Tuning of photocatalytic activity by creating a tridentate coordination sphere for palladium

Article abstract of DOI:10.1039/c4dt01034e

The synthesis and characterisation of an asymmetric potential bridging ligand bmptpphz (bmptpphz = 2,17-bis(4-methoxyphenyl)tetrapyrido[3,2-a:2′, 3′-c:3′′,2′′-h:2′′′, 3′′′-j] phenazine) is presented. This ligand contains a 1,10-phenanthroline (phen) and a 2,9-disubstituted phen sphere and possesses a strong absorbance in the visible. Facile coordination of the phen sphere to a Ru(tbbpy)₂ core leads to Ru(bmptpphz) ([(tbbpy) ₂Ru(bmptpphz)](PF₆)₂; tbbpy = 4,4′-di-tert-butyl-2,2′-bipyridine). UV-vis, emission, resonance Raman and theoretical investigations show that this complex possesses all properties associated with a Ru(tpphz) ([(tbbpy)₂Ru(tpphz)](PF ₆)₂; tpphz = tetrapyrido[3,2-a:2′,3′-c: 3′′,2′′-h:2′′′,3′′′- j] phenazine) moiety and that the ligand based absorbances in the vis-part also populate an MLCT like state. The coordination of a Pd-core in the new 2,9-disubstituted phen sphere is possible, leading to a cyclometallation. The tridentate complexation leads to changes in the UV-vis and emission behaviour. Furthermore, the stability of the Pd-coordination is significantly enhanced if compared to the unsubstituted Ru(tpphz). Ru(bmptpphz)PdCl proved to be an active photocatalyst for H₂ evolution, albeit with lower activity than the mother compound Ru(tpphz)PdCl₂. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4dt01034e

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Tuning of photocatalytic activity by creating a
tridentate coordination sphere for palladium†
Cite this: Dalton Trans., 2014, 43,
11676
M. G. Pfeffer,a L. Zedler,b S. Kupfer,b M. Paul,b M. Schwalbe,c K. Peuntinger,d
D. M. Guldi,d J. Guthmuller,e J. Popp,f S. Gräfe,f B. Dietzekf and S. Rau*a
The synthesis and characterisation of an asymmetric potential bridging ligand bmptpphz (bmptpphz =
2,17-bis(4-methoxyphenyl)tetrapyrido[3,2-a:2’,3’-c:3’’,2’’-h:2’’’,3’’’-j] phenazine) is presented. This ligand
contains a 1,10-phenanthroline (phen) and a 2,9-disubstituted phen sphere and possesses a strong absor-
bance in the visible. Facile coordination of the phen sphere to a Ru(tbbpy)2 core leads to Ru(bmptpphz)
([(tbbpy)2Ru(bmptpphz)](PF6)2; tbbpy = 4,4’-di-tert-butyl-2,2’-bipyridine). UV-vis, emission, resonance
Raman and theoretical investigations show that this complex possesses all properties associated with a
Ru(tpphz) ([(tbbpy)2Ru(tpphz)](PF6)2; tpphz = tetrapyrido[3,2-a:2’,3’-c:3’’,2’’-h:2’’’,3’’’-j] phenazine) moiety
and that the ligand based absorbances in the vis-part also populate an MLCT like state. The coordination
of a Pd-core in the new 2,9-disubstituted phen sphere is possible, leading to a cyclometallation. The tri-
dentate complexation leads to changes in the UV-vis and emission behaviour. Furthermore, the stability
of the Pd-coordination is significantly enhanced if compared to the unsubstituted Ru(tpphz).
Ru(bmptpphz)PdCl proved to be an active photocatalyst for H2 evolution, albeit with lower activity than
the mother compound Ru(tpphz)PdCl2.
Received 11th April 2014,
Accepted 16th June 2014
DOI: 10.1039/c4dt01034e
www.rsc.org/dalton
One of the first homogeneous systems for light-driven
hydrogen generation consisted of Ru(bpy)3 (bpy = 2,2′-bipyri-
Introduction
2+
The photocatalytic splitting of water into molecular hydrogen dine) as photosensitizer for light absorption, colloidal plati-
3+
and oxygen could provide the key to clean and unlimited num as hydrogen producing catalyst and Rh(bpy)3
as
energy.1 Over the last decades, different concepts to convert electron relay to store and transfer electrons from the ruthe-
light energy into chemical energy were published. Beside nium chromophore to the platinum catalyst.17 Relatively
heterogeneous systems, which usually require inorganic2 or recent developments are intramolecular systems, in which the
organic3 semiconductors with an expedient band gap match- photosensitizer, the electron relay and the catalyst are com-
ing the redox potentials for water splitting, also homogeneous bined in one single molecule forming a photochemical mole-
catalysts revealed promising properties to produce hydrogen4–9 cular device (PMD).18,19
and oxygen10–16 from water using sunlight as energy source.
Both, intermolecular as well as intramolecular assemblies
can serve as highly efficient hydrogen evolving systems.
However, in order to perform detailed investigations on the
actual reaction mechanism, intramolecular catalysts are
directly accessible for mechanistic investigations, as their func-
tionality does not depend on diffusion and collision processes.
Thus, by modifying the individual subunits of an intramole-
cular photocatalyst, i.e. either the photocentre, the bridging
ligand or the catalytic centre, structure–activity relationships
can be established and the photocatalyst’s performance, e.g.
catalytic efficiency can be tuned.20–25
aUniversity of Ulm, Institute of Inorganic Chemistry Materials and Catalysis,
Albert-Einstein-Allee 11, 89081 Ulm, Germany. E-mail: sven.rau@uni-ulm.de;
Fax: +49 731 5023039; Tel: +49 731 5023900
bInstitute of Physical Chemistry, Friedrich-Schiller-University Jena, Helmholtzweg 4,
07743 Jena, Germany
cDepartment of Chemistry, Humboldt-Universität zu Berlin, Metalorganic Chemistry
and Photochemistry, Brook-Taylor-Strasse 2, 12489 Berlin, Germany
dPhysikalische Chemie I, Department of Chemistry and Pharmacy and
Interdisciplinary Center for Molecular Materials, Friedrich-Alexander-Universität
Erlangen-Nürnberg, Egerlandstr. 3, 91058 Erlangen, Germany
eFaculty of Applied Physics and Mathematics, Gdansk University of Technology,
Narutowicza 11/12, 80233 Gdansk, Poland
fInstitute of Physical Chemistry and Abbe Center of Photonics, Friedrich-Schiller-
University Jena, Helmholtzweg 4, D-07743 Jena, Germany
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c4dt01034e
The intensively studied intramolecular photocatalyst
[(tbbpy)2Ru(tpphz)PdCl2](PF6)2 (Fig. 1, Ru(tpphz)PdCl2; tbbpy
=
4,4′-di-tert-butyl-2,2′-bipyridine, tpphz = tetrapyrido[3,2-
a:2′,3′-c:3″,2″-h:2′′′,3′′′-j]phenazine) is an active hydrogen evol-
ving PDM generating H2 with a maximum turnover number
(TON) of 238.26
11676 | Dalton Trans., 2014, 43, 11676–11686
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
been suggested to be related to the photodecomposition and
formation of colloidal catalyst. Thus, it must be assumed that
a heterogeneous pathway through palladium colloids might
have a contribution to the photocatalytic formation of hydro-
gen. We have recently shown that also in Ru(tpphz)PdCl2
colloid formation takes place during catalytic turnover.33 By
X-ray absorption spectroscopy (XAS), a fast and significant
accumulation of Pd(0) colloids as a function of irradiation
time was detected. However, from these experiments it is not
finally clarified whether Pd(0) colloids function as an essential
intermediate for the evolution of molecular hydrogen or are
just a catalytically inactive decomposition product.
