Enhancing Photocatalytic Hydrogen Generation: the Impact of the Peripheral Ligands in Ru/Pd and Ru/Pt Complexes

Article abstract of DOI:10.1002/chem.201605980

The synthesis, photophysical properties and photocatalytic efficiency of a range of novel supramolecular assemblies of the type [Ru(dceb)₂(μ-bisbpy)MCl₂][PF₆]₂ and [Ru(bpy)₂(μ-bisbpy)MCl₂][PF₆]₂ (M=Pd or Pt, dceb=diethyl 2,2’-bipyridine-4,4’-dicarboxylate, bpy=2,2’-bipyridine and bisbpy=2,2’:5’,3’’:6’’,2’’’-quaterpyridine) are reported. Photocatalytic hydrogen generation was dependent on the nature of the peripheral ligand, on the catalytic centre and on the amount of water present in the photocatalytic mixture. The best catalytic conditions were obtained with the dceb peripheral ligand (turnover numbers up to 513 after 18 h). The experimental data and DFT calculations on both the bpy- and dceb-based compounds indicated that the peripheral dceb ligands participated in the photocatalytic process.

Full text of DOI:10.1002/chem.201605980

A Journal of
Accepted Article
Title: Enhancing Photocatalytic Hydrogen Generation: The Impact of
the Peripheral Ligands in Ru/Pd and Ru/Pt complexes
Authors: Nivedita Das, Gurmeet Singh Bindra, Avishek Paul, Johannes
G. Vos, Martin Schulz, and Mary T. Pryce
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.201605980
Link to VoR: http://dx.doi.org/10.1002/chem.201605980
Supported by

Chemistry - A European Journal
10.1002/chem.201605980
Enhancing Photocatalytic Hydrogen Generation: The Impact of the
Peripheral Ligands in Ru/Pd and Ru/Pt complexes
[
a]
[a]#
[a]
[a]
[b]
Nivedita Das , Gurmeet Singh Bindra , Avishek Paul , Johannes G. Vos , Martin Schulz* and
[
a]
5
Mary T. Pryce*
[
[
a]
b]
SRC for Solar Energy Conversion, School of Chemical Sciences, Dublin City University, Dublin 9, Ireland
Friedrich Schiller University Jena, Institute of Physical Chemistry, Helmholtzweg 4, 07743 Jena, Leibniz Institute of
Photonic Technology (IPHT), Albert-Einstein-Straße 9, 07745 Jena, Germany.
#
Current address: Exigence Technologies Inc., 200-135 Innovation Drive, Winnipeg, Manitoba Canada R3T 6A8
1
1
2
2
3
0
Abstract
The synthesis, photophysical properties and photocatalytic efficiency of a range of novel
supramolecular assemblies of the type [Ru(dceb) (µ-bisbpy)MCl ](PF ) and [Ru(bpy) (µ-
2
2
6 2
2
bisbpy)MCl ](PF ) (where M = Pd or Pt, dceb = diethyl 2,2'-bipyridine-4,4'-dicarboxylate, bpy = 2,2'-
2
6 2
5
0
5
0
bipyridine and bisbpy = 2,2’:5’,5”:2”,2’’’-quaterpyridine) are reported. Photocatalytic studies for
hydrogen generation were found to be dependent on the nature of the peripheral ligand, on the
catalytic centre and on the quantity of water present in the photocatalytic mixtures. The best catalytic
conditions were obtained with dceb as the peripheral ligand, with turnover numbers up to 513 after 18
hours. The experimental data together with DFT calculations on both the bpy and dceb based
compounds indicate that the peripheral dceb ligands participate in the photocatalytic process.
Introduction
The observation that weather patterns are now becoming more and more erratic together with the fact
that CO levels in the atmosphere have surpassed the 400 ppm mark requires the development of
2
novel environmental friendly energy resources to counteract climate change. Hydrogen as an energy
source is an attractive option, since it is carbon neutral and its application as a fuel in various energy
sectors including transport is widely promoted. However, at present the vast percentage of hydrogen
is obtained from fossil fuels, using methods that produce large amounts of CO . An attractive and
2
greener approach is the direct generation of hydrogen from water using intramolecular photocatalytic
assemblies capable of absorbing solar light.
1
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
Figure 1. Intramolecular photocatalytic assembly. S = sacrificial reductant, PS = photosensitiser, Cat
catalytic centre and B = bridging ligand.
=
5
Therefore, molecular photocatalysts involving a photosensitiser and a catalytic centre have been
1
studied using both, inter-molecular or intra-molecular pathways. In the intramolecular approach, a
photosensitiser is connected to the hydrogen generating catalytic centre via a bridging ligand as
2,3,4
shown in Figure 1.
Following excitation of the sensitiser two electrons are transferred via the
bridging ligand to the catalytic centre most likely in a two-step process leading to the generation of
hydrogen. One problem with the homogeneous approach is that sacrificial agents are required (Figure
1
0
5,6
1
), that can lead to the decomposition of the photocatalytic assemblies. The use of sacrificial
agents can be circumvented by the immobilisation of photocatalytic assemblies onto p-NiO electrode
7
surfaces via the attachment by carboxylate based linkers.
1
5
Figure 2. Bridging ligands discussed in the text; 2,2’-bipyridine (bpy), 4,4’-diethylcarboxy-2,2’-
4a
2b
bipyridine (dceb), 2,5-di(pyridin-2-yl)pyrazine (2,5-dpp) , 2,2’:5’,2’’-terpyridine, (2,5-bpp) , 2,3-
3b
di(pyridin-2-yl)pyrazine (2,3-dpp) , 2,2':5',3'':6'',2'''-quaterpyridine (bisbpy) this work.
2
0
In a series of studies, we have investigated the introduction of carboxylate groupings at the peripheral
ligands [Ru(bpy) -based photosensitisers on the photocatalytic hydrogen generating capacity of
2
3
molecular assemblies. The results obtained show that the peripheral ligands are actively involved in
2
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
the catalytic activity-determining early time excited state processes and dramatically change the
8
electron flows in the photocatalysts. In these studies, the bridging ligands, 2,5-dpp, 2,3-dpp and 2,5-
bpp (Figure 2) were used and upon the introduction of carboxylate moieties onto the peripheral
3a
3b
ligands the TON values increased from 0 to 400 (18h) (PdCl catalyst) , 0 to 44 (18h) (PtCl2
2
8
5
0
5
0
5
0
5
catalyst) and 80 to 650 (6h) (PtI catalyst) respectively. What these ligands (2,5-dpp, 2,3-bpp and
2
2
,5-bpp) have in common is that the photosensitiser and the catalytic centre are both coordinated to
the central ring. It could be argued that this may lead to a strong interaction between the two centres
and that this may instigate a fast back reaction thereby limiting the photocatalytic process.
1
1
2
2
3
3
In this contribution we report our results on hydrogen evolving intramolecular assemblies based on
ruthenium as the photosensitiser, together with Pd or Pt as the catalytic centres. The peripheral bpy
ligands are either unsubstituted or modified with the electron-withdrawing 4-carboxyethyl substituent,
as indicated in Figure 3. 2,2’:5’,5”:2”,2’’’-quarterpyridine (bisbpy) is introduced as the bridging ligand.
