Strongly Coupled Cyclometalated Ruthenium Triarylamine Chromophores as Sensitizers for DSSCs

Article abstract of DOI:10.1002/chem.201601001

A series of anchor-functionalized cyclometalated bis(tridentate) ruthenium(II) triarylamine hybrids [Ru(dbp-X)(tctpy)]^2?[2 a]^2?–[2 c]^2?(H₃tctpy=2,2′;6′,2′′-terpyridine-4,4′,4′′-tricarboxylic acid; dpbH=1,3-dipyr

Full text of DOI:10.1002/chem.201601001

DOI: 10.1002/chem.201601001
Full Paper
&
Cyclometalated Complexes
Strongly Coupled Cyclometalated Ruthenium Triarylamine
Chromophores as Sensitizers for DSSCs
Christoph Kreitner+,[a, d] Andreas K. C. Mengel+,[a] Tae Kyung Lee,[b] Woohyung Cho,[b]
Kookheon Char,[c] Yong Soo Kang,[b] and Katja Heinze*[a]
Abstract: A series of anchor-functionalized cyclometalated
bis(tridentate) ruthenium(II) triarylamine hybrids [Ru(dbp-
X)(tctpy)]2À [2a]2À–[2c]2À (H3tctpy=2,2’;6’,2’’-terpyridine-
4,4’,4’’-tricarboxylic acid; dpbH=1,3-dipyridylbenzene; X=
N(4-C6H4OMe)2 ([2a]2À), NPh2 ([2b]2À), N-carbazolyl [2c]2À)
was synthesized and characterized. All complexes show
broad absorption bands in the range 300–700 nm with
increasing spin delocalization between the metal center and
the triarylamine unit in the order [1a]2+ <[1b]2+ <[1c]2+
.
Solar cells were prepared with the saponified complexes
[2a]2À–[2c]2À and the reference dye N719 as sensitizers
using the I3À/IÀ couple and [Co(bpy)3]3+/2+ and
[Co(ddpd)2]3+/2+ couples as [B(C6F5)4]À salts as electrolytes
(bpy=2,2’-bipyridine; ddpd=N,N’-dimethyl-N,N’-dipyridin-2-
À
a
maximum
at
about
545 nm.
Methyl
esters
yl-pyridine-2,6-diamine). Cells with [2c]2À and I3 /IÀ electro-
[Ru(Me3tctpy)(dpb-X)]+ [1a]+–[1c]+ are oxidized to the
strongly coupled mixed-valent species [1a]2+–[1c]2+ and
the RuIII(aminium) complexes [1a]3+–[1c]3+ at comparably
low oxidation potentials. Theoretical calculations suggest an
lyte perform similarly to cells with N719. In the presence of
cobalt electrolytes, all efficiencies are reduced, yet under
these conditions [2c]2À outperforms N719.
Introduction
electrode. The major advantages of dye sensitized solar cells
over conventional silicon-based or inorganic thin film solar
cells are lower costs and their modular architecture allowing
for systematic optimization of all components (semiconductor,
sensitizer, electrolyte) individually.[3–5]
Pioneered by O’Regan and Grätzel in 1991,[1] the dye-sensitized
solar cell (DSSC) has emerged as a promising light-to-energy
conversion device.[2,3] Its setup has been optimized and stand-
ardized over the past 25 years. Typically, its central component
is a molecular dye that is absorbed onto a mesoporous wide-
bandgap semiconducting electrode, such as TiO2 or ZnO.[3,4]
Upon excitation by visible light, electrons are injected from the
excited state of the dye into the conduction band of the semi-
conductor. The oxidized dye is then regenerated by a redox
mediator, which transports the positive charge to the counter
Tremendous efforts have been put particularly into the de-
velopment of new molecular dyes to optimize cell per-
formance. An ideal sensitizer should be thermally and photo-
chemically stable under working conditions, should rapidly
inject electrons into the conduction band of the semiconduc-
tor after excitation and, most importantly, should efficiently
absorb light between 400 and 900 nm. Among others, polypyr-
idine complexes of iron,[6] copper,[7–9] platinum,[10,11] iridium,[12,13]
and rhenium[14] as well as polyaromatic and conjugated organ-
ic compounds,[15,16] porphyrins,[17–19] and quantum dots[20] have
proven suitable for sensitization. Particularly, polypyridine com-
plexes of ruthenium and osmium have emerged as a promising
class of sensitizers due to their suitable photophysical proper-
ties.[21–24] The visible range of the electromagnetic spectrum of
these complexes is dominated by characteristic metal-to-ligand
charge transfer (MLCT) absorptions.[25–27] In these transitions,
metal orbitals of the t2g set serve as electron donors, while the
polypyridine p* orbitals function as electron acceptors.
[a] C. Kreitner,+ A. K. C. Mengel,+ Prof. Dr. K. Heinze
Institute of Inorganic Chemistry and Analytical Chemistry
Johannes Gutenberg-University of Mainz
Duesbergweg 10–14, 55128 Mainz (Germany)
Fax: (+49)6131-39-27-277
E-mail: katja.heinze@uni-mainz.de
[b] T. K. Lee, W. Cho, Prof. Y. S. Kang
The Department of Energy Engineering and Center for
Next Generation Dye-Sensitized Solar Cells, Hanyang University
222 Wangsimni-ro, Seongdong-gu, Seoul 133-791 (Korea)
[c] Prof. Dr. K. Char
The National Creative Research Initiative Center for Intelligent
Hybrids, School of Chemical and Biological Engineering, Seoul
National University, 1 Gwanak-ro, Gwanak-gu, Seoul 151-744 (Korea)
The most prominent and well-established sensitizers are the
complexes [nBu4N]2[Ru(Hdcbpy)2(NCS)2], N719 (H2dcbpy=2,2’-
bipyridine-4,4’-dicarboxylic acid, Scheme 1),[28,29] and the
so-called “black dye” [nBu4N]3[Ru(Htctpy)(NCS)3] (H3tctpy=
[d] C. Kreitner+
Graduate School Materials Science in Mainz
Staudingerweg 9, 55128 Mainz (Germany)
[+] These authors contributed equally to the work.
2,2’;6’,2’’-terpyridine-4,4’,4’’-tricarboxylic
acid)[22]
reaching
power conversion efficiencies (PCE, h) of 10–11% under full air
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201601001.
mass 1.5 (AM 1.5) irradiation. In these complexes, the carboxy
Chem. Eur. J. 2016, 22, 8915 – 8928
8915
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
clometalating ligands as electron acceptors.[37,38,43–48] Addition-
ally, cyclometalation substantially increases the energy of the
polypyridine-centered lowest unoccupied molecular orbital
(LUMO) compared to non-cyclometalated counterparts. This
should potentially accelerate charge injection into the TiO2
conduction band.[37,38,48] The highest occupied molecular orbi-
tal (HOMO) on the other side typically extends over the metal
center and the anionic part of the cyclometalating ligand. This
should facilitate dye regeneration after charge injection.[38,46,48]
Furthermore, the high s-donating strength destabilizes the in-
herently photochemically reactive metal-centered (3MC) excit-
ed states.[38,48–52] The electron donating or withdrawing charac-
ter of the cyclometalating ligands are easily tuned by further
substitution (for example [AX,Me]+, Scheme 1) including hole-
transport facilities (X=amines).[38,48–52] Indeed, several ap-
proaches have been developed to incorporate electron donors
into the dye structure to rapidly detract the positive charge re-
maining on the sensitizer after electron injection away from
the semiconductor surface.[53–56] Attaching the reversible tri-
phenylamine radical cation/triphenylamine redox couple
0
(TPA+ /TPA ) proved particularly successful in conjunction with
C
several porphyrin dyes, for example, YD2-o-C8, yielding solar
cells with h>12%.[19,57] Berlinguette and co-workers demon-
strated that the overall cell performance can benefit from
a TPA unit linked to a [Ru(pbpy)(tpy)]+ complex via a thiophene
spacer. This architecture yields cell efficiencies of up to 8.0%
(Scheme 1, [BH]+, [CH]+).[38] Through clever dye design and ad-
justment of relative oxidation potentials of RuIII/II and TPA+ /0
C
Scheme 1. N719 reference dye and amine-substituted cyclometalated ruthe-
nium(II) dyes (top) and cobalt-based electrolytes (bottom).
an efficient transfer of the electron hole from the ruthenium
center to the TPA unit is achieved after charge injection. This
retards parasitic electron recombination processes with oxi-
dized dyes in the DSSC.[38,58–60] The mixed-valent complexes
[BR]2+ are valence-localized and assigned to Robin–Day class II
with measurable electronic coupling between the metal center
and the TPA unit.[60,61] Recently, Zhong and co-workers present-
ed a structurally related series of complexes combining bis(tri-
dentate) cyclometalated ruthenium complexes with TPA units
groups serve as anchors to the TiO2 surface while the [NCS]À li-
gands are responsible for an efficient charge transfer from the
redox mediator onto the dye after charge injection (dye regen-
eration).[29] However, a major drawback of complexes contain-
ing monodentate ligands is their high lability towards [NCS]À
substitution in photoexcited or oxidized states hampering
long-term application in photovoltaic devices.[30–33]
(Scheme 1, [AX,Me 2+
]
/[AX,Me]+) lacking the thiophene unit.[62,63]
is valence-delocalized (Robin–
The mixed-valent state [AX,Me 2+
]
Recently, bi- and tridentate cyclometalating ligands emerged
as viable and more robust alternatives for the labile [NCS]À li-
gands. In 2007, van Koten and co-workers reported the suc-
cessful sensitization of TiO2 by bis(tridentate) [Ru(pbpy)(tpy)]+
complexes (tpy=2,2’;6’,2’’-terpyridine, Hpbpy=6-phenyl-2,2’-
bipyridine).[34] Shortly thereafter, Grätzel and co-workers
Day class III) between the metal center and the amine moiety
as evidenced by the shape and bandwidth of the near infrared
absorption band and by density functional theoretical calcula-
+
tions.[62–65] The parent complex [AH,H
]
lacking the amine sub-
stituent (X=H) has been reported recently as well.[66]
In contrast to reported dyes [BR]+, featuring a valence-iso-
published
a
dye with record-breaking characteristics
meric description of the [BR]2+ state (Robin–Day class II), po-
[Ru(H2dcbpy)2(ppy-F2)]+ (h>10%) based on a tris(bidentate)
cyclometalating motif (Hppy-F2 =2-(2,4-difluorophenyl)pyri-
dine).[35] Since then, much work has been dedicated towards
the development of new cyclometalated ruthenium dyes both
in the field of tris(bidentate) and bis(tridentate) complex archi-
tectures.[36–42] These studies indeed reveal several key benefits
of the cyclometalating motif. The introduction of a Ru-C
s bond in the coordination environment reduces the local
symmetry around the metal center. This yields a broad absorp-
tion band in the visible range resulting from multiple close-
lying MLCT transitions involving both the polypyridine and cy-
tential DSSC sensitizers [AX,H
]
with X=amine that provide
+
a means of detracting the electron hole away from the semi-
conductor surface in a resonant fashion ([AX,H 2+
; Robin–Day
]
class III) have not yet been reported. Saponifying the three
+
methyl esters of [AX,Me
]
type complexes should provide suita-
+
ble sensitizers [AX,H
]
with a class III mixed-valent state. Herein,
we present a series of three complexes of the general structure
[nBu4N]2[Ru(dpb-X)(tctpy)] (Hdpb-X=5-substituted 1,3-di-(2-
pyridyl)benzene) with different amine substituents X of increas-
ing electron withdrawing power, namely N,N-bis(4-methoxy-
phenyl)amine (X=N(4-C6H4OMe)2; [nBu4N]2[2a]), N,N-diphenyl-
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8916
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
amine (X=N(C6H5)2; [nBu4N]2[2b]) and carbazole (X=N-carba-
zolyl; [nBu4N]2[2c]). We will discuss how the substituents at the
dpb ligand affect the degree of valence-delocalization in the
mixed-valent state [2]À and to what extent such delocalization
is beneficial for the application of such sensitizers in DSSCs.
