Thesis
Photophysical and photochemical properties of tetrathiafulvalene
derivatives and their complexes
DUPONT, Nathalie
Abstract
Le TTF (tétrathiafulvalène) est l'une des molécules les plus étudiées de ces quarante
dernières années, grâce à ses propriétés physiques et chimiques. Il a tout d'abord été étudié
comme un composant de superconducteurs organiques comme le TTF-TCNQ, et plus
récemment pour quelques-unes de ses propriétés chimiques et photochimiques. En effet, le
TTF est bien connu comme extincteur de luminescence et comme donneur d'électron. Le
transfert d'électron photoinduit, quant à lui, est un des plus importants processus en
photophysique et en photochimie. Il est utilisé par la nature, par exemple, pour le mécanisme
de photosynthèse, ainsi que dans l'industrie en ce qui concerne la photographie, des
réactions photocatalytiques organiques, des dispositifs optoélectroniques et la conversion
d'énergie solaire. Dans le but d'améliorer de telles applications, le transfert de charge
photoinduit est également étudié intensivement en recherche fondamentale. Les molécules
étudiées dans cette thèse ont plusieurs caractéristiques communes. L'une d'elles est leur
comportement en tant qu'ensembles de [...]
Reference
DUPONT, Nathalie. Photophysical and photochemical properties of tetrathiafulvalene
derivatives and their complexes. Thèse de doctorat : Univ. Genève, 2010, no. Sc. 4273
URN : urn:nbn:ch:unige-145405
DOI : 10.13097/archive-ouverte/unige:14540
Available at:
http://archive-ouverte.unige.ch/unige:14540
Disclaimer: layout of this document may differ from the published version.
UNIVERSITE DE GENEVE
FACULTE DES SCIENCES
Section de chimie et biochimie
Département de chimie physique
Professeur Andreas Hauser
!
!
!
Photophysical and photochemical properties
of tetrathiafulvalene derivatives
and their complexes
THESE
Présentée à la Faculté des sciences de l!Université de Genève
pour obtenir le grade de Docteur ès sciences, mention chimie
par
Nathalie Dupont
de Paris (France)
Thèse N°4273
Genève
Atelier Repromail
2011
Remerciements
Ce travail de thèse a été effectué dans le département de chimie physique du l!Université de Genève,
sous la direction du professeur Andreas Hauser, de septembre 2006 à décembre 2010. Je le remercie
vivement de m!avoir accueillie au sein de son groupe de recherche et pour m!avoir formée à la
spectroscopie optique. Je le remercie pour toutes les discussions et explications scientifiques que nous
avons eues, de m!avoir appris à tenir un discours scientifique et à bien choisir les termes employés lors
de ce dernier. Je le remercie aussi pour la bonne ambiance et la convivialité qu!il a su instaurer dans son
groupe.
Je tiens à remercier les Professeurs Silvio Decurtins et Eric Vauthey pour avoir accepté de juger ce
travail de thèse.
Je voudrais également remercier le Professeur Claude Piguet, de l!université de Genève, pour m!avoir
permis de collaborer avec son groupe de recherche et plus particulièrement avec Thomas RiisJohannessen. Je remercie de même les Docteurs Narcis Avarvari, de l!Université d!Angers et Josef
Hamacek, de l!Université de Genève, pour les collaborations avec leurs équipes respectives. Enfin, je
remercie le Professeur Silvio Decurtins pour m!avoir permis de collaborer avec son groupe de recherche
et tout particulièrement avec Shi-Xia Liu.
Mes remerciements vont également au membre du groupe, sans ordre particulier : Claudia, Mia, Enza,
William, Max, Jie, Prodita, Pradip, Juan Carlos, Christophe, Ahmed et Hans. Au cours de ces quatre
années de thèse, j!ai tout particulièrement apprécié les différents moments passés avec eux, autant
scientifiques que plus personnels. Je remercie le docteur Dominique Lovy pour toutes les discussions
que nous avons eues, ainsi que pour sa patience lors des nombreuses explications et de son aide lors de
mes différents problèmes électroniques au cours de mes expériences. Je remercie Nahid Amstutz pour
sa constante bonne humeur, sa gentillesse et son aide technique précieuse. Je remercie Isabelle Garin
pour son aide dans toutes les démarches administratives. Finalement, je remercie Patrick Barman pour
son travail technique exceptionnel et son ingéniosité dans la réalisation des diverses pièces qui m!ont été
nécessaires lors de cette thèse.
Un merci tout particulier à ma famille et à mes amis proches pour leur soutien, leurs encouragements et
toute leur affection. Merci également à Louis qui m!a épaulée sans fléchir une seule fois.
Résumé
Le TTF (tétrathiafulvalène, Figure 1) est l!une des molécules les plus étudiées de ces
quarante dernières années,[1, 2] grâce à ses propriétés physiques et chimiques. Il a tout
d!abord été étudié comme un composant de superconducteurs organiques comme le
TTF-TCNQ,[3] et plus récemment pour quelques-unes de ses propriétés chimiques et
photochimiques. En effet, le TTF est bien connu comme extincteur de luminescence et
comme donneur d!électron.
Figure 1: Tétrathiafulvalène (TTF)
Le transfert d!électron photoinduit, quant à lui, est un des plus importants processus en
photophysique et en photochimie. Il est utilisé par la nature, par exemple, pour le
mécanisme de photosynthèse, ainsi que dans l!industrie en ce qui concerne la
photographie,
des
réactions
photocatalytiques
organiques,
des
dispositifs
optoélectroniques et la conversion d!énergie solaire. Dans le but d!améliorer de telles
applications, le transfert de charge photoinduit est également étudié intensivement en
recherche fondamentale.
Les molécules étudiées dans cette thèse ont plusieurs caractéristiques communes.
L!une d!elles est leur comportement en tant qu!ensembles de donneur-accepteur
d!électron, une autre caractéristique est la présence du TTF comme sous-unité des
ligands ou molécules. De même, les complexes présentés ici contiennent un ou
plusieurs ions de métaux de transition tels que Ru2+, Co2+ et Fe2+, coordinés à un ou
plusieurs ligands contenant du TTF. Tous les composés de ce travail ont été
synthétisés et caractérisés par les collaborateurs du groupe du Pr. S. Decurtins, à
l!Université de Berne. Cette thèse va donc être focalisée sur les propriétés
photophysiques et photochimiques de ces composés, les parties évoquant la synthèse
ainsi que la caractérisation des composés se trouvant dans les différents articles.
Les propriétés photophysiques et photochimiques de molécules contenant du TTF ont
donc été étudiées et plus particulièrement celles de transfert de charge intraligand
photoinduit entre le TTF et d!autres sous-unités (cf. Figure 2). L!étude de différentes
molécules a permis d!effectuer une comparaison de l!impact de la substitution des
ligands sur le phénomène de séparation de charges observé après photoexcitation des
molécules.
Figure 2 : Schéma représentant le ligand TTF-dppz ainsi que le transfert de charge intraligand
obtenu par irradiation entre le TTF et le dppz et spectres d!absorption du ligand, ainsi que du
TTF et du dppz libres.
L!étude des complexes de la forme [{Ru(bpy)2}n(TTF-ppb)](PF6)2n (ppb = dipyrido[2,3a:3",2"-c]phénazine et n = 1, 2) nous a permis de montrer que le premier état excité
dans ces molécules correspond à un état de transfert de charge intramoléculaire avec
le TTF comme donneur d!électron et le ppb comme accepteur.[4] Ceci se traduit sur le
spectre d!absorption par une large bande d!absorption centrée à environ 15000 cm-1.
Cette bande d!absorption pourra être retrouvée dans les spectres d!absorption d!autres
molécules (à différentes énergies) et est représentative du transfert de charge
intraligand photoinduit. Ces composés ont été comparés avec des complexes
analogues définis par la formule [Ru(bpy)3-n(TTF-dppz)n]2+ (dppz = dipyrido-[3,2-a:2!,3!c]phenazine et n = 1, 2, 3).[5] Un état excité de charges séparées a été mis en évidence
pour ces complexes (cf. Figure 3).
Figure 3 : Spectre d!absorption de [Ru(TTF-dppz)3]2+ dans CH2Cl2 (----) avec sa forme obtenue
par oxydation chimique (....), et le spectre d!absorption transitoire (___) obtenu sous irradiation
d!un laser pulse à 16000 cm-1. Dans l!encadré: relaxation de l!état transitoire observe à 12500
cm-1.
Cet état excité de charges séparées est obtenu par irradiation de la bande d!absorption
correspondant au transfert de charge du métal (Ru2+) vers le dppz, qui peut être décrit
par [(TTF+-dppz)Ru(dppz--TTF)(bpy)]2+ pour n = 2. Le temps de vie de cet état est de
2.5 µs dans CH2Cl2. Un tel état de charges séparées photoinduit n!a pas été observé
dans [{Ru(bpy)2}n(TTF-ppb)](PF6)2n. A la place, la photoexcitation donne lieu à un état
triplet de transfert de charge intraligand (3ILCT) avec un temps de vie d!environ 200 ns
pour n = 1 et 50 ns pour n = 2 (cf. Figure 4).
Figure 4. Spectres d!absorption transitoire de [{Ru(bpy)2}n(TTF-ppb)](PF6)2n pour n=1 (__) et
n=2 (__) en comparaison avec les specters d!absorption pour n=1 (…) et n=2 (…) dans CH2Cl2
à temperature ambiante.
Pour accéder à une meilleure compréhension du comportement de l!état de charges
séparées obtenu par photoexcitation dans les molécules du type [Ru(bpy)3-n(TTFdppz)n]2+, un groupe anthraquinone a été rattaché à une unité phénanthroline dans le
but d!obtenir le complexe [Ru(TTF-dppz)2(Aqphen)]2+(Aqphen = anthraquinone reliée à
une phénanthroline via un pont phénazine). Dans cette triade, l!excitation dans les
bandes de transfert de charge métal-ligand résulte en la création d!un état de charge
séparées avec un long temps de vie de 400 ns, impliquant le TTF comme donneur
d!électron et l!anthraquinone comme l!accepteur d!électron final.[6]
Ensuite, une étude sur les interactions électroniques dans les composés de forme
donneur-accepteur a été réalisée en fonction du pH sur des dérivés imidazole-TTF (cf.
Figure 5). Il a été démontré que le fait de relier les dérivés imidazoles au TTF a un effet
sur le transfert de proton ayant lieu sur la partie imidazole des molécules, avec une
augmentation de la valeur du pKa.[7] De même, le transfert de charge intraligand
photoinduit provenant du TTF se trouve plus favorisé (état excité plus bas en énergie)
lors des protonations successives
Figure 5 : Spectres d!absorption de PITTF dans CH2Cl2 à température ambiante en fonction
des équivalents molaires ajoutés de HCl. Le changement de couleur obtenu pour les différentes
protonations de PITTF sont montrées dans l!encadré : A représente une solution de PITTF, B
celle de [PITTFH]+ et C celle de [PITTFH2]2+ dans CH2Cl2 à température ambiante.
Le transfert de charge intraligand est aussi présent dans ces molécules, ainsi que dans
la molécule étudiée de TTF relié à du pérylènediimide (cf. Figure 6).[8] Cette molécule
met également en évidence le rôle de extincteur du TTF de par l!extinction de
luminescence observée pour l!unité PDI quand elle est reliée au TTF.
Figure 6 : Schéma de TTF-PDI.
Enfin, combiné avec les ions de métaux de transition Fe2+ et Co2+, les propriétés de
donneur-accepteur d!électron des ligands contenant du TTF peuvent être associées aux
propriétés de transition de spin des ions métalliques. Dans [Fe(phen)2(TTF-dppz)]2+, il a
été montré que les deux processus de transfert de charge intraligand photoinduit et de
transition de spin photoinduite étaient indépendants. En effet, l!état haut spin ne peut
être atteint que par une irradiation de la bande d!absorption correspondant au transfert
de charge entre le métal et les ligands, l!irradiation dans la bande d!absorption du
transfert de charge intraligand ne donnant accès qu!à l!état excité de ce dernier. Les
expériences correspondantes sur [Co(phen)2(TTF-dppz)]2+ montrent, elles, que pour ce
complexe une étude plus approfondie est nécessaire afin de pouvoir attribuer les
différentes transitions et processus de relaxation mis en jeu lors de la photoexcitation.
Figure 7: (a) Absorption transitoire mesurée à l!échelle des picosecondes, [Fe(phen)2(TTFdppz)]2+ avec une longueur d!onde d!excitation !ex= 400 nm (ce qui correspond à une excitation
dans la bande de transfert de charge metal-ligand), dans CH2Cl2. (b) Décomposition de
l!absorption transitoire obtenue par un fit bi-exponentiel A(t) = A0 + A1exp(-t//"1) + A2exp(-t//"2).
Ceci permet de distinguer très clairement l!absorption transiroire due au transfert de charge
intraligand photoinduit (en rouge) du blanchiment dû à la transition de spin, elle aussi
photoinduite. Dans le spectre (a), la region spectrale à proximité du laser est coupée.
Références :
[1] F. Wudl, G. M. Smith, E. J. Hufnagel J. Chem. Soc. D-Chem. Commun. 1970, 1453.
[2] Q. Y. Zhu, L. B. Huo, Y. R. Qin, Y. P. Zhang, Z. J. Lu, J. P. Wang, J. Dai J. Phys.
Chem. B. 2010, 114, 361-367.
[3] J. Ferraris, V. Walatka, Perlstei.Jh, D. O. Cowan J. Am. Chem. Soc. 1973, 95, 948949.
[4] C. Goze, N. Dupont, E. Beitler, C. Leiggener, H. Jia, P. Monbaron, S. X. Liu, A.
Neels, A. Hauser, S. Decurtins Inorg. Chem. 2008, 47, 11010-11017.
[5] C. Leiggener, N. Dupont, S. X. Liu, C. Goze, S. Decurtins, E. Beitler, A. Hauser
Chimia. 2007, 61, 621-625.
[6] Y.-F. R. Nathalie Dupont, Hong-Peng Jia, Jakob Grilj, Shi-Xia Liu, Sivio Decurtins,
Andreas Hauser Inorg. Chem. 2010, submitted.
[7] J. C. Wu, N. Dupont, S. X. Liu, A. Neels, A. Hauser, S. Decurtins Chem.- Asian J.
2009, 4, 392-399.
[8] M. Jaggi, C. Blum, N. Dupont, J. Grilj, S. X. Liu, J. Hauser, A. Hauser, S. Decurtins
Org. Lett. 2009, 11, 3096-3099.
TABLE OF CONTENTS
1. Introduction
7
!
1.1. The basics of spectroscopy
7
!
1.2. Electron transfer
13
1.3. Tetrathiafulvalene (TTF)
20
!
1.3.1. Introduction
20
1.3.2. General aspects
21
1.3.2.1. History
21
1.3.2.2. Structure and chemical bonding
22
1.3.3. TTF as an intramolecular !-electron donor
27
1.3.4. TTF-"-A
32
1.3.4.1. TCNQ acceptors
32
1.3.4.2. Pyridinium and bipyridinium acceptors
34
1.3.4.3. A fluorescent acceptor
37
1.3.5. TTF-!-A and TTF-A
39
1.3.5.1. Quinone acceptors
39
1.3.5.2. Fullerene acceptors
41
1.3.5.3. Porphyrin acceptors
46
1.3.6. Polymer acceptors
47
!
1.4. Outline of the thesis
50
1.5. References
51
!
!
!
!
!
!
1
2. Experimental part
57
!
2.1. Absorption spectra
57
2.2. Chemical and electrochemical oxidation
58
2.3. Emission and excitation spectra
60
2.4. Lifetime measurements and transient absorption
63
2.5. References
68
!
!
!
!
!
!
3. Ru(II) Coordination Chemistry of a Fused Donor-Acceptor Ligand: Synthesis,
Characterization and Photoinduced Electron Transfer Reactions of
[{Ru(bpy)2}n(TTF-ppb)](PF6)2n (n = 1, 2)
69
!
3.1. Introduction
72
3.2. Experimental Section
73
!
3.2.1. General.
73
3.2.2. Synthesis of 4#,5#-bis(propylthio)tetrathiafulvenyl[i]dipyrido[2,3-a:3#,2#c]phenazine (L).
74
3.2.3. Synthesis of [Ru(bpy)2L](PF6)2 (1).
75
3.2.4. Synthesis of [Ru(bpy)2(µ-L)Ru(bpy)2](PF6)4 (2).
75
3.2.5. Cyclic Voltammetry.
76
3.2.6. Photophysical Measurements:
76
3.2.7. X-ray Crystallography.
77
!
3.3. Results and Discussion
!
78
3.3.1. Synthesis and Characterization.
78
3.3.2. Solid-State Structure of the Complex 2.
79
2
3.3.3. Electrochemical Properties.
82
3.3.4. Optical Properties.
84
!
3.4. Conclusions
93
3.5. References
93
3.6. Supporting Information:
98
!
!
!
!
4. Dual Luminescence and Long-lived Charge Separated states in DonorAcceptor Assemblies based on Tetrathiafulvalene Fused Ruthenium(II)Polypyridine Complexes
100
!
4.1. Introduction
103
4.2. Results and Discussion
106
4.2.1. The free ligands
106
4.2.2. Coordination to innocent transition metal ions
107
4.2.3. Coordination to ruthenium(II)
108
4.2.4. Dual luminescence and long-lived charge-separated states
111
!
!
4.3. Conclusions
115
4.4. References
116
!
!
!
5. Effect of the Addition of a Fused Donor-Acceptor Ligand on a RuII Complex:
Synthesis Characterization and Photo-induced Electron Transfer Reactions of
[Ru(TTF-dppz)2(Aqphen)]2+
118
!
5.1. Introduction
!
121
3
5.2. Experimental Methods
123
5.2.1. General
123
5.2.2. Synthesis
123
5.2.2.1. Synthesis of [Ru(phendion)2(Aqphen)](PF6)2.
123
5.2.2.2. Synthesis of [Ru(TTF-dppz)2(Aqphen)](PF6)2.
124
5.2.3. Physical Methods
124
5.3. Results and Discussion
126
!
5.3.1. Synthesis and Characterization
126
5.3.2. Electrochemistry
126
5.3.3. Photophysical properties
128
!
5.4. Conclusions
137
5.5. References
138
!
!
!
6. Imidazole-Annulated Tetrathiafulvalenes exhibting pH-Tuneable
Intramolecular Charge Transfer and Redox Properties
141
!
6.1. Introduction
144
6.2. Results and Discussion
146
6.3. Conclusion
157
6.4. Experimental Section
157
!
!
!
6.4.1. General
157
6.4.2. Materials
158
6.4.3. Synthesis
158
6.4.4. Crystallography
159
!
!
4
6.5. References
160
!
!
7. A Compactly Fused !-Conjugated Tetrathiafulvalene-Perylenediimide DonorAcceptor Dyad
165
!
7.1. References
175
7.2. Supporting Informatiom:
178
7.3. Experimental section
179
!
!
7.3.1. Materials
179
7.3.2. Physical measurements
179
7.3.3. Preparation of 1
180
7.3.4. 1H NMR Spectrum of 1 in CDCl3
181
!
!
8. A Donor–Acceptor Tetrathiafulvalene Ligand Complexed to Iron(II) or
Cobalt(II): Synthesis, Electrochemistry and Spectroscopy of [M(phen)2(TTFdppz)](PF6)2
185
!
8.1. Introduction
187
8.2. Experimental Section
188
!
8.2.1. Synthesis of [Co(phen)2(TTF-dppz)](PF6)2.
189
8.2.2. Synthesis of [Fe(phen)2(TTF-dppz)](PF6)2.
189
8.2.3. Methods:
190
!
8.3. Results and Discussion
191
8.3.1. Electrochemical properties
191
!
8.4. Photophysical properties
!
193
5
8.5. References
211
!
!
9. Conclusions
214
!
!
6
1. Introduction
Photo-induced electron transfer is one of the most important processes in photophysics
and photochemistry. It is used by nature, for instance in photosynthesis, and in industry
for photo-imaging, photo-catalytic organic reactions, opto-electronic devices, and solar
energy conversion. In order to optimise such applications, photo-induced charge
transfer is also intensively studied in fundamental research. The molecules studied in
this thesis have several common characteristics, one of them being their behaviour as
electron
donor-acceptor
assemblies,
another
one
being
the
presence
of
tetrathiafulvalene (TTF) as a subunit of the ligands or molecules. TTF is well known as
an efficient luminescence quencher as well as an electron donor. The complexes shown
in this thesis contain one or several transition metal ions coordinated to the ligands,
such as Ru2+, Co2+ and Fe2+. All compounds used in this thesis were synthesised by the
collaborators of Prof. S. Decurtins at the University of Bern, so this thesis will be
focussed on the large variety in their photophysical and photochemical properties. In the
following introduction, some basics of spectroscopy as well as the TTF moiety present
in all the molecules will be introduced. This is followed by a presentation of the
experimental methods and setups used. Finally, the results and discussion on the
ligands and complexes will be presented in the form of published papers, submitted
manuscripts and manuscripts to be submitted.
1.1. The basics of spectroscopy
Most of the molecules presented here have strong colours, so one of their properties is
that they absorb in a certain region of the visible light. This absorption also extends into
the UV region, and, for some complexes, into the IR region.
In order to understand their behaviour, the compounds and complexes were studied by
optical spectroscopy. Thus, the photochemical and photophysical properties of the
different compounds were obtained through the interaction between the electromagnetic
radiation, that is, light, and the molecules.
!
7
When molecular entities interact with electromagnetic radiation, three main processes
can occur:
"
absorption of energy resulting in a transition between two energy levels, the latter
being higher in energy, such that the molecule ends up in an excited state.
"
stimulated emission or spontaneous emission of energy, that is, radiative
relaxation of the molecules back to the ground state.
In most of the cases, before any interaction with light, the molecules are in a stable
state, namely the ground state, which is the one with the lowest possible total energy. In
that case, the principal process when the interaction with light is switched on is the
absorption of radiation. When atoms or molecules absorb light, incident photons allow
the excitation from the ground state at an energy E0, to a higher energy level at Ei. In
that process, the photon energy must be equal to h" = (Ei - E0).
After the interaction with light and the absorption of a photon, the molecule has received
energy and takes on an electronic configuration of an excited state, which in an orbital
picture means that one of its electrons has been promoted to a molecular orbital higher
in energy. The excited state is usually photochemically or photophysically unstable. In
order to come back to the ground state configuration, the molecule can dissipate the
excess energy in different ways: [1-5]
i) the disappearance of the original molecule, by photochemical reactions
ii) by radiative relaxation leading to the emission of a photon usually at a lower
energy than the one originally absorbed by the molecule (Stokes shift,
Kasha's rule).
iii) by non-radiative relaxation, such as internal conversion and intersystem crossing
followed by vibrational relaxation.
iv) if the molecule is in solution, it may interact with other species also present in
solution, for example by luminescence quenching processes such as
excitation energy transfer and light-induced electron transfer.
!
8
Figure 1.1: Scheme of different excited state deactivation processes. [6]
Spectroscopy involves the absorption of the electromagnetic radiation (from a lamp
and/or a laser) by the molecules, as well as the one eventually emitted after excitation.
This includes different spectroscopic methods, such as absorption or emission
spectroscopies. The different peaks in the obtained spectra represent transitions
between different energy levels of the molecules. The strong light of Lasers, in particular
pulsed Lasers, can be used to induce and probe the transformation of matter in real
time on different timescales ranging from sub-picoseconds to hours and days.
The energy levels of a molecule represent the characteristic states of this molecule,
which allow the identification of the transitions occurring in it. The energy is absorbed by
quanta, in a discontinuous way and the energy of a photon is:
E = h" = hc = hc˜"
#
(1)
h is Planck#s constant, h = 6.626 "10#34 J $ s , " the frequency, " the wavelength, and "˜
!
!
is the wavenumber.
! a given absorption
! intensity,
! band is its integrated
One of the most!important aspects of
which is proportional to the oscillator strength containing information on the allowed or
forbidden character of the transition. Transitions from ground to excited states having
the same spin value are spin-allowed and if they have opposite parity they are also
!
9
electric dipole allowed, thus giving rise to intense absorption bands. Transition between
states of the same parity, such as dd transitions in transition metal complexes, are
electric dipole forbidden and give rise to only weak absorption bands. Transitions to
excited states of different spin values are likewise forbidden and give very weak
absorption bands. They acquire their intensity via spin-orbit coupling and may thus have
a certain non-negligible intensity for complexes of heavier transition metal ions.
Transitions, which are both spin and parity forbidden, are very weak indeed and only
observable in very special cases.
The photochemical and photophysical processes and most of all the states involved in
those processes, can be illustrated in a so-called Jablonski diagram. The Jablonski
diagram represents the different energy levels of a molecule. The total energy of the
molecule can be described as the sum of the vibrational, rotational and electronic
energies.
Figure 1.2: Schematic energy level diagram (Jablonski diagram)
According to the Jablonski diagram, two different relaxation pathways lead to
luminescence (emission of light): it is called fluorescence when the excited and the
ground state have the same spin and phosphorescence when their spins are different.
!
10
In the same way, non-radiative processes are called either internal conversion or
intersystem crossing when they occur between states of the same or different spins,
respectively.
Fluorescence
and
internal
conversion
are
spin-allowed
processes,
versus
phosphorescence and intersystem crossing which are spin-forbidden.
Every intramolecular decay process is characterised by a rate constant and the excited
states can be characterized by lifetimes: [3, 4, 7, 8]
!
" (S1 ) =
1
k ic + k fl + kisc
(2)
" (T1) =
1
k + k ph
(3)
'
isc
For luminescence processes, a quantum yield can be defined:
!
" fl =
k fl
kic + k fl + kisc
" ph =
!
(k
(4)
k ph # k isc
'
isc
+ k ph )( k ic + k fl + k isc )
(5)
A molecule is a multielectron system, which can be described by molecular orbitals. An
!
approximate wavefunction for a molecule is given by the antisymmetrised product
function:
" = #S = $ % is i
(6)
i
Where " i is a molecular orbital (MO) and si is a spin eigenfunction. The orbital part of
!
the wavefunction represents the electronic configuration. In a zero order description, the
energy associated with an electronic configuration is given by the sum of the energies of
!
!
the occupied MOs. [9] However, in order to obtain a more realistic description of the
energy states of the molecules, two elements should be taken in account. The spin
functions must be added to orbital functions for the description of the electronic
configuration, and the interelectronic repulsion should be taken into account. [9]
!
11
For metal complexes, different MOs can be assigned according to their atomic orbital
contributions. Two groups can be distinguished: the orbitals centred principally on the
ligands (#L, !L, #n* and !L*) and the partially occupied orbitals centred predominantly on
the metal (t2g*(!M*), eg*(#M*)).
Figure 1.3: Molecular orbital diagram for an octahedral complex of a transition
metal.[9]
Principal excited configurations of metal complexes can thus be classified as:
- metal centred (MC) transitions (!M*$#M*)
- ligand centred (LC) transitions (!L$!L*)
- ligand to metal charge transfer (LMCT) transitions (!L$!M*, #M*)
- metal to ligand charge transfer (MLCT) transitions (!M$!L*)
!
12
1.2. Electron transfer
As mentioned above photo-induced electron transfer is an important process in
photophysics and photochemistry. It is a process in which an electron is transferred
from an electron donating species (D) to an electron accepting species (A). The first
step of a photochemical reaction is the interaction between light and molecule, which
brings it to an excited state [D-A]*, in which depending upon the individual properties of
D and A either the donor or the acceptor moiety is in the excited state. The excited
molecules have different properties than in the ground state: the electron donor moieties
become a stronger reducing agent and the electron acceptor moieties become a
stronger oxidising agent than in the ground state.
Thus when the excited states are involved, the excited state redox potentials of the
corresponding redox couples have to be used. These can be calculated from the redox
potentials of the ground state couple and the one electron potential corresponding to the
zero-zero excitation energy.[9]
E 0 (D+ /D* ) " E 0 (D+ /D) # E 00 (D/D* )
(7)
E 0 (A * /A - ) " E 0 (A/A - ) + E 00 (A/A * )
(8)
!
By neglecting the electrostatic interaction, the Gibbs free energy of the ground state
!
reaction of the donor-acceptor pair
AD $ A-D+.
is given by:
(
0
0
"G 0 = #nF E red
(A) # E ox
(D)
)
(9)
Two different photo-induced electron transfer processes can take place in a molecular
!
dyad A-D:
*A-D$A--D+
(oxidative electron transfer)
A-*D$A--D+
(reductive electron transfer)
!
13
The Gibbs free energy of the excited state electron transfer reaction is given by:
(
)
0
0
"G 0* = #nF E ox
(D) # E red
(A) # E 00 (X/X * )
X / X * can be either D / D* or A / A* .
where
!
(10)
In the absence of chemical complications photo-induced electron transfer processes are
back-electron
transfer reactions, the so-called dark reaction or
! followed by spontaneous
!
!
recombination reaction, that regenerate the system in its ground state:
A--D+$A-D
For the dark reaction the Gibbs free energy is given by
"G 0 (dark reaction) = #"G 0
(11)
Kinetically, electron transfer processes involving excited states and those involving
!
ground state molecule can be described within the framework of the Marcus theory [10]
and of successive, more sophisticated theoretical models.[11, 12] Quantum mechanically,
both the photo-induced and back-electron-transfer processes can be viewed as
radiationless transitions between different, weakly interacting electronic states of the AD molecular dyad. The rate constant of such processes is given by an appropriate
Fermi «Golden Rule» expression:
ket =
4 " el 2
H FC
h
Where
(12)
H el is the electronic coupling and FC the Franck-Condon density of states.
!
Electron transfer can be regarded as an extra deactivation path of the locally excited
! (singlet) state that can exist in !addition to internal conversion, inter system crossing to
the triplet manifold (both iso-energetic) and emission (cf. Jablonski diagram). In that
case, the new deactivation path has to be taken into account for the fluorescence
quantum yield and excited state lifetime. If that extra deactivation path is introduced, for
instance by making an electron transfer energetically favourable (e.g. by a change of
solvent), these expressions become:
"'fl =
!
!
k fl
k fl + k ic + kisc + kcs
(13)
14
" 'fl =
!
1
k fl + k ic + kisc + kcs
(14)
The lifetime and quantum yield of the excited state in the absence of electron transfer
can be regarded as reference value, and we can thus determine the charge separation
rate constant ( kcs ) with the following equations:
kcs =
1
1
#
'
!"
" fl
fl
" fl
!
kcs =
"'fl
$ fl
(15)
#1
(16)
The rate constants of charge separation and charge recombination processes can also
!
be probed by using the absorption of the excited state via pulsed excitation and
transient absorption spectroscopy.
As mentioned above, electron transfer between a donor D and an acceptor A can be
described by the Marcus theory. Either A or D can be in an excited state (A*D or AD*),
and the potential energies of the ensuing pair states are represented by parabolas
along the reaction coordinate. Classically, the reaction takes place when the system is
at the intersection (point I in Figure 1.4), which means that the reactants and the
products including the solvent rearrangement are at the same total energy. Thus, an
electron is transferred adiabatically, while according to the principle of Franck Condon
the nuclei remain fixed during the actual process.
!
15
Figure 1.4: Scheme of the energy potential of reactants and products
Photo-induced electron transfer can be illustrated as follow (see Figure 1.5):
1) After photo-excitation, the electron is still mainly localised on D, but there is already a
little probability to find it on A.
2) At the crossing point I, there is an equal chance of finding the electron on both sides
($E = 0). The electron is transferred from D to A.
3) After relaxation into the well of the charge-transfer state, the probability to find the
electron on the A side is highest, and $E has decreases sharply.
!
16
Figure 1.5: Scheme of different steps during the electron transfer
According to the Marcus theory, the rate constant for an electron transfer process can
be expressed as:
' %G & *
ket = " N# et exp)$
,
( RT +
(17)
where " N is the average nuclear frequency factor, " et is the electronic transmission
!
coefficient, and
"G# is the Gibbs free energy of activation. This last term can be
! expressed by the Marcus quadratic relationship:
!
2
$ % "G 0 (
!
"G = '1+
*
#
4&
where
!
(18)
$ )
"G 0 is the standard Gibbs free energy change of the reaction, and " is the
nuclear reorganization energy:
! !
17
!
" = "i + " o
(19)
where "i is the reorganization energy of the molecules themselves and "0 the
!
reorganization energy of the solvent shell.
!The reorganization energy of the molecules is the inner reorganization
! energy and is
described by a change of geometry between the state of the reagents (R) and the state
of the products (P):
$ f ( R) # f i ( P ) '
2
"i = * & i
) # [+x i ]
% f i ( R) + f i ( P ) (
(20)
f is the force constant of the ith normal mode of the reagents R and the products P
!
"x i is the shift of the equilibrium position.
!
!
The reorganization energy of the solvent is the outer reorganization energy, which is
defined by the reorientation of the dipoles of the solvent molecules in answer to the new
distribution of the charges:
"o =
#q 2 ' 1
1
1 *' 1 1 *
+
& ,) 2 & ,
)
4 $%0 ( 2rD 2rA d +( n %s +
(21)
"q =transferred charge
!
rD and rA = radii of the molecules ! d = rD + rA
!
!
n =refractive index (dielectric optical constant)
!
!
"s = permittivity (static dielectric constant) of the solvent
!
!
Equation 19 predicts that for a homogenous series of reactions (same " and " et
values), an
ln k et vs "G 0 plot is a bell-shaped curve involving:
!
1) a region of normal regime, for small driving forces ( " >
!
!
!
"#G 0 > 0), in which the
process
! is activated thermally. The rate increases with the driving force.
18
!
!
2) an activationless regime ( " %
"#G 0 ), where there are only small changes in the
reaction rate as function of driving force and temperature
!
3) a region of an inverted
! regime, for strongly exergonic processes ( " <
"#G 0 ),
where the reaction rate constant decreases when the driving force increases.
! been observed in
In photo-induced electron transfer reactions, the inversed area has
[13]
very few cases.
!
On the other hand, it has been observed for an increasing number of
charge recombination reactions and is well known for intersystem crossing and internal
conversion.
Figure 1.6: Scheme representing the different regions of the Marcus theory. $Q is
supposed to be constant here.
!
19
1.3. Tetrathiafulvalene (TTF)
1.3.1. Introduction
TTF (tetrathiafulvalene), shown in Figure 1.7, is one of the most studied molecules of
the last 40 years
[14, 15]
(cf. Figure 1.8), because of its physical and chemical properties.
It was first studied as a component of the first organic superconductor
[16]
and more
recently for some of its chemical and photochemical properties. Indeed, it is a very well
known luminescence quencher and electron donor.
Figure 1.7: Tetrathiafulvalene (TTF)
The studies and the knowledge thus gained about TTF play an important role in
numerous research fields, such as solid state physics, photochemistry and
photophysics, chemistry and biology.
!
20
Figure 1.8: Graph representing the number of publications on and with TTF between
1920 and 2010
!
!
1.3.2. General aspects
1.3.2.1.
History
Investigations on TTF began some 80 years ago with a report by Hurtley and Smiles.[17,
18]
However, articles on TTF and TTF derivatives really began to appear regularly
around 1970,[14] when Wudl et al. found that TTF was an “unusually stable organic
radical cation”. Moreover, TTF in its neutral form is an orange organic molecule and
readily loses electrons in the presence of oxidizing agents (or by electrochemistry) to
form the purple radical cation,[14] and then a yellow dication. Both cationic species are
aromatic in the Hückel sense and are quite stable. So, TTF has good electron donor
properties.
!
21
Figure 1.9: Oxidation of TTF gives rise to stable cationic species[18]
In 1972, the electrical conductivity of TTF was established, also by Wudl et al.,[20] and in
1973,
the
first
“organic
metal”
TTF-TCNQ
(tetracyanoquinodimethane)
was
discovered.[16] TCNQ is an electron acceptor and gives a black crystal when precipitated
with TTF. A partial charge transfer is observed between TTF and TCNQ, resulting in
electrical conductance several orders of magnitude higher than for other organic
compounds. The band structure of TTF-TCNQ showed some features usually found in
metals, so it was called an “organic metal”.
1.3.2.2.
Structure and chemical bonding
Since the discovery of TTF, a lot of investigations and studies were performed on its
synthesis in order to make efficient derivatives for better superconductors, !-donors or
sensors. Despite the large number of studies on TTF, very few concerned its structure.
Indeed, TTF was assumed to have a D2h symmetry, that is, to be planar. In 1994, a gas
phase electron diffraction study showed that the planar structure was not the best
according the experiments but that a non-planar boat structure fits better.[21]
In fact, it was realized that TTF is a very flexible molecule and can have different
conformations depending of the type of interactions it encounters. In 1999, Katan
published an article discussing the structure of TTF as a function of its oxidation
state.[22] It appears that in principle the planar confirmation should be more stable than
the effectively encountered boat-like conformation (C2v). However, the difference in
energy between the planar and the boat structure is quite small: 0.02 eV for DFT
calculations at the LDA level and 0.04eV with a BP gradient correction.[22] The results of
these calculations are summarized in Table 1.1 (cf. also Figure 1.10). This small energy
difference shows that the TTF molecule is very flexible and, depending on
!
22
intermolecular interactions, TTF can adopt a planar or a boat conformation. Moreover,
according to the calculations, the TTF radical cation seems to be much more stable in
the planar conformation.
Figure 1.10: Definition of the atomic numbering for the atoms of the TTF molecule[22]
Table 1.1: Calculated Bond Lengths (Å) and Angles (deg) for the TTF Molecule in Comparison
with Gas Phase Electron Diffraction Resultsa. This table is taken from [22].
a
& corresponds to the dihedral angle between SCS and SCCS planes.
b
Spin-polarized calculations lead to the same values.
!
!
23
A model for electron transfer processes from a TTF molecule in a charge transfer
complex was obtained by ab-initio calculations.[22] The results of the first principles
calculations for a planar monomer in different oxidation state are given in Table 1.2.
Figure 1.11: Isodensity representation of the HOMO of the TTF molecule.[22]
Table 1.2: PAW-LDA Calculated Total Energy and Energy of the HOMO State for
a
Different Total or Spin-Up and Spin-Down Occupations of the HOMO State(s) .
This table is taken from [22] .
a
Deduced values for the Coulomb repulsion U and the spin-splitting parameter J.
All energies are given in eV.
The ionization energy of TTF can be estimated by two different ways: by the difference
in total energy for the neutral molecule and the radical cation
Eionization = Etot(TTF+ ) – Etot(TTF),
or by the mean values of the HOMO energies for TTF and TTF+ . The same result is
found for both methods, that is, 6.3 eV.[22] This value is in good agreement with those
obtained by photoelectron spectroscopy in the gas phase (6.7 eV [23]).
TTF being well known for its ability to donate electrons, reduction and oxidation are
quite important elementary steps. Its redox potential is known experimentally and
!