In general, the formation of colloidal palladium takes place
if the stability of the zero valent metal centre in the corres-
ponding ligand sphere is too low. This problem can be
addressed by manipulating the corresponding ligand environ-
ment, as frequently utilised in organometallic catalysis. Within
the context of phen like ligands for organometallic catalysts
detailed studies have shown that catalysts based on palladium
complexes with 2,9-diaryl substituted phen ligands induce
increased selectivity and specificity in the ring opening copoly-
merisation of alkenes and carbon monoxide compared to
unsubstituted phen catalysts.34 We therefore were intrigued by
the possibility to stabilise the palladium centre in tpphz based
dinuclear complexes by a similar change of substitution
pattern.
Here, we report on the synthesis, as well as on the charac-
terisation of two novel complexes [(tbbpy)2Ru(bmptpphz)]-
(PF6)2 Ru(bmptpphz) and [(tbbpy)2Ru(bmptpphz)PdCl](PF6)2
Ru(bmptpphz)PdCl (Fig. 1). Steady-state absorption-, emission-
and resonance Raman spectroscopy as well as cyclic voltamme-
try, were applied to gain information on the effect of the novel
substitution pattern on the photophysical and electrochemical
properties of Ru(bmptpphz). Finally, photocatalysis and
mercury poisoning experiments with Ru(bmptpphz)PdCl were
investigated to examine the contribution of Pd(0) colloids to
the formation of molecular hydrogen.
Fig. 1 Top: Mononuclear complexes [(tbbpy)2Ru(tpphz)PdCl2](PF6)2
(Ru(tpphz), left) and [(tbbpy)2Ru(bmptpphz)](PF6)2 (Ru(bmptpphz), right).
Bottom:
(Ru(tpphz)PdCl2,
dinuclear
photocatalysts
left) and
[(tbbpy)2Ru(tpphz)PdCl2](PF6)2
[(tbbpy)2Ru(bmptpphz)PdCl](PF6)2
(Ru(bmptpphz)PdCl, right).
Hereby, the tpphz bridging ligand does not only act as a
link of two metal centres and as electron reservoir,27,28 but
also as tuneable electron relay. For instance, it could be shown
that the introduction of bromo substituents at the ruthenium
faced 1,10-phenanthroline (phen) part of the tpphz ligand sig-
nificantly affects the catalytic activity, by means of a decrease
of the maximum TON from 238 to 94.29
Typical for PMDs, tuning of the bridging ligand does not
change the mechanism and general function of the photo-
catalyst.18 Independent from the substitution pattern of the
tpphz bridging ligand, the photocatalytic reaction starts with
an excitation of the ruthenium photocentre by visible light
irradiation (470 nm) to a 1MLCT (metal to ligand charge trans-
fer) state, which relaxes within less than 1 ps to the corres-
ponding 3MLCT state located on the phen part of tpphz facing
the ruthenium centre. This is followed by an electron transfer
to the phenazine moiety of the bridging ligand. Upon a simul-
taneous dissociation of a chloride from the palladium centre,
the electron is finally transferred to the catalytic metal centre,
which is reduced from Pd(II) to Pd(I).30,31 So far, subsequent
mechanistic steps taking place until molecular hydrogen is
finally abstracted from the catalytic metal centre have
remained vague.
Hammarström et al. studied a structurally related dinuclear
Ru–Pd complex containing a PdCl2 unit bound to a bipyridine
ligand, which is capable of generating hydrogen from a
mixture of acetonitrile and triethylamine, the latter function-
ing as both, electron donor and proton source.32 Via gas
chromatography (GC), X-ray photoelectron spectroscopy (XPS)
and transmission electron spectroscopy (TEM) experiments
they could show that the Ru–Pd complex is an active photo-
catalyst, however the hydrogen evolution goes along with the
formation of colloidal Pd(0). Moreover an induction period
was observed without any formation of hydrogen, which has
Results and discussion
Synthesis and characterization
Since a direct regioselective substitution of tpphz is hard to
realize due to its insolubility, the novel 2,17-arylated bridging
ligand bmptpphz was built up in a stepwise manner (see ESI
Scheme S1†).
Therefore, first the 2,9-bis(4-methoxyphenyl)phenanthro-
line-5,6-dione (bmpphenO2) building block was synthesised,
mostly according to a synthetic route that Sauvage et al.
recently published.35 To obtain the bmptpphz ligand,
bmpphenO2 was reacted with phenanthroline-5,6-diamine
(phen(NH2)2) (76% yield, Scheme 1). Therefore, to a solution
of bmpphenO2 in dry methanol, phen(NH2)2 was added in por-
tions to avoid a possible self-condensation of phen(NH2)2 to
tpphz. Finally, the complexation of the obtained ligand
bmptpphz with [(tbbpy)2RuCl2] (tbbpy = 4,4′-di-tert-butyl-2,2′-
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 11676–11686 | 11677

View Article Online
Paper
Dalton Transactions
observed, leading to a tridentate NNC-coordination motif.
Similar behaviour was observed for Pd-chloro complexes with
6-phenyl-bipyridine.36–39 As the cyclometallation leads to a loss
1
of the C2v symmetry, the aromatic proton signals in the NMR
spectrum are split into a new set of signals for Ru(bmptpphz)-
PdCl (Fig. 2).
The 1H-NMR signals were assigned via H,H-COSY NMR
spectroscopy. The structure of the dinuclear complex was
additionally confirmed via ESI mass spectrometry.
Scheme 1 Synthesis of bmptpphz: methanol, reflux, 8 h (76%).
Steady state UV-vis and emission spectroscopy
Fig. 3 (dotted line) and Table 1 show the absorption properties
of bmptpphz in acetonitrile. The novel bridging ligand has two
main maxima in the UV region at 305 and 341 nm and also a
pronounced absorption feature in the visible region with the
maximum at 413 nm (ε = 32 × 103 M−1 cm−1), these bands
have been assigned based on TDDFT simulations to bright
excitations into the states S9 and S8, the S4, and the S1 of intra-
ligand charge transfer (ILCT) character (see Fig. S8a, Table S5
and Fig. S9† for more details). This is remarkable, as the
unsubstituted tpphz only absorbs at wavelengths below
400 nm.40,41 Even the bis(4-butylphenyl)tpphz shows an
absorption maximum only at 393 nm. These data suggest that
the absorption of visible light is associated with the novel sub-
stitution pattern, which is supported by TDDFT calculations
on the mononuclear ruthenium complex, described in the fol-
lowing paragraph.
Fig. 2 Aromatic region of the 1H-NMR spectra in acetonitrile-d6 of Ru-
(bmptpphz) (top, c = 5 × 10−4 M) and Ru(bmptpphz)PdCl (bottom, c =
8.5 × 10−3 M).
The complexation of bmptpphz with (tbbpy)2RuCl2 forming
Ru(bmptpphz) leads to a combination of ligand and ruthe-
nium based absorption properties (Fig. 3, solid line). Thereby,
three main absorption bands are detected in the UV region,
which can be assigned based on the quantum chemical simu-
lations to first, excitations of the π system of the tbbpy ligands
(287 nm). Second, the phen part of the bmptpphz ligand
causes a shoulder at 312 nm and can be assigned to the
excited states S29, S30, S32, and S50. Third, the phenazine
moiety at 351 nm associated to S20.29,40–42 Further details with
respect to the excited states are provided in the ESI (Table S6,
Fig. S8b and S10†). In comparison to Ru(tpphz) (Fig. 3, dashed
line), which shows a typical broad MLCT band at 445 nm (ε =
20 × 103 M−1 cm−1), the MLCT band of Ru(bmptpphz) is only
visible as a shoulder at 470 nm (ε = 23 × 103 M−1 cm−1). This
shoulder is assigned to excitations into the four bright states,
S13, S14, S15, and S16. The S13 and S16, of MLCT character, lead
to a population of both ligand spheres (bmptpphz and tbbpy),
the excited state S14 refers to a MLCT excitation into the tbbpy
ligand, while the S15 is associated with an ILCT from the
p-anisyl-substituents to the phenazine moiety. Obviously, the
novel bmptpphz ligand strongly contributes to the absorption
in the visible region with a maximum at 416 nm (ε = 50 × 103
M−1 cm−1). TDDFT calculations unravel the nature of this
absorption band to be mainly correlated to the S2 state; an
ILCT transition with small contributions of a MLCT character.