This compound was chosen for two reasons, firstly electrochemical measurements have shown that
4+
for [Ru(bpy) (µ-bisbpy)Ru(bpy) ] , (Ru(bisbpy)Ru) the bisbpy ligand is first reduced and, therefore,
2
2
9,10
its MLCT levels are expected to be at lower energy than those of the peripheral bpy ligands.
Secondly, the bisbpy ligand conceptually consists of two bipyridine subunits connected by a freely
rotating single bond. Coordination of metal ions to these bipyridine subunits does not force the bisbpy
ligand into a planar geometry, which is in contrast to the previously investigated bridging ligands 2,5-
dpp, 2,3-dpp and 2,5-bpp. Consequently, the free rotation of the bipyridine subunits may either allow
electron delocalisation over both bipyridine subunits of the bridge or it may be localised on a single
bipyridine moiety. In the latter case the formation of localised bipyridine Pd- or Pt-hybrid based
orbitals would be possible, potentially allowing for reduced electron back-transfer from the Pd- or Pt-
hybrid orbital to the Ru-centre and hence improved hydrogen evolution.
The results presented in this contribution demonstrate that with dceb peripheral ligands higher TON
values of up to 500 for hydrogen generation are obtained compared with those obtained for
unsubstituted peripheral ligands. DFT studies on the Ru/Pd compounds suggest that the increased
3
catalytic efficiency of the dceb type compounds may be related to a MLCT state where the singly
occupied orbitals are localised on a bipyridine-Pd moiety located on the bridge and the dceb ligand.
1
H NMR spectroscopy indicates that upon mixing of compound 1b with the catalytic centre
[
Pd(CH CN) Cl ] in situ, in a 1:1 ratio results in the formation of a dinuclear TON values obtained from
3 2 2
this mixture are almost identical to those obtained from the isolated supramolecular assembly 2b,
illustrating that “inter-molecular” photocatalytic experiments may in fact lead to “intra-molecular”
photocatalytic assemblies.
3
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
Figure 3: Molecular structures [Ru(bpy) (bisbpy)](PF6)2 (1a), [Ru(dceb) (bisbpy)](PF ) (1b),
2
2
6 2
[
Ru(bpy) (µ-bisbpy)PdCl ](PF )
(2a),
[Ru(dceb) (µ-bisbpy)PdCl ](PF )
(2b),
[Ru(bpy) (µ-
2
2
6 2
2
2
6 2
2
5
0
5
0
bisbpy)PtCl ](PF ) (3a) and [Ru(dceb) (µ-bisbpy)PtCl ](PF ) (3b)
2 6 2 2 2 6 2
Results and Discussion
1
Synthesis and H NMR characterisation
Recently we reported a multi-step synthesis to obtain Ru and Os based bimetallic complexes based
on the bisbpy ligand. The synthesis was based on a Ni(0) catalysed cross-coupling reaction of
1
1
2
9,10
brominated Ru and Os polypyridine precursors.
In the present contribution this ligand is prepared
in two steps using a Negishi coupling protocol. In the first step 2,5-dibromopyridine is coupled with 2-
11
pyridylzinc bromide yielding 5-bromo-2,2’-bipyridine followed by a Ni(0) catalysed homo-coupling of
the 5-bromo species producing bisbpy with an overall yield of 60%. The presence of excess triphenyl
phosphine requires careful purification (see Experimental Part). The mononuclear ruthenium
2+
2+
complexes [Ru(bpy) (bisbpy)] , (1a) [Ru(dceb) (bisbpy)] , (1b) and the deuteriated analogue [Ru(d -
2
2
8
2+
3
bpy) (bisbpy)] (1c) were obtained using standard procedures. To avoid the formation of bimetallic
2
compounds a diluted solution containing the metal centre was added slowly to a 1.5-fold excess of the
bridging ligand. All compounds were fully characterised by spectroscopic methods and their purity
was furthermore confirmed by elemental analysis (see Experimental Section and ESI). Some typical
1
H NMR spectra are shown in Figure 4 outlining the usefulness of partially deuteriated complexes in
1
the assignment of complicated H NMR spectra. Assignments and additional spectra are shown in the
experimental section and ESI respectively.
4
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
1
Figure 4. H NMR (dmso-d , 400 MHz) of (top) [Ru(bpy) (bisbpy)](PF ) (bottom) [Ru(d8-
6
2
6 2,
bpy) (bisbpy)](PF6)2
2
5
0
5
0
5
Photophysical characterisation
1
2+
The MCLT transitions of 1a closely resemble those of [Ru(bpy) ] , however, a red-shifted shoulder is
3
evident indicating the involvement of bisbpy, which is in agreement with similar absorption features
4+
9
previous observed for [Ru(bpy) (µ-bisbpy)Ru(bpy) ] , (Ru(bisbpy)Ru) (see Table 1). Substitution of
2
2
1
1
2
2
bpy to form dceb leads to a 0.31 eV redshift of the dceb-based π-π*-transition at 307 nm. The lowest
1
energy MLCT-transition is similarly red-shifted by about 0.16 eV. The introduction of palladium (2a-b)
or platinum (3a-b) results in relatively small changes in the spectral features (ESI. Figure S4-S6 ). The
emission maxima for the bimetallic compounds do not vary greatly when compared with the precursor
compounds (Table 1, ESI. Figure S4-S6). Compound 1a emits at 633 nm, while the emission
maximum of the RuPd analogues 2a and 3a is shifted to lower energy by between 0.07-0.09 eV.
Similar emission profiles were obtained for compounds 1b, 2b and 3b.
The observed emission quantum yields of all investigated compounds are low and in the range
between about 2-10%. However, for the dinuclear compounds 2b, 3a and 3b reduced lifetimes are
observed which correlate with decreased emission quantum yields. Compound 2a is an exception,
showing a decreased lifetime but without significant change of the quantum yield. This indicates an
increased radiative rate, probably due to increased spin-orbit coupling and was previously observed
2+ 3a,12
for [Ru(bpy) (µ-2,5-bpp)PdClSolvent] .
Furthermore, as shown in Table 1, mono-exponential
2
decays were observed for 1a and 1b,and for the Pd based photocatalysts 2a and 2b dual exponential
decays are observed (see ESI, Figure S8-S10 ). A dual emission decay is unusual, but we have
previously observed this for another Ru-Pd type photocatalyst involving the 2,5-bpp bridging ligand
5
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
12
(
cf. Fig. 2). For this compound two lifetimes of 400 (37%) and 88 (63%) ns were observed, which
3
are most likely associated with emission from the MLCT based on bpy peripheral and the bridging
ligand respectively. Further experiments are needed study the reasons for the dual exponential decay
observed for compounds 2a and 2b.
