Efficient electron transfer through a triazole link in ruthenium(ii) polypyridine type complexes

Article abstract of DOI:10.1039/c1cc13683f

Spectroscopic, electrochemical and theoretical characterisations of photoactive systems readily assembled via click-chemistry show an efficient bi-directional charge shift through the triazole link.

Full text of DOI:10.1039/c1cc13683f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 11011–11013
www.rsc.org/chemcomm
COMMUNICATION
Efficient electron transfer through a triazole link in ruthenium(II)
polypyridine type complexesw
ab
Aurelie Baron, Christian Herrero,ab Annamaria Quaranta,b Marie-France Charlot,a
´
Winfried Leibl,b Boris Vauzeilles*a and Ally Aukauloo*ab
Received 20th June 2011, Accepted 30th August 2011
DOI: 10.1039/c1cc13683f
Spectroscopic, electrochemical and theoretical characterisations of
photoactive systems readily assembled via click-chemistry show an
efficient bi-directional charge shift through the triazole link.
Ru–carbohydrate dendrimers.8 Reports in the literature have
presented the triazole as a ‘‘poor conduit for electrons’’ in
polyoxometalate–porphyrin systems,9 or as a photoluminescence
quencher when incorporated directly into the coordination sphere
of ruthenium ions.10 Other studies, however, have shown the
existence of photo-induced electron transfer in porphyrin–
ferrocene conjugates,11 and a pivotal role of triazoles as
non-conjugating connectors favouring and stabilising charge
separation has also been described in clicked push–pull systems.12
We report here on the synthesis of bipyridine ligands which can
be easily linked by means of click chemistry to organic fragments,
acting either as electron donors or acceptors.
The use of light energy to perform catalysis is a challenging
task facing chemists. One particularly interesting application is
related to conversion of solar energy into fuel, a process
inspired by natural photosynthesis, where light energy is used
to remove protons and electrons from water to fix carbon
dioxide into energy-rich carbohydrates. The basic chemical
modules needed to elaborate an artificial photosynthetic device
are a photosensitiser and a catalytic unit. Ruthenium(II)
trisbipyridine complexes are the most widely used photoactive
chromophores in this field. Although ruthenium is not an
abundant metal, derivatives of ruthenium(II)-polypyridine
complexes have been implemented by Graetzel and his group
in dye sensitised solar cells, which are currently proposed as
cheaper alternatives for silicon based solar cells.1
Our strategy is based on the introduction of a substituted
triazole group onto the 4 position of a bipyridine ligand in the
coordination sphere of
a ruthenium trisbipyridine-type
chromophore (Scheme 1).13 The azido derivative (3) reacts
efficiently with different arylacetylenes (4a–4c), in the presence
of catalytic copper sulfate and sodium ascorbate in a methylene
chloride/water biphasic medium.14 After N-oxide reduction,
the resulting modified bipyridines (6a–6c) were incorporated
into ruthenium complexes by reaction with [Ru(bpy)2Cl2].15
Complexes Ru–H, Ru–NMe2, Ru–NDI were isolated as the
corresponding hexafluorophosphate salts (7a–7c), and
characterised by electrochemical and spectroscopic techniques.
The electrochemical properties of the series of complexes
were studied by cyclic voltammetry and show typical features
of ruthenium complexes, namely, a reversible oxidation potential
around +1.3 V vs SCE for the Ru(II) to Ru(III) process, and
three quasi reversible reduction waves in the range of ꢀ1.20
Their unique photophysical and redox properties, high chemical
stability and easy synthetic access are recognised assets.2
However, chemically assembling a catalytic unit and the
photoactive chromophore remains an important synthetic
difficulty. The actual reaction schemes are long and tedious
and therefore the development of efficient and versatile syn-
thetic methodologies is needed. More importantly, the nature
of the chemical link should not significantly alter the photo-
physical and redox properties of the modules, while it should allow
charge shifts between the constitutive units under light excitation.
Copper-catalyzed Azide–Alkyne Cycloaddition (CuAAC),3,4
a
simple and efficient covalent ligation methodology known as
click chemistry,5 has actually found applications in almost every
domain of synthetic chemistry, including coordination chemistry.6
This synthetic tactic has been efficiently used in modular
assemblies such as
a
copper(I)–rotaxane complex7 and
a
´
Institut de Chimie Moleculaire et des Materiaux d’Orsay,
UMR-CNRS 8182, Universite Paris-Sud, F-91405 Orsay, France.
´
´
E-mail: boris.vauzeilles@u-psud.fr, ally.aukauloo@u-psud.fr
b
Scheme 1 Reagents and conditions: (i) TFA, H2O2 (88%);16 (ii)
HNO3, H2SO4 (40%);17 (iii) NaN3, DMF (91%); (iv) 4a-c, CuSO4
(5 mol%), sodium ascorbate (10 mol%), CH2Cl2/H2O 1 : 1 (96–98%);
(v) PCl3, CH2Cl2 (64–87%);18 (vi) Ru(bpy)2Cl2, AgNO3, MeOH;
NH4PF6 aq., MeOH (40–84%).