The main ILCT transition refers to an excitation from the
π system of the 4-methoxyphenyl moieties (HOMO) to the
π* system of the phenazine (LUMO); detailed information
bipyridine) was carried out in ethanol–water at reflux to obtain
the brownish target molecule Ru(bmptpphz) in 78% yield. In
comparison, the complexation of the unsubstituted tpphz with
[(tbbpy)2RuCl2] has to be performed at 200 °C in refluxing
ethylene glycol with a three-fold excess of the ligand to guaran-
tee a sufficient solvation of tpphz and to avoid a complexation
of the ruthenium moiety on both coordination sites.30
In the case of the methoxyphenyl-substituted tpphz ligand
much milder conditions, i.e. a 1 : 1 ratio of the reactants and a
lower reaction temperature, are sufficient, since the likelihood
of a double complexation is strongly reduced due to steric
hindrance of the bulky substituents in ortho-position to the
nitrogens.
To confirm the structure of the novel ruthenium complex
Ru(bmptpphz)
a series of characterization methods was
applied. The set of proton signals in the 1H-NMR spectrum
was in good agreement with the compound’s structure and
proved its purity (see Fig. 2 and Experimental section).
Additionally, recording of a two-dimensional H,H-COSY spec-
trum allowed for the clear assignment of all signals. Further-
more, 13C-NMR spectroscopy and mass spectrometry provided
extra convincing evidence for the structure of Ru(bmptpphz)
(Scheme 2).
Finally, the mononuclear ruthenium complex Ru-
(bmptpphz) was reacted with (acn)2PdCl2 to obtain the dinuc-
lear complex Ru(bmptpphz)PdCl. Thereby, a subsequent ortho
C–H activation by the introduced palladium centre was
11678 | Dalton Trans., 2014, 43, 11676–11686
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Scheme 2 Synthetic route for the introduction of the metal centres: (i) [(tbbpy)2RuCl2], ethanol–water, reflux, 18 h (78%); (ii) (acn)2PdCl2, dichloro-
methane, r. t., 4 d (61%).
spectrum is based on the calculated oscillator strengths, reso-
nance Raman simulations, as well as the absorption properties
of the free bmptpphz ligand.
Emission spectra of the novel ligand bmptpphz were
recorded in acetonitrile and were found to be independent of
the excitation wavelength (see Table 1). Excitation at 320, as
well as, 420 nm results in an emission maximum at 500 nm.
The ruthenium complex Ru(bmptpphz) emits at 632 nm,
which is almost identical to the unsubstituted Ru(tpphz)
(638 nm) (see Table 1 and ESI Fig. S2†).
In addition, an excitation spectrum of Ru(bmptpphz) in
acetonitrile was recorded to explore, whether the additional
absorption at 416 nm, related to the arylated tpphz ligand,
3
contributes to the population of the luminescent MLCT state
(ESI Fig. S3†). Apparently, this is the case as the obtained exci-
tation spectrum shows an maximum at 416 nm in the visible
range identical to the absorption spectrum. Taking this experi-
Fig. 3 UV-vis absorption spectra of bmptpphz (dotted), Ru(tpphz)
(dashed) and Ru(bmptpphz) (solid) in acetonitrile.
mental evidence and the results of the theoretical calculations
into account an interesting conclusion can be drawn:
obviously the excitation at 416 nm not only leads to the con-
Table 1 Absorption maxima, molar extinction coefficients, emission
ventional MLCT states transferring electron density to the
maxima and lifetimes of the ligand bmptpphz and the complex Ru-
phen moiety,43 but also the ILCT state significantly contributes
(bmptpphz) in comparison to tpphz and Ru(tpphz)PdCl2. All measure-
to the population of the emissive MLCT state.
ments were carried out in acetonitrile under aerated conditions
The emission lifetime of the excited state of Ru(bmptpphz)
in acetonitrile of 152 ns is nearly identical to that of Ru(tpphz)
(150 ns) under aerated conditions.30
Life
Absorption λmax/nm
(ε/103 M−1 cm−1
Emissionb time
λmax/nm
)
τc/ns
Furthermore, we have investigated the impact of the
addition of the second metal centre on the photophysics.
Regarding the absorption features of Ru(bmptpphz)PdCl in
comparison with the mononuclear Ru(bmptpphz), the absorp-
tion maximum in the visible range is shifted to a higher wave-
length (447 nm) (see Table 1 and ESI Fig. S2†). This
absorption wavelength is almost identical to that of the un-
substituted dinuclear Ru(tpphz)PdCl2, which has a maximum
bmptpphz
265 (27), 305 (34), 341 (37),
395 (25), 413 (32)
287 (104), 312 (61), 351 (69), 632
414 (50), 470 (23)
285 (105), 314 (45), 381 (31), 638
445 (20)
500
—
Ru(bmptpphz)
Ru(tpphz)a
152
150
Ru(bmptpphz)PdCl 286 (87), 330 (60), 447 (39)
Ru(tpphz)PdCl2
664
—
—
—
a
283 (105), 379 (33), 444 (20)
a Absorption and life time data were taken from Rau et al.29,30
b λex/bmptpphz = 340 nm and 420 nm, λex/Ru(bmptpphz) = 470 nm,
λex/Ru(tpphz) = 445 nm. λex/Ru(bmptpphz) = 470 nm, λex/Ru(tpphz) =
1
at 444 nm attributed to the population of the MLCT state.31
c
However, the extinction of Ru(bmptpphz)PdCl at 447 nm (ε =
39 × 103 M−1 cm−1) is almost twice as high as the extinction of
Ru(tpphz)PdCl2 at 444 nm (ε = 20 × 103 M−1 cm−1), indicating
that the enhanced absorption of visible light due to the novel
445 nm.
concerning the TDDFT results are given in the ESI (Table S6 substitution pattern of the bridging ligand is also maintained
for the dinuclear complex. As mentioned above, the main
in calculations is mainly caused by the different CT nature absorption band in the visible range is shifted from 414 nm
and Fig. S10†). The inversion of the MLCT and the ILCT states
(Ru(bmptpphz)) to 447 nm (Ru(bmptpphz)PdCl) (see ESI
of these states, while the assignment to the experimental
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 11676–11686 | 11679

View Article Online
Paper
Dalton Transactions
Table 2 Electrochemical data: half-wave potentials E (in V) (vs. Fc/Fc+) in acetonitrile. The electrochemical measurements were carried out in anhy-
drous and argon purged solvent with a complex concentration of 1.0 mM and tetrabutylammonium hexafluorophosphate (0.1 mM) as supporting
electrolyte
II
red
III
red
IV
red
VI
Substance
Eox
EIred
E
E
E
EVred
E
red
Ru(bmptpphz)
Ru(tpphz)
0.82
0.80
−1.44
−1.41
−1.90
−1.92
−2.10
−2.12
−2.27
−2.23
−2.44
−2.48
−2.90
−2.90
Fig. S2†). A possible explanation for this bathochromic
shift might be the induced in-plane alignment of one
anisyl-substituent with respect to the tpphz-plane upon
complexation and the consequential enhanced electron
delocalisation.