24
provides a direct measure of the Gibb#s free energy of adding or removing electrons
from TTF. Experimentally, TTF has two oxidation waves: E11/2 = 0.34 V for the couple
TTF/TTF+ and E21/2 = 0.78 V for the couple TTF+/TTF2+, vs Ag/AgCl in acetonitrile.
Moreover, a lot of TTF based compounds are studied in solution and it is known that the
solvent has a big influence on some processes as demonstrated by Marcus in his
treatment of electron transfer reactions.[24-26] Ab-initio molecular dynamic can give an
idea of the solvent effect for electron transfer in TTF
[27]
and is illustrated with
acetonitrile in Figure 1.12. For TTF in the neutral form, a preferential orientation of the
acetonitrile methyl group toward the sulphur atom with the nitrogen pointing away from it
was observed.[27] This preference is reduced near the carbon atoms. In contrast, the
average orientation of acetonitrile is reversed upon oxidation of TTF (the nitrogen is
turned toward the sulphur atom). This is consistent with the fact that the largest amount
of spin density and therefore electron density is located near the sulphur atoms.
!
25
Figure 1.12: «Snapshots of TTF in solution […], with only the closest solvent
molecules visualized. The spin density (upper panels) and the difference electron
density (lower panels, see text for the definition) are shown at contour levels of
+0.002 (green) and '0.002 (pink) au; i.e., green implies an increase of electron
density in the reduced state. The difference density illustrates the electronic
polarization induced by the cation, in particular of the C'H bonds and first solvation
shell. For each solute, the molecular configuration in the upper panel is identical to the
configuration in the lower panel. However, the TTF configuration is taken from a
trajectory of the neutral molecule […]. The images are generated with VMD». Figure
taken from [27].
!
26
1.3.3. TTF as an intramolecular !-electron donor
!
TTF and its derivatives have been extensively studied for more than 35 years as !electron donors in intermolecular charge transfer materials. The most famous example
for that is given by the number of studies on TTF-TCNQ over the past three decades.[16,
28-33]
The majority of the applications of TTF are determined by its donor abilities, which
are the result of the high-lying HOMO. Nevertheless, the potential of TTF as donor in an
intramolecular sense has only recently been developed. Indeed, as Bryce mentioned in
one of his reviews,[34] TTF has received attention as a donor moiety in intramolecular
charge transfer (ICT) systems principally only over the last decade because of the
synthetic challenge of obtaining functionalized TTFs in reasonable quantities.[35]
Progress of the last ten years in the branches of TTF chemistry has changed this
situation and TTF can now be synthesized in 20 g batches from readily available
starting materials.[34] Fanhaenal et al have reviewed different synthetic methods
[36-38]
which are summarized in the following scheme shown in Figure 1.13:[36]
Figure 1.13: Scheme of synthetic methods for construction of the TTF skeleton.
This is taken from [36].
!
27
TTF as an electron donor evokes the topic of intramolecular donor-acceptor (D-A)
molecules. This includes different fields like molecular electronic devices,[39, 40] organic
metals,[41] chromophores for dyes[42] and non linear optics,[43, 44] and also excited state
energy and electron transfer processes,[45,
46]
and theoretical aspects of charge
transport at the molecular level, conjugation and aromaticity.[47] Different applications
are summarized in Figure 1.14.
Figure 1.14: Different fields of application. This Figure is taken from [35].
The concept of D-A molecules is linked with electron transfer processes. They depend
principally on the energy of the highest occupied molecular orbital (HOMO) and the
lowest unoccupied molecular orbital (LUMO) levels and the orbital interactions. An
enhanced redox activity is observed for donor and acceptor molecules upon photoexcitation. Indeed, donor molecule D becomes D* when excited and has an electron
transferred to a higher-lying LUMO. As a result it becomes a better reducing agent. In
the same way, an acceptor A* is a stronger oxidizing agent than A because the hole
created by photo-excitation in the ground state HOMO increases the electron affinity of
the excited state.[48]
In the 1970s and 1980s, the search for molecular metals, after the discovery of TTF!
28
TCNQ, gave rise to the idea of extending the !-conjugated system of TTF in order to
decrease the Coulomb repulsion between the charges on the molecules in the solid
state and thus to improve the intermolecular interactions.
The first !-extended derivative of TTF, dibenzo-TTF, was, in fact, synthesized 50 years
before its parent molecule.[17] However, its electron-donor properties were studied much
later. The benzene rings have an electron withdrawing influence, which increases the
oxidation potential by 250 mV compared to TTF.[49] Numerous homologues of dibenzoTTF were then synthesized, including polymers.
Wudl et. al. presented an exhaustive review on that subject in 2004.[50] The authors
notice that molecules with low HOMO/LUMO gap are of particular importance due to
their ability to easily donate or accept an electron. Some molecules, called amphoteric
compounds, can even act as an electron donor and as an electron acceptor at the same
time, forming stable redox states within the same molecule. Molecules with a controlled
HOMO/LUMO gap are the prime targets for electronic applications at the single
molecule level. Two strategies can be applied in order to obtain a lower HOMO/LUMO
gap:[50] (i) by extending the !-conjugation in the molecule and (ii) by the construction of
covalent D-A compounds (D being a !-electron donor and A a !-electron acceptor) in
which the HOMO and the LUMO can be tuned relatively independently of each other.
The first method was successful in the design of !-conjugated polymers, fullerene
materials, etc. The second method gave rise to various TTF-A derivatives.
The major problem for the synthesis of TTF-spacer-A compounds is the need to prevent
the formation of a stable intermolecular CT complex between the TTF and the acceptor
moieties, prior to their covalent coupling. Three conceptually different ways of
preventing this unwanted interaction have been explored and reported by Bryce:[34]
• “A TTF derivative can be linked to a group that is either a weak electron acceptor
(e.g., a quinone) or not an acceptor at all (e.g., a pyridine moiety), which is
subsequently converted into a stronger acceptor (TCNQ, or pyridinium cation,
respectively). The conversion of a TTF-quinol derivative into its corresponding
TTF-quinone has been demonstrated on numerous occasions […] but in spite of
several attempts, there appears to be no example of a TTF-quinone being
converted into the corresponding TCNQ or DCNQI (N,N-dicyanoquinodiimide)
derivative although TTF-aldehydes have been converted cleanly into the
!
29
corresponding dicyanomethylene derivatives […]. The converse disconnection is
also possible: an acceptor unit is linked to a 1,3-dithiole-2-thione unit (very weak
donor), which is then converted into a TTF group, by standard coupling
methodology […].
• Lithiated TTF can be substituted by an electrophilic reagent, which, after covalent
bond formation, is transformed into an acceptor group in situ […].
• Steric hindrance in either the TTF or acceptor unit (or both) can be exploited to
prevent their close !-! association, thereby allowing pendant substituent groups
to couple […]”.[34]
In 1994, Jorgensen and co-workers
[19]
reported the principal landmarks in TTF
chemistry. It is always evolving and always involving a lot of different disciplines in
science, such as chemistry, biology or physics, depending upon the area of application,
which is very vast.
Table 1.3: Landmarks in TTF chemistry. Table taken from [19].
1926:
The first TTF derivative, dibenzo-TTF was synthesized as part of a
general study of five-membered ring systems[17, 18]
1965 : Deprotonation of 1,3-dithiolium salts afforded TTF derivatives for the
first time[51]
1970:
First observation of parent TTF. TTF is found to form a stable purple
radical cation on reaction with chlorine gas[14]
1973:
First observation of metallic conductivity in an organic solid
(TTF)(TCNQ)[16]. TCNQ = tetracyanoquinodimethane. Conductivity: 500
1980:
Superconductivity
observed in a TTF derivative at 1.4 K:
!-1cm-1
tetramethyltetraselenafulvalene hexafluorophosphated ([TMTSF]2PF6)[52]
1980:
TTF was derived extensively in the search for organic (super-)
conductors[19]
1985:
Macrocyclic TTF-based systems investigated with the aim of making
molecular devices, sensor, switches and shuttles[19]
Initially, TTF and its derivatives were prepared in order to develop electrically
conducting materials. TTF moieties have also been used in the construction of redox!
30
active supramolecular systems, and chemical sensors and redox-switchable ligands
have been prepared with TTF containing rotaxanes and catetanes.[19] TTF has also an
important role in molecular electronics. Indeed, TTF containing D-"-A molecules have
allowed the preparation of the first confirmed unimolecular rectifier. TTF can also have
non linear optic (NLO) responses in second and third harmonic generation and a good
thermal stability.[53] Segura and co-workers have summarized the reviews on TTF up to
2001, thus giving further examples for the many fields of applications of TTF.
Table 1.4: Recent reviews on specific aspects of TTF. Table taken from [53].
Main author
Title
M. R. Bryce
Recent progress on conducting CA salts[54]
V.
Khodorkovsky
M. R. Bryce
Increasing dimensionality in the solid state[55]
G. Schukat
TTF chemistry[38]
J. Garín
Reactivity of TTF and TSeF[56]
K. B. Simonsen
Functionalisation of TTF[57]
T. Otsubo
TTF dimers[58]
M. Adam
TTF oligomers[59]
J. Becher
TTF oligomers[60]
M. R. Bryce
Macromolecular TTF chemistry[61]
T. Jørgensen
Supramolecular TTF chemistry[19]
K. B. Simonsen
Macrocyclic and Supramolecular TTF chemistry[62]
M. B. Nielsen
Two- and three-dimensional TTF macrocycles[63]
E. Coronado
Hybrid polyoxometalates-TTF materials[64]
J. Roncali
Linearly !-extended TTF derivatives[65]
P. Day
M. R. Bryce
M. B. Nielsen
M. R. Bryce
!
Molecular design of organic conductors[41]
Molecular magnetic semiconductors, metals and
superconductors[66]
TTF as !-donors in intramolecular CT-materials[34]
Tetrathiafulvalenes as building blocks in supramolecular
chemistry[67]
Functionalised tetrathiafulvalenes: new applications as versatile
!-electron systems in materials chemistry[68]
31
One of the key reasons for the synthesis of TTF containing D-A systems was the search
for new charge-transfer materials with a well-defined ratio of D and A molecules. Some
of these systems are discussed in detail below.
1.3.4. TTF-"-A
1.3.4.1.
TCNQ acceptors
Aviram and Ratner proposed for the first time that TTF could be considered as an
electron donor moiety for ICT in D-"-A molecules, where D is an organic one-electron
donor, " is a covalent and non-conjugated bridge and A is an organic one-electron
acceptor.[39] TTF and TCNQ were known to act as donor and acceptor units. Aviram and
Ratner proposed a compound TTF-"-TCNQ, 1 (cf. Figure 1.15), never synthesized, but
source of inspiration for the first TTF-"-A compound to be studied experimentally, 2[69]
(cf. Figure 1.16). The powder EPR spectrum of 2 at room temperature showed a broad
signal from a ground state biradical.
1
Figure 1.15: Scheme of molecule 1, from [39].
2
Figure 1.16: Scheme of molecule 2, from [69].
!
32
Compound 3 (cf. Figure 1.17) was reported at round about the same time.[70] The cyclic
voltammogram (cf. Figure 1.18) of this compound exhibits two reversible one-electron
oxidation waves, typical of TTF, and one reversible two-electron wave, corresponding to
the reduction of TCNAQ. Unlike compound 2, compound 3 is neutral in its ground state.
CT in compound 3 can be readily assigned in its UV-vis absorption spectrum. Indeed, it
shows, in addition to the absorption bands of TCNAQ and TTF, a weak and broad band
centred around 450 nm, assigned to ICT band (cf. Figure 1.19).
3
Figure 1.17: Scheme of molecule 3, from [70].
Figure 1.18: Cyclic voltammogram of compound 3 (in MeCN solution, vs. Ag/AgCl, Pt
working electrodes, electrolyte Bu4N+ClO4- at 20°C). Taken from [70].
!
33
Figure 1.19: UV-vis spectrum of compound 3 in MeCN. The inset shows an
expansion of the ICT band observed in the 420 to 600 nm region (from [70]).
1.3.4.2.
Pyridinium and bipyridinium acceptors
Pyridinium cations and bipyridinium dications have been studied as acceptor groups in
TTF-"-A systems. It is known that the pyridinium cation is a strong acceptor.[71]
Molecule 4, represented in Figure 1.20, was synthesized by Becker and co-workers by
methylation of the corresponding pyridine system. The pyridinium moiety is thus
covalently linked to the TTF unit, making the charge transfer possible due to the
proximity of the donor and acceptor groups. A weak ICT absorption band (Figure 1.21)
is found for compound 4, centred at ca. 665 nm. The solution of 4 was stable without
light, but even with the beam of the spectrophotometer, the ICT process is triggered and
a strong absorption band centred at ca. 675 nm, characteristic of the TTF+" radical
cation appeared.[72]
!
34
4
Figure 1.20: Scheme of molecule 4[73]
Figure 1.21: UV-vis spectra for molecule 4 in acetone: i) a freshly prepared solution,
and consecutive runs after ii) 1h, iii) 2h, iv) 3h[72]
The triad, A-"-TTF-"-A 5 (cf. Figure 1.22)[74] has also been studied. Inspired by this
work, Becher and co-workers[75] have obtained a prototype thermally controlled TTF
based molecular switch. Equilibrium between 6 and 6! (cf. Figure 1.23) shows a CT
absorption band at 785 nm (cf. Figure 1.24), corresponding to the TTF unit being
inserted into the cyclophane cavity.[76] The reflux of the solution for 45 min causes the
disappearance of this band. So, by refluxing, the decomplexed form 6! is obtained. The
complexed form 6 is re-established after 20 hours at room temperature and the CT
absorption reappears.
!
35
5
Figure 1.22: Scheme of compound 5[74]
6
6!
Figure 1. 23: Scheme of compound 6 and its decomplexed form 6!.[76]
!
36
Figure 1. 24: a) UV-vis spectrum of the initially decomplexed 6! in MeCN at i) 0h, ii)
3h and iii) 19h. b) Variation of the maximum absorbance ((max = 785 nm) of 6! with
time.[76]
1.3.4.3.
A fluorescent acceptor
D-A molecules with TTF as electron donor can also be used for redox-fluorescence
switch molecules, as reported by Zhang et al.[77] (Figure 1.25). In this case, the acceptor
unit fluoresces (here, anthracene). But, because it is linked to TTF, only a weak
fluorescence can be observed before oxidation of the molecule. After oxidation, TTF
becomes TTF+" and loses its ability to donate an electron and thus quench the
fluorescence. As a result, a fluorescence increase can be observed. When reduced
back to the neutral form, TTF quenches the fluorescence of the acceptor again. Zhang
et al. show that this quenching is indeed due to a photo-induced electron transfer
reaction. The molecule was oxidised both by Fe(ClO4)3 (see Figure 1.27) as well as
electrochemistry. Spectroelectrochemistry (see Figure 1.26 for the cyclic voltammogram
and Figure 1.27 for the corresponding emission spectra) also served to demonstrate
reversibility of the fluorescent redox switch.[77]
!
37
7
Figure 1.25: Scheme of molecule 7, and of the redox switch fluorescence, taken from
[77].
Figure 1.26: Cyclic voltammogram of 7 (scanning rate 50 mV/s) with Pt wires as
working and counter electrodes, Ag wire as a reference electrode, and n-Bu4PF6 as a
supporting electrolyte. Figure taken from [77].
!
38
Figure 1.27: (A) Fluorescence spectra of 7 (5.8x10-5M) in THF in the presence of
different amounts of Fe(ClO4)3. (B) Absorption spectra of 7 (5.8x10-5M) in THF in the
presence of different amounts of Fe(ClO4)3. Figure taken from [77].
Figure 1.28: (A) Fluorescence spectra of the solution of 7 in THF (4.94x10-5M)
containing n-Bu4NPF6 (27.8 mM) after applying an oxidation potential of 0.7V (vs Ag
wire). (B) Fluorescence spectra of the solution of 7 that has been oxidized
electrochemically for 3 min after applying potential of 0.2 V (vs Ag wire). Figure taken
from [77].
1.3.5. TTF-!-A and TTF-A
1.3.5.1.
Quinone acceptors
Dumur and co-workers reported TTF-benzoquinone diad (8) and triad (9).[78] The
compounds showed pronounced amphotericity in cyclic voltammetry (Table 1.5). A
!
39
weak and broad intramolecular CT band (around 900 nm for 8, see Figure 1.29) was
observed in these compounds.
8
9
10
Figure 1.29: Scheme of compounds 8, 9[78] and 10.[79]
Figure 1.30: Evolution of the UV'vis'NIR spectrum during reduction of D'A 8, at
different times of reaction. The inset is the enlargement of the vis'NIR part. Figure
taken from [78].
!
40
Table 1.5: Summary of the oxidation and reduction potentials of compounds 8 and 9
vs. Ag/AgCl.[78] In the triad 9, successive reduction steps of the quinones are
observed.
Compound
Eox1
Eox2
Ered1
Ered2
8
0.74
1.11
-0.21
-1.20
9
0.99
1.36
-0.20
-0.28
Thioindigo possesses a similar electron affinity to that of benzoquinone. It was attached
to TTF by Aqad and co-workers (10).[79] This dyad has a long wavelength absorption
band at 760 nm, which can be attributed to an intramolecular charge transfer transition
(ICT).
1.3.5.2.
Fullerene acceptors
!
Llacay and co-workers[80] and Boulle and co-workers[81] and have performed studies on
the series of TTFs attached to C60: 11, 13, 14 and 15. Electronic spectra of 13 and 14
indicated weak inter- or intramolecular interactions between the TTF and C60 moieties
around 600 nm (Figure 1.32). Photo-induced ICT in 11, 13 and 14 was studied by
nanosecond time-scale transient absorption. The excited triplet state of C60 was rapidly
quenched by ICT from the TTF with ) = 2.9 *s for 11 and 12, and 0.25 *s for 13.[80] The
transient absorption UV-vis spectra of 13 and 14 show absorption at the same
wavelength than those of the TTF radical cations generated electrochemically of these
compounds (Figure 1.33). This provided direct evidence for the formation of photoinduced charge separated state upon irradiation into the ICT absorption band.
!
41
11:
12:
13:
14:
15:
R=CO2Me; n=1
R=SMe; n=1
R=H; n=1
R=Me; n=1
R=CO2Me; n=2
Figure 1.31: Scheme of compounds 11-15[80, 81]
Figure 1.32: UV'vis spectra in CH2Cl2 of neutral dyad 13 (dashed) and its radicalcation derivative 13+" (solid) as the PF6- salt.
!
42
Figure 1.33: Transient absorption spectra at different times obtained by laser flash
photolysis of 0.1 mM solutions of 12 in benzonitrile after a 532 nm pulsed (9 ns) laser
excitation. Times after pulse: (1) 2 + 10-6 s ; (2) 4 + 10-6 s : (3) 8 + 10-6 s ; (4) 1.6 + 105
s; (5) 3.2 + 10-5; (6) 6.4 + 10-5; (7) 1.3 + 10-4 s; (8) 2.5 + 10-4 s. Inset shows the
corresponding spectra of the reference compound in the same conditions. Taken from
[80].
Figure 1.34: Scheme of C60-exTTF-exTTF (16) and C60-exTTF-TTF (17)[81]
!
43
Illescas and co-workers[82] have reported an example of a polymer including TTF and
C60. They synthesized two rigid and conjugated electron donor-acceptor systems: 16
and 17 (cf. Figure 1.34). Quantum chemical calculations show differences between 16
and 17. Indeed, analysis of molecular orbitals of 16 revealed that the two exTTFs are
almost equivalent with the HOMO and HOMO-1 delocalized over both moieties and very
close in energy to each other. In contrast for 17, in which the degeneracy is cancelled
with the HOMO localized on TTF and HOMO-1 localized on exTTF (Figure 1.35). It can
be deduced that ICT in those molecules occurs between the C60 unit and exTTF or TTF
units. Femtosecond experiments were performed. The transient species are formed
after 1.2 ps and only one metastable transient state of the C60 singlet excited state was
observed (Figure 1.36) with a decay of 28 ps for 16 and 54 ps for 17. Transient
absorption spectra show two maxima. The one at 670 nm agrees with the one-electron
oxidized form of exTTF, and the one at 1010 nm corresponds to the one-electron
reduced form of C60.
Figure 1.35: HOMO-1 (left) and HOMO (right) of 16 (upper part) and 17 (lower part)
resulting from DFT calculations (B3PW91/6-31G*).[82]
!
44
Figure 1.36: Up: differential absorption spectra (visible and near-infrared region)
obtained upon femtosecond flash photolysis (387 nm, 150 mJ) of C60-exTTF-TTF
(17) (10-6 M) in deoxygenated THF with several time delays between 0 and 500 ps at
room temperature. Lower part: time–absorption profiles of the spectra shown above at
1000 nm, monitoring the charge separation.
!
45
1.3.5.3.
Porphyrin acceptors
Porphyrins play an important role in biology and are known for their role in electron
transfer processes. Association of TTF with a porphyrin ring system could produce a
dyad system expected to show light-induced electron transfer from the TTF donor
subunit to the porphyrin ring.
Figure 1.37: Scheme of compound 18[83]
Li
and
co-workers[83]
studied
a
mono-TTF-porphyrin
18.
The
deconvoluted
voltammogram of 18 revealed four reversible redox processes (Table 1.6). The waves
at -1.385 and -1.650 V can be associated with the first and second reduction of the
porphyrin ring system while the waves at +0.040 and +0.405 V can be associated with
the first and second oxidation of the TTF unit. The electrochemical study of 18 shows
electrochemical characteristics and there are only weak interactions between the two
components. The absorption spectrum of compound 18 shows no interaction in the
ground state between TTF and the porphyrin. However, comparison between the
emission spectra of the porphyrin unit and of compound 18 indicates that the
fluorescence of the porphyrin unit is quenched by 98% in compound 18.[83] This
observation demonstrates that electron transfer from TTF to the porphyrin occurs in the
emitting excited state of 18.
Compound 18 was oxidized by FeCl3. In the absorption spectrum of the oxidized
compound 18, a new absorption band appears at 810 nm, typical of a TTF radical
cation. Moreover, the fluorescence intensity of oxidized compound 18 increased upon
addition of FeCl3. So, an electron transfer from the TTF to the porphyrin system occurs
in the emitting excited state of 18 and its emission intensity can be controlled by the
oxidation state of TTF.[83]
!
46
Table 1.6: Electrochemical data for the mono-TTF-porphyrin 18 determined by cyclic
voltammetry in CH2Cl2, at 298K, n-Bu4PF6 (0.1M) as supporting electrolyte.[83]
Porphyrin
TTF
Compound
Ered2(V)
Ered1(V)
Eox1(V)
Eox2(V)
18
-1.650
-1.385
0.040
0.405
Figure 1.38: (a) Absorption (THF, 298 K, full line) and emission (THF, 298 K, dotted
line) spectra of the model porphyrin. Excitation was performed at 419 nm.
(b) Emission (THF, 298 K) spectra of the mono-TTF-porphyrin 18 upon addition of
increasing amounts of the oxidant FeCl3. Excitation was performed at 425 nm. Figure
taken from [83].
1.3.6. Polymer acceptors
“The design and synthesis of novel conjugated polymers attract great attention in the
field of organic semiconductors due to their ease of preparation, low processing
temperature, and nearly unlimited variability”.[84] TTF-fused polymers constitute highly
polarisable species and may lead to a significant increase of their charge storage
capacity.[20] Moreover, the propensity of TTF to self-assemble into !-stacks is interesting
to improve the electron mobility of the TTF-polymer via ! conjugation.[53] Hou and coworkers[84] reported a study on TTF fused to poly(p-aryleneethylene), PAE-TTF. The
linear chain of PAE acts as an electron acceptor and the ! conjugated TTF units act as
the electron donors.
!
47
Figure 1.39: Scheme of PAE-TTF[84]
The UV-vis absorption spectra of PAE-TTF in solution and as a thin film, in Figure 1.40,
show a broad absorption band (located at 460 nm for the solution spectrum), indicating
the intramolecular charge transfer between TTF and PAE. The spectrum of PAE-TTF•+
in solution is also shown in Figure 1.40. In this latter spectrum, a new absorption band
appears between 700 and 1000 nm, representative of the TTF•+ radical cation. Figure
1.41 reports the cyclic voltammogram of a PAE-TTF thin film. Two reversible oneelectron oxidation waves are assigned to the oxidation of TTF (E1/2ox1 = 0.47 V and
E1/2ox2 = 0.78V). Moreover, the polymer has a quasi-reversible reductions process
attributed to the reduction of the main chain of PAE (E1/2red = -1.7 V).
!
48
Figure 1.40: UV-vis absorption spectra of PAE-TTF, PAE-TTF"+ solution in odichlorobenzene (3x10-5M of the repeat unit) and PAE-TTF film on quartz plates (film
from o-dichlorobenzene saturated solution). The PAE-TTF"+ solution was obtained by
chemical oxidation of PAE-TTF by [Fe(bpy)3](PF6)3. Figure taken from [84].
Figure 1.41: Cyclic voltammogram of a PAE-TTF film on platinum electrodes in
acetonitrile solution of 0.1 M Bu4NPF6 with a potential scanning rate of 0.1V/s at room
temperature and potential vs. Fc/Fc+. The PAE-TTF film was prepared via dipping the
platinum-working electrode into the polymer-saturated solution in o-dichlorobenzene
and then drying under an infrared lamp. Figure taken from [84].
!
49
1.4. Outline of the thesis
The principal theme evoked in this thesis is photo-induced electron transfer, as well as
intramolecular and intraligand charge transfer. These processes are studied in different
molecules and transition metal complexes. The motivation for studying the
photophysical and photochemical properties of such complexes, in particular those
containing bidentante ligands such as 2-2#-bipyridine (bpy) or 1,10-phenanthroline
(phen) or their derivatives,[85-91] is their stability and their use for different applications,
as for example in solar energy conversion [92-94] and molecular electronic devices.[95-97] It
is also very interesting to study the effects of such systems combined with TTF as a
subunit, the latter being involved in different processes, as exemplified above.
In Chapter 2 following the introduction, the main experimental methods used for the
different processes and systems studied are described, such as absorption
spectroscopy, techniques of oxidation and spectroelectrochemistry, luminescence
spectroscopy as well as luminescence lifetime and transient absorption measurements.
This thesis being the result of a collaboration with Prof. Silvio Decurtins and Dr Shi-Xi
Liu of the University of Bern, the synthesis and elementary characterisation of the
compounds performed by them and their co-workers are not explicitly described in the
general experimental section but are contained in the experimental parts of the
individual manuscripts and chapters of the thesis.
Results on the series of ruthenium(II) compounds, [{Ru(bpy)2}n(TTF-ppb)](PF6)2n (n = 1,
2, ppb = dipyrido[2,3-a:3,,2,-c]phenazine), that were published in Inorganic Chemistry[98]
in 2008 are presented in Chapter 3. The intraligand charge transfer absorption band
specific of TTF-ppb donor-acceptor molecule is discussed in conjunction with MLCT
transitions and a variety of relaxation processes taking place in those complexes. In
Chapter 4, the specific behaviour of these molecules is compared in an article published
in Chimia[99] in 2007 with that of a similar but still different series of complexes:
[Ru(bpy)3-n(TTF-dppz)n]2+ (n = 1-3, dppz=dipyrido-[3,2-a:2#,3#-c]phenazine), which show
a long-lived photo-induced charge separated state. In an attempt to better understand
the long-lived charge transfer state and the mechanism by which it is created, another
molecule in which a quinone unit was added as a ligand moiety: [Ru(TTFdppz)2(Aqphen)]2+ (Aqphen = anthraquinone liked to a phenanthroline via a phenazine
!
50
bridge). This work is described in chapter 5 in an article submitted to Inorganic
Chemistry.[100] The work on donor-acceptor dyads containing TTF as electron donor and
other organic moieties as acceptors but without metal coordination are presented in
Chapters 6 and 7, in particular with regard to the specific absorption band of the
intraligand charge transfer transition between TTF as electron donor and the other
ligand moiety. In an article published in Chemistry – An Asian Journal
[101]
in 2009, the
electronic interactions in the donor-acceptor assemblies with imidazole moieties were
studied as a function of pH. The absorption spectra allowed the determination of the
pKa for the different imidazole-annulated TTF systems in non-aqueous solvents. In the
other article, published in Organic Letters
[102]
in 2009, the intraligand charge transfer in
a TTF-perylenediimide (TTF-PDI) molecule was highlighted. In the last chapter, more
recent studies on other transition metal ion complexes, namely of iron(II) and cobalt(II)
complexed to TTF donor-acceptor ligands are presented, in order to study systems
which combine photo-induced charge transfer and spin crossover phenomena. To
finish, general conclusions summarize the work of this thesis. Some suggestions for
further work will also be given.
1.5. References
[1] H. H. Jaffe, M. Orchin, Theory and Applications of Ultraviolet Spectroscopy, New
York, 1962.
[2] V. Balzani, V. Carassiti, Photochemistry of Coordination Compoounds, Academic
Press, London, 1970.
[3] J. B. Birks, Photophysics of Aromatics Molecules, London, 1970.
[4] N. J. Turro, Modern Molecular Photochemistry, Menlo Park, California, 1978.
[5] A. B. P. Lever, Inorganic Electronic Spectroscopy, Elsevier, Amsterdam, 1984.
[6] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. v. Zelewsky Coord.
Chem. Rev. 1988, 84, 85-277.
[7] G. B. Porter, V. Balzani, L. Moggi Adv. Photochem. 1974, 9, 147.
[8] V. Balzani, L. Moggi, M. F. Manfrin, F. Bolletta Coord. Chem. Rev. 1975, 15, 321433.
!
51
[9] V. Balzani, G. Bergamini, S. Campagna, F. Puntoriero Top. Curr. Chem. 2007, 280,
1-36.
[10] R. A. Marcus, N. Sutin Biochim. Biophys. Acta. 1985, 811, 265-322.
[11] M. D. Newton, Electron transfer in chemistry, Wiley-VCH, Weinheim, 2001.
[12] F. Scandola, C. Chiorboli, M. T. Indelli, M. A. Rampi, Electron transfer in chemistry,
Wiley-VCH, Weinheim, 2001.
[13] J. R. Miller, L. T. Calcaterra, G. L. Closs J. Am. Chem. Soc. 1984, 106, 3047-3049.
[14] F. Wudl, G. M. Smith, E. J. Hufnagel J. Chem. Soc. D-Chem. Commun. 1970,
1453.
[15] Q. Y. Zhu, L. B. Huo, Y. R. Qin, Y. P. Zhang, Z. J. Lu, J. P. Wang, J. Dai J. Phys.
Chem. B. 2010, 114, 361-367.
[16] J. Ferraris, V. Walatka, J. Perlstei, D. O. Cowan J. Am. Chem. Soc. 1973, 95, 948949.
[17] W. R. H. Hurtley, S. Smiles J. Chem. Soc. 1926, 2263-2270.
[18] W. R. H. Hurtley, S. Smiles J. Chem. Soc. 1926, 1821-1828.
[19] T. Jorgensen, T. K. Hansen, J. Becher Chem. Soc. Rev. 1994, 23, 41-51.
[20] F. Wudl, D. Wobschal, E. J. Hufnagel J. Am. Chem. Soc. 1972, 94, 670-&.
[21] I. Hargittai, J. Brunvoll, M. Kolonits, V. Khodorkovsky J. Mol. Struct. 1994, 317,
273-277.
[22] C. Katan J. Phys. Chem. A. 1999, 103, 1407-1413.
[23] D. L. Lichtenberger, R. L. Johnston, K. Hinkelmann, T. Suzuki, F. Wudl J. Am.
Chem. Soc. 1990, 112, 3302-3307.
[24] R. A. Marcus J. Chem. Phys. 1956, 24, 966-978.
[25] R. A. Marcus Rev. Modern Phys. 1993, 65, 599-610.
[26] R. A. Marcus Angew. Chem.-Int. Ed. Eng. 1993, 32, 1111-1121.
[27] J. VandeVondele, R. Lynden-Bell, E. J. Meijer, M. Sprik J. Phys. Chem. B. 2006,
110, 3614-3623.
[28] D. E. Schafer, F. Wudl, G. A. Thomas, J. P. Ferraris, D. O. Cowan Solid State
Commun. 1974, 14, 347-351.
[29] C. D. Jaeger, A. J. Bard J. Am. Chem. Soc. 1979, 101, 1690-1699.
[30] U. E. Spichiger-Keller, Chemical Sensors and Biosensors for Medical and
Biological Applications, Weinheim, 1998.
[31] P. A. Dowben Surface Science Reports. 2000, 40, 151-245.
!
52
[32] D. Jerome, H. J. Schulz Adv. Phys. 2002, 51, 293-479.
[33] M. Sakai, M. Iizuka, M. Nakamura, K. Kudo Synth. Met. 2005, 153, 293-296.
[34] M. R. Bryce Adv. Mater. 1999, 11, 11-23.
[35] J. Becher, Z. T. Li, P. Blanchard, N. Svenstrup, J. Lau, M. B. Nielsen, P. Leriche
Pure Appl. Chem. 1997, 69, 465-470.
[36] G. Schukat, A. M. Richter, E. Fanghaenel Sulfur Rep. 1987, 7, 155-231.
[37] G. Schukat, E. Fanghanel Sulfur Rep. 1993, 14, 245-383.
[38] G. Schukat, E. Fanghaenel Sulfur Rep. 1996, 18, 1-294.
[39] A. Aviram, M. A. Ratner Chem. Phys. Lett. 1974, 29, 277-283.
[40] R. M. Metzger, B. Chen, U. Hopfner, M. V. Lakshmikantham, D. Vuillaume, T.
Kawai, X. L. Wu, H. Tachibana, T. V. Hughes, H. Sakurai, J. W. Baldwin, C. Hosch, M.
P. Cava, L. Brehmer, G. J. Ashwell J. Am. Chem. Soc. 1997, 119, 10455-10466.
[41] V. Khodorkovsky, J. Y. Becker, Organic Conductors: Fundamentals and
Applications, Marcel Dekker, New York, 1994.
[42] P. F. Gordon, P. Gregory, Organic chemistry in Colour, Berlin, 1983.
[43] S. R. Marder, B. Kippelen, A. K. Y. Jen, N. Peyghambarian Nature. 1997, 388, 845851.
[44] T. Verbiest, S. Houbrechts, M. Kauranen, K. Clays, A. Persoons J. Mater. Chem.
1997, 7, 2175-2189.
[45] H. Imahori, Y. Sakata Adv. Mater. 1997, 9, 537-&.
[46] M. S. Vollmer, F. Wurthner, F. Effenberger, P. Emele, D. U. Meyer, T. Stumpfig, H.
Port, H. C. Wolf Chemistry-a European Journal. 1998, 4, 260-269.
[47] R. Gompper, H.-U. Wagner Angewandte Chemie International Edition in English.
1988, 27, 1437.
[48] M. Julliard, M. Chanon Chem. Rev. 1983, 83, 425-506.
[49] T. Jigami, K. Takimiya, T. Otsubo, Y. Aso J. Org. Chem. 1998, 63, 8865-8872.
[50] M. Bendikov, F. Wudl, D. F. Perepichka Chem. Rev. 2004, 104, 4891-4945.
[51] Prinzbac.H, H. Berger, Luttring.A Angew. Chem.-Int. Ed. 1965, 4, 435-&.
[52] K. Bechgaard, D. Jerome Scientific American. 1982, 247, 52-61.
[53] J. L. Segura, N. Martin Angew. Chem.-Int. Ed. 2001, 40, 1372-1409.
[54] M. R. Bryce Chem. Soc. Rev. 1991, 20, 355-390.
[55] M. R. Bryce J. Mater. Chem. 1995, 5, 1481-1496.
[56] J. Garin Adv. Heterocycl. Chem., Vol 62. 1995, 62, 249-304.
!
53
[57] K. B. Simonsen, N. Svenstrup, J. Lau, O. Simonsen, P. Mork, G. J. Kristensen, J.
Becher Synthesis-Stuttgart. 1996, 407-&.
[58] T. Otsubo, Y. Aso, K. Takimiya Adv. Mater. 1996, 8, 203-&.
[59] M. Adam, K. Mullen Adv. Mater. 1994, 6, 439-459.
[60] J. Becher, J. Lau, P. Mørk, in Electronic Materials:The Oligomer Approach, WileyVCH Weinheim,, 1998.
[61] M. R. Bryce, W. Devonport, L. M. Goldenberg, C. S. Wang Chem. Commun. 1998,
945-951.
[62] K. B. Simonsen, J. Becher Syn. Lett. 1997, 1211.
[63] M. B. Nielsen, J. Becher Liebigs Ann. 1997, 2177.
[64] E. Coronado, C. J. Gomez-Garcia Chem. Rev. 1998, 98, 273-296.
[65] J. Roncali Chem. Rev. 1997, 97, 173-206.
[66] P. Day, M. Kurmoo J. Mater. Chem. 1997, 7, 1291-1295.
[67] M. B. Nielsen, C. Lomholt, J. Becher Chem. Soc. Rev. 2000, 29, 153-164.
[68] M. R. Bryce J. Mater. Chem. 2000, 10, 589-598.
[69] R. Wolf, M. Asakawa, P. R. Ashton, M. Gomez-Lopez, C. Hamers, S. Menzer, I. W.
Parsons, N. Spencer, J. F. Stoddart, M. S. Tolley, D. J. Williams Angew. Chem.-Int. Ed.
1998, 37, 975-979.
[70] P. de Miguel, M. R. Bryce, L. M. Goldenberg, A. Beeby, V. Khodorkovsky, L.
Shapiro, A. Niemz, A. O. Cuello, V. Rotello J. Mater. Chem. 1998, 8, 71-76.
[71] V. E. Kampar Uspekhi Khimii. 1982, 51, 185-206.
[72] L. M. Goldenberg, J. Y. Becker, O. P. T. Levi, V. Y. Khodorkovsky, L. M. Shapiro,
M. R. Bryce, J. P. Cresswell, M. C. Petty J. Mater. Chem. 1997, 7, 901-907.
[73] L. M. Goldenberg, J. Y. Becker, O. P. T. Levi, V. Y. Khodorkovsky, M. R. Bryce, M.
C. Petty J. Chem. Soc.-Chem. Commun. 1995, 475-476.
[74] K. B. Simonsen, N. Thorup, M. P. Cava, J. Becher Chem. Commun. 1998, 901-902.
[75] M. B. Nielsen, S. B. Nielsen, J. Becher Chem. Commun. 1998, 475-476.
[76] D. Philp, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J. Williams J. Chem. Soc.Chem. Commun. 1991, 1584-1586.
[77] G. Zhang, D. Zhang, X. Guo, D. Zhu Org. Lett. 2004, 6, 1209-1212.
[78] F. Dumur, N. Gautier, N. Gallego-Planas, Y. Sahin, E. Levillain, N. Mercier, P.