´
CEA, iBiTecS, URA2096, Service de Bioenergetique, Biologie
Structurale et Mecanismes, 91191 Gif-sur-Yvette, France
´
´
w Electronic supplementary information (ESI) available: Experimental
procedures and characterisations for synthesis, as well as 1H, 13C, and
2D NMR spectra for compounds 1–7c, DFT, emission and flash
photolysis. See DOI: 10.1039/c1cc13683f
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 11011–11013 11011

View Article Online
Table 1 Cyclic voltammetry data for ruthenium complexes
Compound E1/2 (bipyridine)
E1/2 (ligand)
E1/2 (RuIII/II
)
Ru–H
Ru–NMe2
Ru–NDI
ꢀ1.71, ꢀ1.48, ꢀ1.20
ꢀ1.74, ꢀ1.50, ꢀ1.22
—
0.77
1.30
1.35
ꢀ1.68b, ꢀ1.44b, ꢀ1.20b ꢀ0.88a, ꢀ0.53 1.36
Values are reported vs SCE as E1/2 = (Epa + Epc)/2.a Values obtained
b
from the cathodic wave due to adsorption. Values obtained from the
anodic wave due to adsorption.
Fig. 1 Spin density distribution for the lowest triplet state of Ru–H
(left) and Ru–NMe2 (right).
to ꢀ1.74 V attributed to the one electron reduction of each of
the bipyridines in the coordination sphere. Additional redox
features are observed for Ru–NMe2 and Ru–NDI derivatives.
A quasi-reversible oxidation wave is noticed at +0.77 V for
Ru–NMe2 assigned to the tertiary amino group, while Ru–NDI
presents two reversible reduction waves at ꢀ0.53 V and
ꢀ0.88 V corresponding to single and double reduction of the
NDI moiety (Table 1). It is noteworthy that no additional
electrochemical process was observed for the reference
compound Ru–H, indicating that the triazole group is electro-
chemically inert. Ground state absorption spectra (ESIw) of
complexes Ru–H and Ru–NMe2 exhibit a MLCT band peaking
at 460 nm and p–p* (bipy) transitions at B290 nm, while
Ru–NDI also presents NDI p–p* transitions at B340, 360 and
380 nm. For all the studied complexes, the emission (Table 2)
is red-shifted by about 20 nm as compared to that of the
[Ru(bpy)3]2+ (ESIw), suggesting that functionalisation by a
triazole group induces a slight (B0.1 V) decrease of the
HOMO–LUMO energy gap.
effect of the triazole ring, this LUMO is developed on the
corresponding bipyridine and is partially delocalised on the
triazole. The calculated lowest triplet state corroborates this
3
finding. It can be described as a MLCT state generated by
transfer of electron density from the ruthenium to this LUMO,
as evidenced by the spin density distribution map (Fig. 1a).
For Ru–NMe2, DFT calculations show that the dimethyl-
amino substituent exerts an important electron-donor effect
on the phenyl ring and destabilises all the orbitals developed in
this region. It results that the HOMO of the complex is no
longer a dp ruthenium orbital but a molecular orbital localised
on the Me2N–C6H4 fragment (ESIw). As a consequence, the
computed lowest triplet state is now generated by electron
transfer from the HOMO to the LUMO (similar to the LUMO
of Ru–H) located on the extended ligand (see ESIw, Fig. S6).
This 3ILCT state is reminiscent of a triplet state of the
corresponding ligand (ESIw) and, as evidenced on the spin
density map, no longer involves the ruthenium centre
(Fig. 1b). For this reason, this state is non-emissive and the
emission of Ru–NMe2 is quenched, as already observed in
previous examples.15,20 In this context, protonation of the
amino group readily explains the recovery of emission. These
results indicate that the triazole link on the bipyridine ligand
does not alter the intrinsic photophysical properties of the
ruthenium(II) chromophore.
The emission quantum yields (Fem) of complexes Ru–H and
Ru–NDI are increased compared to [Ru(bpy)3]2+, whereas for
Ru–NMe2 Fem is considerably reduced. The observed emission
lifetimes were generally in good agreement with the steady
state results except for Ru–NMe2 where a fast component
(o20 ns, out of detection range) is dominant. The emission
data suggest that for Ru–NMe2 the lowest excited state is not
3
the usual emissive MLCT. Transient absorption data (ESIw)
confirm this result insomuch as the typical MLCT bleaching
A crucial point to investigate concerns the feasibility of
efficient electron transfer through the triazole in our synthetic
systems holding either an electron donor or acceptor unit.