In comparison to the mononuclear Ru(bmptpphz) a signifi-
cant quenching of the emission intensity and a redshift of the
emission maximum to 664 nm was observed for Ru-
(bmptpphz)PdCl (see ESI Fig. S2†). As typically found for
hetero bimetallic ruthenium tpphz complexes, the second
metal centre induces a MLCTphen → MLCTpz relaxation, which
leads to a dominant population of a low-energy “dark state”
located on the phenazine part, and so causes the decrease of
the emission intensity and the bathochromic shift of the emis-
sion maximum.31,44,45
Electrochemical properties
Fig. 4 Experimental and theoretical RR spectra for Ru(bmptpphz) in
resonance with the states S2, S13, S14 , S15, S16, S19 and S20 at the wave-
length 458 nm. The experimental solvent band is indicated by an aster-
isk. The modes associated with vibrations located on the bipyridine
Cyclic voltammetry was used to obtain the redox potentials of
Ru(bmptpphz) and Ru(tpphz) in acetonitrile (Table 2). Ana-
logously to Ru(tpphz), Ru(bmptpphz) shows the metal based oxi-
dation wave at 0.82 V and multiple waves assignable to ligand ligands are given in underlined numbers, whereas the remaining
numbers belong to modes associated with vibrations located on the
based reductions, which are attributed to the reduction of the
phenazine part (−1.44 V) and the ruthenium bound phen part
of tpphz (−2.27 V), as well as to the first and second reduction
tpphz ligand. A Lorentzian function with a FWHM of 5 cm−1 is employed
to broaden the calculated transitions.
of the terminal tbbpy ligands (−1.90, −2.10, and −2.44,
−2.90 V).29,40 Apparently, the 4-methoxyphenyl-substituents
have no significant effect on the redox potentials of Ru(bmptpphz)
compared to Ru(tpphz).
with detailed information concerning the computation of RR
intensities.
The experimental and theoretical RR intensity patterns are
in good agreement, while a mean absolute deviation from the
calculated modes with respect to the Raman bands of 12 cm−1
was obtained. As can be seen by means of the simulated spec-
trum almost all intense Raman active modes are localized on
the bmptpphz ligand, while only eight modes of small and
medium intensity are centred on the tbbpy sphere. Analogous
to Ru(tpphz) only the Raman features at 1483 cm−1 and
1538 cm−1 are assigned to intense modes centred on the tbbpy
ligands. This result is consistent with the TDDFT results,
showing that merely two of the seven bright excited states in
this region are correlated to a charge transfer onto the tbbpy
ligands, while all other bright states excite onto the bmptpphz
ligand. Hence, a relaxation upon photoexcitation is mainly
centred on the bridging ligand. The quantum chemical calcu-
lations show that the new absorption feature at 416 nm, corre-
lated to the S2 ILCT state, mainly contributes to the RR
intensity pattern at 458 nm with its intense bmptpphz centred
modes 313, 314, 336 and 391 (see ESI Table S7†).
Raman spectroscopy
In order to monitor changes in the electron density distri-
bution upon photoexcitation at the Franck–Condon point, reso-
nance Raman (RR) spectroscopy is the method of choice.46 In
RR spectra normal modes, which are affected by the structural
reorganization upon photoexcitation, are selectively enhanced.
Hence, RR spectroscopy is an ideal tool to study the nature of
the excited states in the Franck–Condon region.
RR spectra have been recorded using excitation wavelengths
of 496, 488, 476, and 458 nm, i.e. in partial resonance with the
MLCT band (at 470 nm) as well as the ILCT band (at 416 nm).
Of particular interest is the investigation of the novel ILCT
band in Ru(bmptpphz) and, hence, the RR spectrum recorded
at 458 nm. This RR spectrum measured at 458 nm is
depicted in Fig. 4 and compared to the calculated one; the RR
spectra recorded at lower excitation energies as well as the
non-resonant spectrum are illustrated in the ESI† along
11680 | Dalton Trans., 2014, 43, 11676–11686
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 6 Turnover frequencies (TOFs) of Ru(bmptpphz)PdCl (solid line)
and Ru(tpphz)PdCl2 (dashed line) in a mixture of acetonitrile, triethyl-
amine and water under LED irradiation (470 nm).
Fig. 5 Turnover numbers (TONs) of Ru(bmptpphz)PdCl (solid line) and
Ru(tpphz)PdCl2 (dashed line) in a mixture of acetonitrile, triethylamine
and water under LED irradiation (470 nm).
centre to Pd(0) is a prerequisite for both, metal colloid for-
mation and catalytic turnover of protons to hydrogen.30 The
cyclometallation in Ru(bmptpphz)PdCl will influence both,
Photocatalytic hydrogen production and mercury poisoning
To investigate the photocatalytic properties of Ru(bmptpphz)- the redox properties of the Pd centre and the stability and reac-
PdCl with regard to light-driven hydrogen production, the tivity of the metal centre in all oxidation states since Pd(0)
complex was dissolved in a catalytic mixture of the following forms stable bonds to organometallic ligands.48 (3) Moreover,
composition: c(photocatalyst) = 70 μM, V(triethylamine) = also the steric shielding by the anisyl-substituents might have
0.6 ml (2.2 M), V(acetonitrile) = 1.2 mL and V(water) = 0.2 mL. an impact on the catalytic performance of the Pd centre. As
Afterwards the solution was irradiated with LED light (470 nm, reported in the literature, the catalytic activity of Pd as catalytic
35 5 mW cm−2) and the amount of produced hydrogen was centre in a phen coordination sphere is significantly affected
determined by headspace gas chromatography. In order to by the steric influence of substituents in 2,9-position
evaluate the effect of the novel substitution pattern of (equivalent to the 2,17-position in bmptpphz).34,49
Ru(bmptpphz)PdCl, also the non-substituted Ru(tpphz)PdCl2
was investigated under identical conditions for comparison.
The effect of the novel substitution pattern is also expressed
by the turnover frequencies (TOF = TON/t) of Ru(bmptpphz)-
Fig. 5 shows the catalytic activities of Ru(bmptpphz)PdCl PdCl illustrated in Fig. 6. For both, Ru(bmptpphz)PdCl and
and Ru(tpphz)PdCl2 in terms of turnover numbers (TONs). For Ru(tpphz)PdCl2, the TOF is constantly increasing with continu-
Ru(tpphz)PdCl2 a significant induction period of roughly ous irradiation, which might correlate to the progressive for-
2 hours was observed until a pronounced formation of hydro- mation of Pd(0) colloids as already reported by Hammarström
gen could be measured, which indicates the formation of a and co-workers.32 In case of Ru(tpphz)PdCl2 the maximum
catalytically active species prior to catalytic turnover (see ESI TOF (7.9 h−1) is reached already after 18 h of irradiation,
Fig. S5†).32,33 It is apparent that Ru(bmptpphz)PdCl is less whereas the TOF of Ru(bmptpphz)PdCl is still increasing with
active giving a maximum TON of 50 after 64 hours of irradiation times over 40 h.
irradiation, compared to Ru(tpphz)PdCl2, which shows a
maximum TON of 187 after 24 hours.