Hudhomme, M. Masino, A. Girlando, V. Lloveras, J. Vidal-Gancedo, J. Veciana, C.
Rovira J. Org. Chem. 2004, 69, 2164-2177.
!
54
[79] E. Aqad, A. Ellern, L. Shapiro, V. Khodorkovsky Tetrahedron Lett. 2000, 41, 29832986.
[80] J. Llacay, J. Veciana, J. Vidal-Gancedo, J. L. Bourdelande, R. Gonzalez-Moreno,
C. Rovira J. Org. Chem. 1998, 63, 9144-9144.
[81] C. Boulle, J. M. Rabreau, P. Hudhomme, M. Cariou, M. Jubault, A. Gorgues, J.
Orduna, J. Garin Tetrahedron Lett. 1997, 38, 3909-3910.
[82] B. M. Illescas, J. Santos, M. Wielopolski, C. M. Atienza, N. Martin, D. M. Guldi
Chem. Commun. 2009, 5374-5376.
[83] H. C. Li, J. O. Jeppesen, E. Levillain, J. Becher Chem. Commun. 2003, 846-847.
[84] Y. H. Hou, Y. S. Chen, Q. Lin, M. Yang, X. J. Wan, S. G. Yin, A. Yu
Macromolecules. 2008, 41, 3114-3119.
[85] G. A. Crosby Acc. Chem. Res. 1975, 8, 231-238.
[86] J. Rusanova, S. Decurtins, E. Rusanov, H. Stoeckli-Evans, S. Delahaye, A. Hauser
J. Chem. Soc.-Dalton Trans. 2002, 4318-4320.
[87] S. Delahaye, C. Loosli, S. X. Liu, S. Decurtins, G. Labat, A. Neels, A. Loosli, T. R.
Ward, A. Hauser Adv. Funct. Mater. 2006, 16, 286-295.
[88] L. Spiccia, G. B. Deacon, C. M. Kepert Coord. Chem. Rev. 2004, 248, 1329-1341.
[89] D. V. Kozlov, D. S. Tyson, C. Goze, R. Ziessel, F. N. Castellano Inorg. Chem.
2004, 43, 6083-6092.
[90] M. Galletta, F. Puntoriero, S. Campagna, C. Chiorboli, M. Quesada, S. Goeb, R.
Ziessel J. Phys. Chem. A. 2006, 110, 4348-4358.
[91] J. G. Vos, J. M. Kelly Dalton Trans. 2006, 4869-4883.
[92] T. J. Meyer Pure Appl. Chem. 1986, 58, 1193-1206.
[93] T. J. Meyer Acc. Chem. Res. 1989, 22, 163-170.
[94] J. Gu, J. Chen, R. H. Schmehl J. Am. Chem. Soc. 2010, 132, 7338-7346.
[95] V. Balzani, M. Gomez-Lopez, J. F. Stoddart Acc. Chem. Res. 1998, 31, 405-414.
[96] E. S. Handy, A. J. Pal, M. F. Rubner J. Am. Chem. Soc. 1999, 121, 3525-3528.
[97] F. G. Gao, A. J. Bard J. Am. Chem. Soc. 2000, 122, 7426-7427.
[98] C. Goze, N. Dupont, E. Beitler, C. Leiggener, H. Jia, P. Monbaron, S. X. Liu, A.
Neels, A. Hauser, S. Decurtins Inorg. Chem. 2008, 47, 11010-11017.
[99] C. Leiggener, N. Dupont, S. X. Liu, C. Goze, S. Decurtins, E. Beitler, A. Hauser
Chimia. 2007, 61, 621-625.
!
55
[100] N. Dupont, Y.-F. Ran, H.-P. Jia, J. Grilj, S.-X. Liu, S. Decurtins, A. Hauser Inorg.
Chem. 2010, submitted.
[101] J. C. Wu, N. Dupont, S. X. Liu, A. Neels, A. Hauser, S. Decurtins Chemistry-an
Asian Journal. 2009, 4, 392-399.
[102] M. Jaggi, C. Blum, N. Dupont, J. Grilj, S. X. Liu, J. Hauser, A. Hauser, S. Decurtins
Org. Lett. 2009, 11, 3096-3099.
!
56
2. Experimental part
In this chapter the different spectroscopic techniques used for the determination of the
photophysical and photochemical properties of the different molecules and complexes
studied in this thesis are presented.
2.1. Absorption spectra
All the photophysical measurements were performed on solutions of the compounds in
different solvents. Absorption spectra were recorded on a Varian Cary 50 UV/Vis/NIR or
a Varian Cary 5000 UV/Vis/NIR spectrophotometer. The measurements were
performed in quartz cells with a path length of 1 cm or 1 mm.
Absorption spectroscopy allows the measurement of the absorption of the light as a
function of wavelength (nm) or wavenumber (cm-1), by a sample of a concentration c
(mol.L-1 or M) and of a thickness l (cm). In the case of solutions, the thickness is equal
to the width of the cell (see Figure 2.1).
Figure 2.1: Scheme of the absorption spectroscopy
If an incident beam of light, with an intensity I0 passes through the sample, the emerging
light, after interaction with the molecules has an intensity I, such as I0 > I (see Figure
2.1). The extinction or optical density is then given by:
"I %
Ext = OD = log$ 0 '
# I &
!
!
(1)
57
For a diluted solution, where the interactions between the molecules is negligible, the
extinction can also be described by the Beer-Lambert law, which will be the case for the
different studies shown in this thesis, that is:
OD = "cd
(2)
where % is the molar decadic extinction coefficient, c the concentration and d the path
!
length.
In a double beam spectrometer such as the Cary 5000, the absorption of the solvent is
usually subtracted from the spectrum by either recording a base line with a cell filled
with pure solvent or by placing a cell with the pure solvent in the reference beam.
2.2. Chemical and electrochemical oxidation
In order to study the effect of an oxidation of the ligands and complexes, two methods
were used. The first one involves a chemical oxidation, with FeCl3, [Fe(bpy)3](PF6)3 or
NOBF4 as oxidising agent. The advantage of this method is that only a small amount of
the compound is needed in order to make the few mL of solution with typical
concentrations of 10-5 M. The disadvantage is that in addition to the simple oxidation of
the target molecule further chemical reactions can occur between the reactants present
in the solution, causing perturbations in the measurements.
The second method, spectroelectrochemistry, can avoid theses perturbations. The cell
for this technique used in this thesis is shown in Figures 2.2 and 2.3. The width of that
cell can vary between 0.2 and 0.7 mm by using a 0.5 mm Teflon spacer that can be
added (see Figure 2.3). With 0.2 mm path length, electrolysis of the probe volume is
faster and contamination by diffusion is slower than for 0.7 mm, but then the
concentration of the solute has to be higher, which is not always possible. The cell is
composed of one platinum grid for the working electrode, one platinum wire for the
counter electrode and one silver wire for the reference electrode. A hole of 1 mm
diameter inside the grid of the working electrode allows the beam of the absorption
!
58
spectrophotometer to pass through it. Platinum electrodes being very sensitive to air,
they need to be washed carefully before each new experiment, by solvent with a
conductive salt in the cell for approximately 30 min. The cleaning process can be
followed by cyclic voltammogram measurements. When the cell is not clean enough,
oxidation or reduction waves of impurities can be observed. The cell is assumed to be
clean when the observed cyclic voltammogram is equal to the one of the pure solvent,
without impurities. A conductive salt is also added in excess to the solution. For all
experiments, tetrabutyl ammonium hexafluorophosphate (TBA(PF6)), c = 0.1 M, was
used. The solutions need to be deoxygenated by bubbling N2(g) through them for at
least 30 min before the measurement. When the cell is cleaned and before it is filled
with the actual solution, an absorption spectrum of the solvent is measured in order to
form the baseline. Then, after applying the appropriate potential to the cell, the oxidation
of the ligands or complexes can be measured by consecutive absorption spectra.
Figure 2.2: Picture of the spectroelectrochemical cell designed by F. Hart [1] based on
a standard OTTLE cell.
!
59
Figure 2.3: Picture of the dismounted spectroelectrochemical cell, with the electrodes
(a), the cover of the cell (b), the spacer (c) and the Infrasil window (d) and the nuts
allowing a good sealing of the mounted cell (e).
2.3. Emission and excitation spectra
For luminescence measurements the solutions were deoxygenated by bubbling N2(g)
through them for 30 min in order to avoid quenching from oxygen. Emission and
excitation spectra were recorded on a Horiba Fluorolog 3 spectrometer. Its spectral
resolution is approximately 0.2 nm.
The sample is excited by a lamp, which is followed by an excitation monochromator.
The emitted light is collected either in back reflection or at right angles and analysed by
a second monochromator and a photomultiplier with photon-counting electronics
connected to a computer (see Figure 2.4). Both, emission or excitation spectra can be
recorded.
!
60
For the excitation spectrum, the emission monochromator selects only one wavelength
to monitor the luminescence. The source excites the sample in the wavelength window
chosen by the user. The obtained spectrum shows the emission intensity for one given
emission wavelength as a function of the excitation wavelength emitted by the source.
For excitation spectra, the intensity of the excitation light is monitored at the exit of the
excitation monochromator by a photodiode and the measured signal is divided by the
signal from the photodiode in order to correct for the spectral distribution of the light
source and the response curve of the excitation monochromator.
For emission spectra, the source emits at one given wavelength, and the intensity of the
light emitted by the sample is recorded as a function of wavelength. For emission
spectra, a built in response curve corrects the spectra for the spectral response of the
system between 250 and 1000 nm.
!
Figure 2.4: Schematic diagram showing the principal elements for luminescence
spectra
!
61
The emission spectra allow the determination of the luminescence quantum yield &r of
the sample (s). In order to determine the quantum yield, a reference compound (r) is
needed, with a well-known quantum yield. Depending on the value of the optical density,
the measurement has to be adjusted.
Figure 2.5: Comparison of the attenuation of the beam depending on the
concentration of the sample. For a quantitative evaluation of the data, the excitation/
detection geometry and the attenuation have to be taken into account.
For an optical density < 0.1 at the excitation wavelength, the acquisition can be
performed with the acquisition of the emission perpendicular to the excitation. Indeed,
the absorbance being low, the attenuation of the excitation intensity in the sample is
negligible (case 1 in Figure 2.5).
If the optical density is bigger than 0.1, the attenuation of the beam through the sample
is not negligible. The consequence is that just one part of the signal is seen by the
detector (case 2). In particular when the absorbance of the sample at the excitation
wavelength is equal or bigger than 0.3, this deviation induces too many errors in the
measurements. Thus, the acquisition of the signal has to be performed in parallel to the
excitation path (back reflection).
!
62
Depending on the known parameters of the experiment, the quantum yield for low
absorbance at the excitation wavelength is given by:
r
"s Is $%ex # c r
= # s
"r Ir $%ex # c s
(3)
-i being the quantum yield of the species i = s, r, Ii the integrated emission intensity of i,
!
.i its extinction coefficient at the excitation wavelength and ci its concentration.
For higher absorbance at the excitation wavelength, the formula for the quantum yield
becomes:[2]
"s =
I s $ 1# e # A r
&
I r % 1# e # A s
'$ n s '
)& )" r
(% n r (
(4)
Where as before Ii, i = s, r, is the integrated emission intensity, and Ai the absorbance
!
Ai = OD x ln10
(5)
at the excitation wavelength.
2.4. Lifetime measurements and transient absorption
For luminescence and transient absorption measurements the solutions were
deoxygenated by bubbling N2(g) through them for 30 min, the quenching from oxygen
perturbing the measurements. Luminescence decay curves were measured by exciting
the samples at different wavelength, depending of the studied molecule, either with the
second harmonic of a pulsed Nd:YAG laser (Quantel Brilliant, 7ns pulse width) at 532
nm, or the light from an OPO (Opotek Magic Prism) pumped be the third harmonic of
the Nd:YAG laser. The system used for detection consisted of a Spex 270M
monochromator, a Hamamatsu R928 photomultiplier and a Tektronix TDS 540B
oscilloscope and has a time-resolution of 15 ns (see Figures 2.7a and 2.8).
!
63
!
!
!
Lifetime experiments allow the determination of the luminescence decay curves. The
sample is excited at a certain wavelength with a pulsed laser (the pulse should be
shorter than the lifetime). The population of one excited state decreases according to:
dN e
= "N e k
dt
Ne
(6)
is the number of molecules in the excited state and
k
the relaxation rate constant.
The intensity of the signal is proportional to the population of the excited state. The
measured signal can thus be described by:
I(t) = I0e
"t
!
#
(7)
Where I(t) is the intensity of the signal as a function of the time, I0 is the intensity of the
signal at t = 0 and ' = k-1 is the lifetime of the luminescent state.
In the case of a non-emissive sample, the excited state decay can be followed via
transient absorption or excited state decay measurements by probing the sample with
light from a tungsten-halogen lamp (see Figure 2.7b). For a transient absorption
spectrum as well as for excited state decay measurements, the absorption of the
photoinduced excited states of the molecule, that is to say the population of the excited
states, is observed. The same formulas are used in order to determine the lifetime (eq.
7). Different contributions can be obtained for those measurements: a ground state
bleach is due to the depletion of the ground state and is represented by a negative
signal, versus an excited state absorption, which corresponds to the absorption of the
white light by the populated excited state (positive signal) (see Figure 2.8).
The same system for detection was used as for the luminescence lifetime
measurements. Transient absorption decay curves were recorded at typical
wavelengths, most of the time 470, 620 and 820 nm. Transient absorption spectra could
be recorded with the same setup using the oscilloscope programmed in boxcar mode
and integration over the decay curves while scanning the detection wavelength.
However, higher quality spectra could be obtained with a second setup using a gated
CCD camera (Andor iStar 720). The difference between the two systems is that
!
64
measurements with the monochromator and the photomultiplier are functions of the
wavelength for different times (see Figure 2.9), versus measurements with the gated
CCD camera (see Figure 2.10), taking the entire absorption spectrum of the excited
species for all the wavelengths between 380 and 850 nm but for one integrated time.
a)
b)
Figure 2.7: Comparison between lifetime measurements for a luminescent sample (a)
and a non-luminescent sample (b). In the latter case, transient absorption
measurements are performed.
!
65
Figure 2.8: Schematic representation of the principal contributions of a transient
absorption spectrum observed during this thesis.
Figure 2.9: Setup for excited state lifetime and transient absorption measurements at
fixed wavelengths of detection; the laser beam is shown by the green line, the probe
light from the W-halogen lamp for transient absorption measurements is indicated by
the yellow line.
!
66
Figure 2.10: Setup for recording transient absorption and luminescence spectra at
given time delays following pulsed excitation.
The femtosecond transient absorption set-up has been described in a previous article [3]
and its scheme is shown in Figure 2.11. The output of a Ti:sapphire amplifier (Spitfire,
Spectra Physics; 800 nm pulses of 150 fs FHWM) is split into two parts; ca. 5 µJ are
focused into a 3 mm thick, constantly moving CaF2 window to generate a white light
continuum for probing, the remainder is sent into a home-built two stage non-collinear
optical parametric amplifier to generate the pump pulses at tunable wavelengths. The
polarization of the pump beam is set to magic angle with respect to the probe beam.
The pump power at the sample is approximately 0.1 mJ/cm2. The sample solution in a 1
mm quartz cell is constantly stirred by nitrogen bubbling. The photochemical stability of
the sample is verified by the steady state absorption spectrum taken before and after
the measurement. The probe beam is dispersed in a spectrograph (Andor, SLR163)
and imaged onto a 512+58 pixel back-thinned CCD (Hamamatsu S07030-09). The
spectra are corrected for the chirp of the white light pulses by standard procedures.[4]
The Instrument Response Function, IRF (as deduced from the electronic Optical Kerr
Effect or OKE signal) has a FHWM of approximately 200 fs, depending on the
wavelength. Due to cross-phase-modulation and the coherent signal, the spectra at
early time-delays cannot be observed.
!
67
Figure 2.11: Scheme of the femtosecond transient absorption experiment with a light
probe [5]
2.5. References
[1] M.Krej/ík, M.Dan0k, F.Hartl J. Electroanal. Chem. 1991, 317
[2] S. Delahaye, PhD thesis, Université de Genève, 2005.
[3] G. Duvanel, N. Banerji, E. Vauthey J. Phys. Chem. A. 2007, 111, 5361-5369.
[4] S. Yamaguchi, H. O. Hamaguchi Appl. Spectrosc. 1995, 49, 1513-1515.
[5] J. Grilj, E. Vauthey, Poster for the Gordon Research Conference, 2009.
!
68
3. Ru(II) Coordination Chemistry of a Fused Donor-Acceptor Ligand:
Synthesis, Characterization and Photoinduced Electron Transfer
Reactions of [{Ru(bpy)2}n(TTF-ppb)](PF6)2n (n = 1, 2)
Published in Inorganic Chemistry, 47 (23): 11010-11017, 2008
!
Christine Goze,†,§ Nathalie Dupont, ‡ Elvira Beitler,† Claudia Leiggener,‡ Hongpeng Jia,†
Philippe Monbaron,† Shi-Xia Liu,*, † Antonia Neels,! Andreas Hauser,*,‡ and Silvio
Decurtins†
!
†,
Departement für Chemie und Biochemie, Universität Bern, Freiestrasse 3, CH-3012
Bern, Switzerland
‡
Département de Chimie Physique, Université de Genève, 30 Quai Ernest-Ansermet,
CH-1211 Genève 4, Switzerland
!
Head of the X-ray Diffraction Laboratory, CSEM Centre Suisse d'Electronique et de
Microtechnique SA, Jaquet-Droz 1, Case postale, CH-2002 Neuchâtel, Switzerland
§
Present address of Christine Goze: Université de Bourgogne, LIMRES-ICMUB,
UMR5260-Faculté des Sciences Mirande, 9 avenue Alain Savary, 21000 Dijon, France.
!
69
Abstract
A
!-extended,
redox-active
bridging
ligand
4#,5#-
bis(propylthio)tetrathiafulvenyl[i]dipyrido[2,3-a:3#,2#-c]phenazine (L) was prepared via
direct Schiff-base condensation of the corresponding diamine-tetrathiafulvalene (TTF)
precursor with 4,7-phenanthroline-5,6-dione. Reactions of L with [Ru(bpy)2Cl2] afforded
its stable mono- and dinuclear Ru(II) complexes 1 and 2. They have been fully
characterized and their photophysical and electrochemical properties are reported
together with those of [Ru(bpy)2(ppb)]2+ and
[Ru(bpy)2(µ-ppb)Ru(bpy)2]4+ (ppb =
dipyrido[2,3-a:3#,2#-c]phenazine) for comparison. In all cases, the first excited state
corresponds to an intramolecular TTF$ppb charge transfer state. Both Ru(II)
complexes show two strong and well-separated metal-to-ligand charge transfer (MLCT)
absorption bands whereas the 3MLCT luminescence is strongly quenched via electron
transfer from the TTF subunit. Clearly, the transient absorption spectra illustrate the role
of the TTF fragment as an electron donor which induces a triplet intraligand charge
transfer state (3ILCT) with a lifetime of approximately 200 ns and 50 ns for mono- and
dinuclear Ru(II) complexes, respectively.
!
70
CONTENT
!
3.1. Introduction
72
!
3.2. Experimental Section
73
3.2.1. General.
73
3.2.2. Synthesis of 4!,5!-bis(propylthio)tetrathiafulvenyl[i]dipyrido[2,3-a:3!,2!c]phenazine (L).
74
3.2.3. Synthesis of [Ru(bpy)2L](PF6)2 (1).
75
3.2.4. Synthesis of [Ru(bpy)2(µ-L)Ru(bpy)2](PF6)4 (2).
75
3.2.5. Cyclic Voltammetry.
76
3.2.6. Photophysical Measurements:
76
3.2.7. X-ray Crystallography.
77
3.3. Results and Discussion
78
!
3.3.1. Synthesis and Characterization.
78
3.3.2. Solid-State Structure of the Complex 2.
79
3.3.3. Electrochemical Properties.
82
3.3.4. Optical Properties.
84
!
3.4. Conclusions
93
!
3.5. References
93
!
3.6. Supporting Information:
98
!
!
71
3.1. Introduction
!
There has been a considerable amount of research into the use of tetrathiafulvalenes
(TTFs) that can behave as strong !–donors capable of forming persistent cation radical
and dication species upon oxidation leading, for instance, to the formation of mixedvalence states for conducting systems.1 On the one hand, there have been many
synthetic attempts at introducing paramagnetic metal ions into TTF conducting
molecular lattices.1-2 Consequently, TTFs have been modified with a variety of
functional groups which are well tailored for a chelating coordination function toward
various transition metal ions.2-6 On the other hand, TTFs are frequently used as donor
units in donor–acceptor (D–A) ensembles, which are of prime importance due to their
potential applications in sensors, molecular electronics and optoelectronics.2a,7-8 Our
interest in conducting magnets and molecular electronics led us to the synthesis,
electrochemical and spectroscopic investigations of the TTF-fused dipyrido[2,3-a:3#,2#c]phenazine (ppb) bridging ligand L (TTF-ppb) (Scheme 1).
Scheme 1. Molecular structure of the bridging ligand L.
Bridging polypyridyl ligands can act as building blocks in the construction of
supramolecular arrays such as grids, helicates, boxes, and cylinders.9 It has also been
demonstrated that such systems can be used in photocatalysis, CO2 remediation, solar
energy systems, molecular electronics, sensors and light-emitting diodes.10-11 In the
present case, the bridging ppb unit (Scheme 1) was chosen in order to combine its
coordination ability with the electronic D-A properties of the fused ligand L.
!
72
The combination of TTFs and Ru(II) chromophores has been stimulated by the
development of new antenna and charge separation systems,12 as well as new
photoredox switches.13 However, so far only a few examples of such systems have
appeared in the literature. We recently reported the synthesis, redox properties and
photophysical behavior of three Ru(II) complexes, bearing 1 to 3 TTF–fused
dipyrido[3,2-a:2#,3#-c]phenazine (TTF–dppz) ligands, [Ru(bpy)3-x(dppz–TTF)x]2+ (x = 13).12a For all three complexes, the lowest excited state is a TTF to dppz intraligand
charge transfer (ILCT) state. In particular, we showed that the complex with only one
TTF–dppz exhibits dual luminescence both from the triplet Ru(II)$dppz metal-to-ligand
charge transfer (3MLCT) state as well as from the lowest energy singlet TTF to dppz
intraligand charge transfer (1ILCT) state, whereas for the other two complexes, a
radiationless pathway via electron transfer from a second TTF–dppz ligand quenches
the 3MLCT luminescence. Remarkably, the TTF fragments as electron donors thus
induce a long-lived ligand-to-ligand charge separated (LLCS) state.12a For the complex
with only one TTF–dppz ligand this state has a lifetime of 2.2 µs and is best described
as [Ru(bpy)(bpy•-)(dppz–TTF•+)]2+.
As a continuation of our study, we report here the synthesis of L, the formation of its
mono- and dinuclear Ru(II) complexes, as well as their electrochemical and
photophysical properties, in order to elucidate how the TTF-fused ppb ligand affects the
redox and photophysical behavior of Ru(II) polypyridyl moieties. As shown below, the
side-on coordination to the ppb ligand results in a very different relaxation pathway
compared to the head-on coordination to dppz.
3.2. Experimental Section
3.2.1. General.
!
Unless otherwise stated, all reagents were purchased from commercial sources and
used without additional purification. 5,6-diamino-2-(4,5-bis(propylthio)-1,3-dithio-2ylidene)-benzo[d]-1,3-dithiole
!
(1),7a
4,7-phenanthroline-5,6-dione,14
73
and
cis-
[Ru(bpy)2Cl2](2H2O
15
were prepared according to literature procedures. Elemental
analyses were performed on a Carlo Erba Instruments EA 1110 Elemental Analyzer
CHN. 1H and
13
C NMR spectra were obtained on a Bruker AC 300 spectrometer
operating at 300.18 and 75.5 MHz, respectively: chemical shifts are reported in ppm
referenced to residual solvent protons (CDCl3 CD2Cl2, DMSO-d6). The following
abbreviations were used s (singlet), d (doublet), t (triplet) and m (multiplet). Infrared
spectra were recorded on a Perkin-Elmer Spectrum One FT-IR spectrometer using KBr
pellets. Mass spectra were recorded using an Auto SpecQ spectrometer for EI and an
Applied Biosystems / Sciex Qstar Pulsar for ESI, respectively.
3.2.2. Synthesis of 4#,5#-bis(propylthio)tetrathiafulvenyl[i]dipyrido[2,3-a:3#,2#c]phenazine (L).
!
A
solution
of
5,6-diamino-2-(4,5-bis(propylthio)-1,3-dithio-2-ylidene)-benzo[d]-1,3-
dithiole (290 mg, 0.67 mmol) and 4,7-phenanthroline-5,6-dione (140 mg, 0.67 mmol) in
ethanol (120 mL) was refluxed for 3 h under Argon. After filtration, the resulting
precipitate was collected and purified by chromatography on basic Al2O3 using
CH2Cl2/CH3OH (20:1) as eluent to give the analytically pure ligand as a deep blue
powder. Yield: 0.31 g (76%). 1H NMR (CDCl3): ) = 1.00 (t, 6H), 1.65-1.72 (m, 4H), 2.80
(t, 4H), 7.77 (dd, J = 4.3 Hz , J = 8.3 Hz, 2H), 8.39 (s, 2H), 8.86 (dd, J = 1.5 Hz , J =
8.4 Hz, 2H), 9.25 (dd, J = 1.5 Hz, J = 4.5 Hz, 2H) ppm.
13
C NMR (CDCl3): ) = 151.4,
146.1, 144.0, 142.3, 141.4, 131.6, 125.8, 124.9, 120.7, 38.7, 23.6, 13.2 ppm. IR (KBr): "
= 2959, 1436, 1357, 1090, 741 cm-1. EIMS: m/z 607 [M+H]+. Anal. Calcd (%) for
C28H22N4S6: C, 55.41; H, 3.65; N, 9.23. Found: C, 55.22; H, 3.50; N, 9.26.
!
74
3.2.3. Synthesis of [Ru(bpy)2L](PF6)2 (1).
In a Schlenk flask, a suspension of cis-[Ru(bpy)2Cl2](2H2O (35 mg, 0.067 mmol) and the
ligand L (50 mg, 0.08 mmol) in ethanol (15 mL) / water (3 mL) was sonicated for 15 min
and then heated at 80°C for 15 h under Argon. After cooling down to room temperature,
the precipitate was filtered off and an excess of aqueous Me4NPF6 was added to the
filtrate. The mixture was stirred for 2 h and then the resulting dark-brown precipitate was
filtered, washed with water and dried in vacuum. The crude product was purified by
chromatography on SiO2 with CH2Cl2/EtOH (20:1) as eluent to give the analytically pure
product as a dark-brown crystalline powder. Yield: 49 mg (54%). 1H NMR (DMSO-d6): )
= 0.94 (dt, 6H), 1.55-1.62 (m, 4H), 2.84 (dt, 4H), 7.09 (s, 1H), 7.24 (d, J = 5.8 Hz, 1H),
7.29-7.35 (m, 2H), 7.56-7.60 (m, 1H), 7.68-7.76 (m, 2H), 7.95-7.99 (m, 1H), 8.02-8.13
(m, 5H), 8.15-8.21 (m, 1H), 8.26-8.37 (m, 2H), 8.57 (s, 1H), 8.73 (d, J = 8.4 Hz, 1H),
8.87 (d, J = 8.3 Hz, 1H), 8.91 (t, J = 9 Hz, 2H), 9.23 (d, J = 2.6 Hz, 1H), 9.41 (dd, J =
1.7 Hz, J = 8.5 Hz, 1H), 9.48 (dd, J = 1.1 Hz, J = 8.9 Hz, 1H) ppm. ESI-MS: m/z
510.03; calcd. for [M-2PF6-]2+: 510.03. Anal. Calcd (%) for C48H38F12N8P2RuS6(EtOH: C,
44.28; H, 3.27; N, 8.26. Found: C, 44.90; H, 2.91; N, 7.87.
3.2.4. Synthesis of [Ru(bpy)2(µ-L)Ru(bpy)2](PF6)4 (2).
In a Schlenk flask, to a stirred solution of cis-[Ru(bpy)2Cl2](2H2O (257 mg, 0.49 mmol)
in ethanol (15 mL) / water (3 mL), the ligand L (100 mg, 0.16 mmol) was added. The
mixture was heated at 80°C for 48 h under Argon until the complete consumption of the
starting material was detected by TLC. After cooling down to room temperature, the
precipitate was filtered off. Aqueous potassium hexafluorophosphate was added to the
filtrate. The crude precipitate was washed twice with water and once with diethyl ether,
and was recrystallized successively by slow evaporation of a solution in acetone/Et2O
(80:20) and in CH2Cl2/hexane (80:20), to give the analytically pure product as a green
powder. Yield: 0.24 g (76 %). 1H NMR (CD2Cl2): ) = 0.98 (t, 6H), 1.63-1.70 (m, 4H),
!
75
2.77-2.83 (m, 4H), 6.70 (s, 1H), 7.02 (s, 1H), 7.16 (d, J = 4.9 Hz, 2H), 7.42-7.48 (m,
4H), 7.55-7.63 (m, 8H), 7.83-7.87 (m, 6H), 8.12-8.19 (m, 7H), 8.39-8.42 (m, 2H), 8.488.52 (m, 4H), 8.57-8.62 (m, 3H), 9.14 (d, J = 8.3 Hz, 1H), 9.21 (d, J = 7.7 Hz, 1H) ppm.
ESI-MS:
m/z
862.02;
calcd.
for
[M-2PF6-]2+:
862.01.
Anal.
Calcd
(%)
for
C68H54F24N12P4Ru2S6: C, 40.56; H, 2.70; N, 8.35. Found: C, 40.49; H, 2.85; N, 8.15.
3.2.5. Cyclic Voltammetry.
Cyclic voltammetry was conducted on a VA-Stand 663 electrochemical analyzer. An
Ag/AgCl electrode containing 2 M LiCl served as reference electrode, a glassy carbon
electrode as counter electrode, and a Pt tip as working electrode. Cyclic voltammetric
measurements were performed at room temperature under N2 in CH2Cl2 with 0.1 M
Bu4NPF6 as supporting electrolyte at a scan rate of 100 mV(s-1.
3.2.6. Photophysical Measurements:
Photophysical measurements were performed on solutions of the compounds in CH3CN
and CH2Cl2 at room temperature. For luminescence and transient absorption
measurements the solutions were degassed by bubbling N2(g) through them for 30 min.
Absorption
spectra
were
recorded
on
a
Varian
Cary
5000
UV/vis/NIR
spectrophotometer. Emission and excitation spectra were measured on a Horiba
Fluorolog 3 instrument. Luminescence lifetimes were measured by exciting the samples
at 532 nm with the second harmonic of a pulsed Nd:YAG laser (Quantel Brilliant, 7ns
pulse width) or at 458 nm using the third harmonic of the pulsed Nd:YAG laser to pump
an OPO (Opotek Magic Prism). The measuring time corresponds to deactivation time of
the sample. The obtained signal is then integrated. The system used for detection
consisted of a Spex 270M monochromator, a Hamamatsu photomultiplier and a
!
76
Tektronix TDS 540B oscilloscope and has a time-resolution of 15 ns. For the transient
absorption measurements, the samples were also excited at 458 or 532 nm and probed
with light from a W-halogen lamp. The same system for detection was used as for the
luminescence lifetime measurements. Transient absorption decay curves were recorded
at 440, 470, 540 and 660 nm. Transient absorption spectra were recorded with the
oscilloscope programmed in boxcar mode and integration over the decay curves.
3.2.7. X-ray Crystallography.
A green crystal of 2 was mounted on a Stoe Mark II-Imaging Plate Diffractometer
System equipped with a graphite-monochromator. Data collection was performed at 100˚C using Mo-K1 radiation (( = 0.71073 Å). 120 exposures (5 min per exposure)
were obtained at an image plate distance of 135 mm, 2 = 0˚ and 0 < 3 < 180˚ with the
crystal oscillating through 1.5˚ in 3. The resolution was Dmax - Dmin: 24.00 - 0.82 Å. The
structure was solved by direct methods using the program SHELXS-9716 and refined by
full matrix least squares on F2 with SHELXL-97.17 The hydrogen atoms were included in
calculated positions and treated as riding atoms using SHELXL-97 default parameters.
All non-hydrogen atoms were refined anisotropically. A semi-empirical absorption
correction was applied using MULABS (PLATON,18 Tmin = 0.767, Tmax = 0.902).
Crystal data and structural refinement parameters: C82.50H72F24N14P4Ru2S6, Mr =
2233.92, monoclinic, space group C2/m, a = 15.6106(19), b = 25.8029(17), c =
23.333(3) Å, ! = 98.558(10)°, V = 9293.9(18) Å3, Z = 4, *calcd = 1.597 g cm-3, µ (MoK+)
= 0.629 mm-1, T = 173(2) K, F(000) = 4500, R1 = 0.0838 (wR2 = 0.1917) for 5211 unique
reflections (Rint = 0.1468) with a GOF of 1.010.
!
77
3.3. Results and Discussion
3.3.1. Synthesis and Characterization.
The fused donor-acceptor ligand (L) can be synthesized in 76% yield via the direct
condensation
reaction
of
4,7-phenanthroline-5,6-dione
bis(propylthio)-1,3-dithio-2-ylidene)-benzo[d]-1,3-dithiole
in
with
5,6-diamino-2-(4,5-
ethanol,
as
shown
in
Scheme 2. Elemental analyses and spectroscopic characterization confirmed the
formation of the TTF-fused bis-bidentate ligand L.
Scheme 2. A synthetic route to the bridging ligand L.
Reaction of cis-[Ru(bpy)2Cl2](2H2O with 1.2 and 0.3 equivalents of L in aqueous ethanol
at reflux gave the mono- and dinuclear Ru(II) complexes 1 and 2, respectively. Both
new Ru(II) compounds were purified by chromatographic separation on silica gel or by
recrystallization. The 1H NMR, ESI-MS spectra and elemental analysis data on these
products (see Experimental section) were consistent with the formation of the target
complexes. In the case of 2, 1H NMR spectral data clearly indicate that the isolated
material is a mixture of two diastereoisomers (meso- and rac- form), approximately with
a ratio of 60:40 as evidenced by the presence of the singlets at 6.70 and 7.02 ppm
corresponding to the resonance of two protons of the benzene ring. Obviously, the
diastereoisomeric forms of the complexes exhibit distinctive resonances due to the
enhanced rigidity of the bridging ligand induced by the fusion of the TTF moiety with the
ppb unit.
!
78
3.3.2. Solid-State Structure of the Complex 2.
Slow evaporation of a solution of 2 in acetonitrile/toluene (1:1) gave green crystals
suitable for X-ray structure analysis, which has unequivocally confirmed the existence of
a dinuclear ruthenium(II) complex. This complex crystallizes as a solvated compound
2·2CH3CN·1.5C7H8 in a centrosymmetric monoclinic space group (C2/m). An ORTEP
plot of the dinuclear cation with the atomic numbering scheme is shown in Figure 1. The
asymmetric unit comprises half of the dinuclear complex, because the mirror plane is
dividing the complex along its long axis (C10–C11). The propyl substituent on S3 is
disordered over two positions (0.35 / 0.65); after initial refinement, the atom positions
participating in the disorder were fixed. Two crystallographically independent
hexafluorophosphate anions are located on three positions, whereby two of them are
half occupied. The solvent molecules lie also on partially occupied positions.
Figure 1. Perspective view of the complex cation in 2·2CH3CN·1.5C7H8; thermal
ellipsoids are set at 30% probability. Hydrogen atoms, solvent molecules and the
anions are omitted for clarity.
!
79
As shown in Figure 1, the bridging ligand L is almost planar with a rms deviation of
0.0774 Å from a least-squares plane through all ligand atoms, excluding the two
peripheral propyl groups. The Ru(II) ion is displaced out of this least-squares plane of L
by 0.489(5) Å. The bond lengths and angles (Table 1) of the TTF moiety are in the
range expected for neutral TTF derivatives.7a,19 The ligand L links the two Ru(II) centers
with a Ru(((Ru separation of 6.908 Å which is good agreement with the value of 6.818 Å
reported for the analogous compound meso-[Ru(bpy)2(µ-ppb)Ru(bpy)2]4+.20 Compound
2 was crystallized in a meso- diastereoisomeric form which contains an axial mirror
plane bisecting the ligand L; the two coordination spheres around the Ru(II) centers
show opposite chirality (, and -). Both the short Ru(((Ru separation and the electronic
delocalization of the !-conjugated bridging ligand might be important for the effective
electronic interaction of the two metal ion centers.
Each Ru(II) ion is bound by two bpy chelates and one imine chelating unit from the
bridging ligand L in a distorted octahedral fashion. As shown in Table 1, the Ru–N bond
lengths are in the range of 2.041–2.090 Å with their respective bite angles in the range
of 79.3–79.6°, which are within normal ranges set by similar compounds.20-21
!
80
Table 1. Selected Bond Lengths (Å) and Bond Angles (°) of Compound 2(2CH3CN(1.5C7H8.
Bond Lengths
C9-C9*
1.407(18)
C10-
1.324(19)
Ru1-N1
2.050(6)
Ru1-N2
2.090(6)
C11
C10-S1
1.758(7)
S1-C9
1.735(9)
Ru1-N3
2.051(7)
Ru1-N4
2.071(7)
S2-C12
1.752(12)
C12-
1.35(3)
Ru1-N5
2.041(7)
Ru1-N6
2.041(7)
C12*
C11-S2
1.733(8)
Bond Angles
C9*-C9-S1
116.6(3)
N1-Ru1-N2
79.3(2)
N3-Ru1-N2
86.3(2)
C11-C10-S1
122.7(4)
N5-Ru1-N2
98.9(3)
N6-Ru1-N2
175.0(2)
S1-C10-S1*
114.6(7)
N5-Ru1-N6
79.6(3)
N5-Ru1-N1
91.7(3)
C10-C11-S2
122.8(4)
N5-Ru1-N3
173.1(3)
N6-Ru1-N1
96.0(2)
S2-C11-S2*
114.3(8)
N6-Ru1-N3
95.6(2)
N1-Ru1-N3
93.7(3)
C12*-C12-
116.6(5)
N5-Ru1-N4
95.2(3)
N3-Ru1-N4
79.4(3)
101.0(2)
N6-Ru1-N4
84.0(2)
N1-Ru1-N4
173.0(3)
S2
N4-Ru1-N2
• Symmetry transformation used to generate equivalent atoms: x, -y-1, z.
!
In the crystal packing of 2, the complex cations are stacked in a head-to-tail manner
leading to the formation of dimers with S(((S close contacts of 3.680 Å (Figure 2).
!