Laser flash photolysis experiments with the reference
3
at 450 nm was not observed for Ru–NMe2 in contrast to both
the Ru–H and Ru–NDI compounds. Interestingly, the emission
for the Ru–NMe2 complex was partially recovered in the
presence of acid (vide infra).
compound Ru–H in the presence of methylviologen (MV2+
,
an electron acceptor) as well as ascorbate (an electron donor)
showed behaviour similar to [Ru(bpy)3]2+. In the presence of
10 mM MV2+, features associated with MVꢁ+ (peaks at 390 nm
and 605 nm) are observed together with bleaching at (400–500)
nm due to Ru(II) disappearance (Fig. 2a). The Ru(III) formed
recombines with MVꢁ+ as indicated by the similar kinetics of
the respective absorption decays. In the presence of 100 mM
ascorbate (Fig. 2b), absorption bands at 510 nm, typical for
Ru(I) formation, and at 360 nm, attributed to the oxidised
form of ascorbate, are observed. Here again, kinetic measurements
indicate the recombination of the geminate charges. It can be
concluded that the triazole does not perturb the electron transfer
properties of the excited state of the ruthenium chromophore.
In the case of Ru–NMe2, upon excitation in the presence of
MV2+, only weak and very short-lived bleaching of the 450 nm
In order to gain more insight into the electronic structures
both in the ground and excited states, we have performed
Density Functional Theory (DFT) calculations using B3LYP
functional as we have previously shown to correctly describe
experimental results on ruthenium based complexes.z Our theoretical
data confirm that in Ru–H, as in [Ru(bpy)3]2+, the HOMO is a
dp ruthenium orbital and the LUMO is a bipyridine-like p*
orbital. However, owing to the slight electron withdrawing
Table 2 Emission data for ruthenium complexes in MeCN
Compound
lmax/nm
t/ns
Fem
[Ru(bpy)3]2+
Ru–H
Ru–NMe2
Ru–NMe2(H)b
Ru–NDI
608
632
627
633
635
900
750
o20
860
750
0.059a
0.097
0.009
0.027
0.140
+
MLCT band of the Ru(II) was observed, although the MVꢁ
radical was formed (Fig. 3a). These observations can be ratio-
3
nalised by a very fast decay of the MLCT state to the lowest
a
b
From ref. 19. In the presence of formic acid.
c
11012 Chem. Commun., 2011, 47, 11011–11013
This journal is The Royal Society of Chemistry 2011

View Article Online
where photoluminescence was fully quenched at room temperature.
Light driven electron transfer studies in the presence of a sacrificial
electron acceptor or donor are in favour of efficient intramolecular
electron transfer through the triazole link, either from the electron
donating or to the electron accepting groups respectively. Further
work to click putative catalytic units to the ruthenium chromophore
is underway in our laboratories.
AB thanks the Life science division of the CEA for a post-
doctoral fellowship. This work was partially supported by the
ANR (projects ANR-05-JCJC-0199 and ANR 2010 BLAN
0926), the EU/Energy SOLAR-H2 project (FP7 contract
212508) and the Conseil General de l’Essonne. Calculations
were performed using HPC resources from GENCI-CINES/
IDRIS (Grant 2010-x2010085020).
Fig. 2 Absorption changes upon laser flash excitation of Ru–H (a) in
MeCN with 10 mM MV2+. Inset: kinetics at 460 and 605 nm (b) in
MeCN/H2O (50 : 50) with 100 mM ascorbate. Inset: kinetics at 360 and
510 nm. Spectra taken 1 ms after the laser flash.
Notes and references
3ILCT as described above, wherefrom charge transfer leads to
the formation of the MVꢁ+ radical and the ammoniumyl radical
cation.y This fast intramolecular conversion (3MLCT - 3ILCT)
is suppressed upon protonation of the amino group yielding, in
the absence of MV2+, transient absorption features similar to
those of Ru–H (ESIw). Furthermore, in the presence of formic
acid and MV2+, oxidation of the ligand occurs via formation of
Ru(III) as shown by bleaching at around 450 nm observed at
100 ns and the decay kinetics of MVꢁ+ (605 nm) and of Ru(III)
(460 nm) (ESIw). DFT calculations on the singly oxidised species
clearly indicate a metal centred oxidised form in the case of
[Ru–H]+ while in the case of [Ru–NMe2]+, the oxidation locus is
situated on the aminophenyl group. Excitation of the Ru–NDI
complex in the presence of ascorbate reveals appearance of
z Geometry optimisations of the singlet ground state of the complexes
were performed in the gas phase and followed by calculations in a
PCM/H2O modelled medium to avoid computation artefacts.22
y EPR measurements on the photolysed solution in the presence of a
Co(III) salt as an electron acceptor were monitored by X-band EPR
spectroscopy. The spectrum confirms the formation of photoreduced
cobalt(II) species with g values around 4 while the generated ammoniumyl
radical was observed at g E 2.