As outer-sphere cobalt-based electrolytes[67–80] should deliver
higher open-circuit voltages VOC due to their more positive
redox potential as compared to the standard triiodide/iodide
couple and as they perform extremely well in conjunction with
TPA-appended porphyrin dyes (YD2-o-C8)[57] as well as with
other potent TPA-appended dyes (Y123, D35),[72,73,74] we study
the
TPA-appended
ruthenium(II)
dyes
[nBu4N]2[2a]–
[nBu4N]2[2c] with cobalt-based electrolytes in addition to the
commonly used triiodide/iodide couple. Specifically, we
employ the [Co(bpy)3]3+/2+ [3]3+/2+ and [Co(ddpd)2]3+/2+
[4]3+/2+ redox mediators (bpy=2,2’-bipyridine, ddpd=N,N’-di-
methyl-N,N’-dipyridin-2-yl-pyridine-2,6-diamine,[80]
Scheme 1).[81–83] The DSSCs are studied by incident photon-to-
current conversion efficiency measurements, by current-volt-
age characteristics under AM 1.5 irradiation and in the dark as
well as by electron lifetime measurements.
Scheme 2. Synthesis of [1a][PF6]–[1c][PF6] and [nBu4N]2[2a]–[nBu4N]2[2c].
[Ru(dpb)(tpy)]+ complexes, suggesting the presence of
a second colored species. Yet, cyclic voltammograms confirm
the purity of the synthesized complexes by absence of redox
waves in the range of À3.0 and 1.5 V other than the five ex-
Results and Discussion
pected reversible waves,[62,63a] namely for the [1]3+/2+, [1]2+/+
,
Synthesis and characterization of chromophores
[1]+/0, [1]0/À and [1]À/2À couples (Figure 1, Supporting Informa-
The 5-substituted 1,3-di-(2-pyridyl)benzene ligands La (R=N(4-
C6H4OMe)2), Lb (R=N(C6H5)2) and Lc (R=N-carbazolyl) were syn-
thesized starting from the previously reported 1-bromo-3,5-di-
(2-pyridyl)benzene under Buchwald–Hartwig cross-coupling re-
action conditions similar to a method we,[48] as well Zhong and
co-workers employed previously.[63a] In the present study, the
dimeric palladium(II) precatalyst bis(m-mesylate)bis[(2-(2’-ami-
tion, Figure S5).
[84]
nophenyl-kN)phenyl-kC1)palladium(II)] [Pd]2 was used along
with the phosphane ligand 2-dicyclohexylphosphano-2’,6’-dii-
sopropoxybiphenyl[85] to provide a catalytically competent cat-
alyst that afforded the ligands in yields of 82–98%. The identi-
1
ty of La was confirmed by comparison of its H NMR spectrum
with that reported before.[62,63a] The purity and integrity of the
new ligands Lb and Lc were ascertained by 1H and 13C NMR
spectroscopy, mass spectrometry, and elemental analyses (Ex-
perimental Section; Supporting Information, Figures S1–S4).
The heteroleptic ester-substituted [Ru(dpb)(tpy)]+ com-
plexes [1a]+–[1c]+ were prepared according to a previously
employed synthetic method starting from RuCl3(Me3tctpy)
(Scheme 2).[22,48] The two-step procedure includes chloride ab-
straction with silver tetrafluoroborate followed by complexa-
tion with the respective dipyridylbenzene ligand La–Lc under
reducing conditions in n-butanol. Similar to observations made
by Zhong and co-workers,[62,63a] we were not able to isolate the
complexes [1a]+–[1c]+ with high purity. Despite the reducing
conditions during their synthesis, substantial amounts of the
open-shell RuIII complexes [1a]2+–[1c]2+ were obtained, as evi-
denced from ESI mass spectra and the NMR silence of all three
compounds (paramagnetic broadening).[62,63a] Additionally, the
isolated products are black in solution and in the solid state in-
stead of the dark purple color typically observed for
Figure 1. Cyclic voltammograms of a) [1b][PF6] (a) and b) [nBu4N]2[2b]
(c) dyes in [nBu4N][PF6]/CH3CN (E vs. FcH/FcH+).
Subsequent saponification of the three methyl ester groups
of [1a]+–[1c]+ in aqueous solution using [nBu4N][OH] as base
and hydrazine as reductant yielded the corresponding carboxy-
lates as tetrabutylammonium salts [nBu4N]2[2a]–[nBu4N]2[2c].
This method affords the fully deprotonated complexes [2a]2À–
[2c]2À, in contrast to Berlinguette’s procedure,[38] which yields
the complexes in their neutral zwitterionic form with two pro-
tonated carboxy groups. Owing to the high solubility of the
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8917
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tetrabutylammonium salts [nBu4N]2[2a]–[nBu4N]2[2c] in organic
solvents, the products are isolated straight-forwardly by extrac-
tion of the aqueous phase with dichloromethane. Co-extracted
[nBu4N][PF6] was removed by subsequent dissolution of the
raw products in acetonitrile and addition of a diethylether/hex-
anes mixture that precipitates the desired complexes. The in-
1
tegrity of [nBu4N]2[2a]–[nBu4N]2[2c] was confirmed by H NMR
and 13C NMR spectra as well as by ESI+ and ESIÀ mass spectra
(Supporting Information, Figures S6–S19). All NMR spectra lack
paramagnetic shifts or broadening, substantiating the absence
1
of RuIII in the pristine samples. The H NMR spectra confirm the
presence of two equivalents of [nBu4N]+ cations per complex
anion in all three cases corroborating the stoichiometry of the
salt. IR spectra as KBr disk of the complexes [nBu4N]2[2a]–
[nBu4N]2[2c] lack the characteristic vibrations of [PF6]À ions at
843 cmÀ1 (asym. stretch) and 588 cmÀ1 (deformation) present
in the parent complexes [1a][PF6]–[1c][PF6] underlining the
quantitative [PF6]À removal (Supporting Information, Fig-
ure S20). Additionally, 19F NMR spectra of [nBu4N]2[2a]–
[nBu4N]2[2c] confirm the absence of [PF6]À. Under the acidic
and ionizing conditions of the ESI+ mass spectrometry tech-
nique, the complexes are observed in their fully protonated
form as monocations [2+3H]+ with RuII metal sites or as dicat-
ions [2+3H]2+ with RuIII centers (Supporting Information, Fig-
ure S18). The ESIÀ mass spectra (Supporting Information, Fig-
ure S19) show mass peaks at the expected m/z values for the
dianions [2]2À (RuII) and anions [2]À (RuIII) with typical rutheni-
um isotope patterns. Furthermore, several m/z peaks of decar-
boxylated complexes are found for all three complexes
[nBu4N]2[2a]–[nBu4N]2[2c], confirming the presence of carbox-
ylate substituents. The carboxylate groups are also evident
from the characteristic IR CO stretching vibrations around
1617 cmÀ1 (Supporting Information, Figure S20).
Figure 2. Normalized absorption and emission spectra of [nBu4N]2[2a] (c),
[nBu4N]2[2b] (g), and [nBu4N]2[2c] (a) in CH3CN.
accepting sites (dRu!p*tpy and dRu!p*dpb).[38,48] Owing to the
low local symmetry around the metal center, the number of
absorption bands is larger than for the more symmetric sys-
tems containing all-nitrogen donor ligands such as [Ru(tpy)2]2+
or [Ru(ddpd)(tpy)]2+, for example.[38,46,47,86] The lower symmetry
yields substantially broadened absorption spectra and a more
efficient light harvesting throughout the visible range of the
electromagnetic spectrum.
The tris(carboxylate) complexes [nBu4N][2a]–[nBu4N][2c] are
very weakly emissive at room temperature (Figure 2, Table 1)
with quantum yields below 5”10À6. The wavelength of the
emission maximum is shifted hypsochromically in the order
[nBu4N][2a]> [nBu4N][2b]> [nBu4N][2c] from 817 to 744 nm.
On the one hand, this trend is due to a more pronounced vi-
brational progression in [nBu4N][2a] and [nBu4N][2b] than in
[nBu4N][2c] similar to that observed for other [Ru(dpb)(tpy)]+
complexes with a strong push–pull substitution.[48] On the
Photophysical and electrochemical behavior
3
The absorption and emission spectra of the complexes [nBu4N]
[2a]–[nBu4N][2c] are depicted in Figure 2 and data are sum-
marized in Table 1 (Supporting Information, Figure S21). In the
spectral range between 300 and 800 nm all dyes exhibit very
similar absorption features. The absorption maximum around
536–549 nm is accompanied by three additional bands around
500, 425, and 375 nm. These bands characteristic for cyclome-
talated [Ru(dpb)(tpy)]+ complexes arise from metal-to-ligand
charge transfer excitations involving both ligands as electron-
other hand, the energy of the emissive MLCT state increases
with decreasing donor strength of the amine substituent, as
this lowers the energy of the metal orbitals involved in the
emission process while essentially maintaining the tctpy-cen-
tered LUMO energy (Figure 3; Supporting Information, Fig-
ure S22).[48]
Cyclovoltammetric studies of the ester-substituted com-
plexes [1a][PF6]–[1c][PF6] reveal multiple reversible redox pro-
cesses (Figure 1, Table 1, Supporting Information, Figure S5).
Table 1. Optical and electrochemical data of [1a][PF6]–[1c][PF6] and [nBu4N]2[2a]–[nBu4N]2[2c].
UV/Vis (CH3CN) lmax/nm (e/103 mÀ1 cmÀ1
)
Emission (CH3CN) lem/nm (lexc/nm) Cyclic voltammetry E/V vs. FcH/FcH+
[b]
[1a][PF6][a]
[1b][PF6]
[1c][PF6]
323 (29), 339 (26), 420 (16), 507 (13), 583 (10)
–
–
–
À1.87, À1.52, À0.05, +0.31
À1.85, À1.49, +0.09, +0.49
À1.83, À1.47, +0.34, +0.88
À2.52, À2.09, À0.21, –[d]
À2.54, À2.10, À0.15, –[d]
À2.51, À2.07, À0.06, –[d]
[b]
[b]
[b]
–
[b]
–
[nBu4N]2[2a] 549 (15.4), 503 (12.1), 424 (9.7), 379 (11.3), 323 (37.5), 289 (62.7)
[nBu4N]2[2b] 543 (15.4), 501 (12.1), 428 (8.7), 374 (10.4), 325 (36.4), 288 (61.6)
[nBu4N]2[2c] 536 (13.9), 499 (12.5), 426 (9.1), 373 (10.0), 328 (29.1), 283 (63.2)
817 (549)[c]
791 (543)[c]
744 (536)[c]
[a] From Ref. [63a]. [b] No optical data measured due to the presence of RuIII species. [c] Quantum yield <5·10À6. [d] No second oxidation potential was ob-
tained due to precipitation of the neutral dye on the electrode surface.