5
Table 1 Photophysical properties of the compounds 1a-b, 2a-b, 3a-b and related compounds.
a
b
c
Compounds
λ,abs, max [nm]
λ,em,max [nm]
τ (ns)
Φ
4
-1
-1
(
ε in 10 M cm )
293 K
293 K
293K
1a
1b
2a
448 (1.16)
475 (1.13)
447 (1.02)
633
647
655
755
825
0.040
0.101
0.044
17 (37%)
218 (63%)
2b
475(1.76)
647
42 (60 %)
0.033
325 (40%)
3
a
b
447 (1.15)
475 (1.63)
440 (2.53)
664
644
663
331
0.017
0.054
3
534
9
Ru(µ-bisbpy)Ru
215*
N/A
2
+
13
[
[
[
[
[
Ru(bpy)3]
450 (1.30)
483 (1.65)
605
662
860
624
-
0.068
-----8
2+
Ru(dceb) (µ-2,5-bpp)PtISolvent]
2
2
+
------3b
------3b
-------3a
------3a
Ru(bpy) (µ-2,3-dpp)PtI ]
2
2
516 (4.32)
778 (vw)
2
+
Ru(dceb) (µ-2,3-dpp)PtI ]
485(4.25)
539(4.00)
526(4.20)
-
-
2
2
2
+
Ru(bpy) (µ-2,5-dpp)PdCl ]
807
778
<0.5
<0.5
2
2
2+
[
Ru(dceb) (µ-2,5- dpp)PdCl ]
2 2
a
b
Excitation at 455 nm. Lifetime determined by TCSPC at 293 K in deaerated acetonitrile solution
c
(
λ . = 360). Quantum Yields determined at 293 K in deaerated acetonitrile at λ . = 455 nm, S =
exc
exc
solvent, * = aerated; N/A = not available.
1
0
Photocatalysis
The complexes 2a-b and 3a-b were tested as photocatalysts for hydrogen production. Mixtures of the
photocatalyst and the sacrificial donor (triethylamine, TEA), were irradiated at 470 nm in acetonitrile
solutions with varying water content (for further details see experimental section). Control experiments
did not produce any hydrogen. Intermolecular studies were also carried out with solutions containing
equimolar mixtures of the photosensitiser 2b and a Pd catalyst. All results obtained are listed in Table
1
5
2. The data listed in Table 2 show that the TONs obtained are dependent on both the nature of the
reaction mixture, i.e. the amount of water present and the moieties present on the peripheral ligands
6
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
of the photocatalysts. They indicate that the best TON values, are obtained in the presence of 5%
water. Therefore, the discussion is mainly focused on these results. The results obtained indicate that
the catalytic process is more efficient for the compounds, 2b and 3b, and that the Pd catalysts are
14
more effective than their Pt analogues, as has also been found by others. Similarly to the
enhancement in the TONs observed for complex 2b (when compared to 2a), the TON value acquired
for the platinum analogue 3b is also substantially greater than that of 3a, thereby further indicating the
strongly enhanced photocatalytic efficiency in the presence of dceb peripheral ligands. We note that
in the presence of 0% and 15% of water the Pt compound 3a (bpy ligands) shows a higher TON value
than the associated Pt compound 3b which contains carboxy based ligands. In addition 3a is also
more efficient than the Pd analogue 2a. The Pd compound 2b remains the better photocatalyst in
solutions containing 15% water. These observations are at present not understood and further studies
are needed to explain these behaviour.
5
1
0
Table 2 Turnover numbers (TONs) for photocatalytic hydrogen generation.
a
Compounds
TON as function of water content (v/v)
0%
5%
10%
15%
2
2
3
3
1
1
1
a
0
23
29
13
n.d.
0
5
513
75
258
489
0
16
360
92
286
321
0
13
198
85
45
n.d.
0
b
a
b
b
b + [Pd(CH CN) Cl ]
3
2
2
a
b
0
0
0
0
[
Pd(CH CN) Cl ]
0
0
0
0
3
2
2
[
Pt(dmso) Cl ]
0
0
0
0
2
2
TEA
0
0
0
0
2+
[
Ru(bpy) (µ-2,5-bpp)PtISolvent]
100
8
2
-
---
2+
8
[
[
[
[
Ru(dceb) (µ-2,5-bpp)PtISolvent]
720
0
----
2
2+
3b
3b
3a
Ru(bpy) (µ-2,3-dpp)PtI ]
----
----
2
2
2+
Ru(dceb) (µ-2,3-dpp)PtI ]
44
2
2
2+
Ru(bpy) (µ-2,5-dpp)PdCl ]
0
-----
2
2
2+
3a
[
Ru(dceb) (µ-2,5-dpp)PdCl ]
400
----
2
2
a
1
5
Determined by GC following irradiation (irradiation wavelength = 470 nm) for 18 h. The TON values
b
are averages of three samples and could be determined with a standard deviation of 12% or better.
7
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
The reaction mixture contains an equimolar mix of the mononuclear complex and Pd catalysts (4.08 ×
–5
1
0 M), and triethylamine 2.15 M and varies percentages of water (0-15%). n.d = not determined.
The TONs obtained for the complexes 2b and 3b containing dceb as peripheral ligands are 513 and
258 respectively. These TONs are considerably higher than obtained for the bpy analogues 2a and 3a
which yielded values of 5 and 75 respectively. A similar dependence of the catalytic performance on
5
0
5
0
5
2+
the peripheral ligand was recently reported for the [Ru(dceb) (2,5-dpp)PdCl ] complex with a value
2
2
2+
of 400 (18 h) with the bpy analogue, [Ru(bpy) (2,5-dpp)PdCl ] not producing any detectable
2
2
3a
amounts of hydrogen (for bridging ligand see Fig. 2).
1
1
2
2
Another important observation is that the photocatalytic efficiency observed for the supramolecular
assembly 2b is very similar to that observed for the mixture of 1b and [Pd(CH CN) Cl ]. To further
3
2
2
investigate this observation the time dependence of both processes was investigated under identical
photocatalytic conditions. The results obtained are shown in Figure 5. This Figure shows that both
systems reach a plateau after ~ 8 h with a TON value of about 480 but are still producing hydrogen
after 18 h of irradiation reaching a TON of about 510. For 2b no lag phase is observed, but for the
mixture of 1b and [Pd(CH CN) Cl ] a small delay at the start of the photocatalytic process is observed.
3
2
2
However, the turnover frequencies obtained after 2 hours (TOF) are similar in both instances with a
-1
-1
value of about 60 h for 2b while for the mixture of 1b and [Pd(CH CN) Cl ] a value of about 50 h is
3
2
2
obtained and at longer times the TOF values remain comparable. The similarity in the quantities of the
hydrogen produced and of the hydrogen generation profile with time suggests the in situ formation of
1
a dinuclear species. This issue is further investigated with H NMR and the results obtained from
these studies are shown in the ESI. It is important to note that the concentrations of the reactants
used in the NMR studies are more than an order of magnitude higher than those in the photocatalytic
studies. As a result in the NMR studies containing a mixture of 1b and [Pd(CH CN) Cl ] the
3
2
2
intramolecular species is formed after ~5 min, while in the photocatalysis experiments formation of
the intramolecular compound is slower.
8
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
Figure 5. Time-dependence of the TON achieved for H evolution for 2b (black line) and 1b +
2
[
Pd(CH CN) Cl ] (red line).
3 2 2
5
0
5
0
5
Computational results
12,15
Recently we have studied photocatalysts containing the 2,5-bpp ligand.