1 M. Graetzel, Acc. Chem. Res., 2009, 42, 1788–1798.
2 J. G. Vos and J. M. Kelly, Dalton Trans., 2006, 4869–4883.
3 V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless,
Angew. Chem., Int. Ed., 2002, 41, 2596–2299.
4 C. W. Tornøe, C. Christensen and M. Meldal, J. Org. Chem., 2002,
67, 3057–3064.
5 H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int.
Ed., 2001, 40, 2004–2021.
6 H. Struthers, T. L. Mindt and R. Schibli, Dalton Trans., 2010, 39,
675–696.
ꢀ
NDIꢁ radical features21 within ca. 30 ns (Fig. 3b). The fact
7 P. Mobian, J. P. Collin and J. P. Sauvage, Tetrahedron Lett., 2006,
47, 4907–4909.
that no spectral features of a Ru(I) intermediate were observed
suggests a very fast (o20 ns) intramolecular electron transfer
from Ru(I) to NDI. This reduced state lasts for few milliseconds.
In conclusion, we have used the mild synthetic procedure
provided by click chemistry to covalently attach different functional
groups on the 4 position of the 2,20-bipyridine ligand. The
corresponding heteroleptic ruthenium(II) complexes were isolated.
We have found that the triazole on the 4 position on the bipyridine
periphery alters neither the photophysical nor the electrochemical
properties of the ruthenium chromophore in comparison with the
parent Ru(II)–tris bipyridine complex. This is in sharp contrast to
examples with the triazole in the ruthenium coordination sphere,
8 R. Kikkeri, X. Y. Liu, A. Adibekian, Y. H. Tsai and
P. H. Seeberger, Chem. Commun., 2010, 46, 2197–2199.
9 A. Harriman, K. J. Elliott, M. A. H. Alamiry, L. Le Pleux,
M. Severac, Y. Pellegrin, E. Blart, C. Fosse, C. Cannizzo,
C. R. Mayer and F. Odobel, J. Phys. Chem. C, 2009, 113, 5834–5842.
10 B. Happ, C. Friebe, A. Winter, M. D. Hager, R. Hoogenboom and
U. S. Schubert, Chem.–Asian J., 2009, 4, 154–163.
11 V. S. Shetti and M. Ravikanth, Eur. J. Org. Chem., 2010, 494–508.
12 P. D. Zoon, I. H. M. van Stokkum, M. Parent, O. Mongin,
M. Blanchard-Desce and A. M. Brouwer, Phys. Chem. Chem.
Phys., 2010, 12, 2706–2715.
13 For examples of clicked terpyridine derivatives, see: (a) A. Winter,
A. Wild, R. Hoogenboom, M. W. M. Fijten, M. D. Hager,
R. A. Fallahpour and U. S. Schubert, Synthesis, 2009,
1506–1512; (b) E. C. Constable, C. E. Housecroft, J. R. Price,
L. Schweighauser and J. A. Zampese, Inorg. Chem. Commun.,
2010, 13, 495–497.
14 B. Y. Lee, S. R. Park, H. B. Jeon and K. S. Kim, Tetrahedron Lett.,
2006, 47, 5105–5109.
15 A. Quaranta, F. Lachaud, C. Herrero, R. Guillot, M. F. Charlot,
W. Leibl and A. Aukauloo, Chem.–Eur. J., 2007, 13, 8201–8211.
16 F. W. J. Demnitz and M. B. d’Heni, Org. Prep. Proced. Int., 1998,
30, 467–469.
17 J. R. Jones and R. Stewart, J. Chem. Soc. B, 1967, 1173–1175.
18 H. Arzoumanian, R. Bakhtchadjian, G. Agrifoglio, R. Atencio
and A. Briceno, Transition Met. Chem., 2006, 31, 681–689.
19 A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and
A. Vonzelewsky, Coord. Chem. Rev., 1988, 84, 85–277.
20 M. F. Charlot, Y. Pellegrin, A. Quaranta, W. Leibl and
A. Aukauloo, Chem.–Eur. J., 2006, 12, 796–812.
21 S. V. Bhosale, C. H. Jani and S. J. Langford, Chem. Soc. Rev.,
2008, 37, 331–342.
22 M. F. Charlot and A. Aukauloo, J. Phys. Chem. A, 2007, 111,
11661–11672.
Fig. 3 Absorption changes upon laser flash excitation of (a) Ru–NMe2
with 10 mM MV2+ in MeCN, (b) Ru–NDI with 100 mM ascorbate in
MeCN/H2O (50 : 50). Spectra taken 100 ns after the laser flash.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 11011–11013 11013