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8918
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
600 mV towards more negative values (Figure 1, Table 1; Sup-
porting Information, Figure S5). This is consistent with the sub-
stantial increase of negative charge density on the tctpy3À
ligand in [2]2À and corroborates the tpy centered reduction.
The oxidation waves shift to lower potentials as well, but to
a lesser extent. This is mainly due to the fact, that the metal or-
bital involved with the oxidation process is orthogonal to the
tpy ligand. However, a trend of the potential shifts is observed.
While the first oxidation wave of [2a]2À occurs 160 mV below
that of [1a]+, the first oxidation potentials of [2c]2À and [1c]+
differ by 400 mV. Consequently, the first oxidation potentials of
complexes [2a]2À–[2c]2À only differ by 150 mV as opposed to
a difference of 390 mV between the esters [1a]+–[1c]+. This
can be understood on the basis of the Mulliken spin popula-
tions of the metal center and the amine nitrogen atom in the
mixed-valent anions [2a]À–[2c]À: These amount to 0.49 (Ru)/
0.17 (N) in [2a]À, 0.57 (Ru)/0.13 (N) in [2b]À, and 0.74 (Ru)/0.03
(N) in [2c]À. Apparently, the charge delocalization over the tri-
arylamine fragment is significantly reduced and the spin densi-
ties of the mixed-valent anions [2]À are more valence-localized
at the electron-rich metal center than their ester counterparts.
Consequently, the resonance stabilization of the mixed-valent
species [2]À is not as pronounced as that of [1]2+ resulting in
similar oxidation potentials for all three complexes. The stron-
gest impact of deprotection and deprotonation is observed for
[2c]2À, since its spin density is basically metal-centered. Ac-
cordingly, oxidation occurs in the closest proximity to the neg-
atively charged tctpy3À ligand and is facilitated to the largest
extent in the dye series [2]À.
Figure 3. a) Frontier molecular orbitals of [1a]+, [1b]+, and [1c]+ (contour
value 0.06 a.u.) and b) spin densities of [1a]2+, [1b]2+, and [1c]2+ (contour
value 0.01 a.u.).
The complexes are oxidized at quite low potentials to the
mixed-valent counterparts [1]2+ (À0.05 to 0.34 V vs.
FcH/FcH+). A second oxidation step occurs at higher poten-
tials yielding the RuIII(aminium) complexes [1]3+ (0.31 to 0.88 V
vs. FcH/FcH+).[62] The trend of the first and the second oxida-
tion potentials towards higher values in the order N(4-
C6H4OCH3)2 <N(C6H5)2 <N-carbazole is in agreement with the
decreasing +I effect of the respective amine substituent X. Ad-
ditionally, the unpaired electron of the mixed-valent com-
pounds [1]2+ is substantially delocalized between the metal
center and the triarylamine fragment via the 1,4-phenylene
bridge.[62,63a] The electron donor strength of the dpb substi-
tuent[62,63a] strongly affects degree of delocalization as evi-
denced by DFT calculations (Figure 3b). Mulliken spin density
analysis of [1a]2+ indicates a balanced spin population of 0.26
and 0.25 at ruthenium and the amine nitrogen atom, respec-
tively, while for [1b]2+, values of 0.43 (Ru) and 0.20 (N) are ob-
tained. In [1c]2+, the spin density is further shifted towards the
ruthenium atom with spin populations of 0.63 (Ru) and 0.07
(N). Thus, the degree of delocalization is reduced in this series
from an essentially delocalized Robin–Day class III system in
Increasing the potential beyond 0.15 V vs FcH/FcH+ results
in a multitude of irreversible redox waves. We ascribe this to
the deposition of the neutral RuIII(aminium) complexes [2]0 on
the platinum electrode surface, which impeded an unambigu-
ous determination of the second oxidation potentials.
Combining all electrochemical and spectroscopic data of the
dyes [1a]+–[1c]+ and [2a]2À–[2c]2À with the redox data of the
À
electrolytes I3 /IÀ, 33+/2+ and 43+/2+ and the conduction band
edge of TiO2 yields the redox potential diagram depicted in
Figure 4. Cyclic voltammetry of the dyes (see above) highlight-
ed the strong dependence of the ground state oxidation po-
tentials of the dyes from the degree of protonation of the car-
boxylic acids of the tpy ligand. For the setup of the DSSCs, the
tris(carboxylate) dyes [2a]2À–[2c]2À were employed. Yet, under
the given experimental conditions, partial protonation from
water at the TiO2 surface is conceivable. Additionally, the coad-
sorbent chenodeoxycholic acid (CDCA), as an organic acid, will
modify the degree of protonation of the sensitizer. Hence,
Figure 4 depicts redox potential ranges instead of distinct
values for the redox potentials of the ruthenium dyes. An anal-
ogous range of ground state potentials is applied for N719
owing to the conceivable variation of the protonation state.[28]
[1a]2+ to a strongly coupled class II compound in [1c]2+ [61,63a]
.
As a consequence, the resonance stabilization within complex
[1a]2+ is the largest, yielding the most pronounced negative
shift of the first oxidation potential followed by complexes
[1b]2+ and [1c]2+. Additionally, two reversible reduction waves
are observed for the ester-substituted complexes [1a][PF6]–
[1c][PF6]. As these reductions are tpy-centered,[33b,46,48a,63a,87,88]
their potentials are essentially independent from the substitu-
tion pattern at the dpb-X ligand with the first reduction occur-
ring at À1.5 V and the second at À1.85 V vs. FcH/FcH+ for all
three complexes.
1
3
In a similar manner, the excited state MLCT and MLCT redox
potentials span ranges.
It is apparent from Figure 4 that, similar to the reference dye
N719, all cyclometalated dyes are thermodynamically capable
of injecting an electron from both excited states into the con-
duction band of TiO2. Regeneration of the oxidized dyes by
Deprotection and deprotonation of the ester functionalities,
however, shifts the first and second reduction potentials by
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8919
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 4. Diagram of the ground-state (1GS) and excited-state (1MLCT and
3MLCT) redox potentials of [1a]+/[2a]À, [1b]+/[2b]À, [1c]+/[2c]À, and N719,
the conduction band edge of TiO2, and the redox potentials of the electro-
lytes.
the employed electrolytes on the other hand is not generally
possible. While the carbazole-substituted ruthenium(III) com-
plex [2c]À is potentially regenerated by iodide even if the diio-
dide radical anion is formed as an intermediate,[89,90] this is not
the case for the diarylamine-substituted dyes [2a]2À and [2b]2À
in this simplified consideration of standard redox potentials.
Hence, in a DSSC [2c]2À is expected to outperform [2a]2À and
[2b]2À. DSSC performances of all cyclometalated dyes in con-
junction with the standard triiodide/iodide electrolyte and
cobalt electrolytes will be discussed in the next section.
Solar cell performance
Figure 5. Photocurrent action spectra of cells with [nBu4N]2[2a],
[nBu4N]2[2b], [nBu4N]2[2c], and N719. a) I3À/IÀ redox mediator with/without
CDCA (c/a); b) [3]3+/2+ (a) and [4]3+/2+ (c) electrolytes.
Dye-sensitized solar cells were prepared from the carboxylate
substituted dyes [nBu4N]2[2a]–[nBu4N]2[2c] and the benchmark
dye N719. The coadsorbent CDCA was employed to protect
the TiO2 surface in several setups. Three different liquid electro-
lytes were utilized, namely the standard triiodide/iodide couple than that found with N719 (Experimental Section). Hence, we
and two cobalt-based redox mediators.[67–69,82,83] The cobalt(III/ assume that the light-harvesting efficiencies of [nBu4N]2[2a]–
II) complexes [Co(bpy)3]3+/2+ and [Co(ddpd)2]3+/2+ were pre- [nBu4N]2[2c] in the cells prepared are in the same range as
pared according to literature procedures.[74,78,82,91] Counter ion that of the reference dye N719.
exchange was accomplished using Li[B(C6F5)4] giving the cobalt
Concerning the injection efficiency, all dyes feature 1MLCT
3
salts [Co(bpy)3][B(C6F5)4]2/[Co(bpy)3][B(C6F5)4]3 [3][B(C6F5)4]2/[3] and MLCT levels well above the Fermi level of TiO2 (Figure 4).
[B(C6F5)4]3 and [Co(ddpd)2][B(C6F5)4]2/[Co(ddpd)2][B(C6F5)4]3 [4] Hence, electron injection from the excited sensitizers should
[B(C6F5)4]2/[4][B(C6F5)4]3. NMR and mass spectrometric data con- be feasible and fast for all dyes.[67] We assume rather similar in-
firm their compositions (Supporting Information, Figure S23– jection efficiencies for all dyes.[67] Hence, the differences in
[82]
S31). The redox potential of [4]3+/2+ (E = =0.52 V vs. NHE) is DSSC performance in terms of power conversion efficiency h
1
2
intermediate of the I3 /IÀ (E = =0.32Æ0.03 V vs. NHE) and should predominantly relate to the dye regeneration efficiency,
À
[89]
1
2
the [3]3+/2+ (E = =0.65 V vs. NHE)[92–94] couples (Figure 4).
the charge collection efficiency (recombination losses) and the
1
2
The incident photon-to-current conversion efficiency (IPCE, open-circuit voltage VOC.
Figure 5) depends on several individual key steps, namely
According to the electrochemical data of the dyes and the
light-harvesting, electron injection, dye regeneration, and redox mediators, the RuIII complex [2b]À cannot be efficiently
charge collection.[67] The extinction coefficients around the regenerated by the bipyridine cobalt(II) complex [3]2+, while
MLCT maxima of [nBu4N]2[2a]–[nBu4N]2[2c] (Figure 2, Experi- [2a]À cannot be regenerated by both cobalt(II) complexes
mental Section) are close to that of N719 (l=535 (14700), 395 [3]2+ and [4]2+ (Figure 4). For all other dye/redox mediator
(14300mÀ1 cmÀ1
)
nm.[28]). The achieved dye loadings of combinations, dye regeneration is thermodynamically possible.
[nBu4N]2[2a]–[nBu4N]2[2c] are consistently somewhat higher In the electrolyte series, the highest driving force for dye re-
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8920
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
generation and hence the highest regeneration rate based on
Table 3. Photovoltaic data of cells using the [3]3+/2+ and [4]3+/2+ redox
mediators under AM1.5 light conditions.
À
Marcus theory (Marcus normal region) is achieved with the I3 /
À
ÀC
IÀ couple (although the [I2] /I couple with a higher potential
Dye
Electrolyte
VOC [V]
JSC [mAcmÀ2
]
FF [%]
h [%]
is responsible for the dye regeneration; Figure 4[89,90]). As the
regeneration rate furthermore depends on the concentration
of the reduced mediator,[67] the regeneration rate should de-
crease in the series IÀ (0.550m), [3]2+ (0.165m), and [4]2+
[nBu4N]2[2a]
[nBu4N]2[2a]
[nBu4N]2[2b]
[nBu4N]2[2b]
[nBu4N]2[2c]
[nBu4N]2[2c]
N719
[3]3+/2+
[4]3+/2+
[3]3+/2+
[4]3+/2+
[3]3+/2+
[4]3+/2+
[3]3+/2+
[4]3+/2+
[3]3+/2+
[3]3+/2+
[3]3+/2+
0.10
0.59
0.61
0.66
0.70
0.66
0.69
0.64
0.58
0.62
0.65
0.30
1.07
0.95
1.90
2.63
3.06
2.80
2.68
3.03
3.80
5.47
21
66
70
73
71
71
57
62
66
76
49
<0.1
0.4
0.4
0.9
1.3
1.4
1.1
1.1
1.1
1.8
1.8
(0.080m). All of these factors strongly favor the I3 /IÀ couple
À
over cobalt-based couples.
Electron recombination with oxidized dyes (driving forces
around 1.0–1.5 eV;[67] Figure 4) is reported to occur in the
Marcus inverted region. However, the rates better correlate
with the inverse distance of the positive charge in the dye and
the TiO2 surface instead with the driving force.[67] In oxidized
N719 the positive charge is delocalized between Ru and the
[NCS]À ligands. In [1a]2+–[1c]2+ the positive charge is efficient-
ly delocalized onto the amine substituents of the dpb ligands
(Robin–Day class III/class II behavior, see above). This is visual-
ized in the DFT-calculated singly occupied molecular orbitals
and corresponding spin densities of [1a]2+–[1c]2+ spreading
over the metal center and the amine substituent (see
Figure 3). In this respect, amine substituted sensitizers should
be somewhat advantageous as compared to N719. On the
other hand, electron recombination with oxidized dyes de-
pends on the lifetime of the oxidized dye and hence on its re-
generation efficiency in the specific cell. Dye regeneration is
very efficient for SCN-based ruthenium dyes and iodide, but
often slowed down with other dye/electrolyte combinations
for various reasons (lower electrolyte concentration, lower driv-
ing force, smaller electronic coupling, inner/outer sphere elec-
tron transfer).[67–69] Considering the slower dye regeneration
using [2a]2À–[2c]2À sensitizers or cobalt redox mediators, the
better performance of cells with the N719/I3À/IÀ combination
is quite expected (Figure 5a, Tables 2, Table 3). Interestingly,
a lower electrolyte concentration decreases the efficiency of
the N719 cell to 5.8% but improves the efficiency of the cell
with the [nBu4N]2[2c] sensitizer to 3.3% (Table 2). Obviously,
the effects of electrolyte concentration on dye regeneration
N719
N719[,70a]
N719[,92b]
N719[,93c]
[a] [Co(bpy)3][B(CN)4]2 (0.22m); [Co(bpy)3][B(CN)4]3 (0.05m); 4-tert-butylpyr-
idine (0.2m) and lithium perchlorate (0.1m) in CH3CN; [b] [Co(bpy)3][PF6]2
(0.2m); [Co(bpy)3][PF6]3 (0.02m); 4-tert-butylpyridine (0.5m) and lithium
perchlorate (0.1m) in CH3CN; [c] [Co(bpy)3][PF6]2 (0.2m); [Co(bpy)3][PF6]3
(0.02m); 4-tert-butylpyridine (0.5m) and lithium perchlorate (0.1m) in
CH3CN.
and electron recombination kinetics differs for both dyes. This
shows that the standard electrolyte concentration is optimized
for N719, but the optimum electrolyte concentration needs to
be determined for each dye individually. However, this is
beyond the scope of this study.
Electron recombination of conduction band electrons with
the oxidized mediator (I3À, [3]3+, [4]3+) is a complex function
of the driving force (Marcus normal or inverted region), the
electronic coupling, and the presence of surface protection.[89]
À
Recombination kinetics with I3 is slow,[89] yet all cells with
À
[2]2À/I3 /IÀ combinations profit from the presence of CDCA as
surface protecting agent and show higher short-circuit current
À
densities (Table 2, Figure 5a, Figure 6a). With the I3 /IÀ couple,
the dark current for [2a]2À–[2c]2À sensitizers is higher than that
of N719 (Figure 6b). However this observation reverses for the
cobalt-based electrolytes (Figure 7b). Obviously, cells with the
tridentate cyclometalated complexes [2b]2À and [2c]2À cope
with the cobalt-based electrolytes, but the cell performance
with N719 dyes and cobalt-based electrolytes is severely re-
duced. The poor performance of N719 in combination with
cobalt-based electrolytes can be traced back to the higher
dark current densities and consequently a strongly diminished
short-circuit current density (Figure 6b, 7b; Table 1 and
Table 2). As a working hypothesis, the N719 dye better shields
TiO2 from I2/I3À, but [2b]2À and [2c]2À better shield TiO2 from
cobalt-based electrolytes. This effect is certainly related to mo-
lecular structure, packing density (cf. dye loadings, Table 4) and
protonation state of the dyes.[28]
Table 2. Photovoltaic data of cells using the I3À/IÀ redox mediator with
and without CDCA under AM1.5 light conditions (the data correspond to
averaged values of several cells).
Dye
CDCA
VOC [V]
JSC [mAcmÀ2
]
FF [%]
h [%]
[nBu4N]2[2a]
[nBu4N]2[2a]
[nBu4N]2[2b]
[nBu4N]2[2b]
[nBu4N]2[2c]
[nBu4N]2[2c]
[nBu4N]2[2c][b]
N719
–
+
–
+
–
+
+
+
+
+
–
0.50
0.54
0.53
0.59
0.57
0.59
0.70
0.74
0.78
0.55
0.54
0.69
1.77
3.37
2.91
5.47
5.06
5.58
6.74
13.4
10.97
8.22
8.49
11.61
63
67
67
73
72
74
70
75
68
69
73
73
0.6
1.2
1.1
2.4
2.1
2.5
3.3
7.3
5.8
3.1
3.4
5.8
Comparing the influence of the two cobalt-based electro-
lytes on the dark current densities of the N719 and [2c]2À
dyes, the bpy-based electrolyte [3]3+/2+ shows the lower dark
current density (Figure 7b). For cobalt-based electrolytes with
potentials above about 0.55 V vs. NHE, recombination with
electrons in the conduction band should be in the Marcus in-
verted region.[67] In accordance with this reported limiting
value, recombination with [3]3+ is hampered as compared to
N719[b]
[AH,H
]
+[66]
[BH]+[38,a]
[CH]+[38,a]
–
[a] Guanidinium thiocyanate was employed additionally in the electrolyte
solution. [b] The electrolyte concentration was reduced to 0.100m 1-
methyl-3-propylimidazolium iodide and 0.020m iodine in CH3CN.
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8921
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 7. a) Current density–voltage characteristics of cells with
[nBu4N]2[2b], [nBu4N]2[2c], and N719 using the [3]3+/2+ (a) and [4]3+/2+
(c) redox mediators under illumination; b) corresponding dark current
density–voltage measurements.
Figure 6. a) Current density–voltage characteristics of cells with
[nBu4N]2[2a], [nBu4N]2[2b], [nBu4N]2[2c], and N719 using the I3À/IÀ redox
mediator with/without CDCA (c/a) under illumination; b) correspond-
ing dark current density–voltage measurements.
bination with the iodine shuttle and increases the lifetime (Fig-
ure 8a). N719 displays the highest lifetime with the I3À/IÀ
couple but performs poorly with the cobalt-based electrolytes
(Figure 8). The slow dye regeneration kinetics of the cobalt-
based electrolytes allowing for recombination with the oxi-
dized dyes easily accounts for this observation. Slightly faster
dye regeneration and/or better surface shielding with the dye/
electrolyte combination [2c]2À/[3]3+/2+ yields a higher electron
lifetime (Figure 8b).
Table 4. Dye loadings.
Dye
CDCA
mol cmÀ2
[nBu4N]2[2a]
[nBu4N]2[2a]
[nBu4N]2[2b]
[nBu4N]2[2b]
[nBu4N]2[2c]
[nBu4N]2[2c]
N719
–
+
–
+
–
1.39”10À7
1.24”10À7
1.08”10À7
0.84”10À7
1.81”10À7
1.42”10À7
0.55”10À7
+
+
Finally, the open-circuit voltages VOC should increase in the
À
electrolyte series I3 /IÀ <[3]3+/2+ <[4]3+/2+ according to the
electrochemical data (Figure 4). Indeed, VOC of the cobalt elec-
À
recombination with [4]3+ accounting for the lower dark current
density. For both low-spin cobalt(III) complexes [3]3+ and [4]3+
recombination with an electron initially yields low-spin co-
balt(II) complexes corresponding to a metastable low-spin
state which then undergoes spin crossover to the high spin
state.[83,95] This effect might account for the favorable dark cur-
rent densities of the cobalt-based electrolytes compared to the
I3À/IÀ couple for [2b]2À and [2c]2À (Figure 6b,7b).
trolytes is somewhat larger than that of the I3 /IÀ couple for
[2a]2À, [2b]2À, and [2c]2À dyes, although the effect is less pro-
nounced than expected from the redox potentials (Figure 4).
Overall, the combination of cobalt-based electrolytes with
TPA-appended dyes, especially [2c]2À outperforms the stan-
dard dye N719 with cobalt-based electrolytes. The absolute
À
performance of N719 and the I3 /IÀ couple with optimized
concentration is still unrivaled with the systems under study.
Electron lifetimes were determined by the photovoltage re-
sponse to a small amplitude light modulation as a function of
the quasi Fermi level of TiO2 (Figure 8). All responses are linear
in the semi-logarithmic plot, suggesting that recombination
depends exponentially on the potential without participation
of surface states.[68] As suggested above, CDCA retards recom-
Conclusion
The bis(tridentate) cylcometalated ruthenium complexes [1a]
[PF6]–[1c][PF6] as well as their saponified counterparts
[nBu4N]2[2a]–[nBu4N]2[2c] were synthesized and characterized
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8922
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
surpassed however (h=5.8% under the same conditions).
Compared to the triiodide/iodide cells, the efficiencies of the
cells containing cobalt electrolytes are smaller by about
a factor of three despite larger open-circuit voltages. We attri-
bute this to the substantially slowed dye regeneration by the
cobalt electrolytes which results in reduced short-circuit cur-
rents. In the presence of cobalt electrolytes, however, the cy-
clometalated dye with the carbazole substituent [2c]2À (h=
1.4%) surpasses the thiocyanato-based dye N719 (h=1.1%).
This once again underlines the exceptional suitability of the
triiodide/iodide electrolyte for thiocyanate-based sensitizers.
We aim to dedicate further work to the understanding of the
opposing trends of overall cell performances with the cobalt
and iodide based electrolytes.
Experimental Section
Chemicals were obtained from commercial suppliers (Acros, Sigma-
Aldrich, Solaronix SA, Wako, TCI, Boulder Scientific) and used with-
out further purification. Air- or moisture-sensitive reactions were
performed in dried glassware in an inert gas atmosphere (argon,
quality 4.6). Acetonitrile was refluxed over CaH2 and distilled under
argon prior to use. Toluene was refluxed over sodium and distilled
under argon prior to use. The ligands trimethyl-2,2’;6’,2’’-terpyri-
dine-4,4’,4’’-tricarboxylate Me3tctpy,[22] 1-bromo-3,5-di(2-pyridyl)-
benzene,[96] ddpd[81] as well as the ruthenium(III) complex
[84]
RuCl3(Me3tctpy)[22] and the palladium precatalyst [Pd]2 were syn-
thesized according to previously reported procedures. IR spectra
were recorded on a Varian Excalibur Series 3100 FTIR spectrometer
using KBr disks. IR absorption band intensities are classified as
s (strong), m (medium), w (weak), and sh (shoulder). UV/Vis spectra
were recorded on a Varian Cary 5000 spectrometer in 1 cm cuv-
ettes. Emission spectra were recorded on a Varian Cary Eclipse
spectrometer. Quantum yields were determined by comparing the
areas under the emission spectra on an energy scale recorded for
solutions of the samples and a reference with matching absorban-
ces (F([Ru(bipy)3]Cl2)=0.094 in deaerated MeCN).[97] Experimental
uncertainty is estimated to be 15%. ESI+ and high-resolution ESI+
mass spectra were recorded on a Micromass QTof Ultima API mass
spectrometer with analyte solutions in acetonitrile. FD mass spec-
tra were recorded on a Thermo Fisher DFS mass spectrometer with
a LIFDI upgrade. Elemental analyses were performed by the micro-
analytical laboratory of the chemical institutes of the University of
Mainz. NMR spectra were obtained with a Bruker Avance II 400
spectrometer at 400.31 (1H), 100.66 (13C), 376.67 (19F) at 258C.