In this contribution we
have extended our studies to include the bisbpy bridging ligand. As outlined above this ligand is
different from the earlier used bpp and dpp type ligands, since the photosensitiser and the catalytic
center are now bound to different parts of the bridging ligand through coordination bonds rather than
through the same ring by cyclometallation. In this section DFT studies are used to further investigate
the electronic properties of the compounds reported here. For details on the methods and parameters
used see the experimental part. To reduce the computational costs the dceb ethyl ester groups were
replaced by methyl esters therefore the compounds 2a and 2b are now labelled as 2a’ and 2b’ (See
Figure 3). In these studies the location of the molecular orbitals associated with the singlet ground
1
1
2
2
3
state (SGS), the lowest energy triplet state ( MLCT) and the singly-reduced doublet (SRDS) states
were investigated. The latter state is essentially a ground state with an added electron and could be
obtained by reductive quenching after photoexcitation. The location of the frontier orbitals was
evaluated by a population analysis of the optimised dinuclear Ru-Pd complexes (see ESI, Tables S2-
S7). In the singlet ground states of both 2a’ and 2b’ the three highest energy occupied orbitals are
ruthenium based orbitals as expected (HOMO, H-1, H-2). The LUMO for 2a’ is bridge-based,
however, for 2b’ the LUMO is a mixed orbital which is located on both the peripheral MeOOC-bpy
ligands (63%) and the bridging ligand (35%). The orbitals associated with the singly reduced species
SRDS of 2a’ (doublet state, overall charge is +1) are based on the bridging ligand (cf. α-MO 197 in
Table S4, ESI), while this orbital is found on the MeOOC-bpy peripheral ligand for 2b’ (cf. α-MO 257
in Table S7, ESI). The lower lying doubly occupied orbitals are Ru-based as expected. Population
analysis of the lowest energy triplet states in 2a’ and 2b’ reveals that for 2a’ both singly occupied
9
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
orbitals are bridge-based (cf. Fig.6a,b and α-MOs 197-196 in Table S3, ESI). However, for 2b’ the
high energy singly occupied orbital is based on the peripheral MeOOC-bpy ligand while the low
energy singly occupied orbital is a hybrid bridging ligand-PdCl orbital (cf. Fig.6c and Fig. 6d
2
respectively see also and α-MO 257-256 in Table S6, ESI). The differences in the electronic situation
are also reflected by the torsion angle between the 2,2’-bipyridine moieties that make up the bisbpy
bridging ligand. For 2a’ the torsion angle of 17.1° is considerably smaller than for 2b’ (35.5°) owing to
the delocalisation of the unpaired electrons on the bridging ligand in the former case and their
5
localisation on the peripheral ligand and on a hybrid bridging ligand/PdCl orbital in the latter.
2
1
0
Figure 6: DFT representation of the two singly occupied orbitals of the lowest energy triplet state of
2a’ (a, b) and 2b’ (c, d)
The complexes reported in this contribution are part of a series of compounds published by our group,
where we have investigated a range of Ru-Pd and Ru-Pt dyads with various bridging (see Figure 2)
1
5
3,8,12,15
and peripheral ligands using the same experimental conditions.
As observed for the compounds
investigated in this contribution, their catalytic activity as hydrogen evolving catalysts was found to be
dependent on changes on the bridging or on the peripheral ligands. Of interest in this respect is a
recent ultrafast transient absorption and resonance-Raman investigation on the hydrogen evolving
2+
12
2
0
[Ru(bpy) (2,5-bpp)-PdCl(acetonitrile)] (for bridging ligand see Fig. 2). In this recent publication we
2-
2+
demonstrated that excitation of [Ru(bpy) (2,5-bpp)-PdCl(acetonitrile)] at 470 nm resulted in
2-
1
0
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
1
population of MLCT states on the peripheral bpy- and the 2,5-bpp bridging ligand, followed by
3
12
intersystem crossing into the corresponding hot MLCT states. Furthermore, it was shown that
electron density localised on the bpy ligand cannot flow to the bridging 2,5-bpp ligand, this will limit
electron transfer to the bridging ligand and will reduce hydrogen generation. This is confirmed by the
detection of two emission channels indicating that electron density remains on the peripheral bpy
ligand. Efficient interligand electron transfer is important for efficient hydrogen evolution but this is not
observed for standard, unsubstituted, bpy ligands. However, it was suggested that this can be
achieved by the introduction of functionalised peripheral ligands. Our later studies involving peripheral
ligands modified with ester groupings have confirmed this.
5
0
5
0
5
0
5
1
1
2
2
3
3
A second aspect raised by the investigation in ref. 12 is the localisation of the electron density after
the interligand electron transfer process and subsequent transfer to the catalytically active metal
3
centre. This ligand-to-metal electron transfer process competes with the decay of the MLCT state of
the bridging 2,5-bpp ligand to the ground state, thus limiting hydrogen evolution. It was argued that
the population of a hybrid orbital formed between the bridging 2,5-bpp ligand and the Pd centre may
be beneficial for intermediate electron storage and subsequent proton reduction.
The DFT results obtained in this contribution provide direct evidence for such an orbital. The observed
superior catalytic activity of the dceb complex 2b, with respect to the bpy complex 2a, is in line with
the DFT results for 2b’ showing localised orbitals based on the peripheral dceb ligand as well as on
the PdCl binding part of the bisbpy ligand. In contrast, the delocalised singly occupied orbitals found
2
3
for the MLCT state of 2a’ are most likely disadvantageous for the catalytic process, since the strong
coupling may also enhance electron recombination to the singlet ground state. The bisbpy bridging
ligand is able to adopt both configurations which allows strong coupling via delocalised electron
density as well as localised electron density by virtue of a free rotating single bond, which connects its
bipyridine subunits. Interestingly, it is the peripheral ligand that causes the formation of either one of
these configurations. Despite the possibility of free rotating subunits in the bisbpy bridge, the impact
on the hydrogen generation performance of complexes with different bridging ligands but with the
dceb peripheral ligands is less dramatic than with the same bridge but different peripheral ligands (cf.
Table 2). However, great care has to be taken in this discussion since the catalytic performance is
certainly influenced by a range of (partly unknown) parameters. Moreover, no information is available
on the nature of the catalytically active species. Such species are of great importance but are
experimentally difficult to detect.
Conclusions
In this contribution we report the synthesis and characterisation of novel intra-molecular Ru/Pd and
1
1
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
Ru/Pt complexes tethered to the bisbpy bridging ligand. This work is part of a series of systematic
studies aimed at investigating the influence of bridging ligands, peripheral ligands and the nature of
3,4,8,12,15
catalytic centres may have on light driven hydrogen generation.
The most striking observation
in the present investigations is that much increased TON numbers are obtained when the peripheral
dceb ligands are introduced irrespective of the nature of the bridging ligand or catalytic centre.
However, emission data indicate that emission decays are different for Pd and Pt containing
5
0
5
0
5
0
5
15
complexes and hence the electron flows in those complexes. In an earlier study transient absorption
measurements indicated the dceb ligands most likely act as electron storage ligands, contrary to what
8,12,15
is expected in the traditional assumption that the electron density is stored on the bridging ligand.
1
1
2
2
3
3
The present contribution supports this perception, based on the obtained experimental and DFT data.
The latter indicate the presence of an electronic configuration with one singly occupied Pd-bridge
hybrid orbital and one on the dceb peripheral ligands. The data presented emphasise, that even
relatively small structural changes can have a large impact on the catalytic efficiency. Finally, as
pointed out in the introduction, for the design of a sustainable technology, immobilisation of
photocatalysts on electrode surfaces is necessary. This contribution shows that the introduction of
anchoring carboxylate moieties onto the peripheral ligands can have a dramatic influence on the
hydrogen generation efficiency and underlines the need for well-balanced energy levels between all
three components of the intra-molecular photocatalyst, i.e. photosensitiser, bridging ligand and
catalytic centre.