81
3.3.3. Electrochemical Properties.
The electrochemical properties of L and its ruthenium(II) complexes 1 and 2 in
dichloromethane were investigated by cyclic voltammetry (CV). Their electrochemical
data are collected in Table 2 together with those of [Ru(bpy)2(ppb)]2+ (3) and
[Ru(bpy)2(µ-ppb)Ru(bpy)2]4+
(4)
(ppb
=
dipyrido[2,3-a:3#,2#-c]phenazine)
for
comparison.
Figure
2.
Packing
diagram
(ac
projection)
of
the
complex
cations
in
2·2CH3CN·1.5C7H8. Hydrogen atoms, solvent molecules and the anions are omitted
for clarity. The S(((S close contacts are depicted.
The bridging ligand L undergoes two well-separated (quasi)reversible single electron
oxidation processes to the radical cation and dication states, corresponding to E1/21 and
E1/22, respectively (Table 2). Several CV measurements have been performed at
different scan rates (See Supporting Information). On the one hand, the peak-to-peak
!
82
separations (,Ep = Epa-Epc) increases at high scan rates indicating the quasi-reversible
nature of the electron transfer processes for the oxidation of the TTF unit. On the other
hand, the intensities of the redox waves increase and concomitantly one new wave
appears at 1.10 V as the potentials are cycled. Finally, the color of the solution changes
from purple to dark green. Thus, the instability of the radical cation and dication in the
vicinity of the working electrode is probably attributable to a cleavage of the conjugation
between the TTF moiety and the ppb unit. Upon coordination, the observed redox
potentials for the TTF oxidation processes remain almost unchanged and the peak-topeak separations ,Ep are smaller than those observed for the free ligand L. It can
therefore be deduced that the electrostatic inductive effect of the ruthenium(II) ion
bound to the imine-chelating unit(s) from the bridging ligand L, seems to have a
negligible influence on the redox potentials of the TTF moiety. Interestingly, it seems
likely that coordination renders the ligand more stable in the course of the successive
oxidation processes of the TTF unit. Moreover, both 1 and 2 do not show the Ru(II)centered oxidation process(es) under the experimental conditions used. Since the TTF
unit is oxidized first, the subsequent oxidation process(es) of Ru(II) may be shifted to
more positive potential(s) compared to 3 and 4, which seems reasonable based on
simple electrostatic arguments.
Table 2. Redox Potentials (V vs. Ag/AgCl) of L, 1, 2 in CH2Cl2 and of the reference
compounds 3, 4 in CH3CN.
Compound
Oxidation
E1/21
E1/22
L
0.80a
1.20a
-1.08
1
0.85
1.20
-0.44
2
0.85
1.20
-0.08
E1/23
3
1.64
4
1.57
a
!
Reduction
E1/24
E1/21
E1/22
-1.21
-0.73
-0.44
1.77
-0.12
quasi-reversible, at a 50 mV/s scan rate.
83
E1/23
-1.41
-1.37
-0.85
-1.42
In the cathodic region, one reversible one-electron reduction wave was observed for the
bridging ligand L, which can be assigned to reduction of the phenazine moiety.
Complexes 1 and 2, respectively, undergo two and three reversible reduction processes
for the reduction of the ppb and bpy moieties. The positive shift in the first bridging
ligand-centered reduction process on going from 1 to 2 is in agreement with a further
decrease in electron density around the ppb unit caused by the participation of the
second ruthenium(II) ion in coordination. Moreover, the reduction potentials of 1 and 2
are quite similar to those of the reference compounds 3 and 4, respectively. Thus, the
presence of the TTF unit does not strongly influence the ligand-centered reduction
processes. Since the first two reduction processes (E1/21 and E1/22) take place at less
negative potentials than those of bpy (E1/23), the lowest unoccupied molecular orbital
(LUMO) in each case must reside on the ppb unit of the bridging ligand L.
3.3.4. Optical Properties.
The absorption spectra of the three compounds L, 1 and 2 dissolved in CH3CN,
together with those of the reference compounds 3 and 4, are presented in Figure 3. The
UV-visible-NIR spectrum of the free ligand L (Figure 3a) shows a broad absorption band
at approximately 19200 cm-1 and a very strong band at 30000 cm-1 with a shoulder.
While ppb and TTF units also exhibit absorption bands above 20000 cm-1 individually,
the band at 19200 cm.-1 is only observed in the fused TTF-ppb. In analogy to the
previously reported TTF-dppz,7a it can be readily attributed to a spin-allowed intraligand
!-!* charge transfer (1ILCT) transition with the TTF subunit as an electron donor and the
ppb subunit as an acceptor. The band centered at 30000 cm-1 is characteristic for a !-!*
charge transfer transition of the benzene annulated TTF moiety of the molecule.
Moreover, a weaker band at approximately 27000 cm-1 is characteristic for a !-!*
transition of the ppb unit. In addition, as depicted in Figure 3a, L shows fluorescence in
solution at room temperature. The fluorescence spectrum is strongly solvent-dependent,
with the maximum shifting monotonically to lower energies with increasing polarity of the
solvent.7a-b In parallel, the fluorescence quantum yield at room temperature decreases
!
84
from 5% in toluene to 0.14% in CH3CN in analogy to the one of the previously discussed
TTF-dppz compound. 7a-b The excitation spectrum of L in CH3CN included in Figure 3a
is identical to the absorption spectrum. The oscillator strengths for the different
transitions of L and TTF-dppz are similar, 7a-b particularly, for the ILCT band (f = 0.24),
corresponding thus to a spin- and parity-allowed transition. The assignment to a 1ILCT
band is supported by the Lippert Mataga plot of the solvent shift (see Supporting
Information) which gives a change in dipole moment of 14 Debye.
Figures 3b and 3c show the absorption spectra of the mononuclear complex 1 and the
dinuclear complex 2 in CH3CN together with those of the reference complexes 3 and 4,
respectively. By comparison with the absorption spectra of the reference complexes as
well as of the free ligand L, the absorption bands can be readily attributed to specific
transitions. The 1ILCT band for 1 is at 15100 cm-1 and for 2 at 11900 cm-1. The red shift
of 4100 cm-1 between L and 1, and of 3200 cm-1 between 1 and 2 is due to the
localization of the LUMO on the ppb unit, the energy of which is lowered upon
coordination to Ru2+. These shifts are in good agreement with those observed in the
similar Zn(II)-coordinated system7a-b and with the electrochemical results given in Table
2. The extinction coefficient of the 1ILCT absorption band decreases from 14.103 to
9.103 M-1(cm-1 on going from the free ligand L to the complexes 1 and 2. As expected,
electric-dipole allowed metal-to-ligand-charge-transfer (MLCT) absorption bands around
23800 cm-1 for complexes 1 and 2 are observed. They correspond to a metal-to(terminal)ligands d!-!* charge transfer Ru2+$bpy (1MLCT2).
!
85
Figure 3. a) Absorption spectra of L (__) and ppb (- - -) in CH3CN in comparison to the
excitation spectrum (…) of L in CH3CN and its emission spectra in CH3CN (__) , CH2Cl2 (__)
and toluene (_ . _). b) Absorption spectrum of 1 (__) and absorption (…) and emission (- - -)
spectra of 3 in CH3CN. c) Absorption spectra of 2 (__) and 4 (…) in CH3CN. All measurements
were performed at room temperature in de-oxygenated solutions.!
!
86
The broad and intense bands observed at approximately 18500 and 15000 cm-1, for 1
and 2, respectively, have corresponding bands in the reference complexes 3 and 4, and
can be assigned to the metal-to-(bridging)ligand !-!* charge transfer Ru2+$ppb
(1MLCT1). In each case, the peak position, the broadness and the intensity of the
1
MLCT1 absorption band are slightly solvent-dependent, whereas the
1
MLCT2
absorption band does not vary noticeably with the solvent. These 1MLCT1 and 1MLCT2
bands are clearly separated as observed for other mixed-ligand diimine complexes with
the ppb unit.22 In comparison with the reference compound 3, 1 exhibits an absorption
band centered at 26900 cm-1, which is comparable to that of the free ligand L and can
be attributed to the above-mentioned !-!* transition located on the ppb subunit. Clearly,
the corresponding band is red-shifted in the case of 2, and overlaps with the 1MLCT2
absorption band. Figure 3b shows that 3 emits from the 3MLCT1 state at 12000 cm-1, in
contrast to 1, which does not emit. In 1, the 3MLCT1 luminescence is quenched by
reductive excited state electron transfer from the TTF subunit, which is efficient due to
the geometry of the complex (see discussion below). Neither complex 2 nor complex 4
show any luminescence above 10000 cm-1. Compound 2 probably doesn't show any
luminescence at all for the same reason as 1. For 4, the 1MLCT1 transition is so much
red-shifted that a possible luminescence would be located at lower energies than
accessible with the available spectrofluorimeter.
!
87
Figure 4. Absorption spectra of 2 (___), 2•+ upon oxidation by 2 equivalents of
[Fe(bpy)3](PF6)3 (___), and upon subsequent addition of 2 equivalents of ferrocene
(…..) in CH3CN at room temperature.
Upon chemical oxidation of the free ligand with [Fe(bpy)3]3+, the 1ILCT band at 19200
cm-1 disappears and two new bands at 11800 and 25000 cm-1 appear (see supporting
material). In analogy to the previously studied TTF-dppz, the new band at 11800 cm-1
corresponds to a ppb to TTF•+ intraligand charge transfer transition.7a-b
Oxidation of the complexes 1 and 2 to the respective radical cation can likewise be
achieved chemically using [Fe(bpy)3]3+ as oxidizing agent. Figure 4 shows the
corresponding absorption spectrum of complex 2•+ together with the spectrum of the
non-oxidized form. The most prominent change in the absorption spectrum is the
disappearance of the 1ILCT absorption band at 11900 cm-1 upon oxidation. In addition,
the 1MLCT1 absorption band is slightly red-shifted. The absorption bands at around
20000 cm-1 are due to the 1MLCT of [Fe(bpy)3]2+ formed during the redox process. With
the addition of ferrocene, Figure 4 shows that the oxidation of the complex by
[Fe(bpy)3]3+ is at least partially chemically reversible. Oxidation does not restitute the
3
!
MLCT luminescence. This is due to oxidative electron transfer quenching by TTF•+, the
88
corresponding driving force being around 0.9 eV based on the redox potentials given in
Table 2 and a zero-point energy of the 3MLCT state of 1.75 eV estimated from the
luminescence spectrum of reference compound 3.
Figure 5. Transient absorption spectra of 1 (__) and 3 (__) and absorption spectra of
1 (…) and 3 (…) in CH2Cl2 at room temperature.
!
89
Figure 6. Transient absorption spectra of 2 (__) and 4 (__) and absorption spectra of
2 (…) and 4 (…) in CH2Cl2 at room temperature.
In order to elucidate the nature of the luminescence quenching in 1 and 2 further, and to
compare the photophysical behavior of the two complexes with the one of the series
[Ru(bpy)3-x(dppz-TTF)x]2+ (x = 1-3) previously reported,12a transient absorption spectra
of 1 to 4 were recorded. Those of 1 and 3 in CH2Cl2 are shown in Figure 5 together with
the ground state absorption spectra of both complexes; those of 2 and 4 are shown in
Figure 6. Besides the bleaching of the 1MLCT1 and 1MLCT2 ground state absorptions,
there are two transient absorption bands around 21000 cm-1 and 14200 cm-1 for
complex 1, and one transient absorption band at 21000 cm-1 for complex 2. The
transient absorption band at 21000 cm-1 is observed for excitation either into the
1
MLCT1 or into the 1MLCT2 for both complexes. The corresponding transient state has a
lifetime of around 200 ns for 1 and 50 ns for 2 in CH2Cl2 at room temperature. This
transient absorption is not present for the two reference complexes 3 and 4. Therefore it
must be due to the presence of the TTF unit in complexes 1 and 2. As mentioned
above, the most likely mechanism for the quenching of the luminescence in 1 and the
population of a transient state, is electron transfer quenching with the TTF unit as donor
and the formal Ru3+ as acceptor. Thus, as sketched in Scheme 3, upon irradiation into
!
90
the 1MLCT2 band, a first very quick step takes the system to the lowest energy 3MLCT1
via very fast intersystem crossing (ISC) and electron hopping corresponding to a ligand
to ligand electron transfer (LLET). The quenching step in which the electron is
transferred from TTF to reduce Ru3+ back to Ru2+ results in the direct formation of the
3
ILCT state in a formally spin-allowed process. This 3ILCT state has a much longer
lifetime than the corresponding singlet state ( ' of 1ILCT < 1 ns) and is at lower energy.
The lifetime of the 3ILCT state in the complex is shorter than it would be in the free
ligand L due to the large spin-orbit coupling constant of the coordinated ruthenium ion.
In addition, we note that the reference complexes without the TTF unit, 3 and 4 (for
which the lowest excited states are the 3MLCT1 states), exhibit longer lifetimes of 668
and 300 ns, respectively (see Table 3), and for complex 3, the transient absorption
decay has the same lifetime as the luminescence decay.
Table 3: Summary of excited state lifetime data obtained for complexes 1 - 4.
' (ns)
1
2
3
4
200(5)
50(9)
668(5)
300(40)
3
3
3
3
ILCT
ILCT
MLCT
MLCT
In the related series of complexes [Ru(bpy)3-x(dppz-TTF)x]2+ (x = 1-3) with the head-on
coordination of the dppz-TTF ligand, we reported an unusual dual luminescence for the
member with x = 1. In addition to the 1ILCT fluorescence with a lifetime < 1 ns, the
typical 3MLCT luminescence of Ru(II)-polypyridyl complexes with a lifetime of 1040 ns in
CH2Cl2 was observed for irradiation into the corresponding 1MLCT absorption.12a In the
complex with x = 1, the MLCT state of lowest energy, even though at higher energy than
in the present systems, corresponds to the Ru$dppz-TTF charge transfer. That
luminescence from this state is observed, was attributed to the electron residing on the
dppz unit of the ligand, which effectively hinders the intramolcular electron transfer
quenching via TTF $ Ru. This was borne out by the fact that for the complexes with x =
2 and 3 the 3MLCT luminescence is fully quenched, as now intramolecular electron
transfer quenching from a second dppz-TTF ligand becomes possible. Such electron
!
91
transfer quenching leads to charge separated states with lifetimes of 2.2 - 2.4 µs. Even
for x = 1, a charge separated state results from irradiation into the Ru$bpy MCLT
transition and is best described as [Ru(bpy)(bpy•-)(dppz–TTF•+)]2+. In contrast, the sideon coordination geometry of complex 1 hinders the intramolecular electron transfer
much less even when the complex is in the MLCT1 state with the electron located on the
ppb ligand. Thus the
3
MLCT luminescence is efficiently quenched. The resulting
intermediate state is an ILCT state, and according to spin selection rules this must result
in the triplet state, which in turn must have a longer lifetime than the corresponding
singlet state. The coordination of the second ruthenium ion to the ligand shifts the ILCT
state to lower energy, thus as expected for the Marcus inverted region, the lifetime
decreases compared to the mono-nuclear complex.
Scheme 3. Energy-level scheme for the excited states and relaxation paths of 1.
!
92
3.4. Conclusions
!
A facile synthetic protocol for the preparation of a !-extended, redox-active and bischelating bridging ligand L has been described. Its coordination ability has been
demonstrated by the formation of stable mono- and dinuclear Ru(II) complexes 1 and 2
based on the [Ru(bpy)2Cl2] precursor complex. These new compounds have an ILCT
state as lowest excited state and display an intense 1ILCT absorption band in the NIR.
Both Ru(II) complexes also show two strong and well-separated 1MLCT absorption
bands. The 3MLCT luminescence is strongly quenched via electron transfer from the
TTF subunit. Interestingly, through this quenching step, the corresponding 3ILCT state is
directly
formed
in
a
spin-allowed
process.
Picosecond
transient
absorption
measurements will give further information on the ultrafast formation of the
corresponding longer lived states.
The binding of the redox-active bridging L to a variety of paramagnetic transition metal
ions and further chemical and electrochemical partial oxidation of the resulting
complexes may pave the way to obtain multifunctional materials, which are currently
under investigation in our laboratory.
Acknowledgment. This work was supported by the Swiss National Science Foundation
(grants No. 200020-116003 and 200020-115867), and COST Action D31.
3.5. References
(1) (a) Yamada, J-I.; Sugimoto, T. Eds. TTF Chemistry: Fundamentals and Applications
of Tetrathiafulvalene; Springer: Berlin, Germany, 2004. (b) Khodorkovsky, V.; Becker,
J. Y. Organic Conductors: Fundamentals and Applications; Farges, J. P., Ed.; Marcel
Dekker: New York, NY, 1994; Chapter 3, p 75. (c) William, J. M.; Ferraro, J. R.; Thorn,
!
93
R. J.; Carlson, K. D.; Geiser, U.; Wang, H. H.; Kini, A. M.; Whangbo, M. H. Organic
Superconductors; Prentice Hall: Englewoods Cliffs, NJ, 1992.
(2)
(a) Batail, P., Ed. Chem. Rev. 2004, 104, 4887, special issue on molecular
conductors. (b) Ouahab, L.; Enoki, T. Eur. J. Inorg. Chem. 2004, 933.
(3) (a) Liu, S-X.; Dolder, S.; Pilkington, M.; Decurtins, S. J. Org. Chem. 2002, 67, 3160.
(b) Liu, S-X.; Dolder, S.; Franz, P.; Neels, A.; Stoeckli-Evans, H.; Decurtins, S. Inorg.
Chem. 2003, 42, 4801. (c) Jia, C.; Liu, S-X.; Ambrus, C.; Neels, A.; Labat, G.; Decurtins,
S. Inorg. Chem. 2006, 45, 3152. (d) Hervé, K.; Liu, S-X.; Cador, O.; Golhen, S.; Le Gal,
Y.; Bousseksou, A.; Stoeckli-Evans, H.; Decurtins, S.; Ouahab, L. Eur. J. Inorg. Chem.
2006, 3498. (e) Asara, J. M.; Uzelmeier, C. E.; Dunbar, K. R.; Allison, J. Inorg. Chem.
1998, 37, 1833.
(4) (a) Setifi, F.; Ouahab, L.; Golhen, S.; Yoshida, Y.; Saito, G. Inorg. Chem. 2003, 42,
1791. (b) Liu, S-X.; Ambrus, C.; Dolder, S.; Neels, A.; Decurtins, S. Inorg. Chem. 2006,
45, 9622. (c) Lu, W.; Zhang, Y.; Dai, J.; Zhu, Q-Y.; Bian, G-Q.; Zhang, D-Q. Eur. J.
Inorg. Chem. 2006, 1629. (d) Avarvari, N.; Fourmigué, M. Chem. Commun. 2004, 1300.
(5)
(a) Massue, J.; Bellec, N.; Chopin, S.; Levillain, E.; Roisnel, T.; Clérac, R.; Lorcy,
D. Inorg. Chem. 2005, 44, 8740 and references therein. (b) Devic, T.; Avarvari, N.;
Batail, P. Chem. Eur. J. 2004, 10, 3697. (c) Mosimann, M.; Liu, S-X.; Labat, G.; Neels,
A.; Decurtins, S. Inorg. Chim. Acta 2007, 360, 3848. (d) Benbellat, N.; Gavrilenko, K. S.;
Le Gal, Y.; Cador, O.; Golhen, S.; Gouasmia, A.; Fabre, J.-M.; Ouahab, L. Inorg. Chem.
2006, 45, 10440. (e) Uzelmeier, C. E.; Smucker, B. W.; Reinheimer, E. W.; Shatruk, M.;
O#Neal, A. W.; Fourmigué, M.; Dunbar, K. R. Dalton Trans. 2006, 5259.
(6) (a) Wu, J. C.; Liu, S.-X.; Keene, T. D.; Neels, A.; Mereacre, V.; Powell, A. K.;
Decurtins, S. Inorg. Chem. 2008, 47, 3452. (b) Dolder, S.; Liu, S-X.; Le Derf, F.; Sallé,
M.; Neels, A.; Decurtins, S. Org. Lett. 2007, 9, 3753. (c) Zhao, Y.-P.; Wu, L.-Z.; Si, G.;
Liu, Y.; Xue, H.; Zhang, L.-P.; Tung, C.-H. J. Org. Chem. 2007, 72, 3632. (d)
Gavrilenko, K. S.; Punin, S.V.; Cador, O.; Golhen, S.; Ouahab, L.; Pavlishchuk, V. V. J.
Am. Chem. Soc. 2005, 127, 12246. (e) Chahma, M.; Hassan, N.; Alberola, A.; StoeckliEvans, H.; Pilkington, M. Inorg. Chem. 2007, 46, 3807. (f) Balandier, J.-Y.; Belyasmine,
A.; Sallé, M. Eur. J. Org. Chem. 2008, 269. (g) Benhaoua, C.; Mazari, M.; Mercier, N.;
!
94
Le Derf, F.; Sallé, M. New J. Chem. 2008, 32, 913. (h) Zhu, Q.-Y.; Liu, Y.; Zhang, Y.;
Bian, G.-Q.; Niu, G.-Y.; Dai, J. Inorg. Chem. 2007, 46,10065.
(7) (a) Jia, C-Y.; Liu, S-X.; Tanner, C.; Leiggener, C.; Neels, A.; Sanguinet, L.; Levillain,
E.; Leutwyler, S.; Hauser, A.; Decurtins, S. Chem. Eur. J. 2007, 13, 3804. (b) Leiggener,
C.; Dupont, N.; Liu, S.-X.; Goze, C.; Decurtins, S.; Beitler, E.; Hauser, A. Chimia 2007,
61, 621. (c) Jia, C-Y.; Liu, S-X.; Tanner, C.; Leiggener, C.; Sanguinet, L.; Levillain, E.;
Leutwyler, S.; Hauser, A.; Decurtins, S. Chem. Commun. 2006, 1878. (d) FuksJanczarek, I.; Luc, J.; Sahraoui, B.; Dumur, F.; Hudhomme, P.; Berdowski, J.; Kityk, I.
V. J. Phys. Chem. B 2005, 109, 10179. (e) Loosli, C.; Jia, C. Y.; Liu, S.-X.; Haas, M.;
Dias, M.; Levillain, E.; Neels, A.; Labat, G.; Hauser, A.; Decurtins, S. J. Org. Chem.
2005, 70, 4988. (f) Dumur, F.; Gautier, N.; Gallego-Planas, N.; Sahin, Y.; Levillain, E.;
Mercier, N.; Hudhomme, P.; Masino, M.; Girlando, A.; Lloveras, V.; Vidal-Gancedo, J.;
Veciana, J.; Rovira, C. J. Org. Chem. 2004, 69, 2164. (g) Segura, J. L.; Martín, N.
Angew. Chem. Int. Ed. 2001, 40, 1372. (h) Gautier, N.; Dumur, F.; Lloveras, V.; VidalGancedo, J.; Veciana, J.; Rovira, C.; Hudhomme, P. Angew. Chem. Int. Ed. 2003, 42,
2765.
(8) (a) Peperichka, D.; Bryce, M. R.; Pearson, C.; Petty, M. C.; McInnes, E. J. L.; Zhao,
J. P. Angew. Chem., Int. Ed. 2003, 42, 4635. (b) Wu, J.-C.; Liu, S.-X.; Neels, A.; Le
Derf, F.; Sallé, M.; Decurtins, S. Tetrahedron 2007, 63, 11282. (c) Diaz, M. C.; Illescas,
B. M.; Martin, N.; Perepichka, I. F.; Bryce, M. R.; Levillain, E.; Viruela, R.; Orti, E. Chem.
Eur. J. 2006, 12, 2709. (d) Bouquin, N.; Malinovskii, V. L.; Guégano, X.; Liu, S.-X.;
Decurtins, S.; Häner, R. Chem. Eur. J. 2008, 14, 5732. (e) Wu, H.; Zhang, D. Q.; Su, L.;
Ohkubo, K.; Zhang, C. X.; Yin, S. W.; Mao, L. Q.; Shuai, Z. G.; Fukuzumi, S.; Zhu, D. B.
J. Am. Chem. Soc. 2007, 129, 6839. (f) Tsiperman, E.; Becker, J. Y.; Khodorkovsky, V.;
Shames, A.; Shapiro, L. Angew. Chem. Int. Ed. 2005, 44, 4015.
(9) (a) Seidel, S. R.; Stang, P. J. Acc. Chem. Res. 2002, 35, 972. (b) Glasson, C. R. K.;
Lindoy, L. F.; Meehan, G, V. Coord. Chem. Rev. 2008, 252, 940. (c) Lindoy, L. F.;
Atkinson, I. M. Self-assembly in Supramolecular Chemistry, Royal Society for
Chemistry, Cambridge, UK, 2000. (d) Scandola, F.; Argazzi, R.; Bignozzi, C. A.;
Chiorboli, C.; Indelli, M. T.; Rampi, M. A. Supramolecular Chemistry; Balzani, V.;
DeCola, L. Eds.; Kluwer Academic Publishers; Dordrecht, 1992. (e) Venturi, M.; Serroni,
!
95
S.; Juris, A.; Campagna, S.; Balzani, V. Top. Curr. Chem. 1998, 197, 193. (f) Harriman,
A.; Ziessel, R. Coord. Chem. Rev. 1998, 171, 331.
(10) (a) Balzani, V.; Scandola, F. Supramolecular Photochemistry, Horwood,
Chichester, 1991. (b) Balzani, V.; Juris, A.; Venturi, M.; Campagna, S.; Serroni, S.
Chem. Rev. 1996, 96, 759. (c) Sun, L.; Hammarstrçm, L.; Tkermark, B.; Styring, S.
Chem. Soc. Rev. 2001, 30, 36. (d) Ballardini, R.; Balzani, V.; Credi, A.; Gandolfi, M. T.;
Venturi, M. Acc. Chem. Res. 2001, 34, 445. (e) Balzani, V.; Gomez-Lopez, M.; Stoddart,
J. F. Acc. Chem. Res. 1998, 31, 405.
(11) (a) Handy, E. S.; Pal, A. J.; Rubner, M. F. J. Am. Chem. Soc. 1999, 121, 3525. (b)
Gao, F. G.; Bard, A. J. J. Am. Chem. Soc. 2000, 122, 7426. (c) Lee, S. J.; Lin, W. J.
Am. Chem. Soc. 2002, 124, 4554. (d) Fujita, M.; Kwon, Y. J.; Washizu, S.; Ogura, K. J.
Am. Chem. Soc. 1994, 116, 1151. (e) van Veggel, F. C. J. M.; Verboom, W.; Reinhoud,
D. N. Chem. Rev. 1994, 94, 279.
(12) (a) Goze, C.; Leiggener, C.; Liu, S-X.; Sanguinet, L.; Levillain, E.; Hauser, A.;
Decurtins, S. ChemPhysChem 2007, 8, 1504. (b) Campagna, S.; Serroni, S.;
Punteriero, F.; Loiseau, F.; De Cola, L.; Kleverlaan, C. J.; Becher, J.; Sorensen, A. P.;
Hascoat, P.; Thorup, N. Chem. Eur. J. 2002, 8, 4461.
(13) (a) Bryce, M. R. Adv. Mater. 1999, 11, 11. (b) Goze, C.; Liu, S-X.; Leiggener, C.;
Sanguinet, L.; Levillain, E.; Hauser, A.; Decurtins, S. Tetrahedron 2008, 64, 1345.
(14) Imor, S.; Morgan, R. J.; Wang, S.; Morgan, O.; Baker, A. D. Synth. Commun. 1996,
26, 2197.
(15) Sprintschnik, G.; Sprintschnik, H. W.; Kirsch, P. P.; Whitten, D. J. Am. Chem. Soc.
1977, 99, 4947.
(16)
Sheldrick, G.M. Acta Cryst. A46, 1990, 467.
(17)
Sheldrick, G.M. SHELXL-97, Program for crystal structure refinement; University
of Göttingen, Göttingen, Germany, 1997.
(18)
Spek, A. L. J. Appl. Cryst. 2003, 36, 7.
(19) (a) Bouguessa, S.; Gouasmia, A. K.; Golhen, S.; Ouahab, L.; Fabre, J. M.
Tetrahedron Lett. 2003, 44, 9275. (b) Liu, S-X.; Dolder, S.; Rusanov, E. B.; Stoeckli!
96
Evans, H.; Decurtins, S. C. R. Acad. Sci. Paris, Chimie 2003, 6, 657. (c) Devic, T.;
Avarvari, N.; Batail, P. Chem. Eur. J. 2004, 10, 3697.
(20) D#Alessandro, D. M.; Junk, P. C.; Keene, F. R. Supramol. Chem. 2005, 17, 529.
(21) (a) Masui, H.; Freda, A. L.; Zerner, M. C.; Lever, A. B. P. Inorg. Chem. 2000, 39,
141. (b) Bardwell, D.; Jeffery, J. C.; Joulie, L.; Ward, M. D. J. Chem. Soc., Dalton Trans.
1993, 2255. (c) Bardwell, D. A.; Horsburgh, L.; Jeffery, J. C.; Joulié, L. F.; Ward, M. D.;
Webster, I.; Yellowlees, L. J. J. Chem. Soc., Dalton Trans. 1996, 2527. (d) Bergman, S.
D.; Goldberg, I.; Barbieri, A.; Barigelletti, F.; Kol, M. Inorg. Chem. 2004, 43, 2355. (e)
Hage, R.; Haasnoot, J. G.; Nieuwenhuis, H. A.; Reedijk, J.; De Ridder, D. J. A.; Vos, J.
G. J. Am. Chem. Soc. 1990, 112, 9245.
(22) D'Alessandro D. M.L; Kelso, L. S; Keene F. R. Inorg. Chem. 2001, 40, 6841.
!
97
3.6. Supporting Information:
Figure S1. Cyclic voltammograms of the ligand L in CH2Cl2/ 0.1 M Bu4NPF6 at
a scan rate of 50 mV(s-1.
Figure S2. Cyclic voltammogram of the ligand L in CH2Cl2/ 0.1 M Bu4NPF6 at
a scan rate of 100 mV(s-1.
!
98
Figure S3. Cyclic voltammograms of the ligand L in CH2Cl2/ 0.1 M Bu4NPF6 at
a scan rate of 500 mV(s-1.
Figure S4. Lippert-Mataga plot for the ligand L.
!
99
4. Dual Luminescence and Long-lived Charge Separated states in
Donor-Acceptor Assemblies based on Tetrathiafulvalene Fused
Ruthenium(II)-Polypyridine Complexes
Published in Chimia, 61 (10): 621-625 2007
Claudia Leiggener,a Nathalie Dupont,a Shi-Xia Liu,b Christine Goze,b Silvio Decurtinsb
and Andreas Hausera*
a
Département de Chimie Physique, Université de Genève, 30 Quai Ernest-Ansermet,
CH-1211 Genève 4, Switzerland
b
Departement für Chemie und Biochemie, Universität Bern, Freiestrasse 3,
CH-3012 Bern, Switzerland
!
100
Abstract
The creation of long-lived charge separated states in donor-acceptor assemblies has
been the goal of many studies aimed at mimicking the primary processes in
photosynthesis. Here we present such assemblies based on tetrathiafulvalene (TTF) as
electron donor and a dipyridophenazine (dppz) unit as electron acceptor in the form of a
fused ligand (TTF-dppz) coordinated to ruthenium(II) via the dipyrido coordination site
and with 2,2'-bipyridine (bpy) as auxiliary ligand, namely [Ru(bpy)3-x(TTF-dppz)x]2+ (x =
1-3). For x = 2, irradiation into the metal to dppz charge transfer transition results in
electron transfer from TTF to ruthenium, thus creating a charge separated state best
described by [(TTF+-dppz)Ru(dppz--TTF)(bpy)]2+ with a lifetime of 2.5 µs in
dichloromethane.
!
101
CONTENT
4.1. Introduction
103
!
4.2. Results and Discussion
106
4.2.1. The free ligands
106
4.2.2. Coordination to innocent transition metal ions
107
4.2.3. Coordination to ruthenium(II)
108
4.2.4. Dual luminescence and long-lived charge-separated states
111
!
4.3. Conclusions
115
!
4.4. References
!
116
102
4.1. Introduction
Molecular photochemistry and photophysics play an important role both in modern
technologies such as in lighting [1], laser applications, electroluminescent materials [2],
sensing devices [3], and solar energy conversion [4], as well as in nature as for instance
in photosynthesis [5]. The complex sequence of events can very often be broken down
into elementary steps. Of course the first step is always given by the absorption of a
photon, which promotes the chromophore to an electronically and often vibrationally
excited state. This initially excited state is most often extremely short-lived, and the
system decays via vibrational relaxation, internal conversion, intersystem crossing,
luminescence, and energy and electron transfer back to the ground state or via some
chemical reaction to a given photochemical product.
A key issue of photophysics and photochemistry is the creation of long-lived charge
separated states using so-called donor-acceptor (DA) assemblies, in which the
excitation of either the donor to D*A or the acceptor to DA* is followed by the transfer of
an electron from D to A to form a state best described by D+A-. As an extension of the
simple dyad, the donor and the acceptor may be linked by a photophysically active
bridge affording the triad DBA. In this case, the excitation of the bridge to DB*A results
in double electron transfer, namely from the HOMO of D to the SOMO-1 of B* and from
the SOMO of B* to the LUMO of A, thus forming a species of the form D+BA-, which in
general has a longer lifetime of the charge separated state than that in the
corresponding dyad due to the weaker electronic coupling between D and A. Such
triads, composed of tetrathiafulvalene, porphyrins and fullerenes have been
successfully used in the creation of artificial photosynthetic systems [6].
Tertrathiafulvalene derivatives, on the one hand, constitute a class of versatile electron
donors for a number of interesting applications [7], in fields such as molecular
electronics [8] and in organic conductors and superconductors [9]. Of particular interest
in the context of this paper is that they are known to quench the luminescence of almost
any chromophore in their vicinity through reductive electron transfer quenching [10].
Ruthenium(II) polypyridine complexes, on the other hand, are known not only for the
luminescence from metal-to-ligand-charge-transfer (MLCT) states but also for their use
as sensitisers for light-induced electron transfer in solar energy conversion [11]. Usually
the incorporation of tetrathiafulvalene derivatives into ruthenium(II) complexes results in
the above mentioned intramolecular electron transfer quenching of the ruthenium
!
103
complex based luminescence [12]. However, in the systems investigated to date, the
resulting
charge
separated
states
were
not
very
long-lived,
because
the
tetrathiafulvalene was linked flexibly to the ruthenium(II) chromophore, and thus the
average spatial separation between the units is too small.
In an attempt to sequentially assemble a rigidly bridged donor acceptor triad combining
a tetrathiafulvalene derivative with a ruthenium(II) polypyridile complex, we first
synthesised the two ligands shown in Figure 1a, where 4#,5#-bis-(propylthio)-substituted
tetrathiafulvalene (TTF) is fused to two different dipyridophenazine units, namely
dipyrido[3,2-a:2#,3#-c]phenazine (dppz) [13] and dipyrido[2,3-a:3#,2#-c]phenazine (dppz').
With their bidentate diimine binding sites the two ligands can easily be coordinated to a
number of transition metal ions. Of particular interest within the context of this special
issue of Chimia is coordination to spectroscopically and photophysically active
transition-metal ions, as for example ruthenium(II). Figure 1b depicts the series of
mononuclear complexes with 1, 2 or 3 TTF-dppz ligands, namely [Ru(bpy)3-x(TTFdppz)x]2+ (bpy = 2,2'-bipyridine, x = 1-3)
[{Ru(bpy)2}2 ("-TTF-dppz')]4+.
!
104
[14] as well as the dinuclear complex
!
!
Figure 1: The ligands TTF-dppz and TTF-dppz'. The mononuclear complexes
[Ru(bpy)2(TTF-dppz)]2+
and
[Ru(TTF-dppz)3]2+,
and
the
dinuclear
complex
[{Ru(bpy)2}2(µ-TTF-dppz')]4+ .
In this paper, we shall begin our discussion on the spectroscopic and photophysical
properties of the free ligands, which are quite unique and interesting in their own right.
We shall then proceed to examine the effect of their coordination to innocent zinc(II)
ions before turning to the main subject of our contribution, namely the discussion of the
luminescence properties of the ruthenium complexes and the creation of long-lived
charge separated states, in which the TTF unit acts as electron donor.
!
105
4.2. Results and Discussion
4.2.1. The free ligands
!
Figure 2 shows the absorption spectra of the two ligands, TTF-dppz and TTF-dppz', in
CH2Cl2. For reference the absorption spectra of dppz and free TTF are included. The
spectra of the two fused ligands are very similar to each other, but they are not at all
equal to the sum of the two reference spectra. Indeed, there is an additional band
centred at 18500 cm-1 with an oscillator strength f % 0.17. Even though neutral TTF is
known to quench all luminescence by electron transfer, both compounds emit quite
strongly in the near infrared (see Figure 2). The exact position of the emission band and
the quantum yield depend on the solvent, ranging from 11600 cm-1 and 0.13%,
respectively for DMF to 16100 cm-1 and 7.6%, respectively, for cyclohexane. The
corresponding values in CH2Cl2 are 12200 cm-1 and 1%, which together with the
experimental luminescence lifetime of 0.4 ns in the same solvent give a radiative lifetime
of the emitting state of ~40 ns. Thus the emission can be clearly identified as the Stokes
shifted fluorescence of the corresponding absorption band at 18500 cm-1.
Figure 2: Absorption and emission spectra of TTF-dppz (___) and TTF-dppz' (---) in
CH2Cl2 as well as the corresponding spectra of the reference compounds dppz (....)
and TTF (- - -). TTF shows no emission.
!
106
The Lippert-Mataga plot of the variation of the Stokes shift as a function of the dielectric
constant of the solvent gives a change in dipole moment from the ground state to the
excited state of ~18 Debye. Even though the Lippart-Mataga treatment tends to
overestimate the changes in dipole moment associated with a spectroscopic transition,
this is a clear indication that the said transition is an intraligand charge transfer (ILCT)
transition corresponding essentially to a transfer of charge from the HOMO centred on
the TTF moiety to the LUMO centred on the dppz unit. This assignment is corroborated
by computational results based on density functional theory. Thus, in the fused donoracceptor compounds TTF-dppz and TTF-dppz', the ILCT luminescence is not quenched
by TTF because TTF actively participates in the lowest energy spectroscopic transition,
the lowest energy singlet state being best described by TTF)+-dppz)-. Electrochemical
data of TTF-dppz support this conclusion, as the difference in redox potentials of 2.2 V
between the first oxidation of TTF and the first reduction of dppz is very close to the
zero-point energy of the ILCT state of ~16000 cm-1 as estimated from the crossing point
of the absorption and the emission spectrum. For more details see Ref. [13].
4.2.2. Coordination to innocent transition metal ions
!
Both TTF-dppz and TTF-dppz' can be coordinated to metal ions. This has an immediate
impact on the position of the ILCT band. The coordination to positively charged metal
ions increases the acceptor properties of the dppz and dppz' moieties and consequently
the ILCT bands shift to lower energies, as borne out by the absorption spectra shown in
Figure 3. Upon addition of Zn2+ to a solution of TTF-dppz, the shift is ~2000 cm-1. The
full shift is reached upon addition of little more than one third mole equivalents of Zn2+,
indicating that initially more than one TTF-dppz ligand is coordinated to one Zn2+ ion.
This, in turn, indicates that the shift is essentially due to electrostatic interactions, and,
indeed, addition of a weak acid to a solution of TTF-dppz has exactly the same effect.
For TTF-dppz' with potentially two coordination sites the shift occurs in two distinct
steps. Addition of about one half of an equivalent again results in a shift this time of
~3000 cm-1. Further addition of Zn2+ enhances the effect and results in a further red
shift of ~2000 cm-1.
!
107
In addition, the mononuclear complex with TTF-dppz also shows the ILCT
luminescence, red-shifted by the same 2000 cm-1 as the absorption and with a slightly
smaller quantum yield. The reduction of the quantum yield is, however, not due to an
active quenching by Zn2+, but merely due to the expected increase in the non-radiative
multiphonon relaxation according to the energy gap law.
Figure 3: The effect on absorption and emission spectra of TTF-dppz and TTFdppz' in CH2Cl2 upon addition of Zn2+ to the solution: top TTF-ddpz' (___), TTF.dppz'
+ 0.5 equiv. Zn2+ (....) , TTF- dppz' + 1.0 equiv. Zn2+ (---); bottom TTF-dppz (___),
TTF-dppz + 0.5 equiv. Zn2+(....) ; absorption left, emission right.
4.2.3. Coordination to ruthenium(II)
!
Figure 4a shows the absorption and emission spectra of [Ru(bpy)2(TTF-dppz)]2+ in
direct comparison with those of the reference complexes [Ru(bpy)2(dppz)]2+ and
!
108
[Ru(bpy)3]2+. The references complexes both have intense absorption bands at 35000
and at 22000 cm-1, which have been attributed to !!* transitions on the ligands and to
the spin-allowed metal-ligand-charge transfer (1MLCT) transition from the metal to bpy
and dppz, respectively. Both show luminescence centred at 16000 cm-1 originating from
the corresponding 3MLCT states. The absorption spectrum of [Ru(bpy)2(TTF-dppz)]2+
shows additional features in the form of a shoulder at 30000 and a strong band at
16000 cm-1, which can be attributed to a !!* transition and the red-shifted ILCT
transition of TTF-dppz.
Coordination of more than one TTF-dppz ligand to one ruthenium(II) ion in the series
[Ru(bpy)3-x(TTF-dppz)x]2+, x = 1 - 3, results in a linear increase of the molar absorption
of the TTF-dppz ILCT band with respect to the concentration of the complex, as shown
in Figure 4b. For x = 3, the low-energy shoulder on the ILCT band indicates a nonnegligible exciton splitting.
Whereas the complex with one TTF-dppz ligand, [Ru(bpy)2(TTF-dppz)]2+, shows
unusual dual luminescence behaviour both from the ILCT state and the 3MLCT state,
the complexes with more than one TTF-dppz ligand show only the weak luminescence
of the ILCT state. This is to be discussed in more detail below.
!
109
Figure 4: a) Absorption and emission spectra of [Ru(bpy)2(TTF-dppz)]2+,
[Ru(bpy)2(dppz)]2+ and [Ru(bpy)3]2+ in degassed CH2Cl2, b) absorption spectra of
[Ru(bpy)2(TTF-dppz)]2+, [Ru(bpy)(TTF-dppz)2]2+ and [Ru(TTF-dppz)3]2+ in CH2Cl2.
The absorption spectrum of the dinuclear complex [{Ru(bpy)2}2("-TTF-dppz')]4+
displayed in Figure 5 together with the absorption spectra of [{Ru(bpy)2}2("-TTFdppz')]4+, [Ru(bpy)2(dppz)]2+ and [Ru(bpy)2(TTF-dppz)]2+ for direct comparison, show
the ILCT band now shifted even further into the red with the maximum at 12000 cm-1. In
contrast to the mononuclear complex with a head-on coordination for which the
Ru$bpy and Ru$dppz 1MLCT transitions are very close in energy, the dinuclear
!
110
complex with a side-on coordination to dppz' displays two distinct MLCT bands, namely
at 15000 cm-1 corresponding to Ru$dppz' CT, and at 23500 cm-1 corresponding to
Ru$bpy CT. The dinuclear complex shows no luminescence in the visible, and if there
is any luminescence of the ILCT state in the NIR, its intensity is below the sensitivity of
the available experimental set-up.
Figure 5: Absorption spectra of [{Ru(bpy)2}2(dppz')]4+ and [{Ru(bpy)2}2(µ-TTFdppz')]4+ in CH2Cl2. For direct comparison the spectra of [Ru(bpy)2(dppz)]2+ and
[Ru(bpy)2(TTF-dppz)]2+ from Figure 4 are included.
4.2.4. Dual luminescence and long-lived charge-separated states
!
Similar to the coordination with Zn2+, [Ru(bpy)2(TTF-dppz)]2+ shows luminescence from
the lowest energy ILCT state with a maximum at ~10000 cm-1 upon irradiation at 16000
cm-1, that is selectively into the ILCT absorption. Most interestingly, irradiation into the
1
MLCT band at 22000 cm-1 results in luminescence, which can be identified as 3MLCT
luminescence from its position at 15300 cm-1 and the associated lifetime of 1040 ns
!
111
(see Figure 4). On the other hand, the two complexes with more than one TTF-dppz
ligand per metal ion only show the weak luminescence of the ILCT state. The behaviour
of the complexes with more than one TTF-dppz ligand is as expected, in so far as TTF
quenches the
3
MLCT luminescence of the ruthenium(II) based chromophore via
reductive electron transfer. So the key questions is, why is the 3MLCT luminescence not
likewise quenched by reductive electron transfer from the TTF unit in the case of the
complex with only one TTF-dppz ligand? As mentioned above, in [Ru(bpy)2(TTFdppz)]2+ the Ru$bpy CT and the Ru$dppz CT states are quite close in energy, so
irradiation at 22000 cm-1 excites both states simultaneously. However, it is generally
acknowledged that the MLCT state with the electron on dppz is lower in energy [15].
Therefore, as depicted in Figure 6a, the lowest energy 3MLCT state is best described as
[Ru3+(bpy)2(TTF-dppz-)]2+. Electron transfer quenching, by the TTF unit would involve
transfer of an electron from the TTF unit to the Ru3+, forming the ILCT state
[Ru2+(bpy)2(TTF+-dppz-)]2+. The driving force for this process is quite substantial, but in
order for it to occur, the transferred electron would have to tunnel through the Coulomb
barrier of the electron in the !* orbital localised on the dppz unit. This slows down the
electron transfer process sufficiently for the intersystem crossing process from the
initially excited 1MLCT state to the 3MLCT followed by 3MLCT luminescence becomes
competitive. In the complexes with more than one TTF-dppz ligand the situation is
different, as shown for [Ru(TTF-dppz)3]2+ in Figure 6b. The initially excited CT state may
be described by [Ru3+(TTF-dppz-)(TTF-dppz)2]2+. In addition to intersystem crossing,
electron transfer from a TTF unit other than the one for which the corresponding dppz
unit is negatively charged is now competitive, resulting in a ligand-to-ligand charge
separated (LLCS) state of the form [Ru2+(TTF-dppz-)(TTF+-dppz)(TTF-dppz)]2+. This is
borne out by the transient absorption spectrum of [Ru(TTF-dppz)3]2+ upon pulsed
excitation at 22000 cm-1 shown in Figure 7 together with the ground state absorption
spectrum and the spectrum of the chemically oxidised species [Ru2+(TTF-dppz)2(TTF+dppz)]3+. The absorption spectrum of the oxidised species shows a reduced absorption
at the energy of the ILCT transition and a strong new absorption band at 12000 cm-1.
The former indicates a loss in the intensity of the TTF$dppz ILCT transition, the latter
can be attributed to a dppz$TTF+ ILCT transition on the oxidised species [13], and thus
constitutes a signature for the presence of a TTF+-dppz unit. Similarly, the transient
difference spectrum shows a bleaching of the ILCT transition and an increased
!
112
absorption characteristic for the TTF+-dppz unit, thus supporting the hypothesis of a fast
formation of the LLCS state as a transient state. In fact, the transient spectra of all three
complexes show these features, and thus in all three complexes such a LLCS state with
a lifetime of 2.5 µs is formed. As a result, the full schemes for the processes in all three
complexes shown in Figures 6a and b can be established, the one for [Ru(bpy)(TTFdppz)2]2+ being a combination of the ones for [Ru(bpy)2(TTF-dppz)]2+ and [Ru(TTFdppz)3]2+. Of course, for [Ru(bpy)2(TTF-dppz)]2+ the charge separated state is
preferentially formed upon excitation into the Ru$bpy 1MLCT absorption, and the
electron is localised on one of the bpy ligands.
!
113
Figure 6: Scheme of photophysical processes in a) [Ru(bpy)2(TTF-dppz)]2+ and b)
[Ru(TTF-dppz)3]2+. L stands for the neutral TTF-dppz - adapted from Reference [14].
!
114
Figure 7: Absorption spectrum of [Ru(TTF-dppz)3]2+ in CH2Cl2 (----), the chemically
oxidised form (....), and the transient difference absorption spectrum (___) upon
pulsed irradiation at 16000 cm-1. Inset: decay of the transient state monitored at
12500 cm-1.
4.3. Conclusions
The creation of functional multi-component assemblies is one of the important goals of
modern chemistry, and systems with long-lived charge separated states at
comparatively high energies are of particular interest for the discussion on solar energy
conversion based on mimicking natural photosynthesis. In the rigid donor-acceptor
assembly with a central ruthenium(II) ion coordinated by TTF-dppz, the TTF unit serves
as electron donor to ruthenium with a formal charge of +3 upon irradiation into the
MLCT bands. In the resulting charge separated state, both the auxiliary bipyridine
ligand in the system with only one TTF-dppz ligand as well as the dppz unit of the TTFdppz ligand itself in the system with more than one such ligand may serve as electron
acceptors. The [Ru(bpy)2(TTF-dppz)]2+ complex, in addition, shows an unusual dual
luminescence due to the fact that in the lowest energy MLCT state, the electron in the
!* orbital of the dppz unit serves as Coulomb barrier for the electron transfer from TTF
!
115
to ruthenium. The processes leading to the formation of the charge separated states are
very fast indeed and will require ultra-fast spectroscopy to elucidate. In order to
increase the lifetime of the charge separated states, the auxiliary ligand is to be
replaced in a further step by a better electron acceptor, for instance with a quinone
derivative as terminal acceptor.
Acknowledgements
We thank the Swiss National Science Foundation and the European Union FP6 Network of Excellence MAGMANet for financial support.
4.4. References
!
[1]
Th. Jüstel, H. Nikol, C. Ronda, Angew. Chem. Int. Ed. 1998, 37, 3084.
[2]
a) H. Xia, C. Zhang, S. Qui, P. Lu, J. Zhang, Y. Ma, Appl. Phys. Lett. 2004, 84,
290, b) H. Xia, C. Zhang, S. Qui, P. Lu, F. Shen, J. Zhang, Y. Ma, J. Phys. Chem
B 2004,108, 3185.
[3]
a) M. O. Wolf, C. W. Rogers, Coord. Chem. Rev. 2002, 233, 341, b) M. Bendikov.
F. Wudl, D. F. Perepichka, Chem. Rev. 2004, 104, 4891.
[4]
a) M. Grätzel, Inorg. Chem. 2005, 44, 6841, b) D. Gust, T. A. Moore, A. L. Moore,
Acc. Chem. Res. 2001, 34, 40, c) M. Grätzel, Nature 2001, 414, 338.
[5]
a) E. Krausz, L. J. Hughes, P. Smith, R. Pace, S. Peterson, Photochem. and
Photobiol. Sci. 2005, 4, 744, b) L. M. Yoder, A. G. Cole, R. J. Sension, Photosynt.
Res. 2002, 72, 147.
[6]
a) M. Di Valentin, A. Bisol, G. Agostini, P. A. Liddell, G Kodis, A. L. Moore, T. A.
Moore, D. Gust, D. Carbonera, J. Phys. Chem. B 2005, 109, 14401, b) H. Imahori,
D. M. Guldi, K. Tamaki, Y. Yoshida, C. Luo, Y. Sakata, S. Fukuzumi, J. Amer.
Chem. Soc. 2001, 123, 6617, c) M. Grätzel, J.-E. Moser, Electron Transfer in
Chemistry 5, p. 589 (editor V. Balzani) Wiley-VCH, Weinheim 2001, d) G. Kodis,
P. A. Liddel, L. de la Garza, A. L. Moore, T. A. Moore, D. Gust, J. Mat. Chem
2002, 12, 2100.
!
116
[7]
J. Yamada, T. Sugimoto, TTF Chemistry: Fundamentals and applications of
Tetrathiafulvalene, Springer, Berlin, 2004.
[8]
R. L. Carroll, C. B. Gorman, Angew. Chem. 2002, 114, 4556; Angew. Chem. Int.
Ed. 2002, 41, 4378.
[9]
a) T. Otsubo, K. Takimiya, Bull. Chem. Soc. Jpn. 2004, 77, 43, b) U. Geiser, A. M.
Kini, J. A. Schlueter, H. H. Wang, J. M. Williams, Mol. Cryst. Liq. Cryst 2002, 380,
29, c) R. Samina, S. S. Turner, P. Day, J. A. K. Howard, P. Guionneau, E. J. L.
McInnes, F. E. Mabbs, R. J. H. Clark, S. Firth, T. Biggs. J. Mat. Chem. 2001, 11,
2095, d) P. Cassoux, Coord. Chem. Rev. 1999, 185-186, 213, d) L. Martin, S. S.
Turner, P. Day, P. Guionneau, J. A. K. Howard, D. E. Hibbs, M. E. Light, M. B.
Hursthouse, M. Uruichi, K. Yakushi, Inorg. Chem. 2001, 40, 1363, e) J. A.
Schlueter, U. Geiser, A. M. Kini, H. H. Wang, J. M. Williams, D. Naumann, T. Roy,
B. Hoge, R. Eujen, Coord. Chem. Rev. 1999, 190-192, 781.
[10] a) C. Loosli, C. Jia, S.-X. Liu, M. Haas, M. Dias, E. Levillain, A. Neels, G. Labat, A.
Hauser, S. Decurtins, J. Org. Chem. 2005, 70, 4988, b) M. A. Herranz, N. Martin,
L. Sanchez, C. Seoane, D. M. Guldi, J. Organometallic Chem. 2000, 599, 2, c)
[11] a) G. J. Meyer, Inorg. Chem. 2005, 44, 6852, b) M. K. Nazeeruddin, S. M.
Zekeeruddin, J.-J. Lagref, P. Liska, P. Comte, C. Barolo, G. Viscardi, K. Schenk,
M. Grätzel, Coord. Chem. Rev. 2004, 248, 1317.
[12] a) S. Campagna, S. Serroni, F. Puntoriero, F. Loiseau, L. De Cola, C. J.
Cleverlaan, J. Becher, A. P. Sorensen, P. Hascoat, N. Thorup, Chem. Eur. J.
2002, 8, 4461, c) F. Voegtle, M. Plevoets, M. Nieger, G. C. Azzellini, A. Credi, L.
De Cola, V. De Marchis, M. Venturi, V. Balzani, J. Am. Chem. Soc. 1999, 121, 26.
[13] C. Jia, S.-X. Liu, C. Tanner, C. Leiggener, L. Sanguinet, E. Levillain, S. Leutwyler,
A. Hauser, S. Decurtins, Chem. Eur. J. 2007, 13, 3804.
[14] C. Goze, C. Leiggener, S.-X. Liu, L. Sanguinet, E. Levillain, A. Hauser, S.
Decurtins, ChemPhysChem in press.
[15] R. M. Hartshorn, J. K. Barton, J. Am. Chem. Soc. 1992, 114, 5919.
!
117
5. Effect of the Addition of a Fused Donor-Acceptor Ligand on a RuII
Complex: Synthesis Characterization and Photo-induced Electron
Transfer Reactions of [Ru(TTF-dppz)2(Aqphen)]2+
Submitted in Inorganic Chemistry, 2010
!
!
Nathalie Dupont,† Ying-Fen Ran,‡ Hong-Peng Jia,‡ Jakob Grilj,† Shi-Xia Liu,*,‡ Silvio
Decurtins,‡ Andreas Hauser*,†
†
Département de Chimie Physique, Université de Genève, 30 Quai Ernest-Ansermet,
CH-1211 Genève 4, Switzerland
‡
Departement für Chemie und Biochemie, Universität Bern, Freiestrasse 3,
CH-3012 Bern, Switzerland.
!
118
Abstract
The
synthesis
and
the
photophysical
properties
of
the
complex
[Ru(TTF-
dppz)2(Aqphen)]2+ (TTF = tetrathiafulvalene, dppz = dipyrido-[3,2-a:2#,3#-c]phenazine,
Aqphen = anthraquinone fused to phenanthroline via a phenazine bridge) are described.
In this molecular triad excitation into the metal-ligand charge transfer bands results in
the creation of a long-lived charge separated state with TTF acting as electron donor
and anthraquinone as terminal acceptor. The lifetime of the charge-separated state is
400 ns in dichloromethane at room temperature. A mechanism for the charge
separation involving an intermediate charge-separated state is proposed based on
transient absorption spectroscopy.
!
119
CONTENT
5.1. Introduction
121
!
5.2. Experimental Methods
123
5.2.1. General
123
5.2.2. Synthesis
123
5.2.3. Physical Methods
124
!
5.3. Results and Discussion
126
5.3.1. Synthesis and Characterization
126
5.3.2. Electrochemistry
126
5.3.3. Photophysical properties
128
!
5.4. Conclusions
137
!
5.5. References
!
138
120
5.1. Introduction
!
There has been a considerable amount of research on the photophysical properties of
metal complexes, in particular of ruthenium(II) complexes, containing bidentate ligands
such as 2-2#-bipyridine (bpy) or 1,10-phenanthroline (phen) and their derivatives.1-7 The
interest in this type of complexes relies on their stability and use in multiple applications,
as for instance in solar energy conversion8-11 and molecular electronic devices.12-14 A
common feature of these complexes is their ability to absorb most of the visible light and
thereby to access excited states by means of metal-to-ligand charge-transfer (MLCT)
transitions. The well-known derivative of phen, dipyrido-[3,2-a:2#,3#-c]phenazine (dppz)
has two low-lying !* acceptor levels, one localized on the phenazine subunit and one on
the phenanthroline subunit of the molecule.15 It is generally acknowledged that in the
mixed ligand complex [Ru(bpy)2(dppz)]2+, the Ru(II)$bpy and Ru(II)$dppz 1MLCT
states are close to each other in energy, but the lowest emissive 3MLCT state in
solution is the Ru(II)$dppz state.16
Tetrathiafulavelene (TTF) can be fused to dppz to result in a TTF-dppz ligand (Scheme
1) with two redox centers and a lowest excited state corresponding to a TTF$dppz
intraligand charge-transfer (ILCT) state. This ligand was studied on its own17, as its
ruthenium(II) complexes, including the series [Ru(bpy)1-x(TTF-dppz)x]2+, x = 1 - 3.18
These complexes show interesting photophysical properties in the form of a nonemissive, long-lived charge-separated state corresponding to an electron transfer from
TTF to dppz, with the TTF and dppz moieties belonging to two different subunits,
resulting from the initial excitation into the 1MLCT bands.
A famous electron acceptor is provided by quinones, widely studied because of their
very important role in nature, as for instance in the photosystems of green plants.
Anthraquinone can be fused to phenanthroline via a phenazine bridge in order to form
the Aqphen ligand shown in Scheme 1.19, 20
The combination of two TTF-dppz ligands and one Aqphen ligand to form a
ruthenium(II) complex gives rise to the title complex [Ru(TTF-dppz)2(Aqphen)]2+ (1) also
shown in Scheme 1. Its synthesis has been stimulated by the search for new antenna
systems capable of charge separation21 as well as for new photoredox switches.22 So
!
121
far only a few papers on Ru(II) coordination complexes with TTF derivatives have
appeared in the literature.18, 23-30 In most cases, the famous luminescence from 3MLCT
states is strongly quenched by intramolecular electron transfer quenching from the
pendant TTF unit. This makes them promising candidates for incorporation into optical
sensors and devices. Herein we report the synthesis of Ru(II) complex 1 and thoroughly
examine its electrochemical and photophysical properties. A direct comparison to the
reference complex [Ru(bpy)(TTF-dppz)2]2+ indicates a new relaxation pathway in 1.
Scheme 1. a) TTF-dppz, b) Aqphen, c) [Ru(TTF-dppz)2 (Aqphen)]2+ (1).
!
122
5.2. Experimental Methods
5.2.1. General
!
Unless otherwise stated, all reagents were purchased from commercial sources and
used without additional purification. 5,6-Diamino-2-(4,5-bis(propylthio)-1,3-dithio-2ylidene)-benzo[d]-1,3-dithiole
(TTF-diamine),17
12,17-dihydronaphtho-[2,3-
h]dipyrido[3,2-a:2#,3#-c]-phenazine-12,17-dione (Aqphen)19 and [Ru(phendione)2Cl2]31
were prepared according to literature procedures. 1H NMR spectra were obtained on a
Bruker AC 300 spectrometer operating at 300.18 MHz: chemical shifts are reported in
ppm referenced to residual solvent protons (CD3CN and DMSO-d6). The following
abbreviations were used: s (singlet), d (doublet), t (triplet), m (multiplet) and br (broad).
Infrared spectra were recorded on a Perkin-Elmer Spectrum One FT-IR spectrometer
using KBr pellets. High-resolution mass spectra were recorded using a LTQ Orbitrap XL
for ESI.
5.2.2. Synthesis
5.2.2.1.
Synthesis of [Ru(phendion)2(Aqphen)](PF6)2.
!
In a Schlenk flask (25 mL), a solution of [Ru(phendione)2Cl2] (208 mg, 0.35 mmol) and
the ligand Aqphen (160 mg, 0.38 mmol) in ethylene glycol (16 mL) was heated at 195°C
for 14 h under Argon. After cooling down to room temperature, water (32 mL) was
added into the resulting solution and a black precipitate was immediately formed. After
filtration, an excess of NH4PF6 (200 mg) was added to the filtrate. The resultant dark
brown precipitate was isolated by suction filtration, washed with water and dried in
vacuum. The crude product was dissolved in CH3CN (15 mL) and purified by
precipitation with the addition of diethyl ether. After filtration, the analytically pure
product was obtained as a black powder. Yield: 221 mg (51%). 1H NMR (CD3CN): ) =
7.55 (d, J = 6.0 Hz, 2H), 7.75 (m, 3H), 7.91-8.04 (m, 4H), 8.20-8.28 (m, 4H), 8.39 (br,
2H), 8.54 (br, 2H), 8.62 (d, J = 7.5 Hz, 2H), 8.75 (d, J = 8.4 Hz, 2H), 9.67 (d, J = 7.5 Hz,
!
123
2H), 9.76 (d, J = 7.8 Hz, 1H) ppm. IR (KBr): 4 1703, 1669, 1627, 1400, 1384, 1290, 843
cm-1. ESI-MS: m/z 1079.0556, calcd. for [M-PF6-]+: 1079.0493; 935.0983, calcd. for
[M+H-2PF6-]+: 935.0941; 468.0532, calcd. for [M+2H-2PF6-]2+: 468.0509.
5.2.2.2.
Synthesis of [Ru(TTF-dppz)2(Aqphen)](PF6)2.
In a Schlenk flask (25 mL), a mixture of [Ru(phendion)2(Aqphen)](PF6)2 (123 mg, 0.10
mmol) and 5,6-diamino-2-(4,5-bis(propylthio)-1,3-dithio-2-ylidene)-benzo[d]-1,3-dithiole
(87 mg, 0.20 mmol) in DMF (4 mL) was heated at 150°C for 5 h under Argon. After
cooling down to room temperature, water (5 mL) was added into the black solution. The
resultant black powder was filtered off, washed with water and dried in vacuum. The
crude product was dissolved in CH3CN (20 mL) and purified by chromatography on
basic Al2O3 with CH2Cl2/MeOH (10:1) as eluent to afford the analytically pure product as
a black powder. Yield: 23 mg (11%). 1H NMR (DMSO-d6): ) = 0.94 (t, J = 5.4 Hz, 12H),
1.62-1.66 (m, 8H), 2.91 (t, J = 5.4 Hz, 8H), 7.92-8.04 (m, 6H), 8.07 (d, J = 4.2 Hz, 1H),
8.09 (d, J = 4.2 Hz, 1H), 8.29 (d, J = 5.7 Hz, 2H), 8.34-8.37 (m, 4H), 8.39 (d, J = 4.5
Hz, 2H), 8.58 (s, 4H), 8.86 (dd, J = 6.6 Hz, J = 4.2 Hz, 2H), 9.55 (d, J = 6.3 Hz, 4H),
9.67 (d, J = 6.0 Hz, 1H), 9.76 (d, J = 6.0 Hz, 1H) ppm. IR (KBr): 4 2953, 2923, 1664,
1630, 1401, 1384, 1356, 1088, 841 cm-1. ESI-HRMS: m/z 863.0177; calcd. for [M-2PF6]2+: 863.0165. Anal. Calcd (%) for C82H56F12N12O2P2RuS12 · 5H2O: C, 46.74; H, 3.16; N,
7.98. Found: C, 46.67; H, 2.96; N, 7.59.
5.2.3. Physical Methods
Cyclic voltammetry was conducted on a VA-Stand 663 electrochemical analyzer. An
Ag/AgCl electrode containing 2 M LiCl served as reference electrode, a glassy carbon
electrode as counter electrode, and a Pt disk as working electrode. Cyclic voltammetric
!
124
measurements were performed at room temperature under N2 in CH2Cl2 with 0.1 M
Bu4NPF6 as supporting electrolyte at a scan rate of 100 mV(s-1.
Photophysical measurements were performed on solutions of the compounds in CH3CN
and CH2Cl2 at room temperature. For luminescence and transient absorption
measurements, the solutions were degassed by bubbling N2 through them for 30 min.
Absorption spectra were recorded on a Varian Cary 5000 UV/vis/near-IR (NIR)
spectrophotometer. Excited state decay curves on the nanosecond timescale were
recorded by exciting the samples at 458 nm using the third harmonic of a pulsed
Nd:YAG laser (Quantel Brilliant) to pump an OPO (Opotek Magic Prism). The probe
light came from a W-halogen lamp. The system used for detection consisted of a single
monochromator (Spex 270 M), a photomultiplier (Hamamatsu R928), and a digital
oscilloscope (Tektronix TDS 540B). The overall time resolution is ~15 ns.
Corresponding transient absorption spectra were recorded with a gated CCD camera
(Andor iStar 720).
The femtosecond transient absorption set-up has been described elsewhere.32 Briefly,
the output of a Ti:sapphire amplifier (Spitfire, Spectra Physics; 800 nm pulses of 150 fs
FHWM) is split into two parts; ca. 5 µJ are focused into a 3 mm thick, constantly moving
CaF2 window to generate a white light continuum for probing, the remainder is sent into
a home-built two stage non-collinear optical parametric amplifier to generate the pump
pulses at 400 nm or at 650 nm. The polarization of the pump beam is set to magic angle
with respect to the probe beam. The pump power at the sample is approximately 0.1
mJ/cm2. The sample solution in a 1 mm quartz cell is constantly stirred by nitrogen
bubbling. The photochemical stability of the sample is verified by the steady state
absorption spectrum. The probe beam is dispersed in a spectrograph (Andor, SLR163)
and imaged onto a 512+58 pixel back-thinned CCD (Hamamatsu S07030-09). The
spectra are corrected for the chirp of the white light pulses by standard procedures.33
The IRF (as deduced from the electronic OKE signal) has a FHWM of approximately
200 fs, depending on the wavelength. Due to cross-phase-modulation and the coherent
signal, the spectra at early time-delays cannot be observed.
!
125
5.3. Results and Discussion
5.3.1. Synthesis and Characterization
!
To circumvent the solubility problems that are often encountered when incorporating
rigid and planar diimine ligands into extended mononuclear and polynuclear Ru(II)
systems, a synthetic approach to the target Ru(II) complex 1 involves direct
condensation of a TTF-diamine ligand with [Ru(phendione)2(Aqphen)]2+ (Scheme 2).
The latter was obtained in reasonable yield by a reaction of [Ru(phendione)2Cl2] with 1.1
equivalents of Aqphen in ethylene glycol at reflux. Both new Ru(II) compounds were
purified by recrystallization or by chromatographic separation on basic Al2O3.
Spectroscopic characterization (1H NMR, IR and ESI-HRMS data in the Experimental
Section) confirmed the formation of the desired complexes.
Scheme 2. Synthesis of the target complex 1.
5.3.2. Electrochemistry
!
The electrochemical properties of 1 were investigated by cyclic voltammetry. The
electrochemical data are collected in Table 1 together with those of [Ru(bpy)2(TTFdppz)]2+, [Ru(bpy)(TTF-dppz)2]2+, [Ru(TTF-dppz)3]2+, [Ru(bpy)2(Aqphen)]2+ and TTFdppz and Aqphen, for comparison. Complex 1 undergoes two reversible multi-electron
oxidation processes for the oxidation of the two TTF fragments. Besides, one
!
126
irreversible oxidation wave, corresponding to the Ru(II/III) redox couple, was observed.
Compared to the free ligand TTF-dppz, the observed redox potentials of 1 for the TTF
oxidation processes remain almost unchanged, indicating that the electrostatic inductive
effect of the Ru(II) ion bound to the imine-chelating units seems to have a negligible
influence on the redox potentials of the TTF moieties.
In the cathodic region, two irreversible reduction waves at -0.75 and -1.53 V were
observed for 1, which can be assigned to reductions of the quinone moiety of Aqphen
and
dppz
units,
respectively.
In
contrast
to
the
reference
compound
[Ru(bpy)2(Aqphen)]2+, 1 undergoes one two-electron reduction process to generate the
hydroquinone form [Ru(bpy)2(HAqphen)]2+. It can be therefore deduced that the
negative shift in the second ligand-centered reduction process is attributable to the
presence of the reduced quinone unit in the molecule. Apparently, the lowest
unoccupied molecular orbital (LUMO) in 1 resides on the quinone unit of Aqphen.
Table 1. Redox Potentials (V vs Ag/AgCl) of 1 in CH2Cl2 and of the reference
compounds
TTF-dppz,17
Aqphen15,
[Ru(bpy)(TTF-dppz)2]2+,18
[Ru(bpy)2(TTF-
dppz)]2+,16 [Ru(TTF-dppz)3]2+,18 and [Ru(bpy)2(Aqphen)]2+.15
reduction
oxidation
Compound
E1/21
E1/22
E1/23
E1/21
E1/22
1
0.73
1.04
1.42a
-0.75b
-1.53a
TTF-dppz
0.73
1.08
-1.17
Aqphenc
!
-0.76
[Ru(bpy)2(TTF-dppz)]2+
0.74
1.05
1.43
-0.91
-1.35
[Ru(bpy)(TTF-dppz)2]2+
0.74
1.09
1.51
-0.93
-1.50
[Ru(TTF-dppz)3]2+
0.75
1.10
1.61
-0.86b
1.24
-0.18
[Ru(bpy)2(Aqphen)]2+, c
a
-0.46
Quasi-reversible. b Irreversible. c In 1,2-dichloroethane
127
-0.80
5.3.3. Photophysical properties
!
The absorption spectra of 1 in CH2Cl2, together with those of [Ru(bpy)(TTF-dppz)2]2+,
[Ru(TTF-dppz)3]2+, TTF-dppz and Aqphen are presented in the Figure 1. By comparison
with the spectra of the free ligands and the reference complexes, the absorption bands
of 1 can be readily attributed to specific transitions. The broad absorption band at 16500
cm-1 corresponds to the 1ILCT transition occurring on the TTF-dppz ligands, as shown
previously for the series of reference complexes [Ru(bpy)3-x(TTF-dppz)x]2+, x = 1 - 3,18
with the TTF subunit acting as an electron donor, while dppz is acting as an electron
acceptor. With respect to the free ligand this 1ILCT absorption band is red-shifted by
2000 cm-1, corresponding to a lowering of the energy of the LUMO located on the dppz
subunit due to the coordination to ruthenium(II). As expected, the extinction coefficient
of the 1ILCT absorption band on TTF-dppz increases from 1.104 to 2.2.104 M-1cm-1
upon coordination of two TTF-dppz subunits and one Aqphen subunit to ruthenium(II).
By comparison with the reference complexes,18 the electric-dipole-allowed Ru2+$dppz
1
MLCT transition is located at around 21900 cm-1. The corresponding Ru2+$ phen
1
MLCT transition to the Aqphen ligand is located at roughly the same energy. The
absorption band centered at 26500 cm-1 can be assigned to the !-!* transitions located
on the dppz subunits in both TTF-dppz as well as in Aqphen. In the UV region, a broad
band is attributed to !-!* transitions located on the different subunits. Similar to
[Ru(TTF-dppz)3]2+ and [Ru(bpy)(TTF-dppz)2]2+,18 no luminescence from any MLCT state
of 1 could be detected in the limit of the sensitivity of the experimental set-up. In
contrast to the TTF-dppz and its complexes,18 for which weak luminescence from the
1
ILCT state of TTF-dppz was observed, 1 doesn't show any luminescence from that
state either.
Upon chemical oxidation (Figure 2) of 1 with [Fe(bpy)3]3+ in CH3CN, the
1
ILCT
absorption band located at 17500 cm-1 decreases in intensity and two new bands
appear at 12000 and 21500 cm-1, respectively. In analogy to the previous study of TTFdppz and its complexes,18 the new absorption spectrum highlights the oxidation of the
TTF subunit, and in particular, the new absorption band in the near IR in turn
corresponds to a dppz$TTF"+ 1ILCT transition. This oxidation seems to be irreversible
chemically.
!
128
Figure 1. Absorption spectra of TTF-dppz (blue), Aqphen (green), [Ru(bpy)(TTFdppz)2]2+ (purple), [Ru(TTF-dppz)3]2+ (black) and 1 (red) in CH2Cl2 at room
temperature.
Figure 2. Chemical oxidation of 1: absorption spectra of 1 in CH3CN as a function of
equivalents of [Fe(bpy)3]3+ added to the solution. The spectra are corrected for the
generated [Fe(bpy)3]2+.
!
129
In order to arrive at a better understanding of the photophysical behavior of 1 and the
role of the Aqphen ligand in the molecule, compared to the reference complex
[Ru(bpy)(TTF-dppz)2]2+, transient absorption spectra were recorded on different
timescales using different excitation wavelengths. Figure 3 shows the transient
absorption spectrum of 1 on the ps time scale upon excitation at 650 nm (15400 cm-1),
that is, into the 1ILCT band. As becomes evident from the global fit, the transient signal
decays mono-exponentially with ' = 4.7(5) ps (Figure 4). The bleaching of the band at
16000 cm-1 and the strong absorption at 20000 cm-1 are a signature of the 1ILCT state.
Thus with an observed lifetime of 4.7 ps compared to an estimated radiative lifetime of
approximately 45 ns from its oscillator strength, the 1ILCT state in the complex decays
to the ground state in an essentially radiationless process.
Figure 3. Transient absorption spectra of [Ru(TTF-dppz)2(Aqphen)]2+ at room
temperature in CH2Cl2 on the ps timescale with an excitation wavelength of 650 nm
(excitation in the 1ILCT band).
!
130
50
400
integrated signal [a. u.]
40
450
500
550
600
650
700
nm
30
20
10
0
-10
0
1
2
3
4
time delay [ps]
5
6
7
Figure 4. Global fit to the transient absorption spectra of Figure 3 with a single
exponential decay resulting in a decay rate constant of 2.1(2)!1011 s-1 (" = 4.7(5) ps).
Figure 5 shows the transient absorption spectrum of 1 likewise on the ps timescale
upon excitation at 400 nm (25000 cm-1), that is, into the 1MLCT absorption bands. The
time evolution is clearly not single exponential. Initially there also is a strong transient
absorption centered at 20000 cm-1, which decays quite rapidly, but as it decays a new
species evolves with a band to the blue in the form of a double hump and a weak
bleaching at 16000 cm-1. This spectrum still persists at 1 ns.
!
131
Figure 5. Transient absorption spectra of [Ru(TTF-dppz)2(Aqphen)]2+ at room
temperature in CH2Cl2 on the ps timescale with an excitation wavelength of 400 nm
(excitation in the 1MLCT bands).
The global fit with a double exponential as shown in Figure 6a, clearly identifies two
time constants, namely a fast process with '1 = 5.4(8) ps (k1 = 1.8(2).1011 s-1) and a
slower process with '2 = 242(8) ps (k2 = 4.1(2).1011 s-1), and characteristic difference
spectra that go with them (Figure 6b). In addition, the offset is very similar to the last
spectrum in Figure 5 recorded at 1000 ps after the pulse. As will be shown below, this
last spectrum persists for quite a long time after the pulse.
Figure 7 shows the transient absorption spectrum on the *s timescale of 1 in CH2Cl2 for
excitation at 458 nm (21800 cm-1), that is, into the 1MLCT transitions of the complex.
Besides a bleaching of the 1ILCT ground state absorption band at 16000 cm-1, there are
two transient absorption bands at around 21000 and 12500 cm-1, respectively. The
corresponding transient state has a lifetime of approximately 400 ns in CH2Cl2 at room
temperature. By comparison with the spectrum of the oxidized form of 1 (Figure 2), the
transient absorption spectrum can be attributed to the formation of a cationic TTF"+
subunit in conjunction with a neutral dppz moiety.17 This implies a long-lived charge-
!
132
separated excited state reached by photo-excitation, the question being where is the
odd electron localized.
a)
30
integrated signal [a. u.]
25
400
450
500
550
nm
600
650
20
15
10
5
0
200
b)
400
600
time delay [ps]
800
50
y0
A1
A2
40
30
amplitude
1000
!1 = 5.4(5) ps
!2 = 242(2) ps
20
10
0
22000
20000
18000
-1
wavenumber [cm ]
16000
Figure 6. Global fit of a double exponential decay and an offset to the data in Figure
5: a) experimental decay curves and corresponding fits at different wavelengths, b)
amplitudes (A1 and A2) and offset (A0) as function of wavelength.
!
133
Figure 7. a) Transient absorption spectrum integrated over 100 ns in 100 ns intervals
after pulsed irradiation at 458 nm, for direct comparison the last spectrum from Figure
5 is included (t = 0), b) decay curves on the µs timescale at 470 nm (21300 cm-1) and
550 nm (18200 cm-1) and 800 nm (12500 cm-1).
!
134
In summary, following the initial excitation into the 1MLCT bands, intersystem crossing
on the femtosecond timescale takes the molecule to the corresponding 3MLCT states.34
This is not resolved in our spectra. From there, relaxation takes place to an intermediate
state with a time constant '1 of 5.4 ps. The intermediate state, in turn, leads to the
charge separated state with a time constant '2 of 242 ps. Finally, the charge-separated
state has a lifetime of 400 ns.
Using spectroscopic and CV data, the energy level scheme shown in Figure 8 can be
constructed. The spectroscopically accessible
1
MLCT states for both ligands are
approximately at the same energy of around 2.5 eV (20200 cm-1), with the one involving
Aqphen at slightly lower energy because of the electron withdrawing effect of
anthraquinone as opposed to the electron donating effect of TTF. As for all ruthenium(II)
polypyridyl complexes, the corresponding 3MLCT states are at around 0.4 eV lower
energies,35 that is, at around 2.1 eV (17000 cm-1). Due to the fact that there are two
TTF-dppz and only one Aqphen ligands coordinated to ruthenium(II), excitation in the
region of the 1MLCT absorption bands creates a species formulated best as [Ru3+(TTFdppz-)(TTF-dppz)(Aqphen)]2+ as majority species. This species is very short lived. The
second TTF-dppz will immediately provide an electron and reestablish the +2 charge on
ruthenium, thus resulting in a first charge separated state (CS1) formulated [Ru2+(TTFdppz-)(TTF+-dppz)(Aqphen)]2+, in analogy to the reference complex [Ru(bpy)(TTFdppz)2]2+.18 From the redox potentials of the reference complex (Table 1), this state is at
an estimated energy of 1.67 eV (13700 cm-1). We attribute the 5.4 ps transient signal to
this process. In the reference complexes, that is, in the absence of any further
acceptors, this charge-separated state has a lifetime of 2.4 µs. In the title complex 1,
the Aqphen acts as comparatively strong electron acceptor, thus instead of relaxing to
the ground state, electron transfer from dppz- to the Aqphen takes place, resulting in the
long-lived charge separated state [Ru2+(TTF+-dppz)(TTF-dppz)(Aq-phen)]2+ (CS2), that
is, with the odd electron localized on the quinone part of Aqphen. From the difference in
the corresponding redox potentials of 1 (Table 1) this state is located at 1.48 eV (11900
cm-1). We attribute the 242 ps transient signal with the build up of a new species to this
process.
Of course excitation in the region of the 1MLCT bands creates an MLCT state with the
electron on the phen unit of Aqphen to be formulated as [Ru3+(TTF-dppz)2(Aqphen-)]2+
!
135
as minority species.15 This state has a very short lifetime as the electron density is
redistributed to the quinone part of the ligand on the femtosecond timescale,36 and
subsequent electron transfer from TTF directly results in the final charge separated
state most probably within the same 5.4 ps as the analogue process for the majority
species.
Figure 8. Energy level scheme for [Ru(TTF-dppz)2(Aqphen)]2+; the most probable
pathway for the creation of the low-energy charge-separated state via excitation into
the 1MLCT band to dppz on TTF-dppz is shown by broken lines, an alternative
pathway for excitation into the 1MLCT band to phen on Aqphen by dotted lines
(Notation: TTF $ T, dppz $ d, Aq $ A, phen $ p).
A few questions need answers: why is the lifetime of the charge-separated state CS2
with the very good acceptor shorter than the one of the reference compound? Two
reasons come to mind: a) the charge recombination is in the Marcus inverted region and
thus is faster for the smaller energy gap, b) the Aqphen ligand is bent and thus through
space interaction increases the electronic coupling. The latter could be tested by using
the linear analogue of Aqphen as electron-accepting ligand. Finally for excitation into
!
136
the 1ILCT state, there is no probable mechanism for the creation of the low-energy
charge separated state and thus because of the intrinsic short lifetime of the 1ILCT
state, it simply decays to the ground state.
5.4. Conclusions
The title complex, [Ru(TTF-dppz)2(Aqphen)]2+, constitutes a triad of chromophores
engineered for directional light-induced electron transfer. The donor (TTF-dppz) and the
acceptor (Aqphen) ligands are held together by the metal center. The latter, together
with the ligating chromophores, acts as sensitizer via 1MLCT transitions. The majority
pathway for the creation of the long-lived light-induced charge-separated state, CS2,
involves several steps with an intermediate state, which is also a charge-separated
state CS1. In the absence of the final acceptor Aqphen, the CS1 state would in turn be
long-lived as observed in the reference complex [Ru(bpy)(TTF-dppz)2]2+.18 In the
reference complex, the light-induced electron transfer is not directional. In contrast, in
the title complex the light-induced electron transfer is directional and this constitutes an
important property for potential applications in solar energy conversion and signal
processing.
Acknowledgments This work was supported by the Swiss National Science
Foundation (Grant Nos. 200020-116003 and 200020-130266/1) and EU(FUNMOLS
FP7-212942-1). We thank E. Vauthey for helping us with his expertise in ultrafast
spectroscopy.
!
137
5.5. References
!
1.
Crosby, G. A., Acc. Chem. Res. 1975, 8, 231-238.
2.
Rusanova, J.; Decurtins, S.; Rusanov, E.; Stoeckli-Evans, H.; Delahaye, S.;
Hauser, A., J. Chem. Soc., Dalton Trans. 2002, 4318-4320.
3.
Delahaye, S.; Loosli, C.; Liu, S. X.; Decurtins, S.; Labat, G.; Neels, A.; Loosli, A.;
Ward, T. R.; Hauser, A., Adv. Funct. Mat. 2006, 16, 286-295.
4.
Spiccia, L.; Deacon, G. B.; Kepert, C. M., Coord. Chem. Rev. 2004, 248, 1329-
1341.
5.
Kozlov, D. V.; Tyson, D. S.; Goze, C.; Ziessel, R.; Castellano, F. N., Inorg. Chem.
2004, 43, 6083-92.
6.
Galletta, M.; Puntoriero, F.; Campagna, S.; Chiorboli, C.; Quesada, M.; Goeb, S.;
Ziessel, R., J. Phys. Chem. A 2006, 110, 4348-58.
7.
Vos, J. G.; Kelly, J. M., J. Chem. Soc., Dalton Trans. 2006, 4869-4883.
8.
Meyer, T. J., Pure Appl. Chem. 1986, 58, 1193-1206.
9.
Meyer, T. J., Acc. Chem. Res. 1989, 22, 163-170.
10.
Gu, J.; Chen, J.; Schmehl, R. H., J. Amer. Chem. Soc. 2010, 132, 7338-7346.
11.
Alstrum-Acevedo, J. H.; Brennaman, M. K.; Meyer, T. J., Inorg. Chem. 2005, 44,
6802-6827.
12.
Balzani, V.; Gomez-Lopez, M.; Stoddart, J. F., Acc. Chem. Res. 1998, 31, 405-
414.
13.
Handy, E. S.; Pal, A. J.; Rubner, M. F., J. Amer. Chem. Soc. 1999, 121, 3525-
3528.
14.
!
Gao, F. G.; Bard, A. J., J. Amer. Chem. Soc. 2000, 122, 7426-7427.
138
15.
Lopez, R.; Leiva, A. M.; Zuloaga, F.; Loeb, B.; Norambuena, E.; Omberg, K. M.;
Schoonover, J. R.; Striplin, D.; Devenney, M.; Meyer, T. J., Inorg. Chem. 1999, 38,
2924-2930.
16.
Delgadillo, A.; Arias, M.; Leiva, A. M.; Loeb, B.; Meyer, G. J., Inorg. Chem. 2006,
45, 5721-5723.
17.
Jia, C.; Liu, S. X.; Tanner, C.; Leiggener, C.; Neels, A.; Sanguinet, L.; Levillain,
E.; Leutwyler, S.; Hauser, A.; Decurtins, S., Chem. - Euro. J. 2007, 13, 3804-3812.
18.
Goze, C.; Leiggener, C.; Liu, S. X.; Sanguinet, L.; Levillain, E.; Hauser, A.;
Decurtins, S., Chemphyschem 2007, 8, 1504-12.
19.
Lopez, R.; Loeb, B.; Boussie, T.; Meyer, T. J., Tetrah. Lett. 1996, 37, 5437-5440.
20.
Arias, M.; Concepcion, J.; Crivelli, I.; Delgadillo, A.; Diaz, R.; Francois, A.;
Gajardo, F.; Lopez, R.; Leiva, A. M.; Loeb, B., Chem. Phys. 2006, 326, 54-70.
21.
Wu, H.; Zhang, D. Q.; Su, L.; Ohkubo, K.; Zhang, C. X.; Yin, S. W.; Mao, L. Q.;
Shuai, Z. G.; Fukuzumi, S.; Zhu, D. B., J. Amer. Chem. Soc. 2007, 129, 6839-6846.
22.
Fang, C. J.; Zhu, Z.; Sun, W.; Xu, C. H.; Yan, C. H., New J. Chem. 2007, 31,
580-586.
23.
Campagna, S.; Serroni, S.; Puntoriero, F.; Loiseau, F.; De Cola, L.; Kleverlaan,
C. L.; Becher, J.; Sorensen, A. P.; Hascoat, P.; Thorup, N., Chem. - Euro. J. 2002, 8,
4461-4469.
24.
Bryce, M. R., Adv. Mat. 1999, 11, 11-23.
25.
Goze, C.; Liu, S. X.; Leiggener, C.; Sanguinet, L.; Levillain, E.; Hauser, A.;
Decurtins, S., Tetrah. 2008, 64, 1345-1350.
26.
Goze, C.; Dupont, N.; Beitler, E.; Leiggener, C.; Jia, H.; Monbaron, P.; Liu, S. X.;
Neels, A.; Hauser, A.; Decurtins, S., Inor. Chem. 2008, 47, 11010-11017.
27.
Keniley, L. K.; Ray, L.; Kovnir, K.; Dellinger, L. A.; Hoyt, J. M.; Shatruk, M., Inorg.
Chem. 2010, 49, 1307-1309.
28.
!
Vacher, A.; Barriere, F.; Roisnel, T.; Lorcy, D., Chem. Comm. 2009, 7200-7202.
139
29.
Pointillart, F.; Le Gal, Y.; Golhen, S.; Cador, O.; Ouahab, L., Inorg. Chem. 2008,
47, 9730-9732.
30.
Leiggener, C.; Dupont, N.; Liu, S. X.; Goze, C.; Decurtins, S.; Beitler, E.; Hauser,
A., Chimia 2007, 61, 621-625.
31.
Goss, C. A.; Abruna, H. D., Inorg. Chem. 1985, 24, 4263-4267.
32.
Duvanel, G.; Banerji, N.; Vauthey, E., J. Phys. Chem. A 2007, 111, 5361-5369.
33.
Yamaguchi, S.; Hamaguchi, H. O., Appl. Spectrosc. 1995, 49, 1513-1515.
34.
Tarnovsky, A. N.; Gawelda, W.; Johnson, M.; Bressler, C.; Chergui, M., J. Phys.
Chem. B 2006, 110, 26497-26505.
35.
Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; von Zelewsky, A.,
Coord. Chem. Rev. 1988, 84, 85-277.
36.
Lünnemann, S; Kuleff, A. I.; Cederbaum, L. S., J. Chem. Phys. 2009, 130,
154305.
!
140
6. Imidazole-Annulated Tetrathiafulvalenes exhibting pH-Tuneable
Intramolecular Charge Transfer and Redox Properties
Published in Chemistry-An asian journal, 4 (3): 392-399, 2009
!
!
Jincai Wu,[a,b] Nathalie Dupont,[c] Shi-Xia Liu,*[a] Antonia Neels,[d] Andreas Hauser[c] and
Silvio Decurtins[a]
[a]
Prof. J. Wu, Dr. S.-X. Liu, Prof. S. Decurtins, Departement für Chemie und
Biochemie, Universität Bern, Freiestrasse 3, CH-3012 Bern, Switzerland, Fax: (+) 41 31
631 4399 , E-mail: [email protected]
[b]
Prof. J. Wu , College of Chemistry and Chemical Engineering , State Key
Laboratory of Applied Organic Chemistry , Lanzhou University, 730000 Lanzhou, P. R.
China
[c]
N. Dupont, Prof. A. Hauser, Département de chimie physique, Université de
Genève, 30 quai Ernest-Ansermet, CH-1211 Genève 4, Switzerland
[d]
Dr. A. Neels, XRD Application LAB, CSEM Centre Suisse d'Electronique et de
Microtechnique SA, Jaquet-Droz 1, Case postale, CH-2002 Neuchâtel, Switzerland
!
141
Abstract
In order to study the electronic interactions in donor–acceptor ensembles as a function
of pH, an efficient synthetic route to three imidazole-annulated tetrathiafulvalene (TTF)
derivatives 1-3 is reported. Their electronic absorption spectra in view of photoinduced
intramolecular charge transfer and their electrochemical behaviour were investigated,
and pKa values for the two protonation processes on the acceptor unit were determined
in organic solvents by photometric titration. The influence of the TTF moiety on these
values is discussed.
!
142
CONTENT
6.1. Introduction
144
!
6.2. Results and Discussion
146
!
6.3. Conclusion
157
!
6.4. Experimental Section
157
6.4.1. General
157
6.4.2. Materials
158
6.4.3. Synthesis
158
6.4.4. Crystallography
159
!
6.5. References
!
160
143
6.1. Introduction
!
Electrochemically amphoteric compounds are of current interest due to their potential
applications in molecular electronics and optoelectronics, whereby organic compounds
possessing a high degree of conjugation are particularly interesting for advanced
electronic applications.[1] Accordingly, efforts have been directed to the design and
synthesis of molecular systems composed of building blocks that give rise to electron
donor (D) and acceptor (A) interactions. Amongst all kinds of molecular electron donor
moieties, tetrathiafulvalene (TTF) and its derivatives are known to be strong !–donors
capable of forming persistent cation radical and dication species upon oxidation, thus
leading to a number of conducting and superconducting materials.[2] In this context, the
effective intermolecular orbital overlap in !-stacked assemblies is highly sensitive to
chemical modifications of the TTF framework and this plays a crucial role in the
electronic conductivity. Thus, in order to achieve a highly ordered molecular
organization in charge-transfer (CT) salts, the self-assembling ability and the
dimensionality of the electronic structure has been enhanced primarily by chalcogenchalcogen interactions[2] and halogen or hydrogen bonds.[3-5] Notably, amongst them,
TTF-imidazole systems exhibit unprecedented electronic and structural modulation
effects of hydrogen bonds giving rise to a number of highly conductive CT complexes
with various acceptors. [5]
With respect to covalently linked D-A ensembles, important variables, besides the
nature of the donor and acceptor components, are their relative distance, orientation
and the degree of electronic coupling between them. Normally, the D and A components
are held together by !-spacers such as oligo(phenylene ethylene), oligo(phenylene
vinylene), oligothiophene, or "-spacers of variable length and flexibility.[6] Alternatively,
only a few examples of annulated TTF-!-acceptor systems have been reported in the
literature so far, as for example TTF-diquinones.[7] For a current overview of D-A
ensembles the reader may consult the recent review by Wudl et al.[8]
Along this line, we recently introduced a synthetic concept for the annulation of TTF
derivatives to a variety of acceptor moieties and reported synthetic routes to intimately
fused and rigid D-!-A ensembles, as exemplified by compounds with i) four TTF
moieties fused to phthalocyanine cores;[9] ii) three TTF moieties fused to a
!
144
hexaazatriphenylene core;[10] iii) one TTF annulated with dipyrido-[3,2-a:2#,3#c]phenazine;[11] and iv) one TTF coupled to N,N#-phenylenebis(salicylideneimine).[12] In
all these examples a main principle emerges, namely the molecular D-A systems are
tailored into planar configurations exhibiting defined symmetries which provides a
defined geometrical control. They are also specifically designed for chelation of various
metal ions. [12-13] As a continuation of our ongoing studies, we present here an efficient
synthetic route to further organic !–conjugated D-A molecules which contain the TTF
unit as a donor and an N,N-diimine chelating ligand – in form of 2-(2pyridyl)benzimidazole (PB), 2-(2-quinolinyl)benzimidazole (QB) or its derivative 2-(6methoxy-2-quinolinyl)benzimidazole (MQB) – as an acceptor (1-3, Figure 1).
The general interest in imidazole compounds stems from their specific structural
features and biological activity,[14] including anti-ulcer, anti-tumour and anti-viral effects.
The heterocyclic aromatic compounds reveal strong and directional hydrogen bond
interactions, and act as Brönsted acids and bases. They thus play an important role as
relay for proton transfer (PT) processes, which are among the most extensively studied
chemical processes, owing to their importance in nature.[15] As a consequence,
imidazole derivatives have attracted much attention in fields such as crystal
engineering,[16] molecule-based magnets,[17] molecular conductors[18] and fluorescence
sensors.[19]
Figure 1. Structures of the D-A molecules 1-3.
!
145
A primary feature of compounds 1-3 is that a range of functionalities on the acceptor
part is combined in close and controlled proximity with the fused TTF donor part. For the
former, the aromatic imidazole with its pyridine- and pyrrole-like N atoms introduces an
amphoteric character as a moderately strong base and a weak acid, and the linked
pyridine or quinoline adds another basic N atom. Moreover, the resulting N^N chelating
site makes these molecules attractive ligands for complexation to various metal ions. As
a most striking feature, discussed in detail below, the annulated TTF donor shows a
photoinduced intramolecular charge-transfer (ICT) transition in its absorption spectrum,
which is sensitive to the different protonation states on the acceptor site.
In this article, we describe the synthesis of three TTF-PB/QB/MQB molecules (1-3) and
the single-crystal X-ray structure for 1. In addition, the results of an electrochemical and
a photophysical investigation at various pH values are discussed.
6.2. Results and Discussion
!
As outlined in Scheme 1, the target compounds 1-3 were obtained via the direct
condensation reaction of the corresponding aldehydes with 5,6-diamino-2-(4,5bis(propylthio)-1,3-dithio-2-ylidene)-benzo [d]-1,3-dithiole in acceptable yields. All
compounds were easily purified by flash chromatography on SiO2 and have been fully
characterized by NMR, EI mass spectrometry and elemental analysis.
Scheme 1. A synthetic route to D-A molecules 1-3.
!
146
Orange needle-shaped single crystals of 1 suitable for X-ray analysis were obtained by
slow evaporation of its solution in CH3CN. The molecule 1 crystallizes in a monoclinic
space group (C2/c) and an ORTEP drawing of the molecule with the atomic numbering
scheme is shown in Figure 2. Apparently, this compound adopts a non-planar
conformation along its long molecular axis, and specifically the TTF moiety shows a
boat conformation, folding along the S1(((S4 and S2(((S3 vectors by 28.9(1)° and
18.24(1)°, respectively. In contrast to the TTF part, all atoms of the PB unit lie almost
perfectly within a plane; the rms deviation from a least-squares plane through all
involved atoms of PB is 0.01 Å. Notice, that the nitrogen atoms N1 and N3 are situated
in a trans orientation within the solid state structure. The bond distances within the TTF
and PB moieties are in the expected ranges in comparison with those of similar
compounds in the literature.[5c,
20-21]
Figure 3 highlights the alternating arrangement of
the molecules in the crystal structure. A noticeable feature is the head-to-tail alignment
caused by !(((! stacking between the PB moieties to afford dimers, which are, in
addition, linked parallel to each other by some conventional N–H(((N hydrogen bonds
and short S(((S contacts.
Figure 2. ORTEP drawing and atomic numbering scheme of the molecule 1 with
thermal ellipsoids at 50% probability level. Hydrogen atoms, except for the one on the
imidazole ring, have been omitted for clarity. Front and side views are presented.
!
147
Figure 3. Crystal packing of 1 showing hydrogen bonds (N–H(((N, 2.960 Å) and close
S(((S contacts of 3.536 Å (dashed lines).
The electrochemical properties of the !-conjugates 1-3 were investigated by cyclic
voltammetry (CV) in dichloromethane. All of them show two reversible single-electron
oxidation waves typical of the TTF system, corresponding to E1/21 and E1/22 in Table 1.
Obviously, the presence of the different substituents at the 2-position of the imidazole
ring has a negligible influence on the electron-donating ability of TTF. Remarkably, with
successive addition of HCl to 1-3, both redox processes of the TTF unit are clearly
shifted suggesting the occurrence of two new redox species. Taking 1 as an example
(Figure 4), both oxidation potentials are at first substantially positively shifted upon the
addition of a small amount of HCl, until, upon the addition of 3 equivalents of HCl, the
original E1/21 wave completely disappears and only two reversible redox waves remain
at 0.63 and 0.98 V, respectively. The large potential shifts correspond to a decrease in
the !-donating ability of TTF unit due to the protonation of the imidazole ring in close
proximity to the TTF core. Interestingly, a large excess of HCl leads to a further positive
shift of E1/21 but to a negative shift of E1/22, which is now almost at the same potential as
that of the unprotonated 1.[22a, 22b] This observation might be attributed to the occurrence
of proton dissociation when the TTF unit undergoes the second oxidation as shown in
green in Scheme 2. It can therefore be deduced that there is a strongly dynamical
difference in the proton dissociation and redox processes, as reported in the
literature.[23] All these results indicate the coexistence of three redox species, 1 (E1/21 =
!
148
0.54 V, E1/22 = 0.94 V ) and two protonated ones, that is, [1(H+] (E1/21 = 0.63 V, E1/22 =
0.98 V) and [1(2H+] (E1/21 = 0.67 V, E1/22 = 0.93 V), which can be described with the
equilibrium processes shown in Scheme 2. It is noteworthy that at particular potentials,
constant currents upon addition of HCl are observed, similar to isosbestic points in
optical spectroscopy. This is indicative of a concomitant appearance of the
corresponding protonated species at the expense of 1. These observations are in good
agreement with the results of UV-vis spectroscopy measurements.
Table 1. Redox Potentials (V vs. Ag/AgCl) of 1-3 in CH2Cl2.
E
1
1(H+ 1(2H+ 2
2(H+ 2(2H+ 3
3(H+ 3(2H+
E1/21 0.54 0.63 0.67
0.53 0.61 0.68
0.52 0.60 0.68
E1/22 0.94 0.98 0.93
0.95 0.99 0.95
0.94 0.98 0.90
Figure 4. Cyclic voltammogram of compound 1 (10-3 M) in the presence of increasing
amounts of HCl; CH2Cl2; Bu4NPF6 (0.1 M); 100 mV/s; Pt working electrode. V vs.
Ag/AgCl.
!
149
Scheme 2. Equilibrium reactions during the electrochemical titration of 1 with H+.
The absorption spectra of 1-3 dissolved in CH2Cl2 are presented in Figure 5. They show
two domains of broad and intense absorption bands centred around 24,000 and 30,000
cm-1, respectively. The absorption bands centered at 25,189 cm-1 (397 nm), 23,753 cm-1
(421 nm) and 24,272 cm-1 (412 nm), respectively for 1-3, result from an ICT transition
from the TTF unit to the substituted benzimidazole moieties. In the UV region, strong
absorption bands are characteristic for !-!* transitions located on both, the TTF and the
substituted benzimidazole subunits.[24]
In order to arrive at a deeper understanding of the fused D-A molecules 1-3, the
protonation characteristics of each compound were experimentally investigated in
CH2Cl2 solution. The molecules have two protonation sites each, namely, the
benzimidazole nitrogen (N1) and the nitrogen on pyridine (N3) or on quinoline,
respectively. It has already been demonstrated experimentally and theoretically that a
benzimidazole nitrogen has a higher proton affinity than the one of pyridine.[25a] It was
on this basis that UV-vis titration experiments were carried out in order to determine the
specific pKa values of molecules 1-3.
!
150
Figure 5. Absorption spectra of 1 (___), 2 (___) and 3 (___) in CH2Cl2 at room
temperature.
The UV-vis spectra of 1 dissolved in CH2Cl2, taken as a function of pH, that is, with
successive addition of HCl equivalents, are depicted in Figure 6. From 0 to 3
equivalents of HCl, for an initial concentration of 10-5 M of 1, the broad absorption band
at 25,189 cm-1 gradually disappears and a new absorption band at 21,368 cm-1
emerges. From 3 to 300 equivalents, this new band decreases in intensity and a third
absorption band concomitantly appears at 16,863 cm-1. A remarkable feature is the
occurrence of two quite well defined isosbestic points, at 23,000 cm-1 for up to 3
equivalents added, and at approximately 18,400 cm-1 above. This indicates the
presence of three species in two comparatively well separated chemical equilibria,
namely compound 1 and the two protonated species [1(H+] and [1(2H+] as shown in red
in Scheme 2. Compounds 2 and 3 show exactly the same behaviour and the absorption
maxima of the species in the neutral and protonated forms are reported in Table 2.
!
151
Figure 6. Absorption spectra of 1 in CH2Cl2 at room temperature as a function of
molar equivalents of HCl added.
In accordance with the absorption spectra, the colour of the solution changes reversibly
between yellow, orange and blue during the acid/base tritration, as visualized within the
TOC Figure. Apparently, protonation of the two nitrogen atoms of the PB unit reduces
the electron density on the aromatic ring system successively, thereby lowering the
energy of the LUMO and increasing its acceptor properties. As a result, the
corresponding spectroscopic TTF $ PB ICT transition moves to lower energies in two
distinct steps of ca. 4,000 and 4,500 cm-1, respectively. The first step corresponds to the
protonation of the imidazole nitrogen, the second step to the protonation of the pyridine
nitrogen.
!
152
Table 2. Maxima of the ICT absorption bands for 1-3 and their protonated forms in
CH2Cl2.
Absorption
Absorption
%
[nm]
[cm-1]
[M-1 cm-1]
1
397
25,189
14,800
[1(H+]
468
21,368
9,800
[1(2H+]
593
16,863
6,400
2
421
23,753
15,500
[2(H+]
564
17,730
12,000
[2(2H+]
691
14,472
9,600
3
412
24,272
20,900
[3(H+]
544
18,382
13,000
[3(2H+]
655
15,267
10,600
Compound
The spectra in Figure 6 can be analysed quantitatively according to the acid base
equilibria shown in Scheme 2. The pKa values for the two protonation steps obtained
via an iterative fitting procedure are given in Table 3 for compounds 1-3. Figures 6a and
b show the experimental evolution of the concentration of the three species as a
function of HCl added to the solution for the two protonation steps, as well as the
resulting best fit.
!
153
Figure 7. Experimental (symbols) and calculated (full lines) titration curves. For the
calculated curves the pKa values of Table 3 were used: (a) first step, 1 (!) and [1(H+]
(/), (b) second step [1(H+] (/) and [1(2H+] (") in CH2Cl2 at room temperature.
!
154
Table 3. Experimental pKa values for 1-3 and the reference compound 2-(2pyridyl)benzimidazole (PB) in CH2Cl2.
pKa
PB
1
2
3
pKa1
4.4(1)
5.5(1)
5.2(1)
5.0(1)
pKa2
1.7(1)
3.2(1)
3.4(1)
3.5(1)
In the literature, experimental pKa values in non-aqueous solvents are quite rare. In
aqueous solutions, the values of pKa1 and pKa2 for PB are given as 4.41 and -1.59,
respectively.[25b] In order to compare the values of the TTF annulated compounds in
CH2Cl2 with the ones for the compounds without TTF, titration curves were also
determined for PB in CH2Cl2. The experimental curves together with the best fit are
shown in Figure 8, and the corresponding pKa values are included in Table 3. Whereas
pKa1 has a value close to the one found in aqueous solution, the pKa2 value of 1.7
shows that the proton on pyridine is much less acidic in CH2Cl2 than in aqueous
solution. In comparison to PB, compounds 1-3 show a slight increase in pKa1 and a
more pronounced one in pKa2. Qualitatively this can be understood as the result of the
electron donating properties of the TTF unit. This results in a higher electron density on
the PB unit as compared to the compound without TTF, and therefore the pKa values
increase substantially. Interestingly, this effect is more pronounced for the protonation of
the pyridine nitrogen.
!
155
Figure 8. Experimental (symbols) and calculated (full lines) titration curves. For the
calculated curves the pKa values of Table 3 were used: (a) first step, PB (!) and
[PB(H+] (/), (b) second step [PB(H+] (/) and [PB(2H+] (") in CH2Cl2 at room
temperature.
!
156
6.3. Conclusion
!
In conclusion, three imidazole-annulated TTF derivatives 1-3 have been prepared and
fully characterized, and the influence of the TTF unit on the pKa values of the acceptor
units as determined by photometric titration has been discussed. The novel feature of
these D-A molecules is that they contain a PB ancillary functionality, which has been
incorporated with the following specific roles in mind: (i) The presence of three nitrogen
atoms as proton donor/acceptors renders them promising in the field of chemosensors.
(ii) Direct annulation of PB to the TTF core is expected to enhance cooperativity
between CT on TTF and PT at hydrogen bonding sites on PB moieties in the resulting
simultaneous CT and PT complexes. (iii) The excellent chelating ability of the PB unit
should provide opportunities for the complexation of a wide range of transition metals to
the donor system giving rise to diverse structural chemistry and appealing photophysics.
The results reported here are part of an initial exploratory study of their potential to
generate a range of well-defined coordination networks as well as to produce the
simultaneous CT and PT complexes. Currently, we are engaged in an investigation on
the ability to bind transition metal ions and to form CT complexes and ion radical salts of
these promising new donors.
6.4. Experimental Section
6.4.1. General
!
Melting points were determined using a Büchi 510 instrument and are uncorrected.
Elemental analyses were performed on a Carlo Erba Instruments EA 1110 Elemental
Analyzer CHN.
1
H and
13
C NMR spectra were obtained on a Bruker AC 300
spectrometer operating at 300.18 and 75.5 MHz, respectively: chemical shifts are
reported in ppm referenced to residual solvent protons (DMSO-d6). The following
abbreviations were used s (singlet), d (doublet), t (triplet), b (broad) and m (multiplet).
Infrared spectra were recorded on a Perkin-Elmer Spectrum One FT-IR spectrometer
!
157
using KBr pellets. EI Mass spectra were recorded using an Auto SpecQ spectrometer.
Cyclic voltammetry was conducted on a VA-Stand 663 electrochemical analyzer. An
Ag/AgCl electrode containing 2 M LiCl (in ethanol) served as reference electrode, a
glassy carbon electrode as counter electrode, and a Pt wire as working electrode. Cyclic
voltammetric measurements were performed at room temperature under N2 in CH2Cl2
with 0.1 M Bu4NPF6 as supporting electrolyte at a scan rate of 100 mV s-1.
Photophysical measurements were performed on solutions of the compounds in CH2Cl2
at room temperature. Absorption spectra were recorded on a Varian Cary 5000
UV/vis/NIR spectrophotometer.
6.4.2. Materials
!
Unless otherwise stated, all reagents were purchased from commercial sources and
used without additional purification. 5,6-Diamino-2-(4,5-bis(propylthio)-1,3-dithio-2ylidene)-benzo[d]-1,3-dithiole was prepared according to literature procedures.[10-11]
6.4.3. Synthesis
!
1: Pyridinecarboxaldehyde (0.1 mL, 1 mmol) was added to a solution of 5,6-diamino-2(4,5-bis(propylthio)-1,3-dithio-2-ylidene)-benzo[d]-1,3-dithiole (0.43 g, 1 mmol) in cold
1,4-dioxane (50 mL). The mixture was open to air and stirred for 24 h. The solvent was
evaporated in vacuum and the resulting crude product was purified by chromatography
on silica gel with CH2Cl2/ethylacetate (2:1) to obtain 1 (0.145 g, 0.28 mmol, 28%) as a
light-orange solid. Mp 145-147 °C; 1H NMR: ) = 0.96 (t, 6H), 1.59 (m, 4H), 2.86 (t, 4H),
7.53 (m, 1H), 7.66 (s, 1H), 7.88 (s, 1H), 8.00 (m, 1H), 8.29 (m, 1H), 8.73 (m, 1H), 13.24
ppm (s, 1H);
13
C NMR: ) = 12.69, 22.58, 37.36, 79.11, 108.31, 112.65, 121.45, 124.84,
126.86, 137.54, 147.94, 149.39, 151.66 ppm; IR (KBr) " = 3431, 2959, 1627, 1596,
!
158
1444, 1392, 1088, 882, 791 cm-1; MS(EI): m/z (%): 519 (45) [M+]; elemental analysis:
calcd (%) for C22H21N3S6: C 50.83, H 4.07, N 8.08; found: C 51.12, H 4.16, N 7.42.
Single crystals were obtained by slowly cooling a hot actetonitrile solution of 1.
General procedure for 2-3: A solution of the corresponding aldehyde (0.1 mmol) and
5,6-diamino-2-(4,5-bis(propylthio)-1,3-dithio-2-ylidene)-benzo[d]-1,3-dithiole (0.04 g, 0.1
mmol) in nitrobenzene (10 mL) was heated up to 160 °C for 20 h. The solvent was
evaporated in vacuum and the resulting crude product was purified by chromatography
on silica gel with CH2Cl2/ethylacetate (2:1) to obtain an analytically pure product.
2: Yield: 0.02 g (40%), orange solid. Mp 208-210 °C; 1H NMR: ) = 0.96 (t, 6H), 1.59 (m,
4H), 2.86 (t, 4H), 7.53 (m, 1H), 7.66 (s, 1H), 7.88 (s, 1H), 8.00 (m, 1H), 8.29 (m, 1H),
8.73 (m, 1H), 13.24 ppm (s, 1H);
13
C NMR: ) = 12.72, 22.63, 37.40, 106.03, 108.51,
112.60, 113.11, 119.13, 126.85, 126.92, 127.37, 128.08, 128.21, 128.69, 129.40,
130.50, 130.95, 134.48, 137.43, 143.35, 147.16, 148.16, 151.63 ppm; IR (KBr) " =
3431, 2959, 1611, 1538, 1450, 1314, 1084, 852 cm-1; MS(EI): m/z (%): 569 (40) [M+];
elemental analysis: calcd (%) for C26H23N3S6: C 54.80, H 4.07, N 7.37; found: C 54.99,
H 4.12, N 6.83.
3: Yield: 0.02 g (38%), yellow solid. Mp 235-237 °C; 1H NMR: ) = 0.96 (t, 6H), 1.60 (m,
4H), 2.88 (t, 4H), 3.94 (s, 3H) 7.47 (m, 2H), 7.80 (s, 2H), 8.03 (d, 1H), 8.40 (m, 2H),
12.76 ppm (b, 1H);
13
C NMR: ) = 12.72, 22.61, 37.39, 55.65, 106.06, 108.34, 112.72,
119.43, 122.94, 126.88, 129.40, 129.74, 130.19, 136.03, 143.13, 145.83, 152.00,
157.95 ppm; IR (KBr) " = 3435, 2961, 1622, 1611, 1600, 1504, 1394, 1377, 1241, 833
cm-1; MS(EI): m/z (%): 599 (40) [M+]; elemental analysis: calcd (%) for C27H25N3OS6: C
54.06, H 4.02, N 7.00; found: C 54.33, H 4.24, N 6.67.
6.4.4. Crystallography
!
An orange crystal of compound 1 was mounted on a Stoe Mark II-Imaging Plate
Diffractometer System (Stoe & Cie, 2002) equipped with a graphite-monochromator.
Data collection was performed at –100 °C using MoK+ radiation (( = 0.71073 Å). 245
!
159
exposures (4 min per exposure) were obtained at an image plate distance of 100 mm,
180 frames with 0 = 0° and 0° < 1 < 180°, and 65 frames with 0 = 90° and 0° < 1 < 97°,
with the crystal oscillating through 1° in 1. The resolution was Dmin 2 Dmax: 0.72 2 17.78
Å. The structure was solved by direct methods using the program SHELXS-97[26] and
refined by full matrix least squares on F2 with SHELXL-97.[27] The NH hydrogen atom
was derived from Fourier difference maps and refined while the remaining hydrogen
atoms were included in calculated positions and treated as riding atoms using SHELXL97 default parameters. All non-hydrogen atoms were refined anisotropically. No
absorption correction was applied.
Crystal data for 1: C22H21N3S6, M = 519.78, 0.45 . 0.20 . 0.15 mm3, monoclinic, space
group C2/c, a = 40.925(3), b = 7.4182(4), c = 15.3202(11) Å, 3 = 100.778(6)°, V =
4569.0(6) Å3, Z = 8, *calcd = 1.511 g cm-3, µ = 0.616 mm-1, T = 173(2) K, F(000) = 2160,
29100 reflections collected, 6183 unique (Rint = 0.0891). Final GOF = 1.029, R1 =
0.0564, wR2 = 0.1465, R indices based on 4693 reflections with I > 2#(I), 287
parameters, 0 restraints. CCDC-699063 contains the supplementary crystallographic
data for 1. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre at http://www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
This work was supported by the Swiss National Science Foundation (grant No. 200020116003 and 200020-115867)
6.5. References
!
[1]
a) R. L. Carroll, C. B. Gorman, Angew. Chem. Int. Ed. 2002, 41, 4378–4400; b)
E. Tsiperman, J. Y. Becker, V. Khodorkovsky, A. Shames, L. Shapiro, Angew. Chem.
Int. Ed. 2005, 44, 4015–4018.
!
160
[2]
a) J-I. Yamada, T. Sugimoto, Eds. TTF Chemistry: Fundamentals and
Applications of Tetrathiafulvalene;
Springer: Berlin, Germany, 2004; b) V.
Khodorkovsky, J. Y. Becker, Organic Conductors: Fundamentals and Applications; J. P.
Farges, Ed.; Marcel Dekker: New York, NY, 1994; Chapter 3; c) J. M. William, J. R.
Ferraro, R. J. Thorn, K. D. Carlson, U. Geiser, H. H. Wang, A. M. Kini, M. H. Whangbo,
Organic Superconductors (Including Fullerenes): Synthesis, Structure, Properties, and
Theory; Prentice Hall: Englewoods Cliffs, NJ, 1992; d) G. Saito, Y. Yoshida, Bull. Chem.
Soc. Jpn. 2007, 80, 1–137.
[3]
H. M. Yamamoto, R. Kato, Chem. Lett. 2000, 970–971.
[4]
a) S.-X. Liu, A. Neels, H. Stoeckli-Evans, M. Pilkington, J. D. Wallis, S. Decurtins,
Polyhedron 2004, 23, 1185–1189; b) M. Fourmigué, P. Batail, Chem. Rev. 2004, 104,
5379–5418; c) S. A. Baudron, P. Batail, C. Coulon, R. Clérac, E. Canadell, V. Laukhin,
R. Melzi, P. Wzietek, D. Jérome, P. Auban-Senzier, S. Ravy, J. Am. Chem. Soc. 2005,
127, 11785–11797; d) Y. Morita, S. Maki, M. Ohmoto, H. Kitagawa, T. Okubo, T. Mitani,
K. Nakasuji, Org. Lett. 2002, 4, 2185–2188; e) P. Blanchard, K. Boubekeur, M. Sallé, G.
Duguay, M. Jubault, A. Gorgues, J. D. Martin, E. Canadell, P. Auban-Senzier, D.
Jérome, P. Batail, Adv. Mater. 1992, 4, 579–581; f) E. Miyazaki, Y. Morita, Y. Yakiyama,
S. Maki, Y. Umemoto, M. Ohmoto, K. Nakasuji, Chem. Lett. 2007, 36, 1102–1103; g) Y.
Morita, E. Miyazaki, Y. Umemoto, K. Fukui, K. Nakasuji, J. Org. Chem. 2006, 71, 5631–
5637.
[5]
a) T. Murata, Y. Morita, Y. Yakiyama, K. Fukui, H. Yamochi, G. Saito, K.
Nakasuji, J. Am. Chem. Soc. 2007, 129, 10837–10846; b) T. Murata, Y. Morita, K.
Fukui, K. Sato, D. Shiomi, T. Takui, M. Maesato, H. Yamochi, G. Saito, K. Nakasuji,
Angew. Chem. Int. Ed.. 2004, 43, 6343–6346; c) Y. Morita, Y. Yamamoto, Y. Yakiyama,
T. Murata, K. Nakasuji, Chem. Lett. 2008, 37, 24–25; d) T. Murata, Y. Morita, Y.
Yakiyama, Y. Nishimura, T. Ise, D. Shiomi, K. Sato, T. Takuic, K. Nakasuji, Chem.
Commun. 2007, 4009–4011.
[6]
a) G. J. Ashwell, W. D. Tyrrell, A. J. Whittam, J. Am. Chem. Soc. 2004, 126,
7102–7110; b) G. Ho, J. R. Heath, M. Kondratenko, D. F. Perepichka, K. Arseneault, M.
Pézolet, M. R. Bryce, Chem. Eur. J. 2005, 11, 2914–2922; c) R. M. Metzger, J. W.
Baldwin, W. J. Shumate, I. R. Peterson, P. Mani, G. J. Mankey, T. Morris, G.
Szulczewski, S. Bosi, M. Prato, A. Comito, Y. Rubin, J. Phys. Chem. B 2003, 107,
!
161
1021–1027; d) R. M. Metzger, Acc. Chem. Res. 1999, 32, 950 –957; e) H. Meier,
Angew. Chem. Int. Ed. 2005, 44, 2482–2506; f) D. M. Guldi, F. Giacalone, G. de la
Torre, J. L. Segura, N. Martin, Chem. Eur. J. 2005, 11, 7199–7210.
[7]
N. Gautier, F. Dumur, V. Lloveras, J. Vidal-Gancedo, J. Veciana, C. Rovira, P.
Hudhomme, Angew. Chem. Int. Ed. 2003, 42, 2765–2768.
[8]
M. Bendikov, F. Wudl, D. F. Perepichka, Chem. Rev. 2004, 104, 4891–4946.
[9]
a) C. Loosli, C. Jia, S.-X. Liu, M. Haas, M. Dias, E. Levillain, A. Neels, G. Labat,
A. Hauser, S. Decurtins, J. Org. Chem. 2005, 70, 4988–4992; b) C. A. Donders, S.-X.
Liu, C. Loosli, L. Sanguinet, A. Neels, S. Decurtins, Tetrahedron 2006, 62, 3543–3549;
c) S. Delahaye, C. Loosli, S.-X. Liu, S. Decurtins, G. Labat, A. Neels, A. Loosli, T. R.
Ward, A. Hauser, Adv. Funct. Mater. 2006, 16, 286–295.
[10]
C-Y. Jia, S.-X. Liu, C. Tanner, C. Leiggener, L. Sanguinet, E. Levillain, S.
Leutwyler, A. Hauser, S. Decurtins, Chem. Commun. 2006, 1878–1880.
[11]
C. Jia, S.-X. Liu, C. Tanner, C. Leiggener, A. Neels, L. Sanguinet, E. Levillain, S.
Leutwyler, A. Hauser, S. Decurtins, Chem. Eur. J. 2007, 13, 3804–3812.
[12]
J. C. Wu, S.-X. Liu, T. D. Keene, A. Neels, V. Mereacre, A. K. Powell, S.
Decurtins, Inorg. Chem. 2008, 47, 3452–3459.
[13]
C. Goze, C. Leiggener, S.-X. Liu, L. Sanguinet, E. Levillain, A. Hauser, S.
Decurtins, ChemPhysChem 2007, 8, 1504–1512.
[14]
a) M. R. Grimmett, Comprehensive Heterocyclic Chemistry; A. R. Katrizky, C. W.
Rees, E. F. V. Scriven, Eds. Pergamon: Oxford, 1996; Vol. 3, 77–220; b) P. N. Preston,
M. F. G. Stevens, G. Tennant, Benzimidazoles and Congeneric Tricyclic Compounds,
Part 2; John Wiley & Sons: New York, 1980; c) G. Navarrete-Vazquez, R. Cedillo, A.
Hernandez-Campos, L. Yepez, F. Hernandez-Luis, J. Valdez, R. Morales, R. Cortes, M.
Hernandez, R. Castillo, Bioorg. Med. Chem. Lett. 2001, 11, 187–190.
[15]
S. Iwata, C. Ostermeier, B. Ludwig, H. Michel, Nature 1995, 376, 660–669.
[16]
a) L. J. Charbonnière, A. F. Williams, C. Piguet, G. Bernardinelli, E. Rivara-
Minten, Chem. Eur. J. 1998, 4, 485–493; b) L. Zhou, Y. Wang, C. Zhou, C. Wang, Q.
Shi, S. Peng, Cryst. Growth Des. 2007, 7, 300–306; c) Y. Morita, T. Murata, S. Yamada,
!
162
M. Tadokoro, A. Ichimura, K. Nakasuji, J. Chem. Soc., Perkin Trans. 2002, 1, 2598–
2600.
[17]
a) D. Luneau, P. Rey, Mol. Cryst. Liq. Cryst. 1995, 273, 81–87; b) J. R. Ferrer,
P. M. Lahti, C. George, G. Antorrena, F. Palacio, Chem. Mater. 1999, 11, 2205–2210.
[18]
a) T. Akutagawa, T. Hasegawa, T. Nakamura, T. Inabe, G. Saito, Chem. Eur. J.
2002, 8, 4402–4411; b) Y. Morita, T. Murata, K. Fukui, S. Yamada, K. Sato, D. Shiomi,
T. Takui, H. Kitagawa, H. Yamochi, G. Saito, K. Nakasuji, J. Org. Chem. 2005, 70,
2739–2744; c) T. Murata, Y. Morita, K. Nakasuji, Tetrahedron 2005, 61, 6056–6063; d)
T. Akutagawa, G. Saito, M. Kusunoki, K.-I. Sakaguchi, Bull. Chem. Soc. Jpn. 1996, 69,
2487–2511.
[19]
a) Q. Zeng, P. Cai, Z. Li, J. Qin, B. Zh. Tang, Chem. Commun. 2008, 1094–1096;
b) A. J. Boydston, P. D. Vu, O. L. Dykhno, V. Chang, A. R. Wyatt, A. S. Stockett, E. T.
Ritschdorff, J. B. Shear, C. W. Bielawski, J. Am. Chem. Soc. 2008, 130, 3143–3156; c)
L. De La Durantaye, T. McCormick, X.-Y. Liu, S. Wang, Dalton Trans. 2006, 5675–
5682.
[20]
a) S.-X. Liu, S. Dolder, P. Franz, A. Neels, H. Stoeckli-Evans, S. Decurtins, Inorg.
Chem. 2003, 42, 4801–4803; b) C. Jia, S.-X. Liu, C. Ambrus, A Neels, G. Labat, S.
Decurtins, Inorg. Chem. 2006, 45, 3152–3154; c) J. Wu, S.-X. Liu, A. Neels, F. Le Derf,
M. Sallé, S. Decurtins, Tetrahedron 2007, 63, 11282–11286.
[21]
a) J. S. Casas, A. Castiñeiras, E. García-Martínez, Y. Parajóa, M. L. Pérez-
Parallé, A. Sánchez-González, J. Sordo, Z. Anorg. Allg. Chem. 2005, 631, 2258–2264;
b) S.-M. Yue, Zh.-M. Su, J.-F. Ma, Y. Liao, Y.-H. Kan, H.-J. Zhang, Jiegou Huaxue
2003, 22, 174–178.
[22]
a) After the addition of a large excess of HCl (300 equiv) for half an hour, the two
reversible redox waves were only observed during the potenial scans with a freshly
polished electrode, and also a concomitant color change occurred. Subsequent multiple
cycling of the electrode potential within the measured window and neutralization did not
reveal any redox features, indicative of the ocurrence of adsorption and decomposition
of the TTF compound under strong acidic conditions. b) Q. Y. Zhu, Y. Liu, W. Lu, Y.
Zhang, G. Q. Bian, G. Y. Niu, J. Dai Inorg Chem. 2007, 46, 10065-10070
!
163
[23]
H.-H. Lin, Z.-M. Yan, J. Dai, D.-Q. Zhang, J.-L. Zuo, Q.-Y. Zhu, D.-X. Jia, New J.
Chem. 2005, 29, 509–513.
[24]
C. J. Chang, C. H. Yang, K. Chen, Y. Chi, C. F. Shu, M. L. Ho, Y. S. Yeh, P. T.
Chou, Dalton Trans. 2007, 1881–1890.
[25]
a) J. Smets, W. McCarthy, G. Maes, L. Adamowicz, J. Mol. Struct. 1999, 476,
27–43; b) M. Novo, M. Mosquera, F. R. Prieto, Can. J. Chem. 1992, 70, 823–827.
[26]
G. M. Sheldrick, Acta Cryst. 2008, A64, 112–122.
[27]
G. M. Sheldrick, SHELXL-97, Program for crystal structure refinement, University
of Göttingen: Göttingen, Germany, 1997.
!
164
7. A Compactly Fused !-Conjugated TetrathiafulvalenePerylenediimide Donor-Acceptor Dyad
Published in Organic Letters, 11 (14): 3096-3099, 2009
Michael Jaggi,† Carmen Blum,† Nathalie Dupont,‡ Jakob Grilj,‡ Shi-Xia Liu,*,† Jürg
Hauser,† Andreas Hauser,‡ Silvio Decurtins†
†
Departement für Chemie und Biochemie, Universität Bern, Freiestrasse 3, CH-3012
Bern, Switzerland.
‡
Département de chimie physique, Université de Genève, 30 quai Ernest-Ansermet,
CH-1211 Genève 4, Switzerland.
!
165
Abstract
The
synthesis
and
structural
characterization
of
a
tetrathiafulvalene
fused
perylenediimide molecular dyad is presented. Its largely extended !-conjugation
provides intense optical absorption bands over a wide spectral range. The planar
functional molecule exhibits a short-lived non-luminescent excited state attributed to
intramolecular charge separation.
!
166
CONTENT
7.1. References
175
!
7.2. Supporting Information:
178
!
7.3. Experimental section
!
179
7.3.1. Materials
179
7.3.2. Physical measurements
179
7.3.3. Preparation of 1
180
7.3.4. 1H NMR Spectrum of 1 in CDCl3
181
167
This paper puts forward a study of a to a large extent !-conjugated molecule, which
results from fusing tetrathiafulvalene (TTF) and perylenediimide (PDI) into a single
planar molecular structure. Both components of this fused molecule possess their
distinct and characteristic properties: the objective of this study is to assess the effects
of their mutual electronic interactions.
TTF and its derivatives are used as strong !-donors in the field of organic conductors
and superconductors,1 and as electron donor units in donor–acceptor (D–A) ensembles,
which are of interest due to their potential applications in molecular electronics and
optoelectronics.2 In such molecular ensembles, donors and acceptors are often
covalently linked by flexible or rigid "-spacers (D-"-A) which keep the moieties apart
from each other. Only few examples of fused D–A systems have been reported in the
literature,3 for instance the TTF donor with TCNQ-type bithienoquinoxaline,3b
hexaazatriphenylene,3c or phthalocyanine3d acceptors. The first case is a striking
example of a fused !-conjugated molecule that exhibits electrochemically amphoteric
properties due to a combination of a high-lying HOMO located on D with a low-lying
LUMO confined on A (Eox2Ered = 0.52 V). The annulation of donors and acceptors into a
planar configuration facilitates photoinduced intramolecular charge-transfer (CT)
processes, a topic that is in the focus of the present study. For more information,
especially on conjugated TTF–acceptor systems, the reader may consult a recent
review by Wudl et al.2a
The interest in perylenediimide (PDI) compounds is due to the unique combination of
good electron accepting ability, high absorption in the visible, and outstanding chemical,
thermal and photochemical stability.4 They have thus been used in a variety of D–A
systems showing photoinduced electron or energy transfer processes, potentially
leading to long-lived charge separated states. Their extended aromaticity together with
the possibilities for functionalization render them good candidates for potential
applications in electronic materials, sensors and photovoltaics.4-6 Examples abound of
the different PDI containing molecular ensembles, including combinations with
[60]fullerenes,5a,b phthalocyanines,5c,d,e porphyrins,5f,g corroles5h and TTFs.6 As
mentioned above, in most cases, these donors and the PDI acceptor units are linked
through flexible spacers. Specially for all reported TTF–PDI dyads,6 the moieties were
connected by #-linkers either to the “imide” region or the “bay” region of PDI, thus giving
!
168
rise to only weak electronic interactions between the two components. In contrast, the
herein reported TTF–PDI compound 1 (Scheme 1), which results from the annulation of
sulfur- and nitrogen-containing polycycles, allows an experimental study of intense
optical CT absorptions and photoinduced CT processes between the fused fragments of
the !-extended polycyclic system as a result of significant intra-molecular electronic
interactions.
Direct condensation of N-(1-octylnonyl)perylene-3,4,9,10-tetracarboxylic acid 3,4anhydride-9,10-imide
(2)7
with
5,6-diamino-2-(4,5-bis(propylthio)-1,3-dithio-2-
ylidene)benzo[d]-1,3-dithiole (3)3e afforded 1 in 54% yield. Its molecular structure was
confirmed by spectroscopic data (NMR, MS, IR), and for the first time, a TTF-PDI dyad
was elucidated by X-ray structure analysis.
Scheme 1. The synthetic route to 1.
Dyad 1 crystallizes as solvate-free purple plates in the triclinic space group
P1.
Its
molecular structure is shown in Figure 1. The skeleton of the !-conjugated molecule is
!
nearly planar and exhibits only a slight undulation along its long molecular axis; the rms
deviation from a least-squares plane through all atoms, excluding the alkyl and thioalkyl
groups, is 0.1596 Å, and for the TTF core alone it amounts to 0.1001 Å, which reflects
its slight boat conformation as observed in other related D-A systems.8 Due to crystal
packing effects, the alkyl substituents on PDI are arranged distinctly in an out-of-plane
!
169
conformation. The bond lengths within the TTF and PDI units are in good agreement
with those of their corresponding neutral species.9
In the crystal lattice, the molecules are stacked in a head-to-tail manner, thus forming a
trans-cofacial mode of association (see Supporting Information). This parallel alignment
is mainly caused by !(((! interactions: the two interplanar separations between the leastsquares planes amount to 3.565 and 3.679 Å; the shortest contact amounts to 3.480(5)
Å between S18B and C31. There is no short S(((S contact less than 4 Å. Some
unconventional intermolecular hydrogen bonds C-H(((O show up at 3.221(6) and
3.277(8) Å. Due to the asymmetry of the single molecular dyad with respect to its short
molecular axis, a positional disorder (1:1) occurs in the crystal lattice.
Figure 1. An ORTEP (30% probability ellipsoids) structure of 1 (one of the two
positionally disordered molecules). Hydrogen atoms and disordered parts of the
terminal alkyl groups are omitted for clarity.
The electrochemical properties of the molecular dyad 1 and of the reference compounds
2 and 3 in CH2Cl2 were investigated by cyclic voltammetry (CV). As shown in Figure 2,
compound 1 undergoes two reversible one-electron oxidations for the successive
generation of the TTF·+ radical cation and the TTF2+ dication, as well as two reversible
one-electron reductions corresponding to the consecutive reduction processes of the
PDI moiety. Compared to the corresponding reference compounds 2 and 3, the
!
170
oxidation waves are substantially positive-shifted (1: E1/2ox1 = 0.58 V, E1/2ox2 = 1.00 V; 3:
E1/2ox1 = 0.38 V, E1/2ox2 = 0.76 V), while only the first reduction wave is slightly negativeshifted (1: E1/2red1= –0.52 V, E1/2red2 = –0.67 V; 2: E1/2red1 = –0.45 V, E1/2red2 = –0.68 V).
This observation reflects the electronic interaction between D and A within the fused
system: the HOMO of 1 is energetically lowered and its LUMO is energetically raised
through chemically bonding 2 to 3.
Figure 2. Cyclic voltammograms of 1 (black line), 2 (red line), and 3 (blue line) in
CH2Cl2; 0.1 M TBAPF6 (TBA = tetrabutylammonium); on platinum electrode; scan rate
0.1 V·s–1 for the reference compounds 2 and 3, 0.5 V·s–1 for 1.
The molecular dyad 1 strongly absorbs in the green spectral region as evidenced by its
deep purple color. The UV-Vis-NIR spectrum of 1 (Figure 3a) recorded in CH2Cl2 shows
a strong and broad absorption band from 12000 cm-1 up to 17000 cm-1, followed by a
very intense and structured band that peaks at 18030 cm-1, with a clear progression in a
1250 cm-1 vibrational mode. Further strong absorptions appear above 23000 cm-1. By
comparison with the spectra of compounds 2 and 3, the new electronic transitions which
can only be observed in the fused molecule 1, are attributed to intramolecular !!*
charge-transfer (ICT) transitions from the TTF unit to the PDI core while the intense
!
171
absorption band of 1 at 18030 cm-1 reflects the !!* orbital excitation corresponding to a
PDI-localized transition. Compared to the unsubstituted PDI 2, which peaks at 19340
cm-1 and exhibits also the vibrational progression, the red-shift of 1310 cm-1 is due to
the extension of the !-conjugation of the PDI unit in 1. The electrochemical HOMOLUMO gap (1.1 eV) is in good agreement with the energy of the lowest ICT excited
state (for the relevant HOMO and LUMO orbitals, see Supporting Information).
!
172
Figure 3. a) Electronic absorption spectrum of 1 (black line) together with those of the
reference compounds 2 (red line) and 3 (blue line), in CH2Cl2 (c = 2.10-5 M) at room
temperature; b) Electronic absorption spectra of 1 upon oxidation with different
equivalents of [Fe(bpy)3]3+, corrected for the absorption of [Fe(bpy)3]2+; c) Transient
absorption spectrum of 1 upon pulsed excitation at 400 nm.
!
173
The absorption spectrum of 1 depends on the oxidation state. Figure 3b shows its
evolution upon chemical oxidation using [Fe(bpy)3]3+ as oxidizing agent. The TTF unit
can be oxidized to the TTF•+ radical as borne out by the isosbestic points at 15000 and
24800 cm-1 and the similarity to the final spectrum obtained upon electrochemical
oxidation (see Supporting Information). Whereas the PDI centered !!* transition is not
much affected by the oxidation, the ICT band at 15000 cm-1 of the neutral form of 1
decreases in intensity and is replaced by a new band centered at 12000 cm-1 for 1·+.
Based on the previous observations,3c,3e this new transition can be ascribed either to an
ICT, however, now in the opposite direction to that of the neutral compound 1, that is,
PDI!TTF·+ or to an alkylthio-TTF·+ transition corresponding to the promotion of an
electron from the fully occupied HOMO-1 to the SOMO.10
In contrast to the TTF fused dppz molecule,3e TTF-PDI in its neutral form does not show
any luminescence, neither from the !!* transition on PDI, known to fluoresce strongly in
the absence of quenchers, nor from the ICT transition itself. The former is reductively
quenched by the TTF unit. The latter would appear strongly Stokes shifted in the near
infrared. The transient absorption spectra obtained upon pulsed laser excitation at 400
(Figure 3c) and 650 nm show that the excited state returns to the ground state within
approximately 10 ps (see Supporting Information). With a radiative lifetime of the lowest
energy ICT state of ~5 ns estimated from the oscillator strength of the corresponding
transition, a maximum luminescence quantum yield of < 0.2% would result. This is too
weak for reliable detection in the near infrared. The transient absorption shows
bleaching of the !!* transition on the PDI as well as of the ICT transitions and a
transient absorption around 13000 cm-1. The former indicates that in the excited state
the PDI unit is strongly perturbed, the latter is characteristic for the presence of a TTF•+
radical. Finally, chemical oxidation of 1 does not restore the PDI centered
luminescence. This is in contrast to TTF-fused phthalocyanine,3d for which the
phthalocyanine based fluorescence could be restored upon oxidation of the TTF. The
key difference is that for PDI, the !!* transition is at higher energy and thus its
luminescence is quenched by either oxidative electron transfer or energy transfer.
In conclusion, we have demonstrated that a redox-active TTF unit can be annulated to a
PDI core, giving rise to a large extended !-conjugation. Therefore, this new dyad 1
shows intense optical absorbances over a wide spectral range. In particular, the new
!
174
molecule also combines in a complementary manner the functional properties of PDI
with the strong electron donating properties of TTF.
Acknowledgment This work was supported by the Swiss National Science Foundation
(grant No. 200020-116003 and 200020-115867). We thank Prof. Peter Bigler for his
assistance in the NMR analysis.
7.1. References
(1) (a) Segura, J. L.; Martín, N. Angew. Chem. Int. Ed. 2001, 40, 1372. (b) Yamada, J.;
Sugimoto, T. TTF Chemistry. Fundamentals and applications of Tetrathiafulvalene,
Springer, Berlin, 2004.
(2) (a) Bendikov, M.; Wudl, F.; Perepichka, D. F. Chem. Rev. 2004, 104, 4891. (b)
Tsiperman, E.; Becker, J. Y.; Khodorkovsky, V.; Shames, A.; Shapiro, L. Angew. Chem.
Int. Ed. 2005, 44, 4015. (c) Díaz, M. C.; Illescas, B. M.; Martín, N.; Perepichka, I. F.;
Bryce, M. R.; Levillain, E.; Viruela, R.; Ortí, E. Chem. Eur. J. 2006, 12, 2709. (d) Wu, JC.; Liu, S-X.; Neels, A.; Le Derf, F.; Sallé, M.; Decurtins, S. Tetrahedron 2007, 63,
11282.
(3) (a) Gautier, N.; Dumur, F.; Lloveras, V.; Vidal-Gancedo, J.; Veciana, J.; Rovira, C.;
Hudhomme, P. Angew. Chem. Int. Ed. 2003, 42, 2765. (b) Guégano, X.; Kanibolotsky,
A. L.; Blum, C.; Mertens, S. F. L.; Liu, S-X.; Neels, A.; Hagemann, H.; Skabara, P. J.;
Leutwyler, S.; Wandlowski, T.; Hauser, A.; Decurtins, S. Chem. Eur. J. 2009, 15, 63. (c)
Jia, C-Y.; Liu, S-X.; Tanner, C.; Leiggener, C.; Sanguinet, L.; Levillain, E.; Leutwyler, S.;
Hauser, A.; Decurtins, S. Chem. Commun. 2006, 1878. (d) Loosli, C.; Jia, C-Y.; Liu, SX.; Haas, M.; Dias, M.; Levillain, E.; Neels, A.; Labat, G.; Hauser, A.; Decurtins, S. J.
Org. Chem. 2005, 70, 4988. (e) Jia, C-Y.; Liu, S-X.; Tanner, C.; Leiggener, C.; Neels,
A.; Sanguinet, L.; Levillain, E.; Leutwyler, S.; Hauser, A.; Decurtins, S. Chem. Eur. J.
2007, 13, 3804.
!
175
(4) (a) Würthner, F. Chem. Commun. 2004, 1564. (b) Ahrens, M. J.; Sinks, L. E.;
Rybtchinski, B.; Liu, W.; Jones, B. A.; Giaimo, J. M.; Gusev, A. V.; Goshe, A. J.; Tiede,
D. M.; Wasielewski, M. R. J. Am. Chem. Soc. 2004, 126, 8284. (c) Langhals, H. Helv.
Chim. Acta 2005, 88, 1309. (d) Wasielewski, M. R. J. Org. Chem. 2006, 71, 5051. (e)
Osswald, P.; Würthner, F. Chem. Eur. J. 2007, 13, 7395. (f) Feng, J.; Liang, B.; Wang,
D.; Xue, L.; Li, X. Org. Lett. 2008, 10, 4437. (g) Fron, E.; Schweitzer, G.; Osswald, P.;
Würthner, F.; Marsal, P.; Beljonne, D.; Müllen, K.; DeSchryver, F. C.; Van der Auweraer,
M. Photochem. Photobiol. Sci. 2008, 7, 1509.
(5) (a) Baffreau, J.; Leroy-Lhez, S.; Gallego-Planas, N.; Hudhomme, P. J. Mol. Struct.
THEOCHEM 2007, 815, 145. (b) Shibano, Y.; Umeyama, T.; Matano, Y.; Tkachenko, N.
V.; Lemmetyinen, H.; Imahori, H. Org. Lett. 2006, 8, 4425. (c) Liu, M. O.; Tai, C. H.; Hu,
A. T. J. Photochem. Photobiol. A: Chem. 2004, 165, 193. (d) Fukuzumi, S.; Ohkubo, K.;
Ortiz, J.; Gutiérrez, A. M.; Fernandez-Lazaro, F.; Sastre-Santos, A. J. Phys. Chem. A
2008, 112, 10744. (e) Jiménez, A. J.; Spänig, F.; Rodriguez-Morgade, M. S.; Ohkubo,
K.; Fukuzumi, S.; Guldi, D. M.; Torres, T. Org. Lett. 2007, 9, 2481. (f) Ghirotti, M.;
Chiorboli, C.; You, C-C.; Würthner, F.; Scandola, F. J. Phys. Chem. A 2008, 112, 3376.
(g) Mathew, S.; Johnston, M. R. Chem. Eur. J. 2009, 15, 248. (h) Flamigni, L.; Ventura,
B.; Tasior, M.; Becherer, T.; Langhals, H.; Gryko, D. T. Chem. Eur. J. 2008, 14, 169.
(6) (a) Zhang, Y.; Cai, L-Z.; Wang, C-Y.; Lai, G-Q.; Shen, Y-J. New J. Chem. 2008, 32,
1968. (b) Gómez, R.; Coya, C.; Segura, J. L. Tetrahedron Lett. 2008, 49, 3225. (c)
Leroy-Lhez, S.; Baffreau, J.; Perrin, L.; Levillain, E.; Allain, M.; Blesa, M-J.; Hudhomme,
P. J. Org. Chem. 2005, 70, 6313. (d) Zheng, X.; Zhanga, D.; Zhu, D. Tetrahedron Lett.
2006, 47, 9083. (e) Zhang, Y.; Xu, Z.; Cai, L.; Lai, G.; Qiu, H.; Shen, Y. J. Photochem.
Photobiol. A: Chem. 2008, 200, 334. (f) Leroy-Lhez, S.; Perrin, L.; Baffreau, J.;
Hudhomme, P. C.R. Chimie. 2006, 9, 240.
(7) Wescott, L. D.; Mattern, D. L. J. Org. Chem. 2003, 68, 10058.
(8) Wu, J-C.; Dupont, N.; Liu, S-X.; Neels, A.; Hauser, A.; Decurtins, S. Chem. Asian. J.
2009, 4, 392.
(9) (a) Liu, S-X.; Dolder, S.; Franz, P.; Neels, A.; Stoeckli-Evans, H.; Decurtins, S. Inorg.
Chem. 2003, 42, 4801. (b) Jia, C.; Liu, S-X.; Ambrus, C.; Neels, A.; Labat, G.;
Decurtins, S. Inorg. Chem. 2006, 45, 3152. (c) Briseno, A. L.; Mannsfeld, S. C. B.;
!
176
Reese, C.; Hancock, J. M.; Xiong, Y.; Jenekhe, S. A.; Bao, Z.; Xia, Y. Nano Lett. 2007,
7, 2847. (d) Mizuguchi, J. Acta Cryst. 1998, C54, 1479.
(10) Li, H.; Lambert, C. Chem. Eur. J. 2006, 12, 1144.
!
177
7.2. Supporting Information:
Table of Contents
Experimental section
S2
Procedure for preparation of compound 1
S3
1
S4
H NMR Spectrum of compound 1
Crystal packing of 1
S5
Selected molecular orbitals for 1
S6
Electronic absorption spectra of 1 upon electrochemical oxidation
S7
Decay curves in transient absorption upon pulsed irradiation at 400 nm
S7
!
178
7.3. Experimental section
7.3.1. Materials
!
The compounds N-(1-octylnonyl)perylene-3,4,9,10-tetracarboxylic acid 3,4-anhydride9,10-imide 27 and 5,6-diamino-2-(4,5-bis(propylthio)-1,3-dithio-2-ylidene)benzo[d]-1,3dithiole 33e were prepared according to the literature procedures. All other chemicals
and solvents were purchased from commercial sources and were used without further
purification.
7.3.2. Physical measurements
1
H NMR spectrum was measured at 400 MHz at 50°C. Chemical shifts ) were
calibrated against TMS as an internal standard. FT-IR data were collected on a PerkinElmer Spectrum One spectrometer. UV-vis spectra were recorded on a Perkin-Elmer
Lambda 900 spectrometer. Elemental analyses were performed on an Carlo Erba EA
1110 CHNS apparatus. Cyclic voltammetry was conducted on a VA-Stand 663
electrochemical analyzer. An Ag/AgCl electrode containing 2 M LiCl (in ethanol) served
as reference electrode, a glassy carbon electrode as counter electrode, and a Pt tip as
working electrode. Cyclic voltammetric measurements were performed at room
temperature under N2 in CH2Cl2 (5.10-4 M) with 0.1 M n-Bu4NPF6 as supporting
electrolyte. Mass spectrum was recorded on a FTMS 4.7T BioAPEX II for MALDI
ionisation method.
!
179
7.3.3. Preparation of 1
!
A mixture of N-(1-octylnonyl)perylene-3,4,9,10-tetracarboxylic acid 3,4-anhydride-9,10imide
2
(280
mg,
445
µmol),
5,6-diamino-2-(4,5-bis(propylthio)-1,3-dithio-2-
ylidene)benzo[d]-1,3-dithiole 3 (117 mg, 270 µmol) and imidazole (700 mg, 10.3mmol)
in pyridine (30 mL) was refluxed for 21 h at 135°C under N2. After evaporation of
pyridine, the purple residue was subjected to column chromatography on silica gel
(CHCl3 then 9:1 CHCl3:MeOH) to yield the crude product. This material was
rechromatographed eluting initially with 8:2 CHCl3:CH2Cl2, then with pure CHCl3 and
finally with 9:1 CHCl3:MeOH to afford analytically pure compound 1 as a deep-purple
solid. Yield 54% (150 mg, 146 µmol). T > 223°C (dec.). Anal. Calc. for C57H59N3O3S6: C,
66.69; H, 5.79; N, 4.1. Found: C, 66.71; H, 5.97; N, 3.9% . 1H NMR (CDCl3, 50°C) )
8.76 (d, J = 8.24 Hz, 2H), 8.58 (m, 6H), 8.22 (s, 1H), 7.50 (s, 1H), 5.11 (br, 1H), 2.70 (t,
4H), 2.17 (br, 2H), 1.83 (br, 2H), 1.59 (m, 4H), 1.19 (m, 24H), 0.95 (t, 6H), 0.77 (m, 6H).
Selected IR data (cm-1, KBr pellet): 3436 (br), 2922 (s), 2851 (m), 1693 (s), 1655 (s),
1591 (s), 1501 (w), 1437 (m), 1353 (m), 1340 (s), 1238 (m), 805 (m), 738 (m). MS
(MALDI-TOF, DCTB as matrix, positive) calcd. for C57H59N3O3S6 1025.29, found
1025.29.
Single crystals suitable for X-ray analysis were obtained by slow evaporation of a
CH2Cl2 solution (~150 µg/mL) of 1.
!
180
7.3.4. 1H NMR Spectrum of 1 in CDCl3
!
181
Figure S1. Crystal packing of 1 showing the hydrogen bondings (dashed lines) within the
dimers.
The DFT method was employed with the B3LYP functional and the TZVP (valence
triple-zeta plus polarization) basis set. The calculations were done with the
TURBOMOLE V5.10 program package. The ground-state molecular geometry of 1 was
constrained to have Cs symmetry.
!
182
Figure S2. Selected molecular orbitals for 1.Ref
Ref: (a) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102, 346. (b) Bauernschmitt, R.;
Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454. (c) Bauernschmitt, R.; Häser, M.;
Treutler, O.; Ahlrichs, R. Chem. Phys. Lett. 1997, 264, 573.
!
183
Figure S3. Electronic absorption spectra of 1 upon electrochemical (V = 0.8 V vs.
Ag/AgCl) oxidation.
Figure S4. Decay curves in transient absorption upon pulsed irradiation at 400 nm,
detected at different wavelengths. The decay is non-exponential with a first time
constant of 1 ps and a second time constant of 10 ps.
!
184
8. A Donor–Acceptor Tetrathiafulvalene Ligand Complexed to Iron(II)
or Cobalt(II): Synthesis, Electrochemistry and Spectroscopy of
[M(phen)2(TTF-dppz)](PF6)2
To be submitted to the Journal of Physical Chemistry B
!
!
Nathalie Dupont,a Ying-Fen Ran,b Shi-Xia Liu,b Jakob Grilj,a Silvio Decurtins,b Andreas
Hauser a
a
Département de Chimie Physique, Université de Genève, 30 Quai Ernest-Ansermet,
CH-1211 Genève 4, Switzerland
b
Departement
für
Chemie
und
Biochemie,
CH-3012 Bern, Switzerland.
!
185
Universität
Bern,
Freiestrasse
3,
CONTENT
8.1. Introduction
187
!
8.2. Experimental Section
188
8.2.1. Synthesis of [Co(phen)2(TTF-dppz)](PF6)2.
189
8.2.2. Synthesis of [Fe(phen)2(TTF-dppz)](PF6)2.
189
8.2.3. Methods
190
!
8.3. Results and Discussion
191
8.3.1. Electrochemical properties
191
!
8.4. Photophysical properties
193
!
8.5. References
!
211
186
8.1. Introduction
!
Spin-crossover phenomena in first row transition metal complexes can be monitored by
a number of different techniques, and there has been a considerable amount of
research for various metal complexes of iron(II), iron(III) and cobalt(II) exhibiting spin
crossover between the low-spin (LS) and the high-spin (HS) isomers.1-3 The switch of
configuration within the partially occupied d shell can be obtained by a variation of
pressure, temperature or light irradiation.4-8 Spin crossover is one of the most famous
examples of bistability at the molecular level.6-8 Examples of spin crossover are less
abundant for cobalt(II) complexes than for iron(II) or iron(III) complexes.9 Fe2+ has a 3d6
electron configuration and Co2+ has a 3d7 configuration. In an octahedral complex, the d
orbitals are split into two subsets, eg and t2g, separated by the energy $0.4,6,8,10 This
energy separation depends on the environment of the metal ion. As a consequence,
Fe2+ in a complex can occur in two different spin states, that is, a low-spin state
described as 1A1g(t2g6 eg0), or a high-spin state described as 5T2g(t2g4 eg2). In the same
way, Co2+ in a complex can have a LS state described as 2E(t2g6 eg1) or a HS state
described as 4T1(t2g5 eg2). Fe2+ thus exhibits an S = 0 $ 2 spin change and Co2+ an S =
1/2 $ 3/2 spin change during the transition from the LS to the HS state.
TTF (tetrathiafulvalene) and its derivatives are well known as ! electron donors capable
of forming stable cation radical and dication species upon oxidation.11 As a
consequence, they are good candidates to be used as donor units in donor-acceptor
compounds, which are of considerable research interest due to their potential
applications in sensors, optoelectronics and molecular devices.12-18
TTF can be fused to an electron acceptor unit, for instance dppz (dipyrido-[3, 2-a: 2#, 3#c] phenazine). The latter has two low lying !* acceptor levels, one localized on the
phenazine subunit and one on the phenanthroline subunit of dppz.19 TTF fused to dppz
results in a TTF-dppz ligand with two redox centers and a lowest excited state
corresponding to an intraligand charge transfer (ILCT) from the TTF to the dppz
moieties. This ligand was studied on its own,13 as well as its ruthenium(II) complexes,
including the series [Ru(bpy)1-x(TTF-dppz)x]2+, x = 1-3.20 These complexes show
interesting photophysical properties with a long-lived, non-emissive charge transfer
state, in which an electron from one TTF moiety of one ligand is transferred to the dppz
!
187
moiety of another one. This charge separated state results from the initial excitation into
the spin-allowed metal to ligand charge transfer (MLCT) bands.
The combination of one TTF-dppz ligand and two phenanthroline (phen) units to form
either a cobalt(II) or an iron(II) complex gives rise to the complexes [M(phen)2(TTFdppz)]2+, M = Co2+, Fe2+ (Scheme 1). Their synthesis has been motivated by the search
for new antenna systems capable of charge separation,21 new photoredox switches
22
as well as the study of the properties in relation with spin crossover. So far, only few
papers on ruthenium(II) complexes with TTF derivatives have appeared in the literature
20,23-29
and none combining iron(II) or cobalt(II) to a TTF derivative. Herein, we report the
synthesis of cobalt(II) and iron(II) complexes as shown in Scheme 1 and thoroughly
examine their electrochemical and photophysical properties.
N
N
N
N
S
S
N
N
S
S
S
M 2+
N
S
N
Scheme 1: [M(phen)2(TTF-dppz)]2+, M2+ = Fe2+, Co2+
8.2. Experimental Section
Unless otherwise stated, all reagents were purchased from commercial sources and
used without additional purification. 4#,5#-Bis-(propylthio)tetrathiafulvenyl[i]dipyrido[3,2a:2#,3#-c]phenazine (TTFdppz),13 Fe(phen)2Cl2,30,31 Co(phen)2Cl2
32
were prepared
according to literature procedures. 1H NMR spectrum was obtained on a Bruker AC 300
spectrometer operating at 300.18 MHz: chemical shifts are reported in ppm referenced
!
188
to residual solvent protons (CDCl3). The following abbreviations were used: s (singlet), d
(doublet), t (triplet), m (multiplet) and br (broad). Infrared spectra were recorded on a
Perkin-Elmer Spectrum One FT-IR spectrometer using KBr pellets. Elemental analyses
were performed on a Carlo Erba Instruments EA 1110 Elemental Analyzer CHN. Mass
spectra were recorded using a LTQ Orbitrap XL for ESI.
