Evaluation of the potential effectiveness of ruthenium(II) complexes with 2,3-disubtituted pyrazino[2,3-f][1,10]phenanthroline anchors, R2ppl (R = CN, COOH, COOEt, OH) as sensitizers for solar cells

Article abstract of DOI:10.1016/j.poly.2012.12.003

The ligands of type pyrazino[2,3-f][1,10]phenanthroline, R₂ppl, with R = CN, COOH, COOEt or OH, were synthesized and used as precursors for obtaining the corresponding series of complexes of type [Ru(dmbpy) ₂R₂ppl](PF₆)₂, where dmbpy is 4,4′-dimethyl-2,2′-bipyridine. The compounds were prepared, characterized, and studied by theoretical DFT calculations in order to evaluate their potentiality as dyes in photoelectrochemical cells. The electron acceptor capacity of the R₂ppl ligands was evaluated by analyzing parameters such as electrophilicity and charge distribution on the reduced ligand. Additionally, the R substituents on R₂ppl were evaluated as anchoring groups, by variables such as highest spin occupied molecular orbital (HSOMO). Finally, the I_T parameter was defined and calculated. This is related to the amount of energy that can be delivered to TiO₂ from the acceptor anchoring ligand in the thexi state. According to this parameter, the [Ru(dmbpy)₂(COOH)₂ppl](PF₆)₂ complex is predicted to have the best response, among the compounds of the series, when used as dye in a solar cell.

Full text of DOI:10.1016/j.poly.2012.12.003

Polyhedron 52 (2013) 62–71
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Evaluation of the potential effectiveness of ruthenium(II) complexes with
2
,3-disubtituted pyrazino[2,3-f][1,10]phenanthroline anchors, R ppl (R = CN, COOH,
2
COOEt, OH) as sensitizers for solar cells
a,
⇑
a
a
b
b
c
Angélica Francois , Ramiro Díaz , Angélica Ramírez , Bárbara Loeb , Mauricio Barrera , Irma Crivelli
a
Facultad de Ciencias, Universidad Católica de Temuco, Rudecindo Ortega 02950, Temuco, Chile
Facultad de Química, Pontificia Universidad Católica de Chile, Vicuña Mackenna 4860, Santiago, Chile
Facultad de Ciencias, Universidad de Chile, Las Palmeras 3425, Santiago, Chile
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 14 December 2012
The ligands of type pyrazino[2,3-f][1,10]phenanthroline, R ppl, with R = CN, COOH, COOEt or OH, were
2
synthesized and used as precursors for obtaining the corresponding series of complexes of type
0
0
[
2 2 6 2
Ru(dmbpy) R ppl](PF ) , where dmbpy is 4,4 -dimethyl-2,2 -bipyridine. The compounds were prepared,
Keywords:
R ppl ligand
2
Ruthenium complexes
Acceptor ligand
TDDFT calculations
characterized, and studied by theoretical DFT calculations in order to evaluate their potentiality as dyes
in photoelectrochemical cells.
The electron acceptor capacity of the R
electrophilicity and charge distribution on the reduced ligand. Additionally, the R substituents on R
were evaluated as anchoring groups, by variables such as highest spin occupied molecular orbital
HSOMO). Finally, the I parameter was defined and calculated. This is related to the amount of energy
that can be delivered to TiO from the acceptor anchoring ligand in the thexi state. According to this
parameter, the [Ru(dmbpy) (COOH) ppl](PF complex is predicted to have the best response, among
the compounds of the series, when used as dye in a solar cell.
2
ppl ligands was evaluated by analyzing parameters such as
2
ppl
(
T
2
2
2
6 2
)
Ó 2012 Elsevier Ltd. All rights reserved.
1
. Introduction
This aim implies to consider not only electronic aspects but also
the efficient and stable adsorption of the dye on the semiconductor
surface [10].
Intense research in the context of renewable energy has led to
the development of a series of metal complexes with potential
application as dyes in solar photoelectrochemical cells (DSSCs).
DSSCs are based on the sensitization to visible light of mesoporous,
nanocrystalline semiconductor metal oxide films achieved by
means of the adsorption of molecular dyes [1–4]. Photoinduced
electron injections from the sensitizer dye into the metal oxide
conduction band generate charge separated states; the electrons
are injected through an anchoring group into the dye [5]. In this
way, the processes of electron injection into the semiconductor
conduction band and of absorption of visible and near infrared
light become independent [6].
One of the important variables to optimize and to control in so-
lar photoelectrochemical cells is the anchor [7], through which the
dye is fixed to the semiconductor. The electron injection to the
semiconductor through the anchor is the most Commonly invoked
mechanism for this process [8]. In this context, several studies have
been reported, where an optimization of the anchor is sought,
mostly to improve the anchor–semiconductor connection [5,7,9].
One of the best solar-to-electric power conversion efficiency in
DSSCs has been achieved with polypyridyl complexes of ruthe-
nium(II) [11] and osmium(II) [12] bearing carboxylated ligands
2
as anchors, which are often employed as TiO sensitizers in such
cells. These species give rise to intense visible metal-to-ligand
charge transfer (MLCT) bands with a favorable energetics for
0
0
charge injection. 4,4 -Dicarboxylic acid-2,2 -bipyridine type
ligands have been mostly used as anchors [13]. As an alternative
possibility, in the present work 2,3-disubstituted pyrazino-[2,3-
f][1,10]phenanthroline, R
were evaluated as potential anchoring groups in Ru complexes.
The synthesized complexes were of [Ru(dmbpy) ppl](PF type,
Scheme 1, where dmbpy is 4,4 -Dimetyl-2,2 -bipyridine.
2
ppl, with R = CN, COOH, COOEt and OH
2
R
2
6 2
)
0
0
Specifically, the spectroscopic and electrochemical properties of
complexes 1–4 were studied in order to evaluate the effects of the
R substituents on the electronic properties of the complexes.
The latter were characterized and studied by commonly used
experimental techniques. Theoretical DFT calculations allowed an
understanding of the electronic distribution in ligands and com-
plexes. Finally, a theoretical parameter, I
to evaluate the amount of electronic density available at the
T
, was introduced, in order
⇑
Corresponding author.
E-mail address: anfranci@uct.cl (A. Francois).
0
277-5387/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2012.12.003

A. Francois et al. / Polyhedron 52 (2013) 62–71
63
CH3'
(
PF6)2
N
N
N
N
R
R
N
H C
3
R
R
N
N
N
N
N
Ru
N
B
Q
N
2
Scheme 2. Fragment decomposition of the R ppl ligands.
H C
3
1
2
3
R = OH
R = COOEt
R = COOH
CH3'
4 R = CN
2
.3. Theoretical studies
Scheme 1.
All the calculations where performed on ADF2010 [16]. The geo-
metrical optimization was done using DZP basis set for C, N, O, H
and TZP basis set for ruthenium in combination with the PW91 ex-
change correlation potential [17]. One-electron properties of the li-
gands were determined in the gas phase using TZP basis set while
charge distributions were obtained by means of the Hirshfeld
method [18]. For complexes, TDDFT calculations [19] were per-
formed using SAOP exchange correlation functional [19], SZ basis
set for C, H, O, N and ZORA-TZ2P basis for ruthenium. Solvent ef-
fects were included through the COSMOS model [20] employing
a Van der Waals surface and Bondi atomic radius.