Chemical shifts d [ppm] are reported with respect to residual sol-
vent signals as internal standards (1H, 13C): CD3CN d(1H)=1.94 ppm,
d(13C)=1.32 and 118.26 ppm;[98] CD2Cl2 d(1H)=5.32 ppm, d(13C)=
53.84 ppm[98] or external CFCl3 (19F: d=0 ppm). Electrochemical ex-
periments were performed with a BioLogic SP-50 voltammetric an-
alyzer at a sample concentration of 10À3 m using platinum wire
working and counter electrodes and a 0.01m Ag/AgNO3 reference
electrode. Measurements were carried out at a scan rate of
100 mVsÀ1 for cyclic voltammetry experiments and at 10 mVsÀ1
for square-wave voltammetry experiments using 0.1m [nBu4N][PF6]
as supporting electrolyte in acetonitrile. Potentials are given rela-
Figure 8. a) Electron recombination lifetimes (t) of cells with [nBu4N]2[2a],
[nBu4N]2[2b], [nBu4N]2[2c], and N719 using a) the I3À/IÀ redox mediator
3+/2+
3+/2+
&
&
( )
*
*
) and [4]
with/without CDCA ( / ) and b) using the [3]
(
redox mediators.
using mass spectrometry, UV/Vis and emission spectroscopy,
and electrochemical methods. Oxidation of [2a]2À–[2c]2À yields
strongly coupled mixed valent species [2a]À–[2c]À with sub-
stantial charge delocalization between the metal center and
the triarylamine fragment as evidenced from TD-DFT calcula-
tions.[62,63] Yet, the degree of this delocalization is reduced in
the saponified complexes [2a]À–[2c]À as compared to the cor-
responding esters [1a]2+–[1c]2+ due to an increased charge
density at the metal center. Concomitantly, all redox potentials
shift to substantially lower values. Charge delocalization
should be beneficial for applications in dye-sensitized solar
cells as it hampers undesired recombination processes. Howev-
er, the low redox potentials, which result from the large reso-
nance stabilization, poses a challenge in the selection of a suita-
ble redox electrolyte for efficient dye regeneration. As a conse-
quence, dye [2c]2À with the highest oxidation potential yields
the best cell performance of the cyclometalated complexes in
this study in conjunction with triiodide/iodide (h=3.3%). The
reference dye N719/triiodide/iodide combination remains un-
tive to the ferrocene/ferrocenium couple (0.40 V vs. SCE, E1/2
0.09Æ5 mV under the given conditions).[99]
=
Current–voltage characteristics of the DSSCs were measured with
a Keithley Model 2400 source meter and a solar simulator with
a 300 W Xenon arc-lamp (Newport) under 1 sun illumination
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8923
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(AM 1.5, 100 mWcmÀ2). A light shading mask, placed on the residu-
al area of the front side of the FTO substrate (except for the
0.16 cm2 TiO2 active area), was employed to prevent overestima-
tion of the power conversion efficiency. The quantum efficiencies
of the DSSCs were measured by incident photon-to-current con-
version efficiency (IPCE) measurements (PV Measurements, Inc.).
UV/Vis spectra of the dye loading solutions were collected on
a Jasco V-670 UV/Vis spectrometer. The electron lifetimes were ob-
tained by intensity modulated photovoltage spectroscopy (IMVS)
under open-circuit conditions as a function of light intensity using
a controlled intensity modulated photo spectroscopy (CIMPS)
system (Zahner).
129.7 (C12), 124.7 (C11), 123.5 (C2), 123.3 (C13), 122.8 (C7), 120.9 (C5),
120.3 (C9). IR (KBr disk): n˜ [cmÀ1]=3054 (w, aromatic C-H), 1584 (s,
C=C), 1566 (s, C=C), 1492 (s), 1352 (s), 1255 (s), 779 (s), 749 (s), 700
(s).
Synthesis of 1-(N-carbazolyl)-3,5-(di-2-pyridyl)benzene (dpbH-N-
carbazole) Lc: 2-Dicyclohexylphosphano-2’,6’-diisopropoxybiphenyl
(12 mg, 0.026 mmol, 0.04 equiv) and the precatalyst [Pd]2 (7 mg,
0.009 mmol, 0.01 equiv) were suspended in dry toluene (10 mL)
and stirred for 10 min followed by the addition of 1-bromo-3,5-
di(2-pyridyl)benzene (200 mg, 0.643 mmol, 1 equiv), carbazole
(161 mg, 0.964 mmol, 1.50 equiv), sodium tert-butoxide (93 mg,
0.968 mmol, 1.51 equiv), and further dry toluene (40 mL). The re-
sulting mixture was refluxed for 12 h. After cooling to room tem-
perature, the solvent was removed under reduced pressure and
the brown residue was purified by column chromatography on
silica gel (eluent: dichloromethane/ethyl acetate 20:1) yielding Lc
as slightly yellow solid. Yield: 251 mg (0.556 mmol, 98%). Anal.
calcd. for C28H19N3 (397.47) ·0.5H2O: C 82.73, H 4.96, N 10.34;
found: C 82.96, H 4.53, N 10.04. MS(FD+): m/z (%)=198.6 (1.5)
Density functional theory calculations: DFT calculations were car-
ried out using the ORCA program package (version 3.0.2).[100] Tight
convergence criteria were chosen for all calculations (Keywords
TightSCF and TightOpt). All calculations employ the resolution of
identity (Split-RI-J) approach for the coulomb term in combination
with the chain-of-spheres approximation for the exchange term
(COSX).[101,102] All calculations were performed using the hybrid
functional B3LYP[103] in combination with Ahlrichs’ split-valence
double-z basis set def2-SV(P), which comprises polarization func-
tions for all non-hydrogen atoms.[104,105] Relativistic effects were cal-
culated at the zeroth-order regular approximation (ZORA)
niveau.[106] The ZORA keyword automatically invokes relativistically
adjusted basis sets.[107] To account for solvent effects a conductor-
like screening model (COSMO) modelling acetonitrile was used in
all calculations.[108] Explicit counterions and/or solvent molecules
were not taken into account in all cases.
1
[Lc]2+, 397.1 (100) [Lc]+. H NMR (CD2Cl2): d [ppm]=8.85 (s, 1H, H9),
3
8.73 (d, JHH =4 Hz, 2H, H8), 8.31 (s, 2H, H2), 8.19 (d, J=8 Hz, 2H,
3
H14), 7.92 (d, J=8 Hz, 2H, H5), 7.83 (t, JHH =8 Hz, 2H, H6), 7.54 (d,
3
3JHH =8 Hz, 2H, H11), 7.44 (t, JHH =7 Hz, 2H, H12), 7.29–7.35 (m, 4H,
H7, H13). 13C{1H} NMR (CD2Cl2): d [ppm]=156.4 (C4), 150.3 (C8), 142.2
(C3), 141.4 (C10), 139.1 (C1), 137.3 (C6), 126.5 (C12), 126.2 (C2), 124.6
(C9), 123.8 (C15), 123.2 (C7), 121.0 (C5), 120.7 (C14), 120.4 (C13), 110.3
(C11). IR (KBr disk): n˜ [cmÀ1]=3049 (w, aromatic C-H), 1587 (s, C=C),
1566 (s, C=C), 1448 (s), 1228 (s), 779 (s), 747 (s), 723 (s).
Synthesis of N,N-bis-(4-methoxyphenyl)-3,5-(di-2-pyridyl)aniline
(dpbH-N(4-C6H4OMe)2) La: 2-Dicyclohexylphosphano-2’,6’-diisopro-
poxybiphenyl (12 mg, 0.026 mmol, 0.04 equiv) and the precatalyst
[Pd]2 (8 mg, 0.011 mmol, 0.02 equiv) were suspended in dry tolu-
ene (10 mL) and stirred for 10 min followed by the addition of 1-
bromo-3,5-di(2-pyridyl)benzene (200 mg, 0.643 mmol, 1 equiv), bis-
(4-methoxyphenyl)amine (221 mg, 0.964 mmol, 1.50 equiv), sodium
tert-butoxide (93 mg, 0.968 mmol, 1.51 equiv), and further dry tolu-
ene (40 mL). The resulting mixture was refluxed for 12 h. After
cooling to room temperature, the solvent was removed under re-
duced pressure and the brown residue was purified by column
chromatography on silica gel (eluent: dichloromethane/ethyl ace-
tate 10:1!7:1), yielding La as slightly yellow solid. Yield: 241 mg
(0.524 mmol, 82%). NMR and mass spectrometric data agree with
reported values.[62]
Synthesis of [Ru(La)(Me3tctpy)][PF6] [1a][PF6]:[63a] RuCl3(Me3tctpy)
(222 mg, 0.361 mmol, 1 equiv) and AgBF4 (204 mg, 1.05 mmol,
2.9 equiv) were dissolved in dry acetonitrile (15 mL) and refluxed in
the dark for 3 h. After cooling to room temperature, the mixture
was filtered through a syringe filter (0.2 mm) and the solvent re-
moved under reduced pressure. The dark residue was dissolved in
deaerated n-butanol (20 mL) and La (200 mg, 0.435 mmol,
1.2 equiv) was added. The mixture was refluxed for 13 h followed
by removal of the solvent under reduced pressure. The raw prod-
uct was redissolved in MeCN (5 mL) and triturated by addition of
a solution of [NH4][PF6] (177 mg, 1.08 mmol, 3 equiv) in H2O (2 mL).
The black precipitate was filtered off and washed with diethyl
ether and hexanes. After purification via column chromatography
on silica gel (eluent: chloroform/methanol 1:0!7:1), [1a][PF6] was
obtained as black powder. Yield: 364 mg (0.327 mmol, 91%).
MS(ESI+): m/z (%)=967.1 (100) [1a]+. Traces of the paramagnetic
RuIII complex [1a][PF6]2 broaden all NMR resonances of [1a][PF6]
due to the presence of a fast self-exchange reaction.[62,63a]
Synthesis of N,N-diphenyl-3,5-(di-2-pyridyl)aniline (dpbH-NPh2)
Lb: 2-Dicyclohexylphosphano-2’,6’-diisopropoxybiphenyl (12 mg,
0.026 mmol, 0.04 equiv) and the precatalyst [Pd]2 (7 mg,
0.009 mmol, 0.01 equiv) were suspended in dry toluene (10 mL)
and stirred for 10 min followed by the addition of 1-bromo-3,5-
di(2-pyridyl)benzene (200 mg, 0.643 mmol, 1 equiv), diphenylamine
(163 mg, 0.964 mmol, 1.50 equiv), sodium tert-butoxide (93 mg,
0.968 mmol, 1.51 equiv) and further dry toluene (40 mL). The re-
sulting mixture was refluxed for 12 h. After cooling to room tem-
perature, the solvent was removed under reduced pressure and
the brown residue was purified by column chromatography on
silica gel (eluent: dichloromethane/ethyl acetate 10:1) yielding Lb
as slightly yellow solid. Yield: 222 mg (0.556 mmol, 86%). Anal.
calcd. for C28H21N3 (399.49): C 84.18, H 5.30, N 10.52; found:
C 83.86, H 5.39, N 10.39. MS(FD+): m/z (%)=399.1 (100) [Lb]+.
Synthesis of [nBu4N]2[Ru(La)(H3tctpy)] [nBu4N]2[2a]: Complex [1a]
[PF6] (360 mg, 0.324 mmol) was suspended in deaerated H2O
(35 mL) and nBu4NOH (1.5m in H2O, 5 mL) and hydrazine (1 mL)
were added. After refluxing for 13 h, the mixture was extracted
with dichloromethane (3”50 mL). The organic phases were com-
bined, dried over MgSO4, and the solvent was removed under re-
duced pressure. The purple residue was dissolved in MeCN (10 mL)
and triturated by adding a 1:1 mixture of diethyl ether and hex-
anes (50 mL), yielding [nBu4N]2[2a] as purple powder. Yield:
347 mg (0.247 mmol, 66%). Anal. calcd. for C80H104N8O8Ru (1406.8)
·8H2O: C 61.95, H 7.80, N 7.22; found: C 62.24, H 7.88, N 7.22.
MS(ESI+): m/z (%)=242.3 (100) [nBu4N]+, 462.6 (12) [2a+3H]2+
,
3
1H NMR (CD2Cl2): d [ppm]=8.64 (d, JHH =5 Hz, 2H, H8), 8.32 (s, 1H,
925.1 (48) [2a+3H]+. HR-MS(ESI+, m/z): Calcd. for C48H35N6O896Ru
[2a+3H]+: 919.1592; Found: 919.1578. MS(ESIÀ): m/z (%)=439.4
(85) [2a-CO2]2À, 461.4 (56) [2a]2À, 834.9 (17) [2a-2CO2]À, 878.8 (100)
[2a-CO2]À, 922.8 (19) [2a]À, 1163.8 (77) ([nBu4N][2a])À. 1H NMR
(CD3CN): d [ppm]=9.25 (s, 2H, H2A), 8.86 (s, 2H, H5A), 8.03 (s, 2H,
3
H9), 7.80 (s, 2H, H2), 7.76–7.71 (m, 4H, H5, H6), 7.29 (t, JHH =8 Hz,
3
3
4H, H12), 7.24 (dd, JHH =9 Hz, 5 Hz, 2H, H7), 7.17 (d, JHH =8 Hz, 4H,
H11), 7.06 (t, JHH =7 Hz, 2H, H13). 13C{1H} NMR (CD2Cl2): d [ppm]=
157.1 (C4), 150.0 (C8), 149.4 (C1), 148.3 (C10), 141.5 (C3), 137.1 (C6),
3
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8924
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3
3
H2B), 7.90 (d, JHH =8 Hz, 2H, H5B), 7.47 (t, JHH =7 Hz, 2H, H6B), 7.37
disk): n˜ [cmÀ1]=3440 (s, O-H crystal water), 3065 (w, aromatic C-H),
2960, 2873 (m, aliphatic C-H), 1617 (s, C=O carboxylate), 1598 (sh,
C=C), 1465 (m, aliphatic C-N), 1351 (s, aromatic C-N).
(d, JHH =6 Hz, 2H, H7A), 7.20 (d, JHH =9 Hz, 4H, H12A), 7.07 (d, JHH
=
3
3
3
6 Hz, 2H, H8A), 7.05 (d, JHH =6 Hz, 2H, H8B), 6.92 (d, JHH =9 Hz, 4H,
3
3
3
H12B), 6.58 (t, JHH =6 Hz, 2H, H7B), 3.79 (s, 6H, H14B), 3.19–3.00 (m,
Synthesis of [Ru(Lc)(Me3tctpy)][PF6] [1c][PF6]: RuCl3(Me3tctpy)
(257 mg, 0.418 mmol, 1 equiv), and AgBF4 (237 mg, 1.22 mmol,
2.9 equiv) were dissolved in dry acetonitrile (15 mL) and refluxed in
the dark for 3 h. After cooling to room temperature, the mixture
was filtered through a syringe filter (0.2 mm) and the solvent re-
moved under reduced pressure. The dark residue was dissolved in
deaerated n-butanol (20 mL) and Lc (200 mg, 0.503 mmol,
1.2 equiv) was added. The mixture was refluxed for 13 h followed
by removal of the solvent under reduced pressure. The raw prod-
uct was redissolved in MeCN (5 mL) and triturated by addition of
a solution of [NH4][PF6] (204 mg, 1.25 mmol, 3 equiv) in H2O (2 mL).
The black precipitate was filtered off and washed with diethyl
ether and hexanes. After purification via column chromatography
on silica gel (eluent: chloroform/methanol 1:0!7:1) [1c][PF6] was
obtained as black powder. Yield: 385 mg (0.366 mmol, 88%).
MS(ESI+): m/z (%)=905.1 (100) [1c]+. HR-MS(ESI+, m/z): Calcd. for
C49H35N6O696Ru [1c]+: 899.1694; Found: 899.1725. Traces of para-
magnetic RuIII complex [1c][PF6]2 broaden all NMR resonances of
[1c][PF6] due to the presence of a fast self-exchange reaction.[62,63a]
IR (KBr disk): n˜ [cmÀ1]=3052 (w, aromatic C-H), 1725 (s, C=O ester),
1599 (m, C=C), 1249 (s), 842 (s, P-F), 588 (m, PFdef).
16H, H1), 1.68–1.44 (m, 16H, H2), 1.38–1.21 (m, 16H, H3), 0.93 (t,
3JHH =7 Hz, 24H, H4). 13C{1H} NMR (CD3CN): d [ppm]=219.0 (C9B),
169.1 (C4B), 168.3 (C10A), 167.1 (C11A), 160.2 (C4A), 155.7 (C13B), 154.5
(C8A), 153.7 (C3A), 152.9 (C8B), 147.9 (C6A), 143.9 (C9B), 143.5 (C3B),
142.8 (C1B), 136.0 (C6B), 126.6 (C7A), 124.8 (C11B), 123.6 (C5A), 123.4
(C2B), 122.9 (C2A), 122.4 (C7B), 120.4 (C5B), 115.6 (C12B), 59.3 (C1), 56.1
(C14B), 24.3 (C2), 20.3 (C3), 13.8 (C4). IR (KBr disk): n˜ [cmÀ1]=3440 (s,
O-H crystal water), 3065 (w, aromatic C-H), 2960, 2873, 2843 (m, ali-
phatic C-H), 1620 (s, C=O carboxylate), 1598 (sh, C=C), 1465 (m, ali-
phatic C-N), 1341 (s, aromatic C-N), 1237 (s, C-O-C).
Synthesis of [Ru(Lb)(Me3tctpy)][PF6] [1b][PF6]: RuCl3(Me3tctpy)
(263 mg, 0.429 mmol, 1 equiv) and AgBF4 (243 mg, 1.25 mmol,
2.9 equiv) were dissolved in dry acetonitrile (15 mL) and refluxed in
the dark for 3 h. After cooling to room temperature, the mixture
was filtered through a syringe filter (0.2 mm) and the solvent re-
moved under reduced pressure. The dark residue was dissolved in
deaerated n-butanol (20 mL) and Lb (206 mg, 0.516 mmol,
1.2 equiv) was added. The mixture was refluxed for 13 h followed
by removal of the solvent under reduced pressure. The raw prod-
uct was redissolved in MeCN (5 mL) and triturated by addition of
a solution of [NH4][PF6] (210 mg, 1.29 mmol, 3 equiv) in H2O (2 mL).
The black precipitate was filtered off and washed with diethyl
ether and hexanes. After purification via column chromatography
on silica gel (eluent: chloroform/methanol 1:0!7:1), [1b][PF6] was
obtained as a black powder. Yield: 389 mg (0.370 mmol, 86%).
MS(ESI+): m/z (%)=453.6 (12) [1b]2+, 907.1 (100) [1b]+. HR-
MS(ESI+, m/z): Calcd. for C49H37N6O696Ru [1b]+: 901.1851; Found:
901.1857. Traces of the paramagnetic RuIII complex [1b][PF6]2
broaden all NMR resonances of [1b][PF6] due to the presence of
a fast self-exchange reaction.[62,63a] IR (KBr disk): n˜ [cmÀ1]=3041 (w,
aromatic C-H), 1725 (s, C=O ester), 1599 (m, C=C), 1243 (s), 843 (s,
P-F), 588 (m, PFdef).
Synthesis of [nBu4N]2[Ru(Lc)(H3tctpy)] [nBu4N]2[2c]: Complex [1c]
[PF6] (233 mg, 0.222 mmol) was suspended in deaerated H2O
(35 mL), and [nBu4N][OH] (1.5m in H2O, 5 mL) and hydrazine (1 mL)
were added. After refluxing for 13 h, the mixture was extracted
with dichloromethane (3”50 mL). The organic phases were com-
bined, dried over MgSO4, and the solvent was removed under re-
duced pressure. The purple residue was dissolved in MeCN (10 mL)
and triturated by adding a 1:1 mixture of diethyl ether and hex-
anes (50 mL), yielding [nBu4N]2[2c] as purple powder. Yield:
159 mg (0.118 mmol, 53%). Anal. calcd. for C78H98N8O6Ru (1344.7)
·8H2O: C 63.69, H 7.68, N 7.62; found: C 63.77, H 8.03, N 7.55.
MS(ESI+): m/z (%)=242.3 (100) [nBu4N]+, 431.6 (3) [2a+3H]2+
,
863.1 (89) [2a+3H]+. HR-MS(ESI+, m/z): Calcd. for C46H29N6O696Ru
[2c+3H]+: 857.1225; Found: 857.1218. MS(ESIÀ): m/z (%)=386.5
(26) [2c-2CO2]2À, 408.4 (74) [2c-CO2]2À, 430.4 (20) [2c]2À, 773.0 (40)
[2c-2CO2]À, 816.9 (80) [2c-CO2]À, 860.9 (13) [2c]À, 1101.7 (100)
Synthesis of [nBu4N]2[Ru(Lb)(H3tctpy)] [nBu4N]2[2b]: Complex [1b]
[PF6] (279 mg, 0.265 mmol) was suspended in deaerated H2O
(35 mL), and [nBu4N][OH] (1.5m in H2O, 5 mL) and hydrazine (1 mL)
were added. After refluxing for 13 h, the mixture was extracted
with dichloromethane (3”50 mL). The organic phases were com-
bined, dried over MgSO4, and the solvent was removed under re-
duced pressure. The purple residue was dissolved in MeCN (10 mL)
and triturated by adding a 1:1 mixture of diethyl ether and hex-
anes (50 mL), yielding [nBu4N]2[2b] as purple powder. Yield:
312 mg (0.323 mmol, 87%). Anal. calcd. for C78H100N8O6Ru (1346.8)
·6H2O: C 64.39, H 7.76, N 7.70; found: C 64.06, H 7.40, N 7.53.
1
([nBu4N][2c])À. H NMR (CD3CN): d [ppm]=9.25 (s, 2H, H2A), 8.85 (s,
3
3
2H, H5A), 8.39 (s, JHH =9 Hz, 2H, H2B), 8.29 (d, JHH =8 Hz, 2H, H14B),
8.11 (d, JHH =8 Hz, 2H, H5B), 7.72 (d, JHH =8 Hz, 2H, H11B), 7.60–7.50
3
3
(m, 4H, H6B, H12B), 7.41–7.31 (m, 4H, H7A, H13B), 7.21–7.12 (m, 4H,
3
H8A, H8B), 6.67 (t, JHH =6 Hz, 2H, H7B), 3.16–3.00 (m, 16H, H1), 1.63–
1.51 (m, 16H, H2), 1.40–1.24 (m, 16H, H3), 0.92 (t, JHH =7 Hz, 24H,
3
H4). 13C{1H} NMR (CD3CN): d [ppm]=223.7 (C9B), 169.0 (C4B), 167.9
(C9A), 166.9 (C10A), 160.2 (C4A), 154.7 (C8A), 153.6 (C3A), 153.0 (C8B),
148.8 (C6A), 146.8 (C1A), 144.1 (C3B), 143.2 (C10B), 136.1 (C6B), 130.3
(C1B), 127.1 (C7A), 126.5 (C15B), 123.8 (C2B), 123.6 (C5A), 123.6 (C12B),
122.8 (C2A), 122.7 (C7B), 121.3 (C14B), 120.7 (C5B), 120.6 (C13B), 111.3
(C11B), 59.2 (C1), 24.3 (C2), 20.3 (C3), 13.8 (C4). IR (KBr disk): n˜ [cmÀ1]=
3440 (s, O-H crystal water), 3065 (w, aromatic C-H), 2960 (m, ali-
phatic C-H), 2873 (m, aliphatic C-H), 1614 (s, C=O carboxylate),
1598 (sh, C=C), 1465 (m, aliphatic C-N), 1353 (s, aromatic C-N).
MS(ESI+): m/z (%)=242.3 (100) [nBu4N]+, 432.6 (9) [2b+3H]2+
,
865.2 (8) [2b+3H]+. HR-MS(ESI+, m/z): Calcd. for C46H31N6O696Ru
[2b+3H]+: 859.1381; Found: 859.1390. MS(ESIÀ): m/z (%)=387.5
(13) [2b-2CO2]2À, 409.4 (47) [2b-CO2]2À, 431.4 (15) [2b]2À, 775.0 (28)
[2b-2CO2]À, 818.9 (67) [2b-CO2]À, 862.9 (13) [2b]À, 1103.7 (100)
([nBu4N][2b])À. H NMR (CD3CN): d [ppm]=9.20 (s, 2H, H2A), 8.80 (s,
1
2H, H5A), 8.10 (s, 2H, H2B), 7.96 (d, 3JHH =8 Hz, 2H, H5B), 7.49 (t,
3JHH =8 Hz, 2H, H6B), 7.40–7.26 (m, 10H, H7A, H11B, H12B), 7.11 (d,
3JHH =5 Hz, 2H, H8B), 7.06–6.95 (m, 4H, H8A, H13B), 6.61 (t, JHH =6 Hz,
Synthesis of [Co(bpy)3][B(C6F5)4]2 [3][B(C6F5)4]2: The bpy ligand
(2.27 g, 14.6 mmol, 3.2 equiv), dissolved in CH3CN (50 mL) was
added to a solution of Co(BF4)2·6H2O (1.55 g, 4.55 mmol, 1.0 equiv)
in CH3CN (50 mL) and the mixture was stirred for 6 h at room tem-
perature. The complex was precipitated by addition of Et2O
(200 mL) and filtered off, washed with Et2O (100 mL), and dried
under reduced pressure. The yellow powder was dissolved in a min-
imum amount of CH3CN and Li[B(C6F5)4]·Et2O (13.8 g, 18.2 mmol,
3
2H, H7B), 3.19–3.00 (m, 16H, H1), 1.67–1.49 (m, 16H, H2), 1.41–1.21
(m, 16H, H3), 0.94 (t, JHH =7 Hz, 24H, H4). 13C{1H} NMR (CD3CN): d
3
[ppm]=218.8 (C9B), 169.1 (C4B), 167.6 (C9A), 166.6 (C10A), 160.2 (C4A),
154.4 (C8A), 153.6 (C3A), 152.9 (C8B), 149.9 (C6A), 149.9 (C10B), 149.4
(C1A), 144.0 (C3B), 141.0 (C1B), 135.9 (C6B), 130.3 (C11B), 126.5 (C7A),
124.7 (C2B), 123.5 (C5A), 123.1 (C12B), 122.7 (C2A), 122.5 (C7B), 122.4
(C13B), 120.4 (C5B), 59.3 (C1), 24.3 (C2), 20.3 (C3), 13.8 (C4). IR (KBr
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8925
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
4.0 equiv) was added. The volume of the solvent was reduced
under reduced pressure and the product was precipitated by addi-
tion of water to give a yellow crystalline solid (6.52 g, 3.46 mmol,
(56) [Co(ddpd)2]2+, 1320.2 (8) [Co(ddpd)2 +B(C6F5)4]+, 1999.2 (100)
[Co(ddpd)2 +2”(B(C6F5)4)]+, 2535.5 (18), 2669.6 (20) [5”
(Co(ddpd)2)+11 ”(B(C6F5)4)]4+
,
2892.8 (7) [4”(Co(ddpd)2)+9”
1
76%). H NMR (CD3CN): d [ppm]=88.3 (bs, 1H), 85.0 (s, 1H), 46.5
(B(C6F5)4)]3+
(B(C6F5)4)]2+
,
.
3071.4 (12), 3339.4 (12) [3”(Co(ddpd)2)+7”
HR-MS(ESI+, m/z): Calcd. for C82H3411B2CoF40N10:
(s, 1H), 14.7 (s, 1H). 19F NMR (CD3CN): d [ppm]=À134.04 (bs, 2F),
3
3
À164.20 (pt, 1F, JFF=19.7 Hz), À168.68 (pt, 2F, JFF =16.8 Hz). The
1H NMR data match literature values of [Co(bpy)2][PF6]2.[91]
MS(ESI+): m/z (%)=185.6 (7) [Co(bpy)2]2+, 1206.2 (100) [Co(bpy)3 +
B(C6F5)4]+, 2149.3 (17) [3”(Co(bpy)3)+4”(B(C6F5)4)]2+, 3092.44 (19)
1999.1847; found: 1999.1809.
TiO2 electrode preparation: Fluorine-doped tin oxide (FTO) glass
plates (Pilkington-TEC8) were cleaned using a ultrasonic bath with
2 vol% of Helmanex in deionized water and ethanol. A doctor-
bladed layer of 35 nm TiO2 particles (PST-35 NR, CCIC) was used as
photoelectrode. A 8 mm thick transparent film and an additional
4 mm scattering TiO2 film (PST-400C, CCIC, particle size ca. 400 nm)
were coated on the top of the conducting glass electrode. The
TiO2 electrodes were heated to 4508C for 30 min, treated with
a 0.5 mm TiCl4 solution in deionized water for 20 min at 708C, fol-
lowed by an annealing process for 30 min at 4508C. Following the
heat treatment, these electrodes were immersed into the sensitiz-
er/CDCA solutions (2.95”10À4 m of [nBu4N][2a], 3.01”10À4 m of
[nBu4N][2b], 3.01”10À4 m of [nBu4N][2c], and 3.11”10À4 m of N719
solution (CH3CN/tBuOH) (1:1, volume ratio) with or without 5.99”
10À4 m CDCA (Dyesol) and kept at room temperature for 6 h, 16 h
or 24 h, respectively. The TiO2 electrodes were rinsed with CH3CN
and dried.
[4”(Co(bpy)3)+6”(B(C6F5)4)]2+
.
HR-MS(ESI+, m/z): Calcd. for
C54H2410BCoF20N6: 1205.1204; found: 1205.1224.
Synthesis of [Co(bpy)3][B(C6F5)4]3 [3][B(C6F5)4]3: [Co(bpy)3](BF4)2
(2.94 g, 4.20 mol, 1.0 equiv) was dissolved in CH3CN (100 mL) and
2,3-dichloro-5,6-dicyano-1,4-benzoquinone (1.43 g, 6.30 mmol,
1.5 equiv) was added. The solution was stirred for 5 h at room tem-
perature. Li[B(C6F5)4]·Et2O (12.8 g, 16.8 mol, 4.0 equiv) was added
and the mixture was diluted with water (200 mL). The precipitate
was filtered off, washed with water (200 mL), and dried under re-
duced pressure to give a slightly orange crystalline solid (7.65 g,
3
2.98 mmol, 71%). 1H NMR (CD3CN): d [ppm]=8.71 (d, 1H, JHH
=
3
3
7.8 Hz), 8.49 (pt, 1H, JHH =7.7 Hz), 7.76 (pt, 1H, JHH =6.5 Hz), 7.31
(d, 1H, JHH =5.5 Hz). 19F NMR (CD3CN): d [ppm]=À134.07 (bs, 2F),
3
À164.22 (pt, 1F, JFF =19.7 Hz), À168.65 (bs, 2F). MS(ESI+): m/z
3
(%)=185.6 (20) [Co(bpy)2]2+, 263.6 (6) [Co(bpy)3]2+, 371.1 (46)
[Co(bpy)2]+, 423.1 (74), 1054.2 (14), 1206.2 (31) [Co(bpy)3 +
B(C6F5)4]+, 1885.2 (100) [Co(bpy)3 +2”(B(C6F5)4)]+, 2526.8 (6)
[Co(bpy)3 +3”(B(C6F5)4-2F)]+, 2740.6 (7), 3168.3 (8) [3”(Co(bpy)3)+
Electrolyte solutions: The triiodide/iodide electrolyte solutions
were prepared from 1-methyl-3-propylimidazolium iodide (0.600m)
and iodine (0.050m) in CH3CN. The CoIII/II electrolytes were em-
ployed as 0.035/0.165m and 0.020/0.080m CH3CN solutions of
[3]3+/[3]2+ and [4]3+/[4]2+, respectively. Owing to the employed
relative concentrations the redox potentials shift to lower values
7”(B(C6F5)4)]2+
.
HR-MS(ESI+, m/z): Calcd. for C78H2410B2CoF40N6:
1883.1014; found: 1883.1052.
by 0.061, 0.039, and 0.036 V for I3À/IÀ, [3]3+/[3]2+ and [4]3+/[4]2+
,
Synthesis of [Co(ddpd)2][B(C6F5)4]2 [4][B(C6F5)4]2: The ddpd ligand
(2.53 g, 8.68 mol, 2.2 equiv), dissolved in CH3CN (50 mL), was
added to a solution of Co(BF4)2·6H2O (1.34 g, 3.95 mmol, 1.0 equiv)
in CH3CN (50 mL) and the mixture was stirred for 6 h at room tem-
perature. The complex was precipitated by addition of Et2O
(400 mL) and filtered off, washed with Et2O (100 mL), and dried
under reduced pressure. The yellow powder was dissolved in a min-
imum amount of CH3CN, and Li[B(C6F5)4]·Et2O (12.0 g, 15.7 mmol,
4.0 equiv) was added. The volume of the solvent was reduced
under reduced pressure and the product was precipitated by addi-
tion of water to give a yellow crystalline solid (7.18 g, 3.51 mmol,
89%). 1H NMR (CD3CN): d [ppm]=75.0 (s, 2H), 69.0 (s, 2H), 34.6
(bs, 2H), 34.2 (s, 2H), 22.5 (s, 6H), 21.9 (s, 1H), 2.7 (s, 2H). 19F NMR
respectively. The absolute concentrations of the redox couples are
0.6m, 0.2m and 0.1m for I3À/IÀ, [3]3+/[3]2+ and [4]3+/[4]2+, respec-
tively. 4-tert-Butylpyridine (0.8m) and lithium perchlorate (0.1m)
were used in all cells.
Counter electrode preparation: The Pt electrode was prepared by
spin-coating of 10 mm H2PtCl6 (Sigma-Aldrich) in 2-propanol and
then sintered at 4508C for 30 min. The cells were sealed using
60 mm Surlyn. The electrolyte solutions were introduced through
holes on the counter electrode.
Dye loading: TiO2/FTO electrodes were immersed in a 0.1m KOH
H2O/CH3CN 1:1 solution for at least 5 min. From the UV/Vis absorp-
tion spectra of the resulting dye solutions, the concentrations of
the attached dyes were calculated (Table 4).
3
(CD3CN): d [ppm]=À134.07 (bs, 2F), À164.22 (pt, 1F, JFF =19.7 Hz),
3
À168.67 (pt, 2F, JFF =16.8 Hz). The 1H NMR data match reported
values of [Co(ddpd)2][BF4]2.[82] MS(ESI+): m/z (%)=292.1 (14)
[ddpd+H]+, 320.6 (18) [Co(ddpd)2]2+, 1320.1 (100) [Co(ddpd)2 +
B(C6F5)4]+, 2320.3 (13) [3”(Co(ddpd)2)+4”(B(C6F5)4)]2+, 3320.3 (18)
Acknowledgements
[4”(Co(dpdd)2)+6”(B(C6F5)4)]2+
.
HR-MS(ESI+, m/z): Calcd. for
A.K.C.M. thanks the International Research Training Group
(IRTG 1404) Self Organized Materials for Optoelectronics sup-
ported by the Deutsche Forschungsgemeinschaft (DFG) for
funding. This work was financially supported by the Deutsche
Forschungsgemeinschaft (GSC 266, Materials Science in Mainz,
scholarship for C.K.). Parts of this research were conducted
using the supercomputer MOGON and advisory services of-
fered by Johannes Gutenberg Univ. Mainz (www.hpc.uni-main-
z.de), which is a member of the AHRP and the Gauss Alliance
e.V. We thank Johannes Moll for preparative assistance.
C58H3410BCoF20N10: 1319.2110; found: 1319.2106.
Synthesis of [Co(ddpd)2][B(C6F5)4]3 [4][B(C6F5)4]3: [Co(ddpd)2](BF4)2
(3.24 g, 3.83 mmol, 1.0 equiv) was dissolved in CH3CN (100 mL) and
2,3-dichloro-5,6-dicyano-1,4-benzoquinone (1.30 g, 5.75 mmol,
1.5 equiv) was added. The solution was stirred for 5 h at room tem-
perature. Li[B(C6F5)4]·Et2O (11.6 g, 15.3 mmol, 4.0 equiv) was added
and the mixture was diluted with water (200 mL). The precipitate
was filtered off, washed with water (200 mL), and dried under re-
duced pressure to give a pinkish powder (8.32 g, 3.06 mmol, 86%).
3
1H NMR (CD3CN): d [ppm]=8.33 (pt, 1H, JHH =8.1 Hz), 8.13 (pt, 2H,
3
3JHH =7.8 Hz), 7.38–7.43 (m, 4H), 7.01 pt, 2H, JHH =6.7 Hz), 6.90 (d,
3
2H, JHH =5.7 Hz), 3.12 (s, 6H). 19F NMR (CD3CN): d [ppm]=À134.07
(s), À164.22 (pt), À168.68 (pt). The 1H NMR data match reported
values of [Co(ddpd)2][BF4]3.[82] MS(ESI+): m/z (%)=175.1 (10)
[Co(ddpd)]2+, 184.1 (10) [Co(ddpd)+F]2+, 291.2 (12) [ddpd]+, 320.6
Keywords: cobalt electrolytes · cyclometalated complexes ·
dye-sensitized solar cells · mixed valency · ruthenium
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8926
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] B. O’Regan, M. Grätzel, Nature 1991, 353, 737–740.
[2] M. A. Green, Solar Energy 2004, 76, 3–8.
[3] M. Grätzel, J. Photochem. Photobiol. C 2003, 4, 145–153.
[4] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem. Rev.
2010, 110, 6595–6663.
[35] T. Bessho, E. Yoneda, J.-H. Yum, M. Guglielmi, I. Tavernelli, H. Imai, U.
Rothlisberger, M. K. Nazeeruddin, M. Grätzel, J. Am. Chem. Soc. 2009,
131, 5930–5934.
[36] B. D. Koivisto, K. C. D. Robson, C. P. Berlinguette, Inorg. Chem. 2009, 48,
9644–9652.
[5] R. McConnell, Renewable Sustainable Energy Rev. 2002, 6, 271–293.
[6] S. Ferrere, B. A. Gregg, J. Am. Chem. Soc. 1998, 120, 843–844.
[7] N. Alonso-Vante, J.-F. Nierengarten, J.-P. Sauvage, J. Chem. Soc. Dalton
Trans. 1994, 1649–1654.
[8] T. Bessho, E. C. Constable, M. Graetzel, A. Hernandez Redondo, C. E.
Housecroft, W. Kylberg, M. K. Nazeeruddin, M. Neuburger, S. Schaffner,
Chem. Commun. 2008, 3717–3719.
[37] S. H. Wadman, J. M. Kroon, K. Bakker, R. W. A. Havenith, G. P. M. van
Klink, G. van Koten, Organometallics 2010, 29, 1569–1579.
[38] a) K. C. D. Robson, B. D. Koivisto, A. Yella, B. Sporinova, M. K. Nazeerud-
din, T. Baumgartner, M. Grätzel, C. P. Berlinguette, Inorg. Chem. 2011,
50, 5494–5508; b) P. G. Bomben, T. J. Gordon, E. Schott, C. P. Berlingu-
ette, Angew. Chem. Int. Ed. 2011, 50, 10682–10685; Angew. Chem.
2011, 123, 10870–10873.
[9] C. E. Housecroft, E. C. Constable, Chem. Soc. Rev. 2015, 44, 8386–8398.
[10] E. A. M. Geary, N. Hirata, J. Clifford, J. R. Durrant, S. Parsons, A. Dawson,
L. J. Yellowlees, N. Robertson, Dalton Trans. 2003, 3757–3762.
[11] E. C.-H. Kwok, M.-Y. Chan, K. M.-C. Wong, W. H. Lam, V. W.-W. Yam,
Chem. Eur. J. 2010, 16, 12244–12254.
[12] E. I. Mayo, K. Kilså, T. Tirrell, P. I. Djurovich, A. Tamayo, M. E. Thompson,
N. S. Lewis, H. B. Gray, Photochem. Photobiol. Sci. 2006, 5, 871–873.
[13] E. Baranoff, J. Yum, I. Jung, R. Vulcano, M. Grätzel, M. K. Nazeeruddin,
Chem. Asian J. 2010, 5, 496–499.
[39] J.-J. Kim, H. Choi, S. Paek, C. Kim, K. Lim, M.-J. Ju, H. S. Kang, M.-S.
Kang, J. Ko, Inorg. Chem. 2011, 50, 11340–11347.
[40] P. G. Bomben, K. C. Robson, B. D. Koivisto, C. P. Berlinguette, Coord.
Chem. Rev. 2012, 256, 1438–1450.
[41] T. Funaki, H. Funakoshi, O. Kitao, N. Onozawa-Komatsuzaki, K. Kasuga,
K. Sayama, H. Sugihara, Angew. Chem. Int. Ed. 2012, 51, 7528–7531;
Angew. Chem. 2012, 124, 7646–7649.
[42] B. Schulze, D. G. Brown, K. C. D. Robson, C. Friebe, M. Jäger, E. Birckner,
C. P. Berlinguette, U. S. Schubert, Chem. Eur. J. 2013, 19, 14171–14180.
[43] E. C. Constable, J. M. Holmes, J. Organomet. Chem. 1986, 301, 203–
208.
[14] J. B. Asbury, E. Hao, Y. Wang, T. Lian, J. Phys. Chem. B 2000, 104,
11957–11964.
[15] S. Kim, J. K. Lee, S. O. Kang, J. Ko, J.-H. Yum, S. Fantacci, F. de Angelis,
D. Di Censo, M. K. Nazeeruddin, M. Grätzel, J. Am. Chem. Soc. 2006,
128, 16701–16707.
[44] E. C. Constable, C. E. Housecroft, Polyhedron 1990, 9, 1939–1947.
[45] M. Beley, J. P. Collin, R. Louis, B. Metz, J. P. Sauvage, J. Am. Chem. Soc.
1991, 113, 8521–8522.
[16] A. Mishra, M. K. R. Fischer, P. Bäuerle, Angew. Chem. Int. Ed. 2009, 48,
2474–2499; Angew. Chem. 2009, 121, 2510–2536.
[46] S. H. Wadman, M. Lutz, D. M. Tooke, A. L. Spek, F. Hartl, R. W. A. Have-
nith, G. P. M. van Klink, G. van Koten, Inorg. Chem. 2009, 48, 1887–
1900.
[17] W. M. Campbell, K. W. Jolley, P. Wagner, K. Wagner, P. J. Walsh, K. C.
Gordon, L. Schmidt-Mende, M. K. Nazeeruddin, Q. Wang, M. Grätzel,
D. L. Officer, J. Phys. Chem. C 2007, 111, 11760–11762.
[18] L.-L. Li, E. W.-G. Diau, Chem. Soc. Rev. 2013, 42, 291–304.
[19] S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. E. Curchod, N.
Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazeeruddin, M. Grät-
zel, Nat. Chem. 2014, 6, 242–247.
[20] S. Rühle, M. Shalom, A. Zaban, ChemPhysChem 2010, 11, 2290–2304.
[21] M. K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Mueller, P.
Liska, N. Vlachopoulos, M. Graetzel, J. Am. Chem. Soc. 1993, 115, 6382–
6390.
[47] P. G. Bomben, K. C. D. Robson, P. A. Sedach, C. P. Berlinguette, Inorg.
Chem. 2009, 48, 9631–9643.
[48] a) C. Kreitner, E. Erdmann, W. W. Seidel, K. Heinze, Inorg. Chem. 2015,
54, 11088–11104; b) C. Kreitner, K. Heinze, Dalton Trans. 2016, 45,
5640–5658.
[49] B. Durham, J. V. Caspar, J. K. Nagle, T. J. Meyer, J. Am. Chem. Soc. 1982,
104, 4803–4810.
[50] K. Kalyanasundaram, Coord. Chem. Rev. 1982, 46, 159–244.
[51] M. Maestri, N. Armaroli, V. Balzani, E. C. Constable, A. M. W. C. Thomp-
son, Inorg. Chem. 1995, 34, 2759–2767.
[22] M. K. Nazeeruddin, P. PØchy, T. Renouard, S. M. Zakeeruddin, R. Humph-
ry-Baker, P. Comte, P. Liska, L. Cevey, E. Costa, V. Shklover, L. Spiccia,
G. B. Deacon, C. A. Bignozzi, M. Grätzel, J. Am. Chem. Soc. 2001, 123,
1613–1624.
[52] V. Balzani, A. Juris, Coord. Chem. Rev. 2001, 211, 97–115.
[53] T. Edvinsson, C. Li, N. Pschirer, J. Schçneboom, F. Eickemeyer, R. Sens,
G. Boschloo, A. Herrmann, K. Müllen, A. Hagfeldt, J. Phys. Chem. C
2007, 111, 15137–15140.
[23] S. Altobello, R. Argazzi, S. Caramori, C. Contado, S. Da FrØ, P. Rubino, C.
ChonØ, G. Larramona, C. A. Bignozzi, J. Am. Chem. