geneous catalysis. New insights
have been revealed by XANES,
EXAFS, and STEM investigations.
For more details, see the Full
Paper by M. Beller et al. on Page
&& ff.
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!