To evaluate the contribution of Pd(0) colloids to the for-
mation of hydrogen, the mercury test was applied, known for
From our point of view there are three possible reasons for its ability to deactivate metal colloids and particles by forming
the reduced activity of Ru(bmptpphz)PdCl: (1) as described amalgams.50 However, the mercury test must be treated with
above, the visible-light-driven ILCT implies an electron trans- caution. As recently observed for a structurally related Ru–Pd
fer from the anisyl-moiety onto the phenazine moiety of the photocatalysts, elemental Hg can decompose the photocatalyst
bmptpphz ligand. Thus, electron density located in the peri- already under non-catalytic conditions in pure acetonitrile.51,52
phery of the catalytic centre is withdrawn, which might lead to Thus, prior to investigations under catalytic conditions, the
a less effective proton reduction. (2) Another explanation is the effect of Hg on the photocatalysts was tested under non-reduc-
influence of the tridentate coordination of the Pd centre. As tive conditions. Therefore both catalysts were dissolved in pure
known from phen(Ph)2PdCl (Ph = phenyl), which contains a acetonitrile, a drop of liquid Hg was added and the suspension
similar tridentate coordination motif compared to Ru- was allowed to stir for 7 h under exclusion of light. 1H-NMR
(bmptpphz)PdCl, the Pd core predominantly coordinates to studies revealed that the unsubstituted Ru(tpphz)PdCl2 reacts
the NC sphere with a C–Pd bond lengths of 1.985 Å, whereas to the mononuclear Ru(tpphz) under the presence of Hg,
the N–Pd distance (2.251 Å) of the opposing nitrogen is signifi- meaning an extraction of the Pd(II) centre by Hg.33 Interest-
cantly larger, indicating the higher stability of the C–Pd ingly, no reaction occurred with Ru(bmptpphz)PdCl under
bond.47 In general, the photochemical reduction of the Pd(II) identical conditions underlining the increased stability of the
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 11676–11686 | 11681

View Article Online
Paper
Dalton Transactions
Pd centre (see ESI Fig. S7†). As a next step, a drop of liquid reduction or oxidation of implemented metal centres to
mercury was added to a catalytic mixture containing Ru- induce movement. In that concern, Ru(bmptpphz) might open
(bmptpphz)PdCl as photocatalyst prior to irradiation. The a new pathway for the design of molecular machines fuelled
result was a complete loss of catalytic activity (see ESI by visible light.
Table S4†). Independent from the irradiation time no hydro-
gen was produced in the presence of mercury. This is a first
hint that photodecomposition and in situ generation of col-
loidal Pd might be an important part of the catalytic mechan-
Experimental
ism with Ru(bmptpphz)PdCl, however further methods of Instrumentation
analysis have to be applied to confirm this assumption.52,53
NMR-spectra were measured with a Jeol EX-270 DELTA, a Jeol
Lambda-400 and a Bruker DRX 400 spectrometer. The NMR
spectra were recorded at 298 K and the NMR chemical shifts
Conclusions
were referenced to the solvent peaks. MS analysis was per-
formed on a Finnigan MAT SSQ 710, MAZ95XL or MAT 95 XP
and on a Shimadzu LC-MS 2020 equipped with an electrospray
ionisation source and a SPD-20A UV-vis detector. The UV-vis-
spectra were recorded with a Varian CARY 50 Conc and a
JASCO Spectrometer V-670. Quartz cells with a 10 mm path
length were used. The emission spectra were recorded with a
JASCO Spectrofluorometer FP-8500. Quartz cells with a 10 mm
path length were used. The hydrogen evolved was measured by
headspace GC on a Bruker Scion GC/MS, with a thermal con-
ductivity detector (column: Mol. Sieve 5A 75 m × 0.53 mm I.D.,
Herein, we reported on a novel structural modification of a
tpphz based ligand. In this context, the first example of a
regioselective asymmetric disubstitution in the 2,17-position
of tpphz selectively influencing one coordination sphere of the
potential bridging ligand was shown. Beside synthesis and
characterisation of the novel ligand bmptpphz, also its com-
plexation forming the mononuclear ruthenium complex Ru-
(bmptpphz) was successfully performed. Finally, the dinuclear
Ru(bmptpphz)PdCl complex was synthesized and cyclometalla-
tion at the Pd centre had occurred. The photophysical and
electrochemical investigations on the mononuclear complex
Ru(bmptpphz) revealed analogies to the well-known Ru(tpphz)
system. Interestingly Ru(bmptpphz) exhibits an enhanced
absorption of visible light, which is also addressed by TDDFT
simulations to the novel substitution pattern. Subsequently
the dinuclear complex was tested towards its performance as
hydrogen evolving photocatalyst. We could show that the novel
substitution pattern has a significant influence on the catalytic
performance of the Pd centre. Based on the results obtained
from mercury poisoning experiments, the dissociation of palla-
dium under photoexcitation and formation of palladium col-
loids as part of the catalysis mechanism must be taken into
consideration.
oven temp. 70 °C, flow rate 22.5 ml min−1
, detector
temp. 200 °C) with argon as carrier gas. The GC was calibrated
by mixing different volumes of pure hydrogen together with
argon into a Schlenk vessel. The obtained signal was plotted
against the calibration curve and multiplied accordingly to
obtain the total produced hydrogen content in the headspace.
Resonance Raman (RR) spectra were excited by the visible
lines of an Ar ion laser (Coherent, Innova 300C) and recorded
with an Acton SpectraPro 2758i spectrometer (entrance slit
width 100 µm, focal length 750 mm, grating 600 mm−1). A 90
degree geometry was employed for the resonance Raman
experiments. The Raman signals were detected using a liquid-
nitrogen cooled CCD detector (SPEC-10, Roper Scientific). The
band of acetonitrile at 1373 cm−1 was used as a reference for
normalizing intensities and wavenumbers. The recorded reso-
nance Raman spectra were background corrected and the
solvent spectrum was subtracted.
Compared to Ru(tpphz)PdCl2, Ru(bmptpphz)PdCl pos-
sesses an increased stability of the Pd(II) coordination sphere,
caused by the tridentate coordination pocket and/or the steric
shielding by the anisyl-rings.34,49 As the 2,9-arylated-phen
sphere of Ru(bmptpphz) should be capable of stabilizing a
large variety of metal cations like Ni2+, Co2+, Cu+ and Zn2+,54
Photocatalysis
which are known to be active centres for catalytic hydrogen for- Photohydrogen production experiments were carried out using
mation, a quite large number of potentially stable heterodinuc- a custom-made air-cooling apparatus for maintaining room
lear photocatalysts is accessible.55–57
temperature (22 °C) under constant irradiation of the sample
Moreover, as the p-anisyl-substituents can easily be con- (ESI Fig. S4†). The solvents used were degassed with argon.
verted to hydroxyphenyl substituents, they provide the oppor- The samples for the LED light irradiation experiments were
tunity for further synthetic coupling, which makes Ru- prepared in GC vials (diameter = 45 × 14.75 mm) with a known
(bmptpphz) an interesting candidate concerning its appli- headspace of 3 mL and a headspace-solution ratio of 3/2. Fur-
cation as building block for artificial molecular machines.58 thermore the GC vials were charged in the dark and under
For instance, an interlock of Ru(bmptpphz) with catenane or argon stream. After addition of acetonitrile containing the
rotaxane like architectures, which are used for the construc- catalyst (V = 1.2 ml), triethylamine (V = 0.6 mL) and 10 vol% of
tion of molecular machines and motors, seems to be feas- water (V = 0.20 mL) the final catalyst concentration in aceto-
ible.59 Instead of a chemically induced exchange of metal ions, nitrile was 70 µM. Subsequently, the GC vials were irradiated
Ru(bmptpphz) could function as module for light-driven with blue LED light (470 nm, P = 35 5 mW cm−2). After
11682 | Dalton Trans., 2014, 43, 11676–11686
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
irradiation, 200 µL samples were drawn from the headspace
and injected immediately into the GC apparatus.
The resonance Raman (RR) spectra were calculated within
the independent mode displaced harmonic oscillator model
(IMDHOM), which assumes the electronic ground and excited
states potentials to be harmonic oscillators with differing equi-
librium positions, sharing the same set of vibrational modes.
Within this approach the RR intensity for a fundamental tran-
sition 0→1l can be calculated from the Raman polarizability
tensor.46,67–69
Stability of Ru(bmptpphz)PdCl towards liquid mercury
To test the stability of Ru(bmptpphz)PdCl towards elemental
Hg under non-catalytic conditions, 2 mg (1.6 μmol) of the
dinuclear complex were dissolved in 1.5 ml degassed deuter-
ated acetonitrile under exclusion of oxygen and 300 μL liquid
Hg were added. The suspension was stirred for 7 hours and
afterwards the supernatant solution was transferred to a NMR
tube.
X
1
Δe;l
2
pffiffiffi
ðααβÞg0!g1
¼
ðμgeÞα ðμgeÞβ
fΦeðωLÞ ꢀ ΦeðωL ꢀ ωlÞg
l
ꢀ
h
e
ð1Þ
Mercury test
where (µge)α is a component of the electronic transition dipole
moment at the ground state equilibrium geometry, Δe,l is the
dimensionless displacement of the eth excited state potential
minimum with respect to the ground state for the lth normal
coordinate and ωL is the frequency of the incident light. The
function Φe(ωL)is approximated by
The experimental setup for the catalyst poisoning with elemen-
tal mercury was analogous to the above described conditions
for photocatalysis. In addition, a drop of elemental mercury
(100 µl) was added to the catalytic mixture (see ESI Fig. S6†).
Moreover, all samples (also those without mercury) were
stirred constantly. Each sample was irradiated (sideward) for
24 hours and 99 hours. After irradiation, 200 µL samples were
drawn from the headspace and injected immediately into the
GC apparatus.
1
ΦeðωLÞ ¼
ð2Þ
ωg;e ꢀ ωL ꢀ iΓ′
whereas ωg,e is the vertical excitation energy from the elec-
tronic ground state g to the excited state e and Γ is a damping
factor describing a homogeneous broadening. A value of Γ
equal to 1750 cm−1 was assumed in the simulations to repro-
duce the experimental broadening. An excitation wavelength of
558 nm was employed to calculate the RR spectra, for which
the vertical excitation energies ωg,e were corrected so that
experimental and theoretical absorption maxima coincide.
The absorption spectrum A(ωL) is given by:
Quantum chemistry
The structural and electronic data of Ru(bmptpphz) were
obtained from quantum chemistry calculations performed
with the GAUSSIAN 09 program.60 The geometry, vibrational
frequencies, and normal coordinates of the ground state were
calculated by means of DFT with the XC functional B3LYP.61
The 28-electron relativistic effective core potential MWB62 was
used with its basis set for the ruthenium atom, that is, 4s, 4p,
4d and 5s electrons are treated explicitly, whereas the three
first inner shells are described by the core pseudopotential.
The 6-31G(d) double-ζ basis set63 was employed for the
ligands. To correct for the lack of anharmonicity and the
approximate treatment of electron correlation, the harmonic
frequencies were scaled by the factor 0.97.64 The vertical exci-
tation energies, oscillator strengths and analytical Cartesian
energy derivatives of the excited states were obtained from
TDDFT calculations within the adiabatic approximation with
the same XC functional, pseudopotential and basis set. It was
shown by several groups that B3LYP provides a balanced
description of the absorption features of such complexes.43,65
The absorption spectra were simulated by determining the
excitation energies and oscillator strengths of the 100 lowest
singlet excited states. The effects of the interaction with a
solvent (acetonitrile, ε = 35.688, n = 1.344) on the geometry,
frequencies, excitation energies and excited state gradients
were taken into account by the integral equation formalism of
the polarizable continuum model.66 The nonequilibrium pro-
cedure of solvation was used for the calculation of the exci-
tation energies and of the excited state gradients. Such
procedure is well adapted for processes where only the fast
X
2
AðωLÞ / ωL
ðμgeÞ Im ΦeðωLÞ
ð3Þ
e
The dimensionless displacement Δe,l of the excited state e
is defined in the IMDHOM by the partial derivative of the
excited state potential energy Ee along the normal mode Ql
evaluated at the ground state equilibrium geometry
ꢀ
ꢁ
1
@Ee
@Ql
pffiffiffi
Δe;l ¼ ꢀ
:
ð4Þ
3
2
ꢀ
h ωl
0
These derivatives were obtained from the analytical deriva-
tives of the excited state electronic energy (Ee) along the Carte-
sian coordinates.
More details about the calculation procedure of RR intensi-
ties can be found in ref. 46, 70 and references therein. The RR
intensity has been calculated for an excitation wavelength of
458 nm and taking into account contributions from the singlet
excited states S2, S13, S14, S15, S16, S19, and S20, while the exci-
tation energies were corrected by +2000 cm−1 (S2) and
−3500 cm−1 (S13, S14, S15, and S16), respectively.
Synthesis and characterisation
reorganization of the electronic distribution of the solvent is The complete synthesis procedure, including the synthesis of
important.
bmpphenO2 is shown in the ESI.†
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 11676–11686 | 11683

View Article Online
Paper
Dalton Transactions
bmptpphz. To a boiling solution of bmpphenO2 (0.46 g, finally separated by filtration and washed with water, diethyl
1.08 mmol) in methanol (250 ml) phen(NH2)2 (0.31 g, ether and dried in vacuo.
1.49 mmol) was added in portions over 2 hours. The mixture
was refluxed for 8 hours, whereby the bright yellow product 1665.3
Yield = 61% (0.013 g); M(C74ClH71F12N10O2P2PdRu) =
mol−1
MS (ESI in acetonitrile): m/z 965.9
g
;
=
precipitated. The product was separated by filtration and [2M − 3PF6]3+, 687.4 [M − 2PF6]2+, 617.4 [M − 2PF6 − PdCl]2+;
washed with water, methanol and acetone and dried in vacuo. 1H-NMR (400 MHz, CD3CN, 295 K, c = 8.5 mM): δ = 9.91 (d,
Yield = 76% (0.49 g); M(C38H24N6O2) = 596.6 g mol−1; MS 1H, J = 8.2 Hz, c″), 9.85 (d, 1H, J = 8.4 Hz, c′′′), 9.79 (d, J =
(DEI in acetonitrile
+ trifluoroacetic acid): m/z = 596 8.3 Hz, c), 9.65 (d, 1H, J = 8.7 Hz, c′), 8.60 (s, 2H, 3/3′′′), 8.55 (s,
[M − H]−; 1H-NMR (270 MHz, CDCl3 + trifluoroacetic acid, 2H, 3′/3″), 8.32 (dd, 1H, J = 5.3, 1.2 Hz, a′), 8.27 (dd, 1H, J = 5.3,
295 K): δ = 10.21 (d, 2 H, J = 7.96 Hz, c), 10.10 (d, 2 H, J = 1.2 Hz, a), 8.24 (d, 1H, J = 8.4 Hz, b′′′), 8.08 (dd, 1H J = 8.2, 5.4
8.87 Hz, c′), 9.36 (d, 2 H, J = 4.87 Hz, a), 8.64 (d, 2 H, J = Hz, b″), 8.02 (dd, 1H, J = 8.2, 5.3 Hz, b), 7.88 (d, 1H, J = 8.9 Hz,
8.86 Hz, b′), 8.36 (dd, 2 H, J = 5.20 and J = 8.33 Hz, b), 8.25 b′), 7.85–7.73 (m, 4H, 6/6′′′/o), 7.70 (d, 1H, J = 6.2 Hz, 6″), 7.66
(d, 4 H, J = 8.69 Hz, o), 7.27 (d, 4 H, J = 8.71 Hz, m), 4.01 (s, (d, 1H, J = 6.1 Hz, 6′), 7.58–7.50 (m, 2H, 5/5′′′), 7.48–7.27 (m,
6 H, OMe) ppm;
3H, 5′/5″/*), 7.24 (d, 2H, J = 8.6 Hz, m), 6.19 (s, 1H, *), 5.99 (s,
Ru(bmptpphz). The ligand bmptpphz (0.49 g, 0.82 mmol) 1H, *), 3.96 (s, 3H, OMe), 3.44 (s, 3H, OMe′), 1.48 (s, 9H, tert
was suspended in ethanol–water (100 ml/100 ml) and heated butyl), 1.47 (s, 9H, tert butyl), 1.35 (s, 18H, tert butyl).
to reflux. Afterwards [(tbbpy)2RuCl2] (0.60 g, 0.85 mmol) in
ethanol–water (50 ml/50 ml) was added during 1 hour. The
orange-brown suspension was refluxed for 18 hours more and
subsequently filtered while hot to remove unreacted ligand.
Acknowledgements
Afterwards the solution was concentrated under reduced This research is supported by German Research Association
pressure using a rotary evaporator and the crude product was (DFG SFB 583, GRK 1626) and the COST Action CM1202,
precipitated by adding 100 ml of an aqueous solution of PERSPECT-H2O. J. G. is grateful to the 7th Framework Pro-
NH4PF6 (0.80 g, 5.0 mmol). The precipitated brown-reddish gramme of the European Union. All the calculations have been
solid was separated by filtration and washed with water, performed at the Universitätsrechenzentrum of the Friedrich-
diethyl ether and dried in vacuo.
Schiller University of Jena and at the HP computers of the
Finally, the ruthenium complex was purified by column Theoretical Chemistry group.
chromatography on silica (dichloromethane–methanol 1 : 1
v/v). The product elutes as first band.
Yield = 78% (1.0 g); M(C74H72F12N10O2P2Ru) = 1524.4 g
Notes and references
mol−1; MS (FD in acetonitrile): m/z = 1380 [M − PF6]+; 1H-NMR
(400 MHz, CD3CN, 295 K, c = 0.01 M): δ = 9.42 (d, 2 H, J = 8.18
Hz, c), 9.19 (d, 2 H, J = 8.82 Hz, c′), 8.63 (d, 2 H, J = 1.86 Hz,
3′), 8.60 (d, 2 H, J = 1.81 Hz, 3), 8.13 (dd, 2 H, J = 4.96, 1.25 Hz,
a), 8.07 (d, 4 H, J = 8.57 Hz, o), 8.01 (d, 2 H, J = 8.54 Hz, b′),
7.92–7.64 (m, 6H, 6′, b, 6), 7.55 (dd, 2 H, J = 5.90, 2.14 Hz, 5′),
7.41 (d, 2 H, J = 6.14 Hz, 5), 6.68 (d, 4 H, J = 8.43 Hz, m), 3.73
(s, 6 H, MeO), 1.51 (s, 18 H, tert butyl), 1.37 (s, 18 H, tert butyl)
ppm; 13C-NMR (100 MHz, CD3CN, 295 K): δ = 29.58, 29.68,
35.53, 35.58, 55.08, 118.84, 118.87, 121.76, 121.82, 123.93,
124.9, 126.45, 128.27, 129.67, 130.37, 133.03, 133.44, 146.73,
149.25, 151.20, 151.29, 152.75, 156.98, 157.12, 157.26, 161.13,
163.09 ppm.
Ru(bmptpphz)PdCl. 20 mg (0.013 mmol) Ru(bmptpphz)
were dissolved in 15 ml dichloromethane and 3.9 mg
(0.015 mmol) Pd(acn)2Cl2 were added. The reaction mixture
was stirred for 4 days at room temperature. After 1 day another
1.95 mg (7.5 μmol) Pd(acn)2Cl2 were added. Subsequently, the
solvent was evaporated at room temperature and the residue
1 N. S. Lewis and D. G. Nocera, Proc. Natl. Acad. Sci. U. S. A.,
2003, 103, 15729.
2 P. Ritterskamp, A. Kuklya, M. A. Wustkamp, K. Kerpen,
C. Weidenthaler and M. Demuth, Angew. Chem., Int. Ed.,
2007, 46, 7770.
3 Z. Zhang, J. Long, L. Yang, W. Chen, W. Dai, X. Fu and
X. Wang, Chem. Sci., 2011, 2, 1826.
4 F. Gärtner, B. Sundararaju, A.-E. Surkus, A. Boddien,
B. Loges, H. Junge, P. H. Dixneuf and M. Beller, Angew.
Chem., Int. Ed., 2009, 48, 9962.
5 S. Oishi, J. Mol. Catal., 1978, 39, 225.
6 R. Bauer and H. A. F. Werner, Int. J. Hydrogen Energy, 1994,
19, 497.
7 J. Hawecker, J.-M. Lehn and R. Ziessel, Nouv. J. Chim.,
1983, 7, 271.
8 Y. Na, M. Wang, J. Pan, P. Zhang, B. Åkermark and L. Sun,
Inorg. Chem., 2008, 47, 2805.
9 H. Ozawa and K. Sakai, Chem. Commun., 2011, 47, 2227.
was dissolved acetonitrile and applied to a column with silica 10 E. Amouyal, Sol. Energy Mater. Sol. Cells, 1995, 38, 249.
as solid phase and washed with water. Afterwards the dinuc- 11 A. Guskov, J. Kern, A. Gabdulkhakov, M. Broser, A. Zouni
lear complex was eluted with acetonitrile–water + KNO3
(saturated) 15 : 1 v/v. The fractions were combined and the 12 L. Duan, Y. Xu, P. Zhang, M. Wang and L. Sun, Inorg.
solution was concentrated under reduced pressure at room Chem., 2010, 49, 209.
temperature. Following this, an aqueous solution of NH4PF6 13 J. T. Muckerman, D. E. Polyansky, T. Wada, K. Tanaka and
and W. Saenger, Nat. Struct. Mol. Biol., 2009, 16, 334.
(0.08 mmol) was added to precipitate the product, which was
E. Fujita, Inorg. Chem., 2008, 47, 1787.
11684 | Dalton Trans., 2014, 43, 11676–11686
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
14 F. Liu, J. J. Concepcion, J. W. Jurss, T. Cardolaccia, 38 F. Neve, M. Ghedini and A. Crispini, Chem. Commun., 1996,
J. L. Templeton and T. J. Meyer, Inorg. Chem., 2008, 47,
1727.
15 S. W. Gersten, G. J. Samuels and T. J. Meyer, J. Am. Chem.
Soc., 1982, 104, 4029.
16 C. Sens, I. Romero, M. Rodríguez, A. Llobet, T. Parella and
J. Benet-Buchholz, J. Am. Chem. Soc., 2004, 126, 7798.
17 J.-M. Lehn, M. Kirch and J.-P. Sauvage, Helv. Chim. Acta,
1979, 62, 1345.
2463.
39 F. Neve, A. Crispini, C. Di Pedro and S. Campagna, Organo-
metallics, 2002, 21, 3511.
40 J. Bolger, A. Gourdon, E. Ishow and J.-P. Launay, Inorg.
Chem., 1996, 35, 2937.
41 S. Bodige, A. S. Torres, D. J. Maloney, D. Tate, G. Kinsel,
A. Walker and F. M. MacDonnell, J. Am. Chem. Soc., 1997,
119, 10364.
18 V. Balzani, Photochem. Photobiol. Sci., 2003, 2, 459.
19 S. Rau, D. Walther and J. G. Vos, Dalton Trans., 2007, 915.
42 C. Kuhnt, M. Karnahl, M. Schmitt, S. Rau, B. Dietzek and
J. Popp, Chem. Commun., 2011, 47, 3820.
20 C. Li, M. Wang, J. Pan, P. Zhang, R. Zhang and L. Sun, 43 S. Tschierlei, M. Karnahl, M. Presselt, B. Dietzek,
J. Organomet. Chem., 2009, 694, 2814.
21 H. Ozawa, M. Haga and K. Sakai, J. Am. Chem. Soc., 2006,
128, 4926.
22 A. Fihri, V. Artero, A. Pereira and M. Fontecave, Dalton
Trans., 2008, 5567.
J. Guthmuller, L. González, M. Schmitt, S. Rau and J. Popp,
Angew. Chem., Int. Ed., 2010, 49, 3981.
44 L. Flamigni, S. Encinas, F. Barigelletti, F. M. MacDonnell,
K.-J. Kim, F. Puntoriero and S. Campagna, Chem. Commun.,
2000, 1185.
23 A. Fihri, V. Artero, M. Razavet, C. Baffert, W. Leibl and 45 C. Chiorboli, M. a J. Rodgers and F. J. Scandola, J. Am.
M. Fontecave, Angew. Chem., Int. Ed., 2008, 47, 564. Chem. Soc., 2003, 125, 483.
24 M. Elvington, J. Brown, S. M. Arachchige and K. J. Brewer, 46 M. Wächtler, J. Guthmuller, L. González and B. Dietzek,
J. Am. Chem. Soc., 2007, 129, 10644. Coord. Chem. Rev., 2012, 256, 1479.
25 M. Karnahl, C. Kuhnt, F. W. Heinemann, M. Schmitt, 47 M. Kuritani, S. Tashiro and M. Shionoya, Chem. – Asian J.,
S. Rau, J. Popp and B. Dietzek, Chem. Phys., 2012,
393, 65.
2013, 8, 1368.
48 M. Schwalbe, H. Schreer, J. Langer, H. Görls and
D. Walther, J. Organomet. Chem., 2006, 691, 4868.
49 G.-J. ten Brink, I. W. C. E. Arends, M. Hoogenraad,
G. Verspui and R. A. Sheldon, Adv. Synth. Catal., 2003, 345,
1341.
26 M. Karnahl, C. Kuhnt, F. Ma, A. Yartsev, M. Schmitt,
B. Dietzek, S. Rau and J. Popp, ChemPhysChem, 2011, 2101.
27 R. Konduri, H. Ye, F. M. MacDonnell, S. Serroni,
S. Campagna and K. Rajeshwar, Angew. Chem., Int. Ed.,
2002, 41, 3185.
50 D. R. Anton and R. H. Crabtree, Organometallics, 1983, 2,
855.
28 L. Zedler, J. Guthmuller, I. Rabelo de Moraes, S. Kupfer,
S. Krieck, M. Schmitt, J. Popp, S. Rau and B. Dietzek, Chem. 51 G. Singh Bindra, M. Schulz, A. Paul, S. Soman, R. Groarke,
Commun., 2014, 50, 5227.
J. Inglis, M. T. Pryce, W. R. Browne, S. Rau, B. J. Maclean
29 M. Karnahl, S. Tschierlei, C. Kuhnt, B. Dietzek, M. Schmitt,
and J. G. Vos, Dalton Trans., 2011, 40, 10812.
J. Popp, M. Schwalbe, S. Krieck, H. Görls, F. W. Heinemann 52 V. Artero and M. Fontecave, Chem. Soc. Rev., 2013, 42, 2338.
and S. Rau, Dalton Trans., 2010, 39, 2359.
30 S. Rau, B. Schäfer, D. Gleich, E. Anders, M. Rudolph,
53 J. A. Widegren and R. G. Finke, J. Mol. Catal. A: Chem.,
2003, 198, 317.
M. Friedrich, H. Gorls, W. Henry and J. G. Vos, Angew. 54 A.-M. Albrecht-Gary, C. Dietrich-Buchecker, Z. Saad and
Chem., Int. Ed., 2006, 45, 6215. J.-P. Sauvage, J. Am. Chem. Soc., 1988, 40, 1467.
31 S. Tschierlei, M. Presselt, C. Kuhnt, A. Yartsev, T. Pascher, 55 M. P. McLaughlin, T. M. McCormick, R. Eisenberg and
V. Sundström, M. Karnahl, M. Schwalbe, B. Schäfer, S. Rau, P. L. Holland, Chem. Commun., 2011, 47, 7989.
M. Schmitt, B. Dietzek and J. Popp, Chem. – Eur. J., 2009, 56 J. Halpern, J. Phys. Chem., 1959, 63, 398.
15, 7678.
57 W. Sattler and G. J. Parkin, Am. Chem. Soc., 2012, 134,
32 P. Lei, M. Hedlund, R. Lomoth, H. Rensmo, O. Johansson
and L. Hammarström, J. Am. Chem. Soc., 2007, 130, 26.
33 M. Pfeffer, B. Schäfer, C. Kuhnt, M. Schmitt, J. Popp,
17462.
58 V. Balzani, A. Credi, F. Raymo and J. Stoddart, Angew.
Chem., Int. Ed., 2000, 39, 3348.
J. Guthmuller, G. Smolentsev, J. Uhlig, E. Nazarenko, 59 C. Dietrich-Buchecker, M. C. Jimenez-Molero, V. Sartor and
V. Sundström, L. González, B. Dietzek and S. Rau, manu-
script in preparation.
34 D. Takeuchi, A. Yasuda and K. Osakada, Dalton Trans.,
2003, 2029.
35 J. Frey, T. Kraus, V. Heitz and J.-P. Sauvage, Chemistry, 2007,
13, 7584.
36 E. C. Constable, R. P. G. Henney, T. A. Leese and
D. A. Tocher, J. Chem. Soc., Dalton Trans., 1990, 443.
37 S. W. Lai, T. C. Cheung, M. C. Chan, K. K. Cheung,
S. M. Peng and C. M. Che, Inorg. Chem., 2000, 39, 255.
J.-P. Sauvage, Pure Appl. Chem., 2003, 75, 1383.
60 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 11676–11686 | 11685

View Article Online
Paper
Dalton Transactions
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, 63 P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, 213–222.
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, 64 J. P. Merrick, D. Moran and L. Radom, J. Phys. Chem., 2007,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, 111, 11683.
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, 65 J. Guthmuller and L. González, Phys. Chem. Chem. Phys.,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, 2010, 12, 14812.
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, 66 J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, 105, 2999.
Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford, CT, 67 J. B. Page and D. L. Tonks, J. Chem. Phys., 1981, 75, 5694.
2009.
68 W. L. Peticolas and T. Rush, J. Comput. Chem., 1995, 16,
61 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648; (b) C. Lee,
1261.
W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter, 1988, 69 J. Guthmuller and L. González, Phys. Chem. Chem. Phys.,
37, 785. 2010, 12, 14812.
62 D. Andrae, U. Häußermann, M. Dolg, H. Stoll and 70 S. Kupfer, M. Wächtler, J. Guthmuller, J. Popp, B. Dietzek
H. Preuß, Theor. Chim. Acta, 1990, 77, 123.
and L. González, J. Phys. Chem. C, 2012, 116, 19968.
11686 | Dalton Trans., 2014, 43, 11676–11686
This journal is © The Royal Society of Chemistry 2014