Ruthenium
Dye
E
A
L-A
TiO2
Φ
T
ζ
E
D(L-A)
I
T
Fig. 1. Schematic representation of the energy flow in a DSSC. Parameters are
defined in Section 4.
Regarding the calculations involving ligands, it has been pro-
posed that R
by two fragments, either (phenanthroline(F) + pyrazine (P)) or
bypiridine (B) + quinoxaline (Q)). The latter ligand partitioning,
Scheme 2, was applied in the calculations, since it permits a clear
understanding of the effect of R, in particular on the electron do-
nor/acceptor capacity of the ligand, and on the composition of
the LUMO orbital for the free and complexed ligands.
2
ppl type ligands can be described as being formed
2
anchoring ligand, which is delivered to the TiO semiconductor
(
conduction band. This parameter should be indicative of the injec-
tion capacity of the proposed dye, as can be seen in Fig. 1, which
describes the energy flow in a typical sensitized solar cell.
In this paper a theoretical model that enables the calculation of
the amount of absorbed energy effectively transmitted to the
acceptor anchoring ligand is proposed. In this ligand, the R substi-
tuent plays two roles: first as an anchoring group, providing a
physical linkage between the ruthenium complex and the semi-
conductor – usually titanium dioxide – surface. Second, it acts as
a wire, conducting electronic charge from the dye to the semicon-
ductor conduction band. The proposed model can be classified as
2
. Experimental
2.1. Materials
Both reagents and solvents were pa grade, laboratories Sigma–
Aldrich or Merck. All of them were used as purchased and, when
needed, were purified and dried by standard methods. Diamino-
maleonitrile and NH
4 6
PF were Sigma–Aldrich and the rest of the
‘
‘non-interacting’’ since it considers that the spectroscopic proper-
solid chemicals were Merck. The precursors 1,10-phenanthroline-
ties of the dye are not modified when it is bound to the semicon-
ductor layer, and that the molecular orbital order and
composition of the complexes will remain unaffected by the pres-
ence of the titanium dioxide orbitals.
The amount of energy (I ) that can be delivered to TiO from the
T 2
acceptor anchoring ligand (LA) in the electronic thermally equili-
5
2
,6-dione (phendione) [14] and dichloro-bis-(4,4-dimethyl-2,
0
2 2
-bipyridine)ruthenium(II) (cis-Ru(dmbpy) Cl ) [15] was synthe-
sized following reported procedures. The corresponding elemental
analysis and H NMR spectra were in agreement with the
published results.
1
brated, thexi, state T, is calculated in terms of absorbed energy
2.2. Physical measurements and instrumentation
A D
(E ), and, the delivered energy (E ) to a specific fragment of the
molecule, Fig. 2. f(LA) and U
T
(A) are distribution functions taking
Elemental analyses were performed on a Fisons Instruments
into account the contribution of the LA ligand to the excited state
density, and the contribution of the anchor to the T excited state
Analyzer, model EA 1108/CHNS–O with PC NCR system 3225.
UV–Vis Spectra were recorded on a Shimadzu, UV-3101 PC Spec-
trophotometer. IR Spectra on Bruker Vector-22FTIR spectrometer
[21]. The value of E
ered to the ligand when composed by two molecular fragments L
and A (in this case B and Q, Scheme 2). I represents the amount
A
f(LA) is the amount of electronic density deliv-
1
using KBr pellets and H NMR Spectra on a Bruker AC/200
T
(
200 MHz) or Bruker Aspect 400 MHz spectrometer, using CD
3
CN
of absorbed energy that is channelled by the complex to the anchor
group, and that can be injected into the conduction band of the
semiconductor. By the MLCT absorption process, electronic density
from populated d-metal orbitals is transferred to the LUMO of the
acceptor anchoring ligand. The distribution of this extra charge
along the different sites of the organic molecule can be studied
by analysing the composition of the highest spin occupied molec-
ular orbital (HSOMO). From this study, the role played by the
different anchoring groups can be evaluated.
or CDCl as solvent and TMS as reference. Cyclic voltammetry
3
experiments were carried out using a BAS CV-50W-2,3-MF-9093
equipment; with a three-electrode arrangement with a platinum
wire, a platinum coil and a Ag/AgCl electrode as working, auxiliary
and reference electrode, respectively. Experiments were carried
out in Argon atmosphere, with tetrabutylammoniumhexafluoro-
phosphate (HBTA) 0.1 M (Aldrich) as supporting electrolyte and a
ꢀ
1
scan speed of 200 mV s
.

6
4
A. Francois et al. / Polyhedron 52 (2013) 62–71
b
a
c
a
c
N
N
N
R
R
b
A
B
N
c
a
b
9
.6
9.4
9.2
9.0
8.8
8.6
8.4
8.2
8.0
7.8
7.6
7.4
ppm
Fig. 2. 1H NMR spectra of (CN)
ꢀ
2
3 2 2
ppl (A) in CDCl and (COOH) ppl (B) in D O/OD .
2
2
.4. Synthesis
2.4.2. Synthesis of ruthenium complexes
All the complexes were synthesized using cis-Ru(dmbpy)
precursor. In a typical experiment, stoichiometrical quantities
(ꢁ0.180 mmol) of the appropriate ligand and cis-Ru(dmbpy) Cl
were mixed in 1:2 ratio, in 25 mL of ethanol as solvent, and
refluxed under N atmosphere [22]. The solvent was reduced by
rotary evaporation and 3 mL of saturated NH PF aqueous solution
were added. The red solid was isolated by filtration and purified by
column chromatography (Al ; eluted with CHCl to eliminate
untreated precursor complex and then with MeOH/CHCl ). Partic-
ular details for the synthesis of complexes are described below.
[Ru(dmbpy) (OH) ppl](PF (1): Reflux time: 8 h. Yield: 0.132 g
(70%). Chromatographic column eluted with CHCl /MeOH 70/30.
: 3426, 3065, 2950–2840, 1621, 1140, 844.
2
2
Cl as
.4.1. Synthesis of ligands
,3-Dicyanopyrazino[2,3-f][1,10]phenanthroline((CN)
2
2
ppl): This
2
2
synthesis was carried out by a reported method for unsubstituted
ppl ligand [22]. The obtained results were consistent with those
published [23].
2
4
6
6 6
Yield: 0.686 g (87%). Anal. Calc. for C16H N (M = 282.26 g/mol):
C, 68.08; H, 2.14; N, 29.57. Found: C, 68.36; H, 2.07; N, 28.87%. IR
2
O
3
3
KBr pellet, cm ): 3060, 2241, 1616. 1H NMR (CDCl
ꢀ1
as solvent,
(
3
3
0
d in ppm and J in Hz): 9.50 (dd, 2H, Ha, Ha , Jab = 4.43, Jac = 1.72),
0
0
7
J
.94 (dd, 2H, Hb, Hb , Jba = 4.43, Jbc = 2,96), 9.60 (dd, 2H, Hc, Hc ,
cb = 2.96, Jca = 1.72).
,3-Dicarboxypyrazino[2,3-f][1,10]phenanthroline((COOH)
.30 g (1.06 mmol) of dcppl were suspended in 30 mL of 5 M sul-
2
2
6 2
)
3
ꢀ
1
2
2
ppl):
IR (KBr pellet, cm ) m
0
d: 1617.
phuric acid and refluxed for 5 h. A white precipitate was collected
by centrifugation. The solid was washed with abundant degassed
water, then with ethanol and finally with ethyl ether and dried un-
[Ru(dmbpy)
2
(COOEt)
2
ppl](PF
6
)
2
(2): Reflux time: 9 h. Yield:
/MeOH
) m: 3078, 2980–2900, 1720, 1612,
0.140 g (66%). Chromatographic column eluted with CHCl
60/40. IR (KBr pellet, cm
1240, 845.
3
ꢀ1
der high vacuum. Yield: 0.334 g (98%) Anal. Calc. for C16
8 4 4
H N O
(
M = 320.26 g/mol): C, 60.00; H, 2.52; N, 17.49. Found: C, 59.88;
2 2 6 2
[Ru(dmbpy) ((COOH) ppl](PF ) (3): Reflux time: 8 h. Yield:
ꢀ1
H, 2.49; N, 17.37%. IR (KBr pellet, cm ): 3500–2600, 1741, 1619,
1
Hz): 8.88 (d, 2H, Ha, Ha , Jab = 4.43), 7.75 (dd, 2H, Hb, Hb ,
0.130 g (68%). Chromatographic column eluted with CHCl
3
/MeOH
1
ꢀ1
415. H NMR (D
2
O (NaOD solution) as solvent, d in ppm and J in
60/40. IR (KBr pellet, cm
1737, 1617, 845.
)
m: 3630, 3428, 3076, 2980–2900,
0
0
0
Jba = 4.43, Jbc = 8.37), 9.50 (d, 2H, Hc, Hc , Jcb = 8.37).
[Ru(dmbpy)
2
(CN)
2
ppl](PF
6
)
2
(4): Reflux time: 6 h. Yield: 0.135 g
/MeOH 90/10.
: 3084, 2917, 2223, 1619, 846.
2,3-Diethoxycarbonylpyrazino[2,3-f][1,10]phenanthroline((COO-
(72%). Chromatographic column eluted with CHCl
3
IR (KBr pellet, cm
ꢀ1
Et)
2
ppl): 0.23 g (0.81 mmol) of(COOH)
) under N
was refluxed during 3.5 h and SOCl
2
ppl were dissolved in 3 mL
)
m
of thionyl chloride (SOCl
2
2
atmosphere. The solution
2
in excess was eliminated un-
3
. Results and discussion
der high vacuum. Then, 8.0 mL of dry ethanol was added and the
mixture was refluxed overnight. A cream color precipitate of colloi-
dal appearance was obtained and separated by centrifugation. The
solid was repeatedly washed with ethyl ether and dried under high
vacuum.
3
.1. Synthesis
+2
The synthesis of the complex [Ru(bpy)
2
(CN)
2
ppl)] has been re-
(phendi-
ported previously by in situ condensation of the [Ru(bpy)
one)]
treatment of [Ru(bpy)
to obtain [Ru(bpy)
2
Yield: 0.225 g (83%), Anal. Calc. for C20
H
16
N
4
O
4
, (M: 376.37 g/
+2
precursor with diaminomaleonitrile [24]. Subsequent
mol): C, 63.82; H, 4.28; N, 14.89. Found: C, 63.41; H, 4.1; N,
+2
2
(CN)
2
2 4
ppl)] with NaOH or H SO , permits
ꢀ1
1
1
4.46%. IR (KBr pellet, cm ): 3087, 3053, 2986, 2937, 2906,
710, 1616.
+2
+2
2
(OH)
2
ppl)]
and [Ru(bpy)
2 2
(COOH) ppl)] ,
ppl free ligands (R = CN,
respectively. In the present work, the R
2
2
,3-Dihydroxopyrazino[2,3-f][1,10]phenanthroline((HO)
2
ppl):
OH and COOH) were synthesized prior to their coordination to
Ruthenium, for their unambiguous characterization, and in order
to understand the effects produced by coordination. Specifically,
the synthesis of 5,6-dicianopyrazino[2,3-f][1,10]phenanthro-
0
.30 g (1.06 mmol) of (CN) ppl were dissolved in a hot solution of
2
NaOH 5 M. The mixture was refluxed during 5 h. The pH was ad-
justed to 8–9 with 5 M H SO , a cream color precipitate of colloidal
2
4
appearance being obtained. The solid was filtered off, washed with
ethanol and ethyl ether and dried under high vacuum.
line(CN)
22,25], but using excess phendione, thus increasing the reaction
yield. The (COOH) ppl ligand was synthesized by hydrolysis of
CN) ppl in 50% sulfuric acid media [26]. The new ligand 2,3-
diethoxycarbonilpyrazino[2,3-f][1,10]phenanthroline, (COOEt) ppl
2
ppl was achieved following a reported procedure
[
Yield: 0.274 g (97.7%). Anal. Calc. for C14
8 4 2
H N O , (M: 264.24 g/
2
mol): C, 63.63; H, 3.05; N, 21.20. Found: C, 62.88; H, 2.97; N,
(
2
ꢀ
1
2
0.86%. IR (KBr pellet, cm ): 3425, 1619, 1594, 1124.
2

A. Francois et al. / Polyhedron 52 (2013) 62–71
65
was synthesized by esterification of (COOH)
2
ppl in ethanol.
a displacement to higher energy with regard to the diaminomaleo-
1
ꢀ1
Scheme 3 shows the synthesis routes for all the ligands. H NMR,
IR and Elemental Analysis results were in good agreement with
the proposed formula for all the ligands.
nitrile reagent (2215 cm ). On the other hand, the
m
(C@O) signal
(N–H) bands at
of the diaminomaleonitrile reactive were
ꢀ1
at 1685 cm of the phendione precursor and the
3440 and 3300 cm
m
ꢀ1
The synthesis of the [Ru(dmbpy)
R = CN, OH, COOH and COOEt) was conducted according to re-
ported procedures for similar ruthenium polypyridine complexes
22,27], implying the substitution of chlorine in Ru(dmbpy) Cl by
the corresponding R -ppl ligands. The yields for these syntheses
were in the 66–72% range.