Experimental Section
The compounds 2,2’-bipyridine (bpy), RuCl ∙ H O and [Pd(CH CN) Cl ] are from Aldrich. The
3
2
3
2
2
16
17
8
8
synthesis of [Ru(bpy) Cl ].2 H O , [Ru(d -bpy) Cl ].2H O [Ru(dceb) Cl ].2H O and [Pt(dmso) Cl ]
2
2
2
8
2
2
2
2
2
2
2
2
are reported elsewhere.
2,2’:5’,5”:2”,2’’’-quaterpyridine (bisbpy)
Step-1: Under nitrogen atmosphere Pd(PPh ) (0.3 g, 0.26 mmol) and 2,5-dibromobipyridine (2.0 g,
3
4
8.44 mmol) were added to a dried two neck round bottom flask and cooled to 0°C. Upon addition of 2-
3
pyridylzinc bromide in THF (19.4 cm , 8.44 mmol) the reaction mixture turned brown in colour and
was then stirred over night at room temperature under inert atmosphere while a white precipitate
3
formed. Subsequently the reaction mixture was poured in to 200 cm of a saturated aqueous solution
of equimolar EDTA and Na CO and stirred until the white solid turned completely yellow. The
2
3
3
aqueous suspension extracted with dichloromethane (3 × 50 cm ), the combined organic layers were
dried over sodium sulphate and the solvent was evaporated by vacuum. The obtained crude product
was purified on an activated neutral alumina (150 mesh size) column using hexane/ethylacetate
1
2
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
(
9.5:0.5 v/v), monitored by TLC (second spot, R = 0.2) and yielded 5-bromo-2,2’-bipyridine (81 %).
f
3
Step-2: [Ni(PPh ) Cl ] (0.557 g, 0.85 mmol) was added in to 10 cm of anhydrous DMF and stirred for
3
2
2
few minutes at room temperature under nitrogen atmosphere, until the mixture turned blue. After
addition of zinc powder (0.056 g, 0.85 mmol) the mixture was stirred at room temperature under
nitrogen atmosphere for 45 min while the colour turned green and ultimately deep brown.
Subsequently, 5-bromo-2,2’-bipyridine (0.200 g, 0.85 mmol) was added and the reaction mixture was
5
0
5
0
5
0
5
3
stirred at room temperature for 10 hours and was then poured into 150 cm of 3 molar aqueous NH
4-
OH solution while a greyish product precipitated. The precipitate was extracted with ethyl acetate (3 ×
3
5
0 cm ) and the combined organic phases were dried over MgSO . Removal of the solvent in vacuo
4
1
1
2
2
3
3
yielded the crude product which was purified by column chromatography (neutral alumina,
hexane/ethyl acetate (9:1 v/v), TLC: R = 0.1) to obtain 2,2’:5’,5”:2”,2’’’-quaterpyridine in 78 % yield.
f
Anal. Calcd. for C H N 0.2 H 0 (313.95): C, 76.51; H 4.62; N, 17.85%. Found: C, 76.84; H, 4.63; N,
20
14
4
2
1
1
7.24%. H-NMR (dmso-d 400 MHz): δ = 7.50 (dd, 2H, J = 7.4 Hz, J = 2.5 Hz, J = 1.1 Hz, H-5, H-
6,
5”’), 8.00 (dd, 2H, J = 7.8 Hz, J = 2.0, H-4, H-4”’), 8.43 (dd, 2H, J = 8.3 Hz, J = 2.5 Hz, H-4’, H-4”),
8.46 (dd, 2H, J = 7.8 Hz, J = 1.0 Hz, H-3, H-3”’), 8.54 (dd, 2H, J = 8.0 Hz, J = 1.0 Hz, H-3’, H-3”), 8.72
dd, 2H, J = 4.8 Hz, J = 1.0 Hz, H-6, H-6”’), 9.18 (d, 2H, J = 2.5 Hz, H-6’, H-6”).
(
[
[
Ru(bpy) (bisbpy)](PF ) ∙ 2 H O (1a)
2 6 2 2
3
Ru(bpy) Cl ] ∙ 2 H O (0.34 g, 0.65 mmol) dissolved in 5 cm of ethanol was added drop-wise to a
2
2
2
3
solution of bisbpy (0.26 g, 0.85 mmol) in 10 cm of ethanol/water (3:1 v/v). The reaction mixture was
refluxed for 8 hours and then allowed to cool to room temperature and the solvent was evaporated
3
under vacuum. To the residue, 2 cm of deionized water was added in order to dissolve the product
and excess bisbpy was filtered off. Subsequently, the product was precipitated by addition of KPF to
6
3
3
the filtered solution. The precipitate was washed twice with a total of 5 cm of water and 10 cm of
diethylether. Recrystallization from acetone/water (3:1 v/v) afforded a red solid. Yield: 0.588 g (0.56
mmol, 86%). Anal. Calcd. for C H F N P Ru ∙ 2 H O (1049.75): C, 45.77; H, 3.26; N, 10.67%.
40
30 12
1
8
2
2
Found: C, 46.15; H, 2.86; N, 10.29%. H-NMR (Acetonitrile-d , 400 MHz): δ = 8.72-8.68 (m, 2H, H-6,
3
H-6’), 8.65 (d, 1H, J = 8.0 Hz, H-3”), 8.58 (d, 1H, J = 7.8 Hz, H-3”’), 8.62-8.50 (m, 4H, bpy (H-3a)),
8
.45- 8.39 (m , 2H, H-3’, H-4”), 8.15-8.05 (m, 5H, bpy (H-4a), H-4”’), 7.97-7.89 (m, 4H, H-3, H-4, H-4’,
H-6”), 7.83 (d, 1H, J = 4.8 Hz, bpy (H-6a), 7.79-7.72 (m, 4H, bpy (H-6a), H-6”’), 7.47-7.39 (m , 6H, bpy
H-5a), H-5, H-5”’).
(
[
[
Ru(dceb) (bisbpy)](PF ) ∙ 2CH COCH (1b)
2 6 2 3 3
3
Ru(dceb) Cl ] ∙ 2 H O (0.52 g, 0.65 mmol) dissolved in 5 cm of ethanol was added drop-wise to a
2
2
2
3
solution of bisbpy (0.26 g, 0.80 mmol) in 10 cm of ethanol/water (3:1 v/v). The reaction mixture was
1
3
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
refluxed for 8 hours and then allowed to cool to room temperature and the solvent was evaporated
under vacuum. Subsequently, the product was precipitated by addition of KPF to the filtered solution.
6
3
3
The precipitate was washed twice with a total of 5 cm of water and 10 cm of diethylether.
Recrystallization from acetone/water (3:1 v/v) afforded a red solid. Yield: 0.765 g (0.54 mmol, 83%).
Anal. Calcd. for C H F N P Ru ∙ 2 CH COCH (1418.26): C, 49.12; H, 4.12; N, 7.90%. Found: C,
5
0
5
0
5
0
5
52
46 12
8
2
3
3
1
4
9.42; H, 4.14; N, 7.84%. H-NMR (Acetonitrile-d , 400 MHz): δ = 9.13-9.08 (m, 4H, bpy (H-3a)), 8.74-
3
8.66 (m, 3H, H-3”, H-6, H-6’), 8.61 (d, 1H, J = 8.0 Hz, H-3”’), 8.46 (m, 1H, H-4”), 8.42 (d, 1H, J = 7.8
Hz, H-3’), 8.16 (ddd, 1H, J = 7.8 Hz, J = 2.0, H-4”’), 8.12 (d, 1H, J = 4.8 Hz, bpy (H-6a)), 8.00-7.82 (m,
1H, bpy (H-6a, H-5a), H-3, H-4, H-4’, H-6”), 7.69 (d, 1H, J = 4.8 Hz, H-6”’), 7.48-7.42 (m, 2H, H-5, H-
1
1
1
2
2
3
3
5”’), 4.52-4.43 (m, 8H, CH ), 1.46-1.40 (m, 12H, CH ), 2.10 (s, 12H, CH acetone).
2
3
3,
[
Ru(d -bpy) (bisbpy)](PF ) .H O (1c)
8
2
6 2
2
This complex was synthesised following the procedure used for complex 1a. 86 mg (0.28 mmol)
bisbpy and 100 mg (0.18 mmol) of [Ru(d -bpy) Cl ]. H O were reacted. Yield: 137 mg (0.133 mmol,
8
2
2
2
1
71%). H NMR (600 MHz, dmso-d6) d ppm 7.49 (dd, J = 7.34, 4.71, Hz, H5’’’), 7.55 (dd, J = 7.15, 7.71
Hz, H5), 7.74 (d, J = 5.65 Hz, H6’’’), 7.94 (s, H6’’), 7.97 (d, J = 7.72, 1.51 Hz, H4’’’), 8.12 (dd, J = 8.47,
2
.45 Hz, H4’’), 8.19 (m, H4), 8.37 (d, J = 7.91 Hz, H3’’’), 8.44 (d, J = 8.28 Hz, H3’’), 8.63 (dd, J = 8.47,
.07 Hz, H4’), 8.70 (d, J = 4.14 Hz, H6), 8.78 (s, H6’), 8.92 (d, J = 8.28 Hz, H3), 8.97 (d, J = 8.28 Hz,
2
H3’). Anal. Calcd. for C H D F N OP Ru: 45.8, H 3.0, N 10.4. Found: C 46.2, H 2.7 and N 10.0%.
40
16 16 12
8
2
[
Ru(bpy) (μ-bisbpy)PdCl ] (PF ) ∙ 2H O (2a)
2 2 6 2 2
3
1
a (0.100 g, 0.095 mmol) dissolved in 5 cm of dichloromethane was added drop-wise to a solution of
3
[
Pd(CH CN) Cl ] (0.025 g, 0.095 mmol) in 5 cm of dichloromethane. The reaction mixture was
3
2
2
refluxed for 24 hours and then allowed to cool to room temperature. Subsequently, the product was
3
3
precipitated by addition of 5 cm of diethylether, filtered and washed twice with a total of 10 cm of
diethylether. Recrystallization from acetone/water (3:1 v/v) afforded a red solid. Yield: 0.115 g (0.094
mmol, 99%). Anal. Calcd. for C H F Cl N P RuPd ∙ 2 H O (1226.5): C, 39.14; H, 2.70; N, 9.13;
40
30 12
2
8
2
2
1
Found: C, 39.29; H, 2.50; N, 8.83. H-NMR (Acetonitrile-d , 400 MHz): δ = 9.21 (d, 1H, J = 4.8 Hz, H-
3
6
), 9.17 (d, 1H, J = 2.5 Hz, H-6’), 8.67 (d, 1H, J = 8.0 Hz, H-3”), 8.63-8.57 (m, 3H, bpy (H-3a), H-3”’),
8.54-8.50 (m, 2H, bpy (H3a)), 8.43-8.37 (m, 2H, H-4’, H-4”), 8.30 (d, 1H, J = 7.8 Hz, H-3’), 8.26-8.21
m, 2H, H-3, H-4), 8.13-8.05 (m, 5H, bpy (H-4a), H-4”’), 8.02 (d, 1H, J = 2.3 Hz, H-6”), 7.89 (d, 1H, J =
(
4
.8 Hz, bpy (H-6a)), 7.79-7.74 (m, 4H, bpy (H-6a), H-6”’), 7.70 (m, 1H, H-5), 7.48-7.39 (m, 5H, bpy (H-
a), H-5”’).
5
[Ru(dceb) (μ-bisbpy)PdCl ] (PF ) ∙ 2 H O (2b)
2
2
6 2
2
1
4
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
3
1
b (0.100 g, 0.071 mmol) dissolved in 5 cm of dichloromethane was added drop-wise to a solution of
3
[
Pd(CH CN) Cl ] (0.018 g, 0.071 mmol) in 5 cm of dichloromethane. The reaction mixture was
3
2
2
refluxed for 24 hours and then allowed to cool to room temperature. Subsequently, the product was
3
3
precipitated by addition of 5 cm of diethylether, filtered and washed twice with a total of 10 cm of
3
5
0
5
0
5
0
5
diethylether and 10 cm of n-hexane. Recrystallization from acetone/water (3:1 v/v) afforded a red
solid. Yield: 0.112 g (0.070 mmol, 98%). Anal. Calcd. for C H F O Cl N P RuPd∙ 2H O (1515.29):
52
46 12
8
2
8
2
2
1
C, 41.21; H, 3.29; N, 7.39; Found: C, 41.12, H, 3.22, N, 6.67. H-NMR (Acetonitrile-d , 400 MHz): δ =
3
9
.21 (d, 1H, J = 4.8 Hz, H-6), 9.19 (d, 1H, J = 2.5 Hz, H-6’), 9.13-9.10 (m, 2H, bpy (H-3a), 9.08-9.05
m, 2H, bpy (H-3a), 8.67 (d, 1H, J = 8.0 Hz, H-3”), 8.62 (m, 1H, J = 7.8 Hz, H-3”’), 8.47-8.43 (m, 2H,
H-4’, H-4”), 8.27 (d, 1H, J = 8.0 Hz, H-3’), 8.26-8.21 (m, 2H, H-3, H-4), 8.16 (ddd, 1H, J = 7.8 Hz, J =
.4 Hz, H-4”’), 8.11 (d, 1H, J = 4.8 Hz, bpy (H-6a)), 8.00 (d, 1H, J = 2.3 Hz, H-6”), 7.97-7.90 (m, 7H,
bpy (H-6a, H-5a)), 7.72-7.66 (m, 2H, H-5, H-6”’), 7.48 (ddd, 1H, J = 4.8 Hz, J = 1.0 Hz H-5”’), 5.40 (s,
H, CH , CH Cl ), 4.52-4.44 (m, 8H, CH , ethyl-ester), 1.46-1.41 (m, 12H, CH , ethyl-ester).
(
1
1
2
2
3
3
2
2
2
2
2
2
3
[Ru(bpy) (μ-bisbpy)PtCl ] (PF ) ∙ 3H O (3a)
2
2
6 2
2
3
1
a (0.100 g, 0.095 mmol) dissolved in 5 cm of dichloromethane was added drop-wise to a solution of
3
[
Pt(dmso) Cl ] (0.042 g, 0.095 mmol) in 5 cm of dichloromethane. The reaction mixture was refluxed
2
2
for 24 hours and then allowed to cool to room temperature. Subsequently, the product was
3
3
precipitated by addition of 5 cm of diethylether, filtered and washed twice with a total of 10 cm of
diethylether. Recrystallization from acetone/water (3:1 v/v) afforded a red solid. Yield: 0.109 g (0.082
mmol, 86%). Anal. Calcd. for C H F Cl N P RuPt ∙ 3 H O (1334.2): C, 35.97; H, 2.60; N, 8.30;
40
30 12
2
8
2
2
1
Found: C, 35.70; H, 2.35; N, 7.95. H-NMR (Acetonitrile-d , 400 MHz): δ = 9.54 (d, 1H, J = 2.5 Hz, H-
3
6
’), 9.51 (d, 1H, J = 4.8 Hz, H-6), 8.69 (d, 1H, J = 8.0 Hz, H-3”), 8.63-8.58 (m, 3H, bpy (H-3a), H-3”’),
8
.55-8.51 (m, 2H, bpy (H-3a)), 8.47-8.40 (m, 2H, H-4’, H-4”), 8.30-8.23 (m, 2H, H-3’, H-4), 8.19 (d,
1H, J = 7.8 Hz, H-3), 8.14-8.04 (m, 6H, bpy (H-4a), H-4”’, H-6”), 7.92 (d, 1H, J = 4.8 Hz, bpy (H-6a)),
.79-7.74 (m, 4H, bpy (H-6a), H-6”’), 7.70 (m, 1H, H-5), 7.48-7.39 (m, 5H, bpy (H-5a), H-5”’).
7
[
Ru(dceb) (μ-bisbpy)PtCl ](PF ) (3b)
2 2 6 2
3
1
b (0.100 g, 0.071 mmol) dissolved in 5 cm of dichloromethane was added drop-wise to a solution of
3
[Pt(dmso) Cl ] (0.030 g, 0.071 mmol) in 5 cm of dichloromethane. The reaction mixture was refluxed
2
2
3
for 24 hours. Subsequently, the product was precipitated by addition of 5 cm of diethylether, filtered
3
3
and washed twice with a total of 10 cm of diethylether and 10 cm of n-hexane. Recrystallization from
acetone/water (3:1 v/v) afforded a red solid. Yield: 0.115 g (0.068 mmol, 96%). Anal. Calcd. for
1
C H F O Cl N P RuPt (1567.95): C, 39.83; H, 2.96; N, 7.15; Found: C, 39.97, H, 3.24; N, 6.25. H-
52
46 12
8
2
8
2
NMR (Acetonitrile-d , 400 MHz): δ = 9.57 (d, 1H, J = 2.5 Hz, H-6’), 9.49 (d, 1H, J = 4.8 Hz, H-6), 9.13-
3
1
5
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
9.10 (m, 2H, bpy (H-3a)), 9.08-9.02 (m, 2H, bpy (H-3a)), 8.72 (d, 1H, J = 8.0 Hz, H-3”), 8.63 (m, 1H, J
= 7.8 Hz, H-3”’), 8.47-8.40 (m, 2H, H-4’, H-4”), 8.23 (t, 1H, J = 7.8 Hz, H-4), 8.19-8.09 (m, 4H, bpy (H-
6a), H-3, H-4”’, H-6”), 8.05 (d, 1H, J = 2.5 Hz, H-3”), 7.95-7.81 (m, 7H, bpy (H-6a, H-5a), 7.71-7.66
(
m, 2H, H-5, H-6”’), 7.48 (ddd, 1H, J = 4.8 Hz, J = 1.0 Hz, H-5”’), 5.40 (s, 2H, CH , CH Cl ), 4.52-4.44
2 2 2
5
0
5
0
5
0
5
(m, 8H, CH , ethyl-ester), 1.46-1.41 (m, 12H, CH , ethyl-ester).
2
3
General Methods
NMR spectra were recorded on a Bruker Advance 600 and 400 spectrometer and referenced to the
solvent signal. Elemental analysis was carried out on an Exador Analytical CE440 by the
Microanalytical Department of the University College Dublin. UV/Vis absorption spectra were
recorded on Varian Cary 50 spectrophotometer at 20 ± 1°C in a 1 cm pathlength quartz cuvette.
Acetonitrile for spectrophotometric measurements was purchased from Aldrich in spectrophotometric
grade and used as received.
1
1
2
2
3
3
Photophysical characterisation
Emission and excitation spectra were obtained on a Perkin Elmer LS 50B at 20 ± 1°C. The emission
quantum yields were measured in de-aerated acetonitrile solution by absorbance matching at the
2+
18
wavelength of excitation respectively to the [Ru(bpy)3] standard with quantum yield, Φ = 0.06.
Excited-state lifetimes were measured by time-correlated single photon counting on an Edinburgh
Analytical Instruments TCSPC instrument (at 293 K) in deaerated acetonitrile solution. Samples were
excited with a LED at 360 nm.
DFT calculations
19
All calculations were carried out with the Gaussian 09 program suite. The compounds, 2a, and 2b
20
were optimized using the M06 functional. The MWB28 basis with an effective core potential was
used for the heavy Ru and Pd atoms while 6-31G(d) was used for the remainder. Tight convergence
criteria were applied for the geometry optimization process and local minima were confirmed by a
frequency calculation. All calculations were carried out in the presence of a solvent sphere, which was
21
modelled by the IEF-PCM method in acetonitrile (ε = 35.688000). Orbital contributions were
22
calculated by a Mulliken population analysis and evaluated using GaussSum.
Photocatalysis
All manipulations were carried out under strictly anaerobic inert conditions. Acetonitrile was dried over
calcium hydride and triethylamine over sodium according to common procedures and freshly distilled
under nitrogen prior to use. Photocatalytic hydrogen production experiments were carried out using a
1
6
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
home-built air-cooled apparatus (at 22 °C) under constant irradiation (LED 470 nm) of the sample. For
3
the photocatalysis experiments 2 cm of the sample solution were added to GC vials (total volume 5
3
3
cm , diameter 13 mm, 3 cm headspace) in the dark and under a stream of nitrogen. The vials were
3
closed with gas-tight septum caps. A typical sample solution was prepared by mixing 0.65 cm of a
-
4
5
0
5
0
5
1.8 × 10 M Ru/Pd complex in acetonitrile, 0.6 ml of triethylamine, 0.0 – 0.2 ml (0 – 10 vol%) of water
and 0.55 – 0.75 ml of anhydrous acetonitrile. Subsequently, the GC vials were irradiated at 470 nm
using an LED for 18 h. After irradiation, 20 μl samples were drawn from the headspace with a gas
tight syringe (50 µl, SGE Analytical Science) and determined by GC, a Varian CP3800
chromatograph, with a thermal conductivity detector and a CP7536 Plot Fused Silica 25 MX 0.32
MMID column (length 25 m, layer thickness 30 μm) with nitrogen carrier gas (purity 99.999 %). The
1
1
2
2
GC was calibrated using 100% hydrogen gas. The obtained signal (retention time for H = 1.58 min)
2
was plotted against the calibration curve and multiplied accordingly to determine the total amount of
hydrogen in the headspace. The LED-torch consists of a stick-shaped printed board (19 × 1 cm) with
3
0 LEDs (Kingbright, type L-7113PBC-G, 470 ± 20 nm) with a luminous efficiency of 2000 mcd per
LED. LEDs are soldered closely on front and backside in a range of 9 cm. The torch was placed
within the home built reactor.
Acknowledgments
This research is supported by the EPA grant 2008-ET-MS-3-S2 and SFI Grants No. 07/SRC/B1160,
08/RFP/CHE1349, SFI/TIDA/E2763, SFI/TIDA/2435.
Keywords: Photocatalytic hydrogen generation, inter-molecular photocatalysis, intra-molecular
photocatalysis, supramolecular assemblies, heterodinuclear photocatalyst
References
[
1]
a) E. Fujita, Coord. Chem. Rev., 1999, 185, 373; b) A. J. Morris, G. J. Meyer and E. Fujita, Acc.
Chem. Res. 2009, 42, 1983; c) T. S. Teets and D. G. Nocera, Chem. Commun. 2011, 47, 9268;
d) S. Rau, D. Walther, J.G. Vos, Dalton Trans. 2007, 915; e) S. Losse, J. G. Vos and S. Rau,
Coord. Chem. Rev. 2010, 254, 2492; f) V. Artero, M. Chavarot-Kerlidou , M. Fontecave, Angew.
Chem., Int. Ed. 2011, 50, 7238; g) H. Ozawa and K. Sakai, Chem.Commun. 2011, 47, 2227; h)
A. Inagaki, M. Akita, Coord. Chem. Rev. 2010, 254, 1220; i) H. Takeda. O. Ishitani, Coord.
Chem. Rev. 2010, 254, 346; j) W. T. Eckenhoff, R. Eisenberg, Dalton Trans. 2012, 41, 13004;
k) P. D. Frischmann, K. Mahata, F. Wurthner, Chem. Soc. Rev. 2013, 42, 1847; l) Y. Halpin,
M.T. Pryce, S. Rau, D. Dini, J.G Vos Dalton Trans. 2013, 42 , 16243.
[
2]
a) M. Schulz, M. Karnahl, M. Schwalbe, J. G. Vos, Coord. Chem. Rev. 2012, 256, 1682; b) G.
Singh. Bindra, M. Schulz, A. Paul, R. Groarke, S. Soman, J. L. Inglis, W., R. Browne, M. Pfeffer,
S. Rau, B. J. MacLean, M. T. Pryce, J. G. Vos Dalton Trans. 2012 41, 13050.
1
7
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
[
[
3]
4]
a) G. Singh Bindra, M. Schulz, A. Paul, S. Soman, R. Groarke, J. Inglis, M.T. Pryce, W. R.
Browne, S. Rau, B. J. Maclean, J. G. Vos Dalton Trans. 2011, 40, 10812; b) T. Kowacs, Q. Pan,
P. Lang, L. O’Reilly, S. Rau, W. R. Browne, M. T. Pryce, A. Huijser, J. G. Vos. Faraday Discuss.
2015,185, 143.
M. G. Pfeffer, T. Kowacs, M. Wächtler, J. Guthmuller, B. Dietzek, J. G. Vos, Sven Rau, Angew.
Chem. Int. Ed. Engl. 2015, 54, 6627.
[
[
[
5]
6]
7]
Y. Pellegrin, F. Odobel, Comptes Rendus Chim. 2016, DOI 10.1016/j.crci.2015.11.026.
A. Paul, D. Connolly, M. Schulz, M. T. Pryce and J. G. Vos, Inorg Chem, 2012, 51, 1977−1979.
a) A. Hagfeldt, M. Grätzel, Acc. Chem. Res. 2000, 33, 269; b) M. Braumüller, M. Schulz, M.
Staniszewska, D. Sorsche, M. Wunderlin, J. Popp, J. Guthmuller, B. Dietzek, S. Rau, Dalton
Trans. 2016, 45, 9216; c) M. Braumüller, M. Schulz, D. Sorsche, M. Pfeffer, M. Schaub, J.
Popp, B.-W. Park, A. Hagfeldt, B. Dietzek, S. Rau, Dalton Trans. 2015, 44, 5577; d) M.
Bräutigam, J. Kubel, M. Schulz, J. G. Vos, B. Dietzek, Phys. Chem. Chem. Phys. 2015, 17,
7823.
[8]
T. Kowacs, L. O’Reilly, Q. Pan, A. Huijser, P. Lang, S. Rau, W. R. Browne, M. T. Pryce, J. G.
Vos, Inorg. Chem. 2016, 55, 2685−2690.
L. Cassidy, S. Horn, L. Cleary, Y. Halpin, W. R. Browne, J. G. Vos, Dalton Trans. 2009, 3923.
Y. Halpin, L. Cleary, L. Cassidy, S. Horne, D. Dini, W. R. Browne, J. G. Vos, Dalton Trans.
[
[
9]
10]
2
009, 4146.
[
[
11]
12]
Y-Q Fang and G.S. Hanan, Synlett. 2003, 34, 852.
Q. Pan, F. Mecozzi, J. P. Korterik, D. Sharma, J. L. Herek, J. G. Vos, W. R. Browne, A. Huijser,
J Phys. Chem. 2014, 118, 20799.
[
[
[
13]
14]
15]
J. M. Calvert, J. V. Caspar, R. A. Binstead, T. D. Westmoreland, T. J. Meyer, J. Am. Chem. Soc.
1982, 104, 6620.
M. G. Pfeffer, B. Schäfer, G. Smolentsev, J. Uhlig, E. Nazarenko, J. Guthmuller, C. Kuhnt, M.
Wächtler, B. Dietzek, V. Sundström, Angew. Chem. Int. Ed. 2015, 54, 5044.
Q. Pan, L. Freitag, T. Kowacs, J. C. Falgenhauer, J. P. Korterik, D.Schlettwein, W. R. Browne,
M. T. Pryce, S. Rau, L. Gonzalez, J G. Vos, A. Huijser, Chem Commun. 2016, 52, 9371.
B.P. Sullivan, D.J. Salmon and T.J. Meyer, Inorg. Chem. 1978, 17, 3334.
[
[
16]
17]
a) T.E. Keyes, E. Müller, P. Pechy, M. Grätzel, J.G. Vos. J. Chem. Soc., Dalton Trans. 1995,
2705; b) W. R. Browne, C. M. O’Connor, J. S. Killeen, A. L. Guckian, M. Burke, P. James, M.
Burke, J.G. Vos, Inorg. Chem. 2002, 41, 4245.
M. Montalti, A. Credi, L. Prodi, M. T. Gandolfi, Handbook of Photochemistry, 2006, CRC Press
[
[
18]
19]
(
Taylor and Francis group), 10, 574.
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. J.
A. Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V.
N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V
Ortiz, J. Cioslowski, D. J. Fox, 2009.
1
8
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201605980
[
[
[
20]
21]
22]
X. Cao, J. Mol. Struct. THEOCHEM, 2002, 581, 139.
J. Tomasi, B. Mennucci, and R. Cammi, Chem. Rev., 2005, 105, 2999.
N. M. O’Boyle, A. L. Tenderholt, K. M. Langner, J. Comput. Chem., 2008, 29, 839.
1
9
This article is protected by copyright. All rights reserved.