8.2.1. Synthesis of [Co(phen)2(TTF-dppz)](PF6)2.
A solution of TTF-dppz (30 mg, 0.49 mmol) in CH2Cl2 (5 mL) was added to the solution
of Fe(phen)2Cl2 (24 mmg, 0.49 mmol) in ethanol (5 mL). The mixture was heated up to
80°C for 2 h. After cooling down to room temperature, aqueous potassium
hexafluorophosphate was added and the precipitate was immediately formed. The dark
blue powder was collected and washed with water and ethanol, and dried in vacuum.
Yield: 48.9 mg (75%). 1H NMR (CDCl3, 300 MHz): ) = 9.53 (d, 4H), 9.22 (d, 2H), 8.528.48 (m, 2H), 8.14 (t, J = 6.9 Hz, 2H), 8.07 (s, 2H), 8.00-7.91 (m, 4H), 7.78-7.77 (m,
8H), 2.83 (t, J = 7.2 Hz, 4H), 1.70-1.66 (m, 4H), 1.03 (t, J = 7.2 Hz, 6H). IR (KBr): " =
2958, 1630, 1426, 1356, 1212, 1122, 1088, 837, 722, 556 cm-1. ESI-MS: m/z: 511.05,
calcd for [M-2PF6]2+: 511.04. Anal. Calcd (%) for C52H38F12FeN8P2S6(2EtOH: C, 47.86;
H, 3.59; N, 7.97. Found: C, 48.21; H, 3.07; N, 7.55.
8.2.2. Synthesis of [Fe(phen)2(TTF-dppz)](PF6)2.
A solution of TTF-dppz (30 mg, 0.49 mmol) in CH2Cl2 (5 mL) was added to the solution
of Co(phen)2Cl2 (24.4 mmg, 0.49 mmol) in ethanol (5 mL). The mixture was heated to
80°C for 2 h. After cooling down to room temperature, aqueous potassium
hexafluorophosphate was added to the solution and the precipitate was immediately
formed. The dark blue powder was collected and washed with water and ethanol, and
!
189
dried in vacuum. Yield: 50.6 mg (78%). IR (KBr): " = 2960, 1625, 1425, 1358, 1216,
1120, 1088, 837, 725, 556 cm-1. ESI-MS: m/z: 512.54, calcd for [M-2PF6]2+: 512.56.
Anal. Calcd (%) for C52H38CoF12N8P2S6: C, 47.45; H, 2.91; N, 8.51. Found: C, 47.11; H,
3.23; N, 8.56.
8.2.3. Methods
Cyclic voltammetry was conducted on a VA-Stand 663 electrochemical analyzer. An
Ag/AgCl electrode containing 2 M LiCl served as reference electrode, a glassy carbon
electrode as counter electrode, and a Pt disk as working electrode. Cyclic voltammetric
measurements were performed at room temperature under N2 in CH2Cl2 with 0.1 M
Bu4NPF6 as supporting electrolyte at a scan rate of 100 mV(s-1.
Photophysical measurements were performed on solutions of the compounds in
CH3CN and CH2Cl2 at room temperature. Spectoelectrochemical measurements were
performed with a spectroelectrochemical cell designed by F. Hart
33
based on a
standard OTTLE cell and TBA(PF6) as conductive salt. For luminescence and transient
absorption measurements, the solutions were degassed by bubbling N2 through them
for 30 min. Absorption spectra were recorded on a Varian Cary 5000 UV/vis/near-IR
(NIR) spectrophotometer. Excited state decay curves on the nanosecond timescale
were recorded by exciting the samples at 532 nm using the second harmonic of a
pulsed Nd:YAG laser (Quantel Brilliant, 7ns pulse width). The probe light came from a
W-halogen lamp. The system used for detection consisted of a single monochromator
(Spex 270 M), a photomultiplier (Hamamatsu R928), and a digital oscilloscope
(Tektronix TDS 540B). The overall time resolution is ~15 ns.
The femtosecond transient absorption set-up has been described earlier elsewhere.34
Briefly, the output of a Ti:sapphire amplifier (Spitfire, Spectra Physics; 800 nm pulses of
150 fs FHWM) is split into two parts; ca. 5 µJ are focused into a 3 mm thick CaF2
window that is constantly moved to generate a white light continuum that is used for
probing. The remainder serves as pump and is either frequency doubled or sent into in
a home-built two stage non-collinear optical parametric amplifier to generate the pump
!
190
pulses at 500 or 650 nm. The pump beam has an intensity of ca. 0.1mJ/cm2 at the
sample, and its polarization is at magic angle with respect to the probe beam. The
sample solution is located in a 1 mm quartz cell and constantly stirred by nitrogen
bubbling. Its photochemical stability is verified by the steady state absorption. After the
sample, the probe beam is dispersed in a spectrograph (Andor, SLR163) and imaged
onto a 512+58 pixel back-thinned CCD (Hamamatsu S07030-09). The spectra are
corrected for the chirp of the white light pulses by standard recipes.35 The IRF has a
FHWM of approximately 200 fs. Due to cross-phase-modulation and the coherent
signal, the spectra around time zero cannot be observed.
8.3. Results and Discussion
8.3.1. Electrochemical properties
!
Figure
1
shows
the
cyclic
voltammograms
of
[Fe(phen)2(TTF-dppz)]2+
and
[Co(phen)2(TTF-dppz)]2+ in CH2Cl2. By comparison with the reference molecules in
Table 1, oxidation and reduction waves can be attributed for both complexes.
[Fe(phen)2(TTF-dppz)]2+ can be oxidized in three distinct reversible waves, at 0.80 V,
1.16 V and 1.35 V vs Ag/AgCl. The free TTF-dppz ligand has two reversible oxidation
waves at 0.73 and 1.08 V and [Fe(phen)3]2+ has an oxidation potential of 1.12 V, that is,
higher than the second oxidation wave of TTF-dppz. The two successive oxidation wave
of TTF can thus be attributed at 0.80 and 1.08 V in [Fe(phen)2(TTF-dppz)]2+, the third
oxidation wave being the one of Fe2+/3+. The reduction waves can be attributed to the
dppz subunit of the TTF-dppz ligand at -0.84 V, the reduction of the phen moieties being
known to be at approximately 300 mV more negative potential,36 here at -1.13 V. The
attribution of the reduction waves of [Co(phen)2(TTF-dppz)]2+ is very similar, with the
reduction
wave
of
the
dppz
subunit
at
-0.74V and the one of the phen subunit still approximately lower by 300 mV, here at 1.02 V. The oxidation waves can be attributed by comparison with the reference
molecules. [Co(phen)]2+ has a very low oxidation potential, at 0.42 V, lower than the fist
oxidation potential of TTF-dppz, which is at 0.73 V. In contrast to [Fe(phen)2(TTF!
191
dppz)]2+ , the first oxidation wave of [Co(phen)2(TTF-dppz)]2+ is thus the oxidation of
Co2+ to Co3+ at 0.66 V, followed by the two successive oxidations on the TTF moieties
at 0.84 V and 1.15 V.
Figure 1: Cyclic voltammograms of [Fe(phen)2(TTF-dppz)]2+ (1.10-4 M, red) and
[Co(phen)2(TTF-dppz)]2+ (1.10-4, black) in CH2Cl2; 0.1 M TBAPF6; on platinum
electrode at 100 mV(s-1.
!
192
Table 1: Redox potentials (V vs Ag/AgCl) of [Fe(phen)2(TTF-dppz)]2+ and
[Co(phen)2(TTF-dppz)]2+ in CH2Cl2 and of the reference compounds and complexes
TTF-dppz,20
[Fe(phen)3]2+,37
[Co(phen)3]2+,37
[Co(phen)2(dppz)]3+
38
and
[Ru(bpy)2(TTF-dppz)]2+.20
Oxidation
a
Compound
E1
E2
[Co(phen)]2+, a
0.42
[Fe(phen)3]2+, a
1.12
TTF-dppz
0.73
[Co(phen)2(dppz)]3+, a
0.44
[Ru(bpy)2(TTF-dppz)]2+
0.74
1.05
[Co(phen)2(TTF-dppz)]2+
0.66
[Fe(phen)2(TTF-dppz)]2+
0.80
Reduction
E3
E2
E3
-0.91
-1.17
-1.79
1.43
-0.91
-1.35
0.84
1.15
-0.74
-1.02
1.16
1.35
-0.84
-1.13
1.08
E1
-1.17
In MeCN, saturated calomel electrode (SCE) (in order to obtain values vs Ag/AgCl,
0.04 V were added)
8.4. Photophysical properties
!
The absorption spectrum of [Fe(phen)2(TTF-dppz)]2+ in CH2Cl2, together with those of
the
reference
compounds
and
complexes
TTF-dppz,
[Fe(phen)3]2+
and
[Fe(phen)2(dppz)]2+ are presented in Figure 2a. The broad absorption band centered at
17000 cm-1 corresponds to the intraligand charge transfer (1ILCT) transition occurring
on the TTF-dppz ligand, as shown previously for the series of ruthenium(II) complexes
!
193
[Ru(bpy)3-x(TTF-dppz)x]2+, x = 1-3,20 with the TTF subunit acting as an electron donor
and the dppz subunit acting as an electron acceptor. The 1ILCT absorption band is redshifted by 2500 cm-1 with respect to the free ligand TTF-dppz, corresponding to a
lowering of the energy of the LUMO located on the dppz subunit due to the coordination
to the iron(II). The extinction coefficient increases from 1x104 to 2.4x104 M-1cm-1 upon
coordination of one TTF-dppz and two phen subunits to iron(II). For coordination to
three diimine ligands, the iron(II) center is expected to have a low-spin ground state.39
By comparison with the reference complexes [Fe(phen)3]2+ and [Fe(phen)2(dppz)]2+,40,41
the characteristic absorption bands corresponding to the 1MLCT transitions of iron(II) in
the low-spin state are expected between 18000 and 26000 cm-1. Thus, the broadening
of the absorption band centered at 17000 cm-1 and the asymmetry towards higher
energy are due to the underlying presence of these 1MLCT bands. This is confirmed by
the
difference
absorption
spectrum
between
[Fe(phen)2(TTF-dppz)]2+
and
[Co(phen)2(TTF-dppz)]2+ shown in Figure 3, the spectrum of the cobalt complex for
which no intense bands apart from the 1ILCT band are expected in this spectral region
(see below) serving as reference. Indeed, this difference spectrum is almost identical to
the absorption spectrum of [Fe(phen)3]2+ with its dominant 1MLCT transition. Last but
not least, the weakly structured absorption band centered at 26500 cm-1 can be
assigned to the lowest energy !-!* transition located on the dppz subunit in TTF-dppz.
In the UV region, the broad band with several maxima and shoulders is attributed to the
!-!* transitions located on the different subunits of [Fe(phen)2(TTF-dppz)]2+. In contrast
to [Ru(bpy)2(TTF-dppz)]2+,29 no luminescence from the 1ILCT state of TTF-dppz could
be detected in the limit of the sensitivity of the experimental setup. [Fe(phen)2(TTFdppz)]2+ doesn#t show any luminescence from any other state either.
!
194
Figure 2: (a) Absorption spectrum of [Fe(phen)2(TTF-dppz)]2+ (purple) in comparison
with those of the reference compounds TTF-dppz (red), [Fe(phen)3]2+ (green),
[Fe(phen)2(dppz)]2+ (blue) in CH2Cl2. (b) Absorption spectrum of [Co(phen)2(TTFdppz)]2+ (purple) in comparison with those of the reference compounds, TTF-dppz
(red), [Co(phen)3]2+ (green), [Co(phen)2(dppz)]2+ (blue) in CH2Cl2.
!
195
Figure 3: Absorption spectra of [Fe(phen)2(TTF-dppz)]2+ (red), [Co(phen)2(TTFdppz)]2+ (green) and the deduced absorption spectrum of FeLS (purple) in comparison
with [Fe(phen)3]2+ (blue) in CH2Cl2 at room temperature. This deduced spectrum is
obtained by subtraction of the absorption spectrum of [Co(phen)2(TTF-dppz)]2+
(green) from the absorption spectrum of [Fe(phen)2(TTF-dppz)]2+ (red).
The absorption spectra of [Co(phen)2(TTF-dppz)]2+ in CH2Cl2 together with those of
TTF-dppz, [Co(phen)3]2+ and [Co(phen)2(dppz)]2+ are shown in Figure 2b. As briefly
mentioned above, the tris-diimine complexes of cobalt(II) are essentially high-spin at
room temperature in solution. In the high-spin state, MLCT transitions are usually very
weak,42 as exemplified by the absorption spectrum of [Co(phen)3]2+ included in Figure
2b, which despite the comparatively low redox potential doesn't show any strong
absorption bands below 27000 cm-1. Therefore the absorption band centered at 16500
cm-1 corresponds solely to the 1ILCT transition of the coordinated TTF-dppz. The higher
energy bands can be attributed to the different !-!* transitions located on the ligands in
analogy to the iron(II) complex. A very weak emission is observed for the cobalt(II)
complex from the 1ILCT state (Figure 4). That it is really due to the complex and not to
some impurity is borne out by the corresponding excitation spectrum, which perfectly
matches the absorption of the 1ILCT band.
!
196
Figure 4: Emission ( red, 4ex = 600 nm) and excitation spectra (green, 4em = 825 nm)
of [Co(phen)2(TTF-dppz)]2+ in comparison to its absorption spectrum (purple), in
CH2Cl2 at room temperature.
Upon electrochemical oxidation at 0.80 V vs Ag/AgCl, that is the first anodic peak
potential of [Fe(phen)2(TTF-dppz)]2+ in CH2Cl2, the 1ILCT absorption band located at
17000 cm-1 decreases in intensity and two new absorption bands appear at 12000 and
22300 cm-1, respectively (see Figure 5). In analogy to previous studies of TTF-dppz and
its complexes,29 the new absorption bands obtained after 20 minutes show the oxidation
of the TTF subunit of the complex, and, in particular, the band in the near IR is due to
the reverse ILCT transition dppz$TTF•+. This confirms the expected oxidation of the
TTF unit to the TTF•+ radical in the first oxidation wave of the CV. Furthermore, this
oxidation is fully reversible electrochemically also on a long time scale (see spectrum
after subsequent reduction at reduction at -0.84 V for for approximately 2 minutes
included in Figure 5).
The absorption spectra during electrochemical oxidation of [Co(phen)2(TTF-dppz)]2+ in
CH2Cl2 at 0.66 V vs Ag/AgCl, that is corresponding to the first anodic peak potential in
the CV, are shown in Figure 6. In contrast to the spectrum of the oxidized iron(II)
!
197
complex, no new band appears and only a slight red-shift is observed for the 1ILCT
absorption band. The absence of any of the characteristic bands of the TTF•+ radical
following oxidation confirms the tentative assignment of the first oxidation wave at E1/2 =
0.66 V in the CV of [Co(phen)2(TTF-dppz)]2+ to the oxidation of Co2+ to Co3+.
Furthermore the oxidation of the cobalt(II) compound is reversible electrochemically.
Figure 5: Absorption spectra of spectroelectrochemical oxidation of [Fe(phen)2(TTFdppz)]2+ in CH2Cl2, at room temperature, E = 0.80 V vs Ag/AgCl in comparison with
difference spectrum (black) and reversibility spectrum (doted line).
!
198
Figure 6: Absorption spectra of spectroelectrochemical oxidation of [Co(phen)2(TTFdppz)]2+ in CH2Cl2, at room temperature, E = 0.66 V vs Ag/AgCl.
In order to arrive at a better understanding of the photophysical behavior of the iron(II)
and cobalt(II) complexes, transient absorption spectra were recorded on different
timescales using different excitation wavelengths. Figure 7a shows the transient
absorption spectra of the reference compound [Fe(phen)2(dppz)]2+ on the ps timescale
upon excitation at 500 nm (20000cm-1), that is, into the 1MLCT absorption band of the
Fe$phen charge transfer. The transient signal essentially consists of a bleaching of the
MLCT band, indicating an ultrafast,43 light-induced population of the high-spin state,39
and according to the global fit shown in Figure 7b, it decays mono-exponentially with a
lifetime of ~1 ns corresponding to the HS$LS relaxation.
!
199
Figure 7: (a) Transient absorption obtained by ps spectroscopy of [Fe(phen)2(dppz)]2+
with 4ex= 500 nm, in CH2Cl2, in comparison with its steady state absorption spectrum
(black line). (b) Decomposition of the transient absorption obtained with a monoexponential global fit A(t) = A0 + A1exp(-t/'1). In (a) and (b), the spectral region around
the laser wavelength are cut.
!
200
Figure 8a shows the transient absorption spectra of [Fe(phen)2(TTF-dppz)]2+ also on the
ps timescale upon excitation at 650 nm (15385 cm-1), that is, into the 1ILCT absorption
band. The transient signal has a mono-exponential decay with ' = 35(3) ps (see result
of the global fit in Figures 8b and c). This transient spectrum of the [Fe(phen)2(TTFdppz)]2+ complex with its prominent absorption band at 20000 cm-1 is identical to the
one of a similar ruthenium complex,20 in which it has simply been attributed to the
transient population of the 1ILCT state itself. Formation of a radical species with TTF•+
can be excluded as, according to the results from spectro-electrochemistry and
transient absorption in an analogous ruthenium complex,44 this would result in transient
absorptions at 22000 and 12000 cm-1 of similar intensities. Thus we may conclude that
upon irradiation into the 1ILCT transition, this state decays in a radiationless process
with a lifetime of approximately 35 ps.
!
201
Figure 8: (a) Transient absorption obtained by ps spectroscopy of [Fe(phen)2(TTFdppz)]2+ with 4ex= 650 nm (excitation into the
1
ILCT band), in CH2Cl2. (b)
Decomposition of the transient absorption obtained with a mono-exponential global fit
A(t) = A0 + A1exp(-t//'). c) Decay curves
!
202
In Figure 9, transient absorption was performed on the ps timescale with an excitation
wavelength of 400 nm (25000 cm-1), that is, with irradiation into the high-energy wing of
the 1MLCT transition as well as other higher energy transitions of [Fe(phen)2(TTFdppz)]2+. In these spectra, the time evolution is clearly not single exponential. Initially,
there is a strong transient absorption band centered at 20000 cm-1, which decays quite
rapidly. At the same time, a bleaching occurs at 19500 cm-1 with a longer decay time.
The global fit with a double exponential, as shown in Figure 9b, clearly identifies two
time constants, namely a fast process with '1 = 30 ps and a slower process with '2 = 1
ns, and their corresponding amplitudes as a function of wavenumber. The wavenumber
dependence of the amplitude of the fast process is identical to the transient absorption
of the 1ILCT state observed upon irradiation at 650 nm; the wavenumber dependence of
the amplitude of the slow process corresponds to the transient bleaching observed in
the reference complex [Fe(phen)2(dppz)]2+ upon irradiation into the 1MLCT band. Thus
the faster process can be readily assigned the decay of the 1ILCT state as also found in
the transient absorption spectra of [Fe(phen)2(TTF-dppz)]2+ excited into the 1ILCT band.
A comparison with the transient absorption spectra of the reference compound
[Fe(phen)2(dppz)]2+ (Figure 7) demonstrates that the bleaching with a decay lifetime of
approximately 1 ns can be attributed to the HS$LS relaxation of iron(II) following the
ultrafast, light-induced population of the HS state. The various processes thus identified
are summarized in Figure 10. The first excited state is the corresponding HS state of
iron(II) with a lifetime of 1 ns. The 1ILCT state in turn has a lifetime of around 30 ps. This
is much shorter than the 400 ns in CH2Cl2, but clearly in the vicinity of the one of the
corresponding ruthenium complex,20 for which in the absence of any lower lying excited
stated, the reduction in lifetime was attributed solely to the increase in non-radiative
decay resulting from the lowering of the excited state energy. This is also the case here
as no noticeable population of the HS state upon irradiation into the 1ILCT state is
observed. This is not surprising as the ILCT state with the iron in the HS state would be
at higher energy and as there is no efficient mechanism for the 1ILCT state to decay
directly to the low-lying HS state. Upon irradiation into the MLCT bands, which does not
differentiate between charge transfer to phen or dppz, populations of both the ILCT
state as well as the iron(II) HS state are created within the first picoseconds and the two
populations essentially decay independent of each other.
!
203
Figure 9: (a) Transient absorption obtained by ps spectroscopy of [Fe(phen)2(TTFdppz)]2+ with 4ex= 400 nm (excitation into the MLCT band), in CH2Cl2. (b)
Decomposition of the transient absorption obtained with a bi-exponential fit A(t) = A0 +
A1exp(-t//'1) + A2exp(-t//'2). In (a), the spectral region around the laser is cut. (c)
Decay curves at different wavelengths according to the color scheme in the inset.
!
204
Figure 10: Energy level scheme for [Fe(phen)2(TTF-dppz)]2+
In Figures 11 and 12, show the transient absorption spectra for [Co(phen)2(TTF-dppz)]2+
recorded on the ps timescale upon excitation at 650 nm (15385 cm-1) and at 400 nm
(25000 cm-1), that is into the spin-allowed ILCT and the !-!* absorption band of the
dppz unit, respectively. For both transient absorption spectra, the initial bleaching of the
band at 17000 cm-1 and the comparatively strong transient absorption at 20000 cm-1 are
the signature of the ILCT state, by comparison with the transient spectrum of the iron(II)
complex for irradiation at 650 nm. Both experimental spectra are well described by
global fits with a double exponential, as shown in Figures 11b and 12b. Two different
processes on quite different timescales can be discerned: a fast process attributed to
the decay of the transient absorption from the ILCT with '1 = 3 - 5 ps based on the
corresponding amplitude spectrum (A1, blue line in Figures 11b and 12b), and a second
process with a longer decay lifetime of around 300 - 400 ps. The amplitude spectrum of
the latter process (A2, green line) with two absorption bands of almost equal intensity at
14000 and 21000 cm-1 is in line with the presence of a TTF•+ radical species. The key
difference between the transient spectra for excitation at 650 and 400 nm is the relative
amplitude of the two amplitude spectra. The one belonging to the longer decay process
is more intense for the excitation at 400 nm than at 650 nm. In addition, for excitation at
!
205
400 nm the transient spectrum after 1 ns shows a residual species (A0, black line in
Figure 11b). The decay of this residual species occurs on a much longer timescale as
shown by the decay curves in Figure 13 obtained for an excitation at 532 nm (18800
cm-1) and probe wavelengths of 470 nm (21280 cm-1) and 810 nm (12345 cm-1), as well
at 640 nm (15625 cm-1). The former show transient absorption, the latter transient
bleaching. Two time constants can be associated with these signals: '3 % 30 µs and '4 %
10 ms.
It is unfortunately not possible to assign all of these observations to specific processes
and meta-stable states. The scheme of possible excited states for [Co(phen)2(TTFdppz)]2+ elaborated in Figure 14 shows a daunting complexity. There being no doubt
about the HS ground state of the complex in solution at room temperature, the
corresponding LS state must nevertheless be quite close in energy.42 Based on the
redox potentials given in table 1, MLCT states have to be situated at approximately 1.4
eV. These correspond to the oxidation of Co2+HS to Co3+LS, as in the oxidized form
Co3+LS is the stable spin-state. This state is not accessible optically but it is at lower
energy than the optical ILCT state at approximately 1.6 eV. In the optical ILCT state,
Co2+ is in the HS state. However, this is not necessarily the lowest excited ILCT state.
With the electron on dppz, the ligand-field strength is substantially larger than in the
ground state. As a result, the corresponding LS state could become sufficiently
stabilized to actually drop below the HS ILCT state. Next in the energetic order comes
the charge separated state written as [Co2+(phen-)(phen)(TTF•+-dppz)]2+ estimated to be
at 1.7 - 2.0 eV again with Co2+ in the HS and in the LS state in a priory unknown order.
The MLCT states of the type Co2+HS/Co3+HS(phen-) or (dppz-), which in principle would
be optically accessible albeit with small extinction coefficients, are estimated to be at 2 2.5 eV, and finally the !!* transition on dppz is at around 3 eV. Even though
experiments suggest the existence of a species with a TTF•+ radical, this is more likely
to be due to a bimolecular process based on the time scale of µs to ms of the
corresponding decay processes. Further experiments with time-resolved spectroscopy
in conjunction with electrochemical or chemical oxidation to form the Co3+ species or a
sacrificial electron donors will be needed to further our understanding of this system.
!
206
Figure 11: (a) Transient absorption obtained by ps spectroscopy of [Co(phen)2(TTFdppz)]2+ with 4ex= 650 nm (excitation in the ILCT), in CH2Cl2. (b) Decomposition of the
transient absorption obtained with a bi-exponential fit. (c) Decay curves.
!
207
Figure 12: (a) Transient absorption obtained by ps spectroscopy of [Co(phen)2(TTFdppz)]2+ with 4ex= 400 nm (excitation in the ILCT), in CH2Cl2. (b) Decomposition of the
transient absorption obtained with a bi-exponential fit. (c) Decay curves.
!
208
Figure 13: (a) Excited state lifetime spectra of [Co(phen)2(TTF-dppz)]2+ obtained by
ns spectroscopy of with 4ex= 532 nm, in CH2Cl2.
!
209
Figure 14: Energy level scheme for [Co(phen)2(TTF-dppz)]2+
!
210
8.5. References
!
(1)
Spin Crossover in Transition Metal Compounds I; Topics in Current
Chemistry Vol. 233 (Gutlich, P.; Goodwin, H. A.; eds) Springer: Berlin, 2004
(2)
Spin Crossover in Transition Metal Compounds II; Topics in Current
Chemistry Vol. 234 (Gutlich, P.; Goodwin, H. A.; eds) Springer: Berlin, 2004
(3)
Spin Crossover in Transition Metal Compounds III; Topics in Current
Chemistry Vol. 235 (Gutlich, P.; Goodwin, H. A.; eds) Springer: Berlin, 2004
(4)
Gütlich, P. Struct. Bond. 1981, 44 83.
(5)
König, E. Struct. Bond. 1991, 76, 51.
(6)
Kahn, O. Molecular Magnetism; VCH Publishers: New York, 1993.
(7)
Gütlich, P.; Hauser, A. Coord. Chem. Rev. 1990, 97, 1.
(8)
Gütlich, P.; Hauser, A.; Spiering, H. Angew. Chem.-Int. Ed. Eng. 1994, 33,
(9)
Zarembowitch, J. New J. Chem. 1992, 16, 255.
(10)
Huheey, J. E. Inorganic Chemistry: Principles of Structure and Reactivity,
2024.
2nd edition ed.; Harper and Rows: New York, 1978.
(11)
Jorgensen, T.; Hansen, T. K.; Becher, J. Chem. Soc. Rev. 1994, 23, 41.
(12)
Jia, C. Y.; Liu, S. X.; Tanner, C.; Leiggener, C.; Sanguinet, L.; Levillain, E.;
Leutwyler, S.; Hauser, A.; Decurtins, S. Chem. Commun. 2006, 1878.
(13)
Jia, C. Y.; Liu, S. X.; Tanner, C.; Leiggener, C.; Neels, A.; Sanguinet, L.;
Levillain, E.; Leutwyler, S.; Hauser, A.; Decurtins, S. Chem.-Eur. J. 2007, 13, 3804.
(14)
Dumur, F.; Gautier, N.; Gallego-Planas, N.; Sahin, Y.; Levillain, E.;
Mercier, N.; Hudhomme, P.; Masino, M.; Girlando, A.; Lloveras, V.; Vidal-Gancedo, J.;
Veciana, J.; Rovira, C. J. Org. Chem. 2004, 69, 2164.
(15)
Fuks-Janczarek, I.; Luc, J.; Sahraoui, B.; Dumur, F.; Hudhomme, P.;
Berdowski, J.; Kityk, I. V. J. Phys. Chem. B 2005, 109, 10179.
(16)
Loosli, C.; Jia, C. Y.; Liu, S. X.; Haas, M.; Dias, M.; Levillain, E.; Neels, A.;
Labat, G.; Hauser, A.; Decurtins, S. J. Org. Chem. 2005, 70, 4988.
(17)
Perepichka, D. F.; Bryce, M. R.; Pearson, C.; Petty, M. C.; McInnes, E. J.;
Zhao, J. P. Angew. Chem. -Int. Ed. Engl. 2003, 42, 4636.
(18)
Wu, J.; Liu, S.-X.; Neels, A.; Derf, F. L.; Sallé, M.; Decurtins, S.
Tetrahedron 2007, 63, 11282.
!
211
(19)
Lopez, R.; Leiva, A. M.; Zuloaga, F.; Loeb, B.; Norambuena, E.; Omberg,
K. M.; Schoonover, J. R.; Striplin, D.; Devenney, M.; Meyer, T. J. Inorg. Chem. 1999,
38, 2924.
(20)
Goze, C.; Leiggener, C.; Liu, S. X.; Sanguinet, L.; Levillain, E.; Hauser, A.;
Decurtins, S. Chemphyschem 2007, 8, 1504.
(21)
Wu, H.; Zhang, D. Q.; Su, L.; Ohkubo, K.; Zhang, C. X.; Yin, S. W.; Mao,
L. Q.; Shuai, Z. G.; Fukuzumi, S.; Zhu, D. B. J. Am. Chem. Soc. 2007, 129, 6839.
(22)
Fang, C. J.; Zhu, Z.; Sun, W.; Xu, C. H.; Yan, C. H. New J. Chem. 2007,
31, 580.
(23)
Campagna, S.; Serroni, S.; Puntoriero, F.; Loiseau, F.; De Cola, L.;
Kleverlaan, C. L.; Becher, J.; Sorensen, A. P.; Hascoat, P.; Thorup, N. Chem.-Eur. J.
2002, 8, 4461.
(24)
Bryce, M. R. Adv. Mater. 1999, 11, 11.
(25)
Goze, C.; Liu, S. X.; Leiggener, C.; Sanguinet, L.; Levillain, E.; Hauser, A.;
Decurtins, S. Tetrahedron 2008, 64, 1345.
(26)
Keniley, L. K.; Ray, L.; Kovnir, K.; Dellinger, L. A.; Hoyt, J. M.; Shatruk, M.
Inorg. Chem. 2010, 49, 1307.
(27)
Vacher, A.; Barriere, F.; Roisnel, T.; Lorcy, D. Chem. Commun. 2009,
(28)
Pointillart, F.; Le Gal, Y.; Golhen, S.; Cador, O.; Ouahab, L. Inorg. Chem.
7200.
2008, 47, 9730.
(29)
Leiggener, C.; Dupont, N.; Liu, S. X.; Goze, C.; Decurtins, S.; Breitler, E.;
Hauser, A. Chimia 2007, 61, 621.
(30)
Fu, X. C.; Li, M. T.; Wang, C. G. Acta Crystallogr. Sect. E-Struct. Rep.
Online 2005, 61, M1221.
(31)
Madeja, K.; Wilke, W.; Schmidt, S. Zeitschrift Für Anorganische Und
Allgemeine Chemie 1966, 346, 306.
(32)
Li, L. L.; Liu, D. X.; Liu, T. F. Acta Crystallogr. Sect. E-Struct. Rep. Online
2007, 63, M1880.
!
(33)
Krej/ík, M.; Dan0k, M.; Hartl, F. J.Electroanal.Chem. 1991 317
(34)
Duvanel, G.; Banerji, N.; Vauthey, E. J. Phys. Chem. A 2007, 111, 5361.
(35)
Yamaguchi, S.; Hamaguchi, H. O. Appl. Spectrosc. 1995, 49, 1513.
212
(36)
Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.;
Vonzelewsky, A. Coord. Chem. Rev. 1988, 84, 85.
(37)
Richert, S. A.; Tsang, P. K. S.; Sawyer, D. T. Inorg. Chem. 1989, 28, 2471.
(38)
Arounaguiri, S.; Easwaramoorthy, D.; Ashokkumar, A.; Dattagupta, A.;
Maiya, B. G. Proceedings of the Indian Academy of Sciences-Chemical Sciences 2000,
112, 1.
(39)
A. Hauser, J. J., H. Romstedt, R. Hinek, H. Spiering. Coord. Chem. Rev.
1999, 471, 190.
(40)
Braterman, P. S.; Song, J. I.; Peacock, R. D. Inorg. Chem. 1992, 31, 555.
(41)
Mudasir; Wijaya, K.; Wahyuni, E. T.; Inoue, H.; Yoshioka, N. Spectrochim.
Acta A 2007, 66, 163.
(42)
Enachescu, C.; Krivokapic, I.; Zerara, M.; Real, J. A.; Amstutz, N.; Hauser,
A. Inorg. Chim. Acta 2007, 360, 3945.
(43)
Cannizzo, A.; Milne, C. J.; Consani, C.; Gawelda, W.; Bressler, C.; van
Mourik, F.; Chergui, M. Coord. Chem. Rev. 2010, 254, 2677.
(44)
Dupont, N.; Ran, Y.-F.; Jia, H.-P.; Grilj, J.; Liu, S.-X.; Decurtins, S.;
Hauser, A. Inorg. Chem. 2010, submitted.
!
213
9. Conclusions
TTF is well known for its electron donor properties, as well as for its luminescence
quenching ability. In this thesis, some photophysical and photochemical properties of
molecules incorporating TTF units have been studied, in particular the optical intraligand
charge and electron transfer between TTF and the other subunits of the molecules and
complexes. The study of the different molecules and complexes allowed a comparison
of the effect of substitution on the ligand moieties for instance on photo-induced charge
separation.
In a first study on the series of complexes of composition [{Ru(bpy)2}n(TTF-ppb)](PF6)2n
(n = 1, 2) it was shown that the first excited state corresponds to an intramolecular
charge transfer state with TTF as an electron donor and ppb as electron acceptor, the
transition to which manifests itself in a strong intraligand charge transfer absorption
band at 15000 cm-1. The series of compounds was compared with the analogue
complexes [Ru(bpy)3-n(TTF-dppz)n]2+ (n = 1, 2, 3). The latter showed a charge
-
separated state, which can be described for n = 2 by [(TTF+-dppz)Ru(dppz TTF)(bpy)]2+ and which has a lifetime of 2.5 µs in CH2Cl2 at room temperature. The
analogous charge separated state was not observed in [{Ru(bpy)2}n(TTF-ppb)](PF6)2n.
Instead, photo-excitation induced a triplet intraligand charge transfer state (3ILCT) via
intersystem crossing with a lifetime of approximately 200 ns for n = 1 and 50 ns for n =
2.
For a better understanding of the behaviour of the long lived charge separated state
obtained by photo-excitation in [Ru(bpy)3-n(TTF-dppz)n]2+, an anthraquinone group was
fused
to
a
phenanthroline
unit
in
order
to
obtain
the
complex
[Ru(TTF-
dppz)2(Aqphen)]2+. In this triad, excitation into the metal-ligand charge transfer bands
results in the creation of a comparatively long-lived charge separated state, with a
lifetime of 400 ns, implying TTF as electron donor and the anthraquinone as the final
electron acceptor. The fact that the charge recombination is faster for this complex than
for the parent complex with the rather poor dppz acceptor was attributed to the effect of
the Marcus inverted region.
!
214
With imidazole-annulated TTF derivatives, it was demonstrated that direct annulation of
imidazole derivatives to TTF influences the proton transfer properties on the imidazole
moiety, namely leading to an increase in pKa values. Conversely the energy of the
excited ILCT state is significantly lowered by the protonation.
In the tightly fused TTF-perylenediimide with the extended ! system, the presence of
the TTF unit quenches the luminescence of the perylenediimine by reductive electron
transfer quenching. But also the lowest energy ILCT state is non-emissive with a lifetime
of only 10 ps due to non-radiative decay.
Finally, combined with other metal ions such as Fe2+ and Co2+, the donor-acceptor
properties of the TTF-dppz ligands can be associated with spin-crossover properties.
For [Fe(phen)2(TTF-dppz)]2+, it was shown that the light-induced population and decay
of the intraligand charge transfer state and the light-induced population of the metastable high-spin state occur independently of each other. The latter of course requires
irradiation into the MLCT absorption band of the complex.
The experimental results on [Co(phen)2(TTF-dppz)]2+, however, show that the
photophysical behaviour of this molecule is much more complex and needs further
investigations in order to arrive at a better understanding of all the processes. One of
the major problems for the identification of the processes in this complex is to separate
possible charge transfer processes from intersystem crossing processes due to the fact
that the complex although with a high-spin ground state is close to the spin-crossover
point. For instance, the absorption spectrum during the reduction of the dppz subunit by
spectroelectrochemistry would reveal the spectroscopic signature of the reduced TTFdppz ligand and allow its identification and possible role in the cascade of relaxation
processes. Likewise, a comparison between the photophysical and photochemical
properties of [CoII(phen)2(TTF-dppz)]2+ and [CoIII(phen)2(TTF-dppz)]3+ would be
interesting. The latter is not easily accessible by chemical synthesis but it could be
produced electrochemically in situ as the oxidation wave of cobalt(II) to cobalt(III) in the
complex is at a lower potential than the first oxidation wave of TTF. Two ways of such
an
oxidation
in
situ
could
be
envisaged:
the
first
one
is
to
use
the
spectroelectrochemical cell in a constant flux during the oxidation in order to avoid the
diffusion of other products, the second one is to make conductive polymer films on an
!
215
ITO surface incorporating [CoII(phen)2(TTF-dppz)]2+ and apply a constant voltage in
order to obtain the cobalt(III) complex.
One of the major problems in the study of the photophysical properties of the complexes
and compounds of this thesis lies in their photochemical instability. It seems that upon
photoexcitation, the C=C double bond of the TTF unit is comparatively easily cleaved. In
our experiments we excluded artefacts from photoproducts by frequent exchange of
samples and by excluding history dependent signals, in particular luminescence.
Integrating the studied complexes or molecules of this thesis in insulating as well as
conductive polymers on an ITO surface could have one significant advantage by limiting
the photochemical decomposition of the TTF ligands. This is essential for the use such
functionalized molecules and complexes for potential applications in devices for solar
energy or optical sensors for instance for determination of the pH using imidazole
annulated TTF molecules.
!
216
During that thesis, other articles were published:
« Towards inert and preorganized d-block-containing receptors for trivalent lanthanides:
The synthesis and characterization of triple-helical monometallic Os-II and bimetallic
Os-II-Ln(III) complexes »
T. Riis-Johannessen, N. Dupont, G. Canard, G. Bernardinelli, A. Hauser, C. Piguet
Dalton Trans. 2008, 3661-3677.
!
!
« Self-Assembly of a Trinuclear Luminescent Europium Complex »!
S. Zebret, N. Dupont, G. Bernardinelli, J. Hamacek Chemistry-a European Journal.
2009, 15, 3355-3358.
« Unsymmetrical
Tripodal
Ligand
for
Lanthanide
Complexation:
Structural,
Thermodynamic, and Photophysical Studies »
B. El Aroussi, N. Dupont, G. Bernardinelli, J. Hamacek Inorg. Chem. 2010, 49, 606-615!
!
217