Heteroleptic Copper Photosensitizers: Why an Extended π-System Does Not Automatically Lead to Enhanced Hydrogen Production

Article abstract of DOI:10.1002/chem.201604005

A series of heteroleptic copper(I) photosensitizers of the type [(P^P)Cu(N^N)]⁺with an extended π-system in the backbone of the diimine ligand has been prepared. The structures of all complexes are completely characterized by NMR spectroscopy, mass spectrometry, and X-ray crystallography. These novel photosensitizers were assessed with respect to the photocatalytic reduction of protons in the presence of triethylamine and [Fe₃(CO)₁₂]. Although the solid-state structures and computational results show no significant impact of the π-extension on the structural properties, decreased activities were observed. To explain this drop, a combination of electrochemical and photophysical measurements including time-resolved emission as well as transient absorption spectroscopy in the femto- to nanosecond time regime was used. Consequently, shortened excited state lifetimes caused by the rapid depopulation of the excited states located at the diimine ligand are identified as a major reason for the low photocatalytic performance.

Full text of DOI:10.1002/chem.201604005

A Journal of
Accepted Article
Title: Heteroleptic Copper Photosensitizers: Why an extended π-
system does not automatically lead to an enhanced hydrogen
production
Authors: Martin Heberle, Stefanie Tschierlei, Nils Rockstroh, Mark
Ringenberg, Wolfgang Frey, Henrik Junge, Matthias Beller,
Stefan Lochbrunner, and Michael Karnahl
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201604005
Link to VoR: http://dx.doi.org/10.1002/chem.201604005
Supported by

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
Heteroleptic Copper Photosensitizers: Why an extended π-system
does not automatically lead to an enhanced hydrogen production
Martin Heberle,[a]+ Stefanie Tschierlei,[a, b]+ Nils Rockstroh,[c] Mark Ringenberg,[d] Wolfgang Frey,[a]
Henrik Junge,[c] Matthias Beller,[c] Stefan Lochbrunner,[b] and Michael Karnahl*[a]
Abstract: A series of heteroleptic copper(I) photosensitizers of the
type [(P^P)Cu(N^N)]+ with an extended π-system in the backbone of
the diimine ligand have been prepared. The structures of all
complexes are completely characterized by NMR spectroscopy, mass
methanol.4-6 These energy carriers are ideally derived by
processes inspired by natural photosynthesis such as the
photocatalytic splitting of water or the light-driven reduction of CO2,
respectively.7-9 In order to realize such reactions light energy has
to be efficiently captured by photosensitizers, which are able to
spectrometry
and
X-ray
crystallography.
These
novel
photosensitizers were assessed with respect to the photocatalytic
reduction of protons in the presence of triethylamine and [Fe3(CO)12].
While the solid state structures and computational results show no
significant impact of the π-extension on the structural properties
transfer this energy to a
suitable catalyst.10-12 Moreover,
replacement of conventional noble metal compounds by cheap
and abundant 3d metals is the key to bring such molecular
systems into practice.13, 14
decreased activities were observed. To explain this drop,
a
Since decades, the study of photosensitizers is primarily
focused on noble metal complexes of ruthenium, iridium, rhenium
or platinum, with Ru(II) polypyridines and cyclometalated Ir(III)
complexes are by far the most popular.10-12 In the past,
luminescent Cu(I) complexes were already used in a wide range
of applications such as organic light-emitting diodes (OLEDs),15-
combination of electrochemical and photophysical measurements
including time-resolved emission as well as transient absorption
spectroscopy in the femto- to nanosecond time regime were used.
Consequently, shortened excited state lifetimes caused by the rapid
depopulation of the excited states located at the diimine ligand are
identified as a major reason for the low photocatalytic performance.
17
19
light-emitting electrochemical cells (LECs),18,
oxygen
sensors20 or dye-sensitized solar cells (DSSCs).21-23 However, the
great potential of this class of compounds for direct photocatalytic
applications has been overlooked for a long time.24, 25
Introduction
In 2013, a fully noble-metal-free system for the light-driven
production of hydrogen from water was reported, by using a
heteroleptic copper(I) photosensitizer (CuPS) in combination with
an iron-based water reduction catalyst (WRC).26, 27, 14 For the first
time, it has been successfully demonstrated that heteroleptic
copper complexes with the general formula [(P^P)Cu(N^N)]+,
merging a diimine N^N and bulky diphosphine P^P chelate ligand,
enable high photocatalytic activities. Since this report, several
heteroleptic CuPS have been prepared and systematically tested
in the photocatalytic reduction of protons from water in the
presence of [Fe3(CO)12] as WRC and triethylamine as sacrificial
electron donor (SR).27, 28, 29 In the best systems, turnover numbers
The increasing world population causes a strongly rising energy
consumption, which is accompanied by vast emissions of CO2
and other greenhouse gases due to the burning of fossil fuels.1-3
Their contribution to environmental damage as well as their
limited availability stimulate the development of alternative and
more sustainable energy carriers like hydrogen, formic acid or
[a]
M. Heberle, Dr. S. Tschierlei, Dr. W. Frey, Dr. M. Karnahl
Institute of Organic Chemistry
University of Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart, Germany
E-mail: michael.karnahl@oc.uni-stuttgart.de
Homepage: www.karnahl-group.de
Dr. S. Tschierlei, Prof. Dr. S. Lochbrunner
Institute of Physics
(TONs) of up to 1330 were obtained, strongly depending on the
28
respective CuPS.27,
It has been found, that the ligand
environment and the substituents on these ligands allow for fine
tuning of the photophysical and electrochemical properties of the
resulting photosensitizers. For instance, introduction of alkyl-
substituents at the 2,9-position of 1,10-phenanthroline (phen)
leads to an increased excited state lifetime due to sterical
hampering of structural relaxation by Jahn-Teller distortion, which
decelerates undesired exciplex quenching.30-32 However, these
heteroleptic Cu(I) complexes still suffer from limited absorption in
the visible region and lower MLCT extinction coefficients,
especially when compared to advanced Ru(II) and Ir(III)
photosensitizers.10, 33, 34
[b]
University of Rostock
Albert-Einstein-Str. 23, 18059 Rostock, Germany
Dr. N. Rockstroh, Dr. H. Junge, Prof. Dr. M. Beller
Leibniz-Institute for Catalysis at the University of Rostock (LIKAT)
Albert-Einstein-Str. 29a, 18059 Rostock, Germany
Dr. M. Ringenberg
Institute of Inorganic Chemistry
University of Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart, Germany
These authors contributed equally to the work.
[c]
[a]
+
Supporting information for this article, including experimental and
synthetic details, crystallographic data, NMR, steady-state and
transient absorption spectra as well as photocatalytic measurements
and (TD)DFT calculations, is given via a link at the end of the
document.
1
This article is protected by copyright. All rights reserved.

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
extended π-system. Upon purification the heteroleptic CuPS 1-3
were obtained as yellow to orange solids in high yields from 65-
90% (for further details see SI). In contrast, the homoleptic Cu(I)
complex 4 (Figure 1), which was required as reference compound
for comparison, had a red color and was prepared in 90% yield
from an ethanol/water (4/1) mixture. All compounds were found to
be stable towards air and moisture under ambient conditions.
Scheme 1. General concept for the extension of the π-system starting from the
2,9-dimethyl-1,10-phenanthroline (Me2phen) moiety, including its numbering
scheme.
A promising method to increase extinction coefficients and to red-
shift the absorption is to extend the conjugated π-system.35-37
Therefore, the present study aims at the investigation of this issue
starting from 2,9-dimethyl-1,10-phenanthroline (Me2phen) and
turning to diimine ligands with an extended π-system, like 3,6-
dimethyl-dipyrido[3,2-f:2′,3′-h]-quinoxaline (Me2dpq) and 3,6-
dimethyl-dipyrido[3,2-a:2',3'-c]phenazine (Me2dppz) (Scheme 1).
In particular, dppz and derivatives thereof already have a long
history and are well-known in the field of ruthenium chemistry.38-
46
Figure 1. Overview of the heteroleptic Cu(I) complexes 1-3 and the homoleptic
reference complex 4 investigated in this study.
Consequently, the synthesis, spectroscopic and electrochemical
characterization as well as the photocatalytic activity of a
systematic series of novel heteroleptic Cu(I) complexes are
reported. All compounds were fully analyzed by NMR, MS, UV/vis
and emission spectroscopy as well as electrochemical
measurements. In addition, the solid-state structures of all three
complexes were determined crystallographically and are
compared with the results of density functional theory (DFT)
calculations. Further, the respective absorption spectra were
calculated by time-dependent DFT methods to assist the
assignment of the electronic transitions. Femto- and nanosecond
transient absorption spectroscopy have been applied to
understand the underlying electron transfer processes within this
class of photosensitizers. The combination of these different
spectroscopic and theoretical techniques helped to generate a
better understanding of the catalytic performance. Finally, based
on these findings future design strategies towards improved
CuPS’s are proposed, that challenge non-abundant noble metal
systems for solar fuels production.
NMR measurements (1H-, 13C- and 31P-NMR) and mass spectra
(see SI) are all in agreement with the proposed structure. In each
case the [M-PF6]+ peak could be assigned to the corresponding
fragment with matching isotopic pattern. 31P{1H}-NMR spectra of
the heteroleptic complexes 1-3 feature a single resonance around
-13 ppm (Figure SI1), which is also characteristic for structurally
related compounds.26, 27 The small differences in the 31P{1H}-NMR
shift (1: -13.44 ppm, 2: -12.75 ppm and 3: -12.63 ppm) indicate
that the extended π-system has only a marginal influence on the
structure around the copper center when changing from Me2phen
(1) to Me2dpq (2) and Me2dppz (3).
Single crystals of 1-3 suitable for X-ray crystallography were
obtained by slow evaporation from saturated dichloromethane/n-
hexane mixtures. The structural analysis (Figure 2 and Figure
SI2) revealed a distorted tetrahedral geometry around the copper
center, which is common for this class of copper complexes.25, 27,
47
This distortion can be explained by the different bite angles of
the bidentate N^N (e.g. N1-Cu-N2 for 1: 80.53(13) °, Table 1) and
the P^P ligand (e.g. P1-Cu-P2 for 1: 112.93(4) °). Furthermore, in
accordance with the larger size of the phosphorus atom the
coordination bond length of Cu-P1 is significantly longer (e.g. for
1: 226.26(11) pm) than that of Cu-N1 (for 1: 208.4(3) pm) in all
complexes. Structural comparison of 1-3 (Table 1) does not show
any major effect on the coordination geometry due to the addition
of the extended π-system, which is in agreement with the results
from NMR spectroscopy. However, a small raise in the P1-Cu-P2
angle can be recognized, when comparing complexes 2 and 3
with 1. This might be explained by a π-π–interaction of one phenyl
ring of the Xantphos ligand with the extended π-system of the
Me2dpq and the Me2dppz ligand, respectively. Interestingly, there
are no stacked dimers observed in the solid state of complex 3
Results and Discussion
Synthesis and structural characterization: The heteroleptic
copper(I) complexes 1-3 (Figure 1) were synthesized starting
from the [Cu(ACN)4]PF6 (ACN = acetonitrile) precursor via an
one-pot two-step procedure as reported previously.26, 27 Under
inert conditions (N2 atmosphere) the diphosphine ligand is
introduced first, forming the proposed [(P^P)Cu(ACN)2]+-
intermediate, followed by fast substitution of the remaining
acetonitrile ligands by the diimine. Xantphos (Xant) was chosen
as suitable diphosphine ligand due to its bulkiness and rigidity,
which were identified as beneficial properties in previous studies26,
27
and kept constant in order to elucidate the sole effect of the
2
This article is protected by copyright. All rights reserved.

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
(Figure SI3), that are typical for other coordination compounds
containing such a planar dppz ligand.36, 48
charge transfer (MLCT) transitions to the phenanthroline sphere
of the diimine ligand (Figure 4) as discussed below.
Table 1. Selected crystallographic bond lengths (pm) and angles (°) of
complexes 1-4. For atom numbering see also Figure 2 and the supporting
information. The CCDC reference numbers are 1500699 (1), 1500700 (2) and
1500702 (3).
1
2
3
4[a]
Cu-N1
208.4(3)
211.2(3)
226.26(11)
228.63(13)
80.53(13)
112.93(4)
108.77(9)
120.33(9)
120.33(9)
118.59(9)
76.9
209.1(2)
208.3(2)
227.46(8)
223.50(8)
80.49(9)
119.33(3)
120.03(7)
102.84(6)
118.31(7)
108.83(7)
76.9
211.7(2)
209.8(2)
227.98(7)
225.98(7)
80.02(8)
117.72(3)
107.72(6)
115.75(6)
105.79(6)
123.37(6)
71.8
205.3(3)
Cu-N2
203.4(3)
Cu-P1
-
Cu-P2
-
N1-Cu-N2
P1-Cu-P2
N1-Cu-P1
N1-Cu-P2
N2-Cu-P1
N2-Cu-P2
P-P-N-N
82.52(12)
-
-
-
-
-
-
[a] Data was taken from reference 40 and the respective structure with the
CCDC number 712222.
Electrochemical
and
Photophysical
studies:
The
electrochemical properties of the compounds 1-4 were studied by
cyclic voltammetry (Figure 3 and Table 2). The respective cyclic
voltammograms of the heteroleptic complexes 1-3 contain a
reversible reduction wave and an irreversible oxidation wave in
acetonitrile solution. The reversible one-electron reduction was
previously assigned to a reduction of the diimine ligand forming a
[(P^P)Cu(I)(N^N)-] species and is in accordance to related copper
complexes.27, 50, 51 The impact of the omitted phenyl groups on the
reduction potential is negligible as that of 1 (−2.10 V) is nearly
equal to that of [(Xant)Cu(Me2phenPh2)]+ (−2.05 V, with
Me2phenPh2 = 2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline). In
contrast, with increasing extension of the π-system the reduction
potentials are anodically shifted in the order of −2.10 V (1) < −1.92
V (2) < −1.49 V (3). A shift of more than 600 mV between 1 and 3
clearly demonstrates the pronounced impact of the extended π-
system on the electronic properties of the Me2dppz ligand in
comparison to Me2phen. Therefore, it is interesting, if this effect is
also reflected in the photocatalytic performance of these
photosensitizers as discussed below.
Figure 2. Solid state structures (ORTEP representation) of complex 2 (A - top)
and 3 (B - bottom) with thermal ellipsoids at a probability level of 50 %. Hydrogen
atoms, counter anions and solvent molecules are omitted for clarity. For
comparison with complex 1 see Figure SI2.
In addition to X-ray analysis density functional theory (DFT)
calculations provide further insight into structural characterization
(for details see SI). For example, the calculated Cu-N bonds of 1-
3 (e.g. for 1: 213 pm, TableSI2 and Figure SI4) are increased
compared to the homoleptic complex 4 (204 pm), while the N-Cu-
N bite angles of 1-3 (e.g. for 1: 79.7°) are smaller as in 4 (82.8°),
supporting the results from X-ray crystallography. This behavior
can be explained by the introduction of the sterically more
demanding Xantphos ligand.10 As a consequence a blue shift of
the MLCT transition band in the absorption spectra of the
complexes 1-3 can be noticed compared to 4 as will be discussed
below (Figure 4). Apart from that, the extended π-system has only
a negligible influence on the coordination of the diimine ligand to
the Cu center where no changes of the Cu-N bond lengths and
the N1-N2 distances were observed (Table SI2). Consequently,
this should merely lead to slight changes in the metal-to-ligand
3
This article is protected by copyright. All rights reserved.

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
Figure 3. Cyclic voltammograms of 1 (black), 2 (dark grey) and 3 (grey) in
acetonitrile solution referenced vs. the Fc/Fc+ couple. Conditions: scan rate of
100 mVs-1, 0.1 M Bu4NPF6 as supporting electrolyte.
Figure 4. Absorption spectra (A – top) and emission spectra (B - bottom) of 1
(black), 2 (dashed dark grey), 3 (dotted grey) and 4 (solid grey) in acetonitrile
solution. The emission spectra are normalized to the Raman peak of acetonitrile
at 439 nm.
In contrast to the cathodic scans, the anodic region in the cyclic
voltammograms of 1-3 (Figure 3) exhibit an irreversible oxidation
wave. This finding is consistent to analogous heteroleptic CuPS
and most likely caused by a dissociation of the P-Cu bond
resulting in a decomposition of the Cu(II) species.29, 50, 52
Under aerobic conditions the emission of the heteroleptic
complexes 1 and 2 is very weak with emission maxima at 564 and
577 nm, while 3 emits no light (Figure 4). The emission behavior
of 1 seems unusual at the first sight as the structurally related
The heteroleptic complexes 1-3 strongly absorb light in the
UV region, but also in the visible up to 450 nm (Figure 4A), which
results from ligand-centered and metal-to-ligand charge transfer
(MLCT) transitions, respectively. The absorption band above 355
nm can be assigned to transitions from the d-orbitals of the copper
center to the π*-orbital of the diimine ligand10,30 and not to the
Xantphos ligand. As a consequence the extinction coefficients in
the range of the MLCT transitions of 1 and 2 are lower than that
of 4 (Table 2). This behavior was also observed for related copper
complexes30 and is further substantiated by time-dependent DFT
calculations (Figure SI6). From the calculated MLCT transitions in
the visible region it became apparent that the involved π*-orbitals
are mainly located at the Me2phen (1) and at the Me2dpq (2)
ligand (Table SI3 and Figure SI7). In the case of 3 the MLCT
transitions either end in the π*-orbital that is delocalized over the
whole Me2dppz ligand or in the phenanthroline sphere of the
Me2dppz ligand. The contribution of the whole Me2dppz ligand to
the absorption is three times smaller than that of the
phenanthroline part of Me2dppz (Table SI3 and Figure SI6),
resulting only in a slightly broadened MLCT transition band.
However, with an increasing extent of the π-system (Me2phen <
Me2dpq < Me2dppz), the calculated absorption band around 380
nm (Figure SI6) is slightly shifted towards higher wavelengths.
This behavior is nearly negligible in the experimental spectra
(Figure 4A) and in 3 the MLCT transitions are overlaid by the π-
π* transitions of the dppz ligand around 370 nm. Further, the
calculations reveal that the MLCT transition to the phosphine
ligand needs much more energy (approx. 308 nm, Table SI3), and
thus, is superimposed by the different ligand-centered transitions
and does not contribute to photocatalysis in the visible.
complex [(Xant)Cu(Me2phenPh2)]+ emits at 569 nm with
a
quantum yield of 8 %.27 However, it was found that the two
additional phenyl groups possess a strong impact on the radiative
rates of the emitting states, which results in a greatly enhanced
emission intensity.54 Consequently, the emission intensity of 1,
which exhibits no phenyl groups at the diimine ligand, is
significantly lower.
In contrast to the electrochemical studies, where the reduction
potential is clearly shifted by the extended π-system, the impact
on the absorption and emission properties is comparatively weak
(Table 2).
Table 2. Summary of the photophysical and electrochemical properties of the
heteroleptic complexes 1-3 and of the homoleptic reference 4 in acetonitrile
solution at room temperature.

abs [nm]

em [nm]
 [ps] [a]
 [ns]
Ered [V]
[b]
( [10³ M-1 cm-1])
1
2
3
378 (3.1)
564
0.9, 8.9
64.0
28.0
14.5
-2.10
-1.92
-1.49
379 (3.1)
577
1.1, 13.7
0.9, 8.7, 130
364 (13.5) [c]
382 (11.7)
n.d.[d]
4
455 (7.0) [e]
700 [e]
0.9, 9.8 [f]
54.0 [e]
-2.17 [g]
[a] Time constants from the nonlinear least-square fitting of the transient
kinetics obtained by femtosecond transient absorption spectroscopy, [b] in
acetonitrile vs. Fc/Fc+, [c] extinction coefficients are higher due to ligand-
centered transitions at the Me2dppz, [d] the emission intensity was below the
detection limit, [e] in dichloromethane, data was taken from reference 10, [f]
data was taken from reference 32, [g] irreversible reduction.
4
This article is protected by copyright. All rights reserved.

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
Photocatalytic hydrogen production: Based on previous
findings26, 27 the newly developed Cu(I) complexes have been
evaluated in the photocatalytic hydrogen production from protons
together with [Fe3(CO)12] as the water reduction catalyst and
triethylamine (TEA) as sacrificial reductant to study the influence
of the extended π-system on the photocatalytic activity.
The hydrogen evolution measurements reveal a strong influence
of the diimine ligand on the overall catalytic yield (Figure 5). While
complex 1 still produces 16.0 mL of hydrogen, which corresponds
to a maximum TON of 305 (Table 3), the complexes 2 and 3
possess a very low activity. With a maximum volume of 3.9 mL
(2) and 4.0 mL (3) the hydrogen evolution of the two copper
complexes with an extended π-system is close to the blank
volume of the system, which is about 2.8 mL. This blank volume
represents the maximum uncertainty of the photocatalytic system
and results from several independent blank measurements
without any catalyst and photosensitizer.
Table 3. Overview of photocatalytic H2 production with the copper-based
photosensitizers 1-4.[a]
[c]
Complex
Vmax [mL]
Vcorr [mL] [b]
TONH,Cu
1
2
3
4
16.0
3.9
13.2
1.1
305
25
4.0
1.2
29
5.4
2.6
60
[a] Conditions: CuPS (ca. 3.5 µmol), [Fe3(CO)12
]
(ca. 5.0 µmol),
THF/TEA/H2O (4:3:1, 10 mL), 25°C, Xe light irradiation (output 1.5 W)
without light filter, 24 h [b] Vmax corrected by a blank volume of 2.8 mL [c]
TONH,Cu = n(H) / n(CuPS) with Vm,H2,25°C = 24.48 mL/mmol.
Beside this, complex 1 exhibits a fairly constant production
of hydrogen for the first couple of hours with a linear rise until 12 h.
After this period the activity is strongly decreased reaching a
plateau after 15 h (Figure 5). This is most likely caused by a
decomposition of the [Fe3(CO)12] due to CO ligand loss and by
dissociation and transformation of the diphosphine from the
heteroleptic CuPS as evident from former studies.29, 50, 52, 55
Ultrafast spectroscopy: From other CuPS it is known that light
irradiation induces the initial MLCT transfer and subsequently a
geometric distortion followed by an intersystem crossing (ISC)
30, 32
occurs.53,
Femtosecond time-resolved spectroscopy of
complexes 1-3 (Figure 6) revealed that the flattening from a
(distorted) tetrahedral to a more square planar structure takes
place in approx. 1 ps (Table 2). The ISC is accomplished within
8.9 (1) and 13.7 ps (2), respectively, which is in the same range
as observed for similar heteroleptic CuPS.30 In 2 the ISC is
decelerated compared to 1 probably due to the extended π-
system and the consequentially changed electron density at the
phenanthroline sphere. While these processes are the only ones
in 1 and 2, there are significant differences for complex 3. In 3,
the second time constant with 8.7 ps is comparable to the required
time for ISC of 1. In this case the extended π-system of the
Me2dppz does not lead to a slowdown of the process. This could
be caused by the fact that the MLCT is mainly located on the
phenanthroline part of the Me2dppz ligand (Scheme 1) and not on
the whole diimine ligand as in 2, where the ISC takes more time.
Additionally, for 3 a third time constant of 130 ps was determined.
The global analysis of the data (Figure SI10) revealed that this
time constant is associated with a depopulation of the MLCT
excited state. As known from related ruthenium photosensitizers
this process can be assigned to the depopulation of the emitting
3MLCT state (the so-called bright state), with the excited electron
located at the phenanthroline part, and the population of a dark
3MLCT state, in which the excited electron is located at the
phenazine part of the dipyridophenazine (dppz) ligand while the
hole is still at the copper atom.44-46 Accordingly, the relaxation
dynamics corresponds to an electron transfer from the
phenanthroline to the phenazine moiety. For [(tbbpy)Ru(dppz)]2+
(with tbbpy = 4,4’-di-tert-butyl-2,2’-bipyridine) this transfer occurs
in 150 ps in acetonitrile solution,36 which is very similar to the
A
possible explanation for the comparatively low
performance of 1-3 in photocatalytic hydrogen evolution
experiments can be seen in the location of the excited electron
and the lifetime of the excited states. In contrast to the previously
investigated heteroleptic CuPS, no or only a weak emission were
detected for 1-3. However, the emission behavior alone does not
provide information about radiationless deactivation and electron
transfer processes. Thus, the excited states of 1-3 have been
studied by transient absorption spectroscopy in the femto- to
nanosecond timescale to reveal the nature of the underlying
transfer steps, especially the deactivation processes.
Figure 5. Hydrogen evolution curves for the photocatalytic reduction of protons
using the different copper photosensitizers (ca. 3.5 µmol) 1 (black), 2 (dashed
dark grey), 3 (dotted grey) and 4 (grey) in the presence of [Fe3(CO)12] (ca. 5.0
µmol) as water reduction catalyst in a solvent mixture of THF/TEA/H2O (4/3/1)
(light source: Xenon lamp, output: 1.5W, no filter, for further information see SI).
process in 3. Moreover, the match of the absorption spectra of the
39-41
electrochemically reduced dppz- and [(bpy)2Ru(dppz)]+
as
well as the similarity of the transient absorption spectra of
[(bpy)2Ru(dppz)]2+ 39 and 3, especially above 500 nm, suggest
that a photoinduced electron transfer process to the phenazine
part occurs. Consequently, this is a strong indication that the dppz
5
This article is protected by copyright. All rights reserved.

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
behaves in a similar way upon coordination to copper or
similarities in the photophysics of Ru(II) and Cu(I) dppz
complexes.
All in all, the obtained excited state lifetimes of these novel
Cu(I) complexes are too short to enable sufficient collisions
between the excited CuPS and the catalyst, providing a potential
explanation for the low catalytic activity.
ruthenium causing
cases.41, 44-46
a bichromophoric characteristic in both
Finally, the population of this non-emitting dark state on the
phenazine moiety of complex 3 can be considered as the reason
for the observed absence of emission. It has a pronounced charge
transfer character with practically no spatial overlap between the
hole located at the Cu atom and the electron at the phenazine
moiety. In other words, the active excited states of 3 are
deactivated by non-radiative processes providing a hint for the low
performance in the photocatalytic hydrogen production.
Figure 7. Transient absorption spectra after 10 ns (A) and kinetics at 520 nm
(B) of 1 (black, black circles), 2 (dashed dark grey, triangles) and 3 (dotted grey,
pluses) in acetonitrile excited with 390 nm nanosecond laser pulses.
Figure 6. Transient absorption spectra (A) of 3 in acetonitrile excited at 388 nm
after 1.5 (black), 5 (red), 50 (green), 100 (blue), 200 (cyan) and 900 ps
(magenta) and kinetics (B) at 370 (black), 430 (red), 520 (green) and 605 nm
(blue). The grey lines represent a triple-exponential fit with 1 = 0.9 ps, 2 = 8.7
ps and 3 = 130 ps.
Conclusions
A series of heteroleptic copper(I) photosensitizers (CuPS) with
the general formula [(P^P)Cu(N^N)]+, where N^N represents a
diimine ligand with an extended π-system at the backbone, were
synthesized for the first time. These novel photosensitizers have
been applied in proton reduction experiments within a fully noble
metal-free system using an iron based catalyst and triethylamine
as sacrificial reductant in order to assess their photocatalytic
Furthermore, transient absorption measurements revealed the
population of triplet excited states. The deactivation of these
states could not be identified by using steady-state or time-
resolved emission spectroscopy because the depopulation
occurs radiationless. Thus, nanosecond transient absorption
spectroscopy has been used to investigate the long-lived
processes. For 1 and 2 the transient spectra (Figure 7A) after
10 ns are similar to that after 100 ps (Figure SI10). For both
complexes the shape of the spectra, with maxima at 524 (1) and
532 nm (2), are equal indicating similar populated triplet excited
states. The depopulation occurs within 64 ns for 1 and 28 ns for 2
according to the fit of the respective kinetics (Figure 7B). The
transient absorption spectrum of 3 after 10 ns is similar to the
spectrum after 500 ps (Figure 6A), but different to those of 1 and
2 due to the localization of the excited state on the phenazine part
of Me2dppz. The depopulation of this dark state occurs within 14.5
ns. Thus, compared to 1 and 2 the deactivation of 3 is the fastest
one.
capability.
Unfortunately,
the
performance
of
these
photosensitizers is lower compared to related copper
photosensitizers.
This behavior is unexpected, because investigation of 1-3
by means of 1H- and 31P-NMR spectroscopy as well as mass
spectrometry and X-ray crystallography revealed no significant
impact of the extended π-system on the structural features
around the Cu(I) ion. Therefore, a comprehensive combination of
electrochemical and photophysical measurements were used to
clarify the bottleneck for catalysis.
For instance, cyclic voltammetry revealed a pronounced
anodic shift of the reduction potential in the order of 1 < 2 < 3. A
difference of more than 600 mV between 1 and 3 clearly
demonstrates the strong influence of the increasing size of the π-
system on the electronic properties of the respective CuPS.
In contrast, the effect on the absorption spectra is
comparatively weak. The low energy absorption above 360 nm is
dominated by the metal-to-ligand charge transfer transitions and
its energy is only slightly lowered with increasing size of the π-
Furthermore, it is worth to mention, that the transient
absorption spectrum of 3 after 10 ns (Figure 7A) is very similar to
the spectra of the related [(bpy)2Ru(dppz)]2+ complex published
by Amouyal and coworkers.39 This underlines again the
6
This article is protected by copyright. All rights reserved.

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
[6]
M. Nielsen, E. Alberico, W. Baumann, H.-J. Drexler, H. Junge, S. Gladiali,
M. Beller Nature 2013, 495(7439), 85–89.
system. This is caused by the contribution of different π*-orbitals
to the relevant optical processes. In 1 and 2 the π*-orbitals are
distributed over the whole diimine ligand, i.e. 2,9-dimethyl-1,10-
[7]
[8]
J. Barber, P. D. Tran J. R. Soc. Interface 2013, 10(81), 20120984.
A. Thapper, S. Styring, G. Saracco, W. A. Rutherford, B. Robert, A.
Magnuson, W. Lubitz, A. Llobet, P. Kurz, A. Holzwarth, S. Fiechter, H.
de Groot, S. Campagna, A. Braun, H. Bercegol, V. Artero 2013, 3(1), 43–
57.
phenanthroline
and
3,6-dimethyl-dipyrido-quinoxaline,
respectively (see Figure SI7 in the SI). Instead, in complex 3 the
metal-to-ligand charge transfer transitions involves either the
phenanthroline or the phenazine sphere of the 3,6-dimethyl-
dipyridophenazine ligand, illustrating its bichromophoric character.
However, the visible absorption is hardly affected by the fused
aromatic rings at the 5,6-position of the phenanthroline. The
extension of the π-system is also responsible for the very weak
emission of the respective copper complexes, which implies rapid
deactivation processes.
Therefore, transient absorption spectroscopy was used to
determine the radiationless deactivation processes of the excited
states, which are clearly accelerated by the extension of the π-
system. In 1 and 2 the spectral shape and position of the excited
states are similar with short lifetimes of 64 and 28 ns, respectively.
In 3 another low lying MLCT state, with the excited electron
located at the phenazine sphere, is populated within 130 ps. From
there the deactivation occurs even faster in only 14.5 ns due to its
reduced energy gap to the ground state. This observation is in
agreement with related Ru(II) dipyridophenazine complexes,
which shows that dppz behaves in a similar way whether it is
coordinated to copper or ruthenium.
[9]
E. E. Benson, C. P. Kubiak, A. J. Sathrum, J. M. Smieja Chem. Soc. Rev.
2009, 38(1), 89–99.
[10] N. Armaroli Chem. Soc. Rev. 2001, 30(2), 113–124.
[11] M. Schulz, M. Karnahl, M. Schwalbe, J. G. Vos Coord. Chem. Rev. 2012,
256, 1682–1705.
[12] P. D. Frischmann, K. Mahata, F. Würthner Chem. Soc. Rev. 2013, 42(4),
1847–1870.
[13] W. T. Eckenhoff, R. Eisenberg Dalton Trans. 2012, 13004–13021.
[14] B. van den Bosch, H.-C. Chen, J. I. van der Vlugt, A. M. Brouwer, J. N.
H. Reek ChemSusChem 2013, 6(5), 790–793.
[15] H. Yersin, A. F. Rausch, R. Czerwieniec, T. Hofbeck, T. Fischer Coord.
Chem. Rev. 2011, 255(21-22), 2622–2652.
[16] M. Hashimoto, S. Igawa, M. Yashima, I. Kawata, M. Hoshino, M. Osawa
J. Am. Chem. Soc. 2011, 133(27), 10348–10351.
[I17] S. Igawa, M. Hashimoto, I. Kawata, M. Yashima, M. Hoshino, M. Osawa
J. Mater. Chem. C 2013, 1(3), 542–551.
[18] N. Armaroli, G. Accorsi, M. Holler, O. Moudam, J.-F. Nierengarten, Z.
Zhou, R. Wegh, R. Welter Adv. Materials 2006, 18(10), 1313–1316.
[19] R. D. Costa, D. Tordera, E. Orti, H. J. Bolink, J. Schonle, S. Graber, C.
E. Housecroft, E. C. Constable, J. A. Zampese J. Mater. Chem. 2011,
21(40), 16108–16118.
[20] X.-d. Wang, O. S. Wolfbeis Chem. Soc. Rev. 2014, 43(10), 3666–3761.
[21] C. Bignozzi, R. Argazzi, R. Boaretto, E. Busatto, S. Carli, F. Ronconi, S.
Caramori Coord. Chem. Rev. 2013, 257(9-10), 1472–1492.
[22] L. N. Ashbrook, C. M. Elliott J. Phys. Chem. C 2013, 117(8), 3853–3864.
[23] C. E. Housecroft, E. C. Constable Chem. Soc. Rev. 2015, 44(23), 8386–
8398.
All in all, the lifetimes of the excited states of these novel
photosensitizers are too short, which hampers an efficient
interaction with the iron based catalyst.
The results show that the performance of a photosensitizer
is not only dependent on the metal-to-ligand charge transfer
transitions (HOMO-LUMO gap). In fact, advanced substitution
schemes have to be developed, which specifically address the
orbitals involved in the lowest absorption band to adjust the
properties of the excited states (energetic position and lifetime) in
order to improve the photocatalytic activity.
[24] A. Edel, P. A. Marnot, J.-P. Sauvage New. J. Chem. 1984, 8, 495-498.
[25] S.-M. Kuang, D. G. Cuttell, D. R. McMillin, P. E. Fanwick, R. A. Walton
Inorg. Chem. 2002, 41(12), 3313–3322.
[26] S.-P. Luo, E. Mejía, A. Friedrich, A. Pazidis, H. Junge, A.-E. Surkus, R.
Jackstell, S. Denurra, S. Gladiali, S. Lochbrunner, M. Beller Angew.
Chem. 2013, 125(1), 437–441; Angew. Chem. Int. Ed. 2013, 52(1), 419–
423.
[27] E. Mejía, S.-P. Luo, M. Karnahl, A. Friedrich, S. Tschierlei, A.-E. Surkus,
H. Junge, S. Gladiali, S. Lochbrunner, M. Beller Chem. Eur. J. 2013,
19(47), 15972–15978.
Acknowledgements
[28] M. Karnahl, E. Mejía, N. Rockstroh, S. Tschierlei, S.-P. Luo, K. Grabow,
A. Kruth, V. Brüser, H. Junge, S. Lochbrunner, M. Beller ChemCatChem
2014, 6, 82–86.
The authors thank Rüya Duran (University of Stuttgart) and Petra
Bartels (LIKAT Rostock) for experimental support. M.K. gratefully
acknowledges funding by the Fonds der Chemischen Industrie
(FCI) and the University of Stuttgart for a starting grant. The
LIKAT Rostock thanks for financial support within a joint
DFG/NFSC project.
[29] A. J. J. Lennox, S. Fischer, M. Jurrat, S.-P. Luo, N. Rockstroh, H. Junge,
R. Ludwig, M. Beller Chem. Eur. J. 2016, 22(4), 1233–1238.
[30] S. Tschierlei, M. Karnahl, N. Rockstroh, H. Junge, M. Beller, S.
Lochbrunner ChemPhysChem 2014, 15(17), 3709–3713.
[31] M. W. Mara, K. A. Fransted, L. X. Chen Coord. Chem. Rev. 2015, 282–
283, 2–18.
Keywords: copper complexes • hydrogen production •
[32] M. Iwamura, S. Takeuchi, T. Tahara Acc. Chem. Res. 2015, 48(3), 782–
791.
photocatalysis • photophysics • transient absorption
[33] M. S. Lazorski, F. N. Castellano Polyhedron 2014, 82, 57–70.
[34] S. Paria, O. Reiser ChemCatChem 2014, 6(9), 2477–2483.
[35] B. Schäfer, H. Görls, S. Meyer, W. Henry, J. G. Vos, S. Rau Eur. J. Inorg.
Chem. 2007, (25), 4056–4063.
[1]
[2]
[3]
N. S. Lewis, D. G. Nocera Proc. Natl. Acad. Sci. U.S.A. 2006, 103(43),
15729–15735.
N. Armaroli, V. Balzani Angew. Chem. Int. Ed. 2007, 119(1-2), 52–67;
Angew. Chem. Int. Ed. 2007, 46(1-2), 52–66.
[36] C. Kuhnt, M. Karnahl, S. Tschierlei, K. Griebenow, M. Schmitt, B. Schafer,
S. Krieck, H. Görls, S. Rau, B. Dietzek, J. Popp Phys. Chem. Chem. Phys.
2010, 12(6), 1357–1368.
Q. Schiermeier, J. Tollefson, T. Scully, A. Witze, O. Morton Nature 2008,
454, 816-823.
[4]
[5]
S. Styring Faraday Discuss. 2012, 357–376.
[37] L. Troian-Gautier, C. Moucheron Molecules 2014, 19(4), 5028–5087.
[38] M. N. Ackermann, L. V. Interrante, Inorg. Chem. 1984, 23, 3904–3911.
A. Boddien, D. Mellmann, F. Gärtner, R. Jackstell, H. Junge, P. J. Dyson,
G. Laurenczy, R. Ludwig, M. Beller Science 2011, 333(6050), 1733-1736.
7
This article is protected by copyright. All rights reserved.

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
[39] E. Amouyal, A. Homsi, J.-C. Chambron, J.-P. Sauvage J. Chem. Soc.
Dalton Trans. 1990, 1841–1845.
[40] J. Fees, W. Kaim, M. Moscberosch, W. Matheis, J. Klíma, M. Krejcik, S.
Zális, Inorg. Chem. 1993, 32, 166–174.
[41] J. Fees, M. Ketterle, A. Klein, J. Fiedler, W. Kaim J. Chem. Soc., Dalton
Trans. 1999, 2595–2599.
[42] A. E. Friedman, J.-C. Chambron, J.-P. Sauvage, N. J. Turro, J. K. Barton
J. Am. Chem. Soc. 1990, 112(12), 4960–4962.
[43] R. M. Hartshorn, J. K. Barton J. Am. Chem. Soc. 1992, 114(15), 5919–
5925.
[44] M. K. Brennaman, J. H. Alstrum-Acevedo, C. N. Fleming, P. Jang, T. J.
Meyer, J. M. Papanikolas, J. Am. Chem. Soc., 2002, 124, 15094–15098.
[45] M. K. Brennaman, T. J. Meyer, J. M. Papanikolas J. Phys. Chem. A 2004,
108, 9938–9944.
[46] J. Olofsson, L. M. Wilhelmsson, P. Lincoln, J. Am. Chem. Soc. 2004, 126
(47), 15458-15465.
[47] D. G. Cuttell, S.-M. Kuang, P. E. Fanwick, D. R. McMillin, R. A. Walton J.
Am. Chem. Soc. 2001, 124(1), 6–7.
[48] R. Liu, M.-M. Huang, X.-X. Yao, H.-H. Li, F.-L. Yang, X.-L. Li Inorg. Chim.
Acta 2015, 434, 172–180.
[49] J. A. Rusanova, O. V. Kozachuk, V. V. Semenaka, V. V. Dyakonenko
Acta Cryst. E 2013, 69(12), m684–m685.
[50] S.Fischer, D. Hollmann, S. Tschierlei, M. Karnahl, N. Rockstroh, E.
Barsch, P. Schwarzbach, S.-P. Luo, H. Junge, M. Beller, S. Lochbrunner,
R. Ludwig, A. Brückner ACS Catal. 2014, 4, 1845–1849.
[51] S. Berger, J. Fiedler, R. Reinhardt, W. Kaim Inorg. Chem. 2004, 43(4),
1530–1538.
[52] A. Kaeser, M. Mohankumar, J. Mohanraj, F. Monti, M. Holler, J.-J. Cid,
O. Moudam, I. Nierengarten, L. Karmazin-Brelot, C. Duhayon, B.
Delavaux-Nicot, N. Armaroli, J.-F. Nierengarten Inorg. Chem. 2013,
52(20), 12140–12151.
[53] M. Iwamura, H. Watanabe, K. Ishii, S. Takeuchi, T. Tahara J. Am. Chem.
Soc. 2011, 133(20), 7728–7736.
[54] T. Tsubomura, K. Kimura, M. Nishikawa, T. Tsukuda Dalton Trans. 2015,
44(16), 7554–7562.
[55] S. Fischer, O. S. Bokareva, E. Barsch, S. I. Bokarev, O. Kühn, R. Ludwig
ChemCatChem 2016, 8(2), 404–411.
.
8
This article is protected by copyright. All rights reserved.

10.1002/chem.201604005
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
Layout 1:
FULL PAPER
The extension of the π-system at
the backbone of the diimine ligand
within a series of heteroleptic copper
photosensitizers has a strong impact
on the light-driven reduction of
protons. The origin of this behaviour
was studied by a combination of
different photophysical and
Martin Heberle, Stefanie Tschierlei,
Nils Rockstroh, Mark Ringenberg,
Wolfgang Frey, Henrik Junge, Matthias
Beller, Stefan Lochbrunner, and
Michael Karnahl*
Page No. – Page No.
Heteroleptic Copper
electrochemical techniques, which
revealed a connection with the
excited state lifetime.
Photosensitizers: Why an extended
π-system does not automatically
lead to an enhanced hydrogen
production
9
This article is protected by copyright. All rights reserved.