2
R
2
ppl](PF
6
)
2
series of complexes
not observed. The displacement of the C„N stretching, and the dis-
appearance of the OH and NH bands evidence the formation of the
(
2 2
(CN) ppl ligand. For the (OH) ppl ligand, the disappearance of the
[
2
2
m
(C„N) signal and the appearance of two intense signals, typical
ꢀ1
2
of the hydroxyl group: m(O–H) at 3425 cm (wide) and d(O–H)
at 1634 cm , demonstrate that the substitution reaction was com-
ꢀ1
plete. In the IR spectra of the (COOH) ppl ligand, the carboxyl group
2
ꢀ1
signals highlight. The bands at 3434 and 3081 cm are assigned to
stretching modes of the free and associated O–H respectively,
while the band at 1741 cm is assigned to the stretching mode
3.2. Spectroscopic properties of the free ligands
ꢀ1
The 1H NMR spectra of ligands (CN)
ꢀ1
2 2
ppl and (COOH) ppl are
of protonated carboxyl group. The band at 1415 cm is assigned
shown in Fig. 2. Both spectra show the same pattern with three
magnetically nonequivalent protons on the bipyridine moiety,
with the expected integration. The lowest field signal is assigned
to the stretching mode of C–O plus deformation mode of O–H. A
of low intensity also appears at 1631 cm , and is assigned to
deprotonated carboxyl groups, indicating that the completely pro-
ꢀ
1
0
to protons in Hc and Hc positions (see Fig. 2), due to the aniso-
tonated ligand predominates. Besides, in the (COOEt) ppl ligand the
2
0
ꢀ1
tropic effect of the pyrazine N-atoms [22,28]. The Ha and Ha pro-
aliphatic
m
(C–H) c.a. at 2990–2850 cm is observed, as well as a
ꢀ1
tons appear at a higher, due to the inductive effect of the
phenanthroline N-atoms. The highest-field signal is assigned to
band at 1730 cm related to the C@O stretching. The disappear-
ance of the OH signals of the carboxyl group and the appearance
of a carboxylate signal indicates the formation of the ester.
The UV–Vis spectra of the free ligands show a series bands in
0
protons Hb and Hb , showing an orto interaction with Ha and Hc.
The (CN)
2
ppl ligand shows a displacement to lower field of protons
0
0
0
Ha, Ha ; Hb, Hb and Hc, Hc when compared to the unsubstituted
the UV region between 280 and 350-nm, and a strong band at
⁄
ppl ligand [28], in agreement with the electron acceptor effect of
260 nm, all assigned to
p?p
transitions.
the cyano groups.
1
The H NMR spectra of the (OH)
2 2
ppl and (COOEt) ppl ligands was
not registered because they turned out to be insoluble in all the
deuterated solvents available.
3.3. Spectroscopic properties of the complexes
1
The IR spectra of the free ligands show the characteristic bands
It was possible to record the H NMR spectra of the four com-
of the corresponding functional groups, in addition to the
m
(C–H)
plexes synthesized, although some with low resolution due to their
low solubility in the available deuterated solvents. All the spectra
exhibit the signals of the dmbpy and R ppl ligands, showing good
2
integrals correlation between the protons of the different ligands;
therefore, the presence of the desired complexes can be inferred.
ꢀ1
band c.a. at 3080–3000 cm , and d (C–H) signal c.a. at 1600–
ꢀ
1
1
620 cm , corresponding to the aromatic hydrogen atoms of
ꢀ1
2
phenantroline. For the (CN) ppl ligand, the band at 2247 cm
,
was assigned to the stretching mode of the C„N group and shows
O
O
N
NC
CN
Ethanol
N
N
N
N
CN
CN
Reflux 5 h, N2
+
N H
2
N H
2
N
(
CN) ppl
2
H SO 5 M
NaOH 5M
2
4
Reflux 5 h
Reflux, 5 h
N
N
N
N
COOH
COOH
N
N
N
N
OH
OH
(
COOH) ppl
(OH) ppl
2
2
1
2
. SOCl , Reflux 3,5 h, N
2
2
. Ethanol, reflux 12h, N2
N
N
N
N
COOEt
COOEt
(
COOEt) ppl
2
Scheme 3. Routes of synthesis of all the ligands.

6
6
A. Francois et al. / Polyhedron 52 (2013) 62–71
CH3'
(
PF6)2
N
3C
H
N
CN
CN
N
N
N
6
Ru
N
5
N
a
N
c
3C
H
b
6'
5'
3
3
’ 3
3
'
6
’
6
a
c
CH3'
b
5
’
5
2
.0
4.2
8.6
2.0
8.4
2.0
8.0
2.1 2.0
7.6
2.1
7.4
2.0
9.6
9.4
9.2
9.0
8.8
8.2
7.8
7.2
ppm
Fig. 3. Aromatic zone of 1H NMR spectra of [Ru(dmbpy)
(CN)
ppl](PF
2
2
6
)
2 3
, at 400 MHz in CD CN.
0
2 2 6 2
Fig. 3 shows the spectrum of complex [Ru(dmbpy) (CN) ppl](PF ) ,
R
2
ppl ligand. The H5 and H5 protons are not equivalent either,
4
, while Table 1 summarizes the data obtained for all the synthe-
for the same reason previously mentioned. These appear as doublet
0
sized complexes. All spectra show nine aromatic signals, corre-
sponding to nine equivalent protons pairs. The signals were
signals split by coupling at long distance with the H3 and H3 pro-
0
tons, respectively. The H3 and H3 protons suffer the meta effect of
assigned by comparing them with the spectra of (COOH)
CN) ppl free ligands, and based on those coupling constants that
were split Three signals are assigned to protons on the R ppl
ligands (see Fig. 3). Since the spectra of all complexes were regis-
tered in acetonitrile-d and those of the (COOH) ppl and (CN) ppl
O/OD and chloroform-d respectively, it is not
possible to analyze the effect of the metal on the chemical shifts
of protons of these ligands in their respective complexes. Despite
this, it can be observed that in complex 4 the Hb and Hc protons
signals shift to lower field: 30 and 33 ppm respectively, and the
Ha protons to 47 ppm, which cannot be attributed only to the
solvent, but also to the deshielding effect exerted by the metal
coordinated to the acceptor ligand.
2
ppl and
the N coordinated to the metal; thus they appear more deshielded
and their coupling with H5 and H5 is slightly lowered.
0
(
2
2
The IR spectra of all complexes show the C–H tension bands of
ꢀ
1
the methyl groups on the dmbpy ligand in the 2980–2850 cm
ꢀ
3
2
2
range, and the typical signals of the PF
6
counter ion (c.a. 842–
848 cm ), confirming the cationic nature of the complexes. By
comparison with the ring-stretching modes reported for
ꢀ
ꢀ1
free ligands in D
2
2
+
ꢀ1
[Ru(bpy)
3
]
[29] the bands at c.a. 1600, 1480, 1440 cm that ap-
pear in the IR spectra of all the complexes were assigned to the
ring-stretching modes of conjugated C@C bonds of the dmbpy li-
gand. Additionally, the IR spectrum of each complex shows the sig-
nals of the functional group substituted on the ppl ligands, with
small variations with regard to the free ligand, due to the coordina-
tion to the metal. It can be seen that the general tendency is a slight
displacement to lower energy for the ligands in the complexes in
regard to the corresponding free ligand, due to decreased electron
density on them when coordinated to the metal. The most marked
The aromatic protons of the dmbpy ligand were assigned based
0
on their coupling constants. In general, the H6 and H6 protons
appear as doublet signals and are not equivalent. This originates
from the fact that the H6 protons are affected by the N of the
0
dmbpy ligand, while the H6 protons are affected by the N of the
effect is observed for [Ru(dmbpy)
2
((CN)
2 6 2
ppl](PF ) , 4, where the
Table 1
1
H NMR chemical shifts (ppm) for all synthesized complexes, in CD
3
CN (200 or 400 MHz).
Position
Complexes
Ru(dmbpy)
+
2
+2
+2
+2
[
2
(OH)
2
ppl] (1)
[Ru(dmbpy)
2
(COOEt)
2
ppl] (2)
[Ru(dmbpy)
2
(COOH)
2
ppl] (3)
2 2
[Ru(dmbpy) (CN) ppl] (4)
0
Ha, Ha
8.64 (s, 2H); Jab = 4.43
8.13 (m, 2H)
9.59 (d, 2H); Jca = 8.61
8.81 (s, 2H)
8.77 (s, 2H)
7.30 (d, 2H); J56 = 4.92
8.53 (d, 2H); Jab = 4.92
8.01 (dd, 2H); Jbc = 8.37, Jba = 4.92
9.48 (d, 2H); Jcb = 8.37
8.58 (s, 2H)
8.63 (s, 2H)
7.09 (d, 2H); J56 = 5.17
8.30 (d, 2H); Jab = 8.12
7.93 (dd, 2H); Jbc = 8.11, Jba = 5.41
9.45 (d, 2H); Jcb = 8.12
8.36 (s, 2H)
8.40 (s, 2H)
7.07 (d, 2H); J56 = 5.66
8.41 (dd, 2H); Jab = 5.41, Jac = 1.23
8.05 (dd, 2H); Jba = 5.41, Jbc = 8.36
9.58 (dd, 2H); Jcb = 8.38, Jca = 1.23
8.57 (s, 2H); J35 = 1.23
0
Hb, Hb
0
Hc, Hc
H3
H3
0
8.60 (s, 2H); J
7.15 (dd, 2H); J53 = 1.23, J56 = 5.91
7.38 (dd, 2H); J = 5.56
7.55 (dd, 2H); J65 = 5.91
3 5
0 0
= 1.23
H5
H5
0
7.54 (d, 2H); J
7.91 (d, 2H); J65 = 4.92
5
0
6
0
= 5.16
7.36 (d, 2H); J
7.68 (d, 2H); J65 = 5.17
5
0
6
0
= 5.41
7.30 (d, 2H); J
7.44 (d, 2H); J65 = 5.66
5
0
6
0
= 5.17
0 0
5 6
H6
H6
0
8.07 (d, 2H); J
2.58 (s, 6H)
2.48 (s, 6H)
6
0
5
0
= 5.16
7.86 (d, 2H); J
2.50 (s, 6H)
2.39 (s, 6H)
4.47 (m, 4H)
1.93 (d, 4H)
6
0
5
0
= 5.41
7.62 (d, 2H); J
2.57 (s, 6H)
2.48 (s, 6H)
6
0
5
0
= 5.17
7.71 (dd, 2H); J
2.61 (s, 6H)
6 5
0 0
= 5.56
CH
CH
CH
CH
3
0
2.52 (s, 6H)
3
2
0
2

A. Francois et al. / Polyhedron 52 (2013) 62–71
ꢀ1 ꢀ1
67
4
Table 2
of 10 M cm ), without any appreciable resolution of transitions
involving different ligands. By comparing complexes 1–4 with sim-
ilar reported complexes, it can be observed that, as is typical in
Ru(II)–polypyridine complexes [22], the energy of the intraligand
bands are not affected by the substituent on the dmbpy or ppl li-
gands. The MLCT bands do not change significantly, although there
is a certain tendency. A more detailed analysis of the absorption
bands and their assignment can be achieved by a theoretical char-
acterization of the complexes, as follows.
Table 3 shows TDDFT results for the simulated absorption spec-
tra. In general all the complexes exhibit two electronic transitions
with oscillator strength between 0.08 and 0.20, with the exception
of complex 3 where three signals are observed. Electronic excita-
tions are displayed in column 6 and their assignments are pro-
posed in column 7.
A transition density study[21] (Tables S1–S4) was performed
using the above data, revealing that the absorption bands of com-
plexes 1–4 contain mostly MLCT bands from the metal to the donor
ligand (M?L). This contribution corresponds to about 50–80% of
the band. Additionally, 20–30% of this band is due to MLCT bands
to the corresponding acceptor ligand (M?LA or M?BQ). Finally,
a small contribution of ligand to ligand (LL) bands also exists,
which only becomes important for complex 3 (R = COOH).
UV–Vis spectroscopic data for [Ru(dmbpy)
2
R
2
ppl](PF
6 2
) complexes in acetonitrile.
Complexes
k
max (nm)
IL
TCML
1
2
3
259, 284, 338
265, 282, 354
260, 283, 337
259, 284, 337
256, 285
433, 457
442
450
446
450
4
a
[
[
[
Ru (dmbpy)
Ru (dmbpy)
Ru(dmbpy)
2
ppl](PF
6
)
2
(5)
a
2
(COOMe)ppl](PF
)
6 2
(6)
259, 284
296
447
377, 560
b
2
Cl
2
(7)
a
Ref. [22].
Ref. [27].
b
tension signal of the CN group at 2223 cm 1, shifts 23 cm- 1 to low-
er energy respect to the free ligand, reflecting a probable loss of the
triple bond character in CN, due to back bonding effects.
ꢀ
The UV–Vis spectra of all complexes show a similar structure,
with bands around 280 and 300 nm, which can be ascribed to
⁄
⁄
spin-allowed ligand-centered (IL,
p?p and n?p ) transitions,
associated with polypyridine units present in the complexes. The
shoulder of lower intensity at 320–380 nm can be assigned to
⁄
intraligand (
p
?p
) transitions in the R
2
ppl moiety overlapped with
intrametal transitions [24].
Table 2 summarizes the data obtained for the UV–Vis spectra of
complexes 1–4. Results from comparable complexes are also
shown. All of the complexes display bands in the 400–500 nm
range, which are of MLCT (metal-to-ligand charge transfer) charac-
3
.4. Electrochemical properties of complexes 1–4
Table 4 shows the electrochemical results for the series of com-
plexes. Data for other Ru(II)–polypyridine complexes are included
for comparison purposes.
ter, with kmax at 450–460 nm (extinction coefficient
e in the order
Table 3
Comparison of TDDFT calculations with UV–Vis spectroscopic data of complexes 1–4, recorded in acetonitrile.
Complexes
k
max (Exp)
k
max (Calc)
Oscillator strength
0.10
Electronic transition
1A
Main excitations
Assignment
1
457
451
41% H2?L1
MBL
ML
MBL
MB
ML
MBL
ML
MB
MB
MQ
ML
MB
ML
ML
ML
MB
QL
ML
ML
MBQ
ML
LL
ML
MBQ
LL
3
9
7
7% H1?L2
% H2?L
% H3?L2
4
43
0.17
1B
34% H2?L2
17% H1?L1
13% H2?L12
11% H1?L
10% H?L3
5% H1?L4
2
3
442
450
433
0.09
0.11
2A
2B
44% H2?L3
3
1
5% H1?L11
6% H1?L2
4
29
37% H1?L3
2
2
5
4
7% H3?L3
1% H2?L11
% H4?L
% H2?L2
433
0.09
0.06
0.07
3A
3B
3C
44% H2?L3
3
1
5% H1?L4
5% H1?L2
4
32
28
48% H3?L3
3
9
4% H1?L3
% H2?L4
4
51% H3?L3
2
1
4
5% H1?L3
5% H2?L4
% H2?L2
ML
MBQ
ML
ML
ML
MB
MB
ML
ML
ML
4
446
436
0.11
0.08
4A
4B
32% H5?L
29% H1?L4
28% H2?L2
9% H?L5
4
32
52% H6?L4
4
2
3% H10?L
% H?L6
H = HOMO, H1 = HOMO ꢀ 1, L = LUMO, L1 = LUMO + 1.
Metal to ligand charge transfer excitations: ML = metal to ligand (L = dmbpy), MB = metal to the bipyridine fragment (B), MQ = metal to quinoxaline fragment (Q),
MBL = metal to B and L, MBQ = metal to B and Q fragments; LL = ligand to ligand excitation.

6
8
A. Francois et al. / Polyhedron 52 (2013) 62–71
Table 4
It is noteworthy that for ligands R = H and R = OH, the LUMO + 1
molecular orbital is centered on the (B) fragment, while for the
other ligands of the series, which contain an acceptor group, the
electron density is centered on the (Q) fragment.
2 2 6 2 3
Electrochemical properties of [Ru(dmbpy) R ppl](PF ) complexes, vs. SCE in CH CN.
Complexesa
E
1/2,ox (V)
E
1/2,red (V)
c
c
1
2
3
4
1.22
1.28
1.29
1.30
1.24
1.29
1.09
ꢀ0.95 , ꢀ1.11
ꢀ0.89, ꢀ1.19, ꢀ1.47
ꢀ0.87, ꢀ1.18
ꢀ0.86
4.2. Electrophilicity index
[Ru (dmbpy)
[Ru (dmbpy)
[Ru (dmbpy)
2
2
3
ppl]2+b (5)
ꢀ1.23
2+b
Since these ligands play the role of anchors when used in solar
cells, it is important to analyze their capacity to concentrate elec-
tron density. This can be achieved by means of the electrophilicity
(COOMe)ppl]
(6)
ꢀ0.96
2+b
]
(7)
ꢀ1.47
a
b
c
6ꢀ
As PF salt.
Ref. [22].
l
2
index [31], E ¼ where
l
is the chemical potential, identified with
is the global hardness [32], which can
2
g
the HOMO eigenvalue and
g
Ep,red, irreversible processes.
be calculated as half of the energy difference between the HOMO
and SOMO molecular orbitals. Results are displayed in Table 5
Comparison of the E1/2,ox and E1/2,red values of complexes 1–4
0
0
and include values for the 4,4 -dicarboxylic acid-2,2 -bipyridine
dcbpy), and dmbpy ligands. The first of them is introduced as a ref-
2
+
with the homoleptic [Ru(dmbpy)
ppl acts as acceptor ligand in these complexes. The electron-
withdrawing effect of R ppl produces a decrease in the electron
3
] compound 7, demonstrate that
(
R
2
erence acceptor compound since it is widely employed as anchor
group while the latter is known for its donating ability.
2
density on the Ru, making its oxidation more difficult. The reduc-
tion process is more sensitive to this effect, decreasing E1/2,red
down to approximately 0.5 V with respect to 7. In complexes 1, 2
and 3 more than one process of reduction was observed. In com-
plex 1, both reduction processes are irreversible. In all cases, the
first and second reduction process was assigned to the R ppl accep-
2
tor ligand. In complex 2 the third process was assigned to the
dmbpy ligand, by comparison with 7.
The values of E1/2,red for the first reduction of complexes 2, 3 and
show no significant difference. Nevertheless, the effect of R sub-
stituents COOEt, COOH, CN is important when compared to com-
plex 5 (R = H). On the other hand, as expected, by comparing the
Fig. 5 confirms the observed trend that dcbpy is a better electron
acceptor than dmbpy, which is reflected by its higher electrophilic-
ity (7.2 versus 5.4 eV). It can also be observed that ppl is located be-
tween these values (6.4 eV) and when the OH moiety is introduced
in the pyrazine ring, a slight decrease (to 6.3 eV) of its acceptor
capacity occurs. On the other hand, COOMe, COOH and CN substi-
tuent groups clearly enhance the acceptor capacity of the ppl li-
gand. Indeed the (CN) ppl is the most electrophilic of the whole
2
series. The general trend observed for the increase of electrophilic-
ity follows the sequence:
4
ðOHÞ ppl < ppl < ðCOOMeÞ ppl < ðCOOHÞ ppl < ðCNÞ ppl
2
2
2
2
E
1/2,red values of 2 and 6, the effect of electron density reduction
produced by two substituent groups on 2 is greater than that of
one substituent in 6; thus 2 is more easily reduced. In complex 1
the reduction process is irreversible, so it is not possible to com-
pare it with other complexes.
When the oxidation processes are considered, the effect of the
different R substituents in the ppl ligand on the values of E1/2,ox is
insignificant, the major difference being 0.08 V. Nevertheless, the
comparison of the E1/2,ox values for complexes 2, 3 and 4 with that
for 5 shows that the electron-acceptor effect of the R substituents
hinders the metal oxidation respect to R = H. In complex 1 the OH
group produces a slight decrease of E1/2,ox due to the electron-
donor character of the OH group relative to H. By comparing 2 with
6
there is no significant effect on the E1/2,ox value associated with
the number of ester substituents in the ppl ligand.
In addition, according to the results discussed above for the cor-
responding complexes, the electron-acceptor ability of the disub-
stituted ppl ligands increases in the following order:
2
Fig. 4. Isodensity plots for HOMO, LUMO and LUMO + 1 molecular orbitals of R ppl
free ligands (R = H, OH, COOMe, COOH, CN).
ppl < ðCOOEtÞ ppl < ðCOOHÞ ppl < ðCNÞ ppl
2
2
2
4
. Theoretical studies
Table 5
4
.1. Molecular orbital calculations of the free ligands
Theoretical characterization of the acceptor capacity of the R
2
ppl ligands.
(eV)
R
2
ppl
eHOMO (eV)
e
SOMO (eV)
l
(eV)
g
E (eV)
As mentioned in Scheme 2 of Section 2.3, the ppl ligand can be
OH
H
COOMe
COOH
CN
ꢀ9.91
ꢀ2.17
ꢀ2.21
ꢀ2.91
ꢀ3.07
ꢀ3.71
ꢀ9.91
ꢀ10.01
ꢀ10.20
ꢀ10.40
ꢀ10.70
7.74
7.80
7.30
7.28
7.00
6.34
6.42
7.14
7.36
8.19
described as the sum of bipyridine and quinoline fragments
27,30]. The molecular orbital calculations performed for the series
ꢀ10.01
ꢀ10.20
ꢀ10.40
ꢀ10.70
a
[
of ligands under study reveals that the HOMO is composed mainly
of atomic orbitals from the nitrogen atom of the bipyridine frag-
ment (B), with the exception of the hydroxo ligand, where the
HOMO is degenerate and contains contributions from the (B) and
Reference ligands
dmbpy
dcbpy
ꢀ9.55
ꢀ1.09
ꢀ2.90
ꢀ9.55
8.46
7.44
5.39
7.19
ꢀ10.34
ꢀ10.34
(Q) fragments. Looking at the LUMO‘s, it can be mentioned that
a
all of them are centered on the quinoxaline fragment (Q), as shown
in Fig. 4.
Although experimentally COOEt was used, for simplicity the calculations were
performed for COOMe.

A. Francois et al. / Polyhedron 52 (2013) 62–71
69
Since the chemical potential remains relatively constant along
the series, the increase of the electron-acceptor capacity must be
governed by the reduction of the band gap caused by a stabiliza-
tion of the LUMO/SOMO molecular orbital.
4.3. Reduced charge distribution of ligands
Another interesting aspect to analyze is the effectiveness of the
substituent in the ppl ligand to concentrate charge and, therefore,
to increase its potentiality to inject electron density to the semi-
conductor conduction band. It should be noted that in this type
of molecules, charge analysis indicates that the nitrogens in bipyr-
idine as well as those in pyrazine are negatively charged, and,
2
Fig. 5. Electrophilicity Index of dcbpy, dmbpy and R ppl ligands (R = H, OH, COOMe,
COOH, CN).
2
therefore, compete with the acceptor group(R) in the R ppl ligand
to attract electronic charge.
It is important to analyse the charge distribution that would be
observed if an extra electron is received by the ligand, as occurs via
MLCT bands, when the ligand forms part of a ruthenium complex.
Fig. 6 shows this distribution along the component fragments
defined in Scheme 2. For example, on the dcbpy ligand, 71% of
the electronic charge is located on the bipyridine fragment (B)
and 29% on the carboxylated moiety (R). A different pattern is ob-
served on ppl (R = H), which contains the quinoxaline fragment (Q)
that stores 40% of the total charge. The hydroxyl moiety of the
(
2
OH) ppl ligand clearly diminishes the acceptor capacity of the
quinoxaline fragment. The ability of carbonyl and cyano R groups
to attract electronic density is evident since they concentrate 34%
and 35% of the total charge. It is interesting to note that a conse-
quence of the increase of the electron density on the quinoxaline
(
as for example in the dicyano ligand) is that its basicity also
increases, opening the possibility of side reactions that could affect
the performance of the dye in the cell, as has been pointed out pre-
viously [33].
Fig. 6. Charge distribution on the fragment components of the reduced free ligands.
Fig. 7. Molecular orbital Scheme for complexes 1–4.

7
0
A. Francois et al. / Polyhedron 52 (2013) 62–71
Fig. 8. (a) correlation between the HOMO energy (eV) and the first oxidation potential (V) and (b) influence of the ligand’s electrophilicity (eV) on the LUMO energy (ꢂ, eV)
o
and on the first reduction potential ( , V).
Table 6
T
It can be thought that a c relatively higher I value can be re-
Calculated values of I
T
for [Ru(dmbpy)
2
R
2
ppl](PF
6
)
2
complexes.
lated to a better performance of the dye in the cell. According to
this, among the complexes studied in this work, complex 3, bearing
the (COOH) ppl ligand, would be the most efficient dye, while in
2
complex 1, it is predicted that no injection would occur. The main
difference between complex 3 and complexes 2 and 4 is the
Complexes
E
A
f(BQ)
U
T
(Q)
I
T
1
2
3
4
1.31
1.12
1.89
1.10
1.87
0.30
0.29
0.23
0.21
0.41
0
0
0.30
0.31
0.52
0.83
0.10
0.14
0.12
0.64
absorbed energy (E
the thexi state, U (Q), with the highest value (0.52) for complex
. The relatively high value of the absorbed energy for complex 3
A
) and also the contribution of the anchor to
[
Ru(dcbpy)(bpy)
2
]2+
T
4
can be attributed to the presence of 33% of ligand to ligand, LL,
electronic transition (Table 3, excitations 3B and 3C), value that
in the case of the other three complexes does not exceed 3% (Tables
S1–S4). Preliminary IPCE results show that compound 3 has the
best response as dye in a solar cell.
4
.4. Molecular orbital calculations of the complexes
The HOMO and LUMO molecular orbitals were calculated for
2
+
2 2
the set of ruthenium compounds [Ru(dmbpy) R ppl] (R = OH,
COOMe, COOH, CN) under study, Fig. 7.
0
It can be seen that all the HOMO s are centered on the metal
orbitals with a small contribution of the ligands (3%) for the case
of R = H and OH. Fig. 8(a) shows the correlation between the en-
ergy of the HOMO level and the first oxidation potential. A linear
correlation coefficient of 0.98 is found. Fig. 7 shows that the LUMO
of the complexes is mainly the result of the contribution of two
molecular fragments, one centered in the bipyridine (B) and the
other on the quinoxaline fragment (Q). Complete data is given in
Tables S5–S8, Supplementary material. Electron donor substituents
such as OH localize electronic density in the bipyridine ring of ppl,
while electron-acceptor substituents such as CN and COOH display
it on the quinoxaline ring. It should be pointed out that the elec-
tronic distribution in the LUMO of the complexes shows some
differences in regard to the free ligands, Fig. 4. Specifically, upon
coordination, the metallic center seems to stabilize the B fragment
5
. Concluding remarks
In this article the synthesis and complete characterization of a
series of ligands of type R
their corresponding [Ru(dmbpy)
ported. Elementary analyses, IR and H NMR spectroscopic results
are consistent with the proposed molecular structures.
Experimental and theoretical studies were carried out in order
to evaluate the effect of the acceptor R substituent on the proper-
ties of the ppl ligands and complexes. In particular, as the com-
plexes are intended to be used as dyes in DSSCs, their potential
injection capacity on the semiconductor surface was estimated.
The results indicate that when R = CN, COOH or COOEt, an increase
in the electron-acceptor capacity of the ppl ligand is observed,
which can be theoretically estimated in terms of the electrophilic-
ity parameter. The general trend shown is:
2
ppl (R = OH, COOEt, COOH and OH) and
ppl](PF complexes is re-
2
R
2
6 2
)
1
(
located mainly on the LUMO + 1 in the free ligand, Fig. 4), turning
it part of the LUMO of the corresponding Ruthenium complexes.
The influence of ligand’s electrophilicity on the LUMO level and
on the first reduction potential is depicted in Fig. 8(b). It can be
seen that the increase of electrophilicity correlates with a stabiliza-
tion of the energy of the LUMO level; accordingly, the reduction
potential of the series of ligands follows the trend expected by
the electrophilic character of the ligand.
ðOHÞ ppl < ppl < ðCOOMeÞ ppl < ðCOOHÞ ppl < ðCNÞ ppl
The same trend was observed experimentally for the oxidation
potential, while the inverse tendency was detected for the reduc-
tion potential.
An analysis of the charge distribution on the SOMO orbitals of
the complexes allowed for an insight on how the extra electron
received by the ligand through MLCT absorption is distributed.
The carbonyl and cyano substituent group son complexes 3 and
4attract electronic density and concentrate 34–35% of the total
2
2
2
2
4.5. Theoretical evaluation of the complexes as dye sensitizer
T
The I parameter i.e. the amount of energy that can be delivered
to TiO from the acceptor anchoring ligand in the thexi state (see
2
T
charge received. This is reflected in the I parameter, that quanti-
Section 2) was calculated for complexes 1–4, Table 6. Also intro-
fies the potential injection capacity of the complexes, and which
has the highest values for these two complexes. Nevertheless in
complex 4, the increase of the electron density on the quinoxaline
opens the possibility of alternative deactivation routes, which
could affect the performance of the dye in the cell. In this way, of
duced in this table are data for a reference model complex,
2
+
[
Ru(dcbpy)(bpy)
parameter with a well known dye, with structural similarities
to the compounds of these series.
2
] , with the aim of checking the response of the
I
T

A. Francois et al. / Polyhedron 52 (2013) 62–71
[11] (a) Y. Qin, Q. Peng, Int. J. Photoenergy 2012 (2012) 21;
71
the four complexes studied in this work, complex 3 appears to be
(
b) M.K. Nazeeruddin, S.M. Zakeeruddin, J.-J. Lagref, P. Liska, P. Comte, C.
Barolo, G. Viscardi, K. Schenk, M. Gräetzel, Coord. Chem. Rev. 248 (2004)
317.
potentially the most promising dye to be used in a photoelectro-
chemical solar cell.
1
[
12] (a) T. Kinoshita, J.-I. Fujisawa, J. Nakazaki, S. Uchida, T. Kubo, H. Segawa, J.
Phys. Chem. Lett. 3 (2012) 394;
Acknowledgement
(
b) S. Altobello, R. Argazzi, S. Caramori, C. Contado, S. Da Fr e´ , P. Rubino, C.
Chon e´ , G. Larramona, C. Contado, C.A. Bignozzi, J. Am. Chem. Soc. 12 (7) (2005)
The support of Fondecyt through Project Grants 1020517 and
110991 is gratefully acknowledged.
15342.
[
13] (a) E. Galoppini, Coord. Chem. Rev. 248 (2004) 1283;
1
(
b) M.K. Nazeeruddin, R. Humphry-Baker, P. Liska, M. Grätzel, J. Phys. Chem. B
07 (2003) 8981.
[14] R. López, B. Loeb, T. Boussie, T.J. Meyer, Tetrahedron Lett. 37 (1996) 5437.
1
Appendix A. Supplementary data
[
15] (a) B.P. Sullivan, D.J. Salmon, T.J. Meyer, Inorg. Chem. 17 (1978) 3334;
b) P.A. Lay, A.M. Sargeson, H. Taube, Inorg. Synth. 24 (1986) 24.
(
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.poly.2012.12.003.
[
16] G. Velde, F.M. Bickelhaupt, S.J.A. Van Gisbergen, C. Fonseca Guerra, E.J.
Baerends, J.G. Snijders, T.J. Ziegler, Comput. Chem. 22 (2002) 931.
17] J.P. Perdew, Y. Wang, Phys. Rev. B 45 (1992) 13244.
18] F.L. Hirshfeld, Theor. Chim. Acta 44 (1977) 129.
19] O. Gritsenko, P. Schipper, E. Baerends, Chem. Phys. Lett. 302 (1999) 199.
20] A. Klamt, G. Schüürmann, J. Chem. Soc., Perkin Trans. 2 (1993) 799.
[
[
[
[
References
[
1] S. Caramori, V. Cristino, R. Boaretto, R. Argazzi, C.A. Bignozzi, A. Di Carlo, Int. J.
Photoenergy 2010 (2010) 1.
[21] F. Gajardo, M. Barrera, R. Vargas, I. Crivelli, B. Loeb, Inorg. Chem. 50 (2011)
5910.
[
[
[
2] M.K. Nazeeruddin, E. Baranoff, M. Grätzel, Sol. Energy 85 (2011) 1172B.
3] M. Grätzel, Inorg. Chem. 44 (2005) 6841.
4] M.K. Nazeeruddin, P. Péchy, T. Renouard, et al., J. Am. Chem. Soc. 123 (2001)
[22] A. Delgadillo, P. Romo, A.M. Leiva, B. Loeb, Helv. Chim. Acta 86 (2003)
2110.
[23] M.A. Ivanov, M.V. Puzyk, K.P. Balashev, Russ. J. Gen. Chem. 76 (2006) 843.
[24] B. Gholamkhass, K. Koike, N. Negishi, H. Hori, K. Takeuchi, Inorg. Chem. 40
(2001) 756.
[25] A. Ambroise, B.G. Maiya, Inorg. Chem. 39 (2000) 4264.
[26] M.D. Stephenson, T.J. Prior, M.J. Hardie, Cryst. Growth Des. 8 (2008) 643.
[27] R. Díaz, A. Francois, M. Barrera, B. Loeb, Polyhedron 39 (2012) 59.
[28] R. Díaz, A. Francois, A.M. Leiva, B. Loeb, E. Norambuena, M. Yañez, Helv. Chim.
Acta 89 (2006) 1220.
1
613.
[
[
[
5] F. Gajardo, A.M. Leiva, B. Loeb, A. Delgadillo, J.R. Stromberg, G.J. Meyer, Inorg.
Chim. Acta 361 (2008) 613.
6] (a) G.J. Meyer, J. Photochem. Photobiol. A 158 (2003) 119;
(
b) B.V. Bergeron, G.J. Meyer, J. Phys. Chem. B 107 (2003) 245.
7] (a) K. Kalyanasundaram, M. Grätzel, Coord. Chem. Rev. 177 (1998) 347;
b) K. Murakoshi, G. Kano, Y. Wada, S. Yanagida, H. Miyazaki, M. Matsumoto, S.
(
Murasawa, J. Electroanal. Chem. 396 (1995) 27.
8] O.V. Prezhdo, W.R. Duncan, V.V. Prezhdo, Acc. Chem. Res. 41 (2008) 339.
9] (a) V. Duprez, M. Biancardo, F.C. Krebs, Sol. Energy Mater. Sol. Cells 91 (2007)
[29] D.P. Strommen, P.K. Mallick, G.D. Danzer, R.S. Lumpkin, J.R. Kincaid, J. Phys.
Chem. 94 (1990) 1357.
[30] E. Norambuena, C. Olea-Azar, A. Delgadillo, M. Barrera, B. Loeb, Chem. Phys.
359 (2009) 92.
[31] R. Parr, L. Szentpaly, S. Liu, J. Am. Chem. Soc. 121 (1999) 1922.
[32] R. Parr, W. Yang, Density-Functional Theory of Atoms and Molecules, Oxford
University Press, New York, 1989.
[33] M. Arias, Complejos polipiridínicos de Ru (II) con ligandos altamente
conjugados con ancla fosfonato y compuestos ciclometalados de Ru (II) con
ancla carboxilato, Thesis to obtain the Dr. Degree in Chemistry, P. Catholic
University of Chile, 2007.
[
[
2
(
1
(
30;
b) P. Piotrowiak, E. Galoppini, Q. Wei, G.J. Meyer, P. Wiewio, J. Am. Chem. Soc.
25 (2003) 5278;
c) O. Taratula, J. Rochford, P. Piotrowiak, E. Galoppini, R.A. Carlisle, G.J. Meyer,
J. Phys. Chem. B 110 (2006) 15734.
[
10] (a) I. Gillaizeau-Gauthier, F. Odobel, M. Alebbi, R. Argazzi, E. Costa, C.A.
Bignozzi, P. Qu, G.J. Meyer, Inorg. Chem. 40 (2001) 6073;
(
b) S. Ardo, G.J. Meyer, Chem. Soc. Rev. 38 (2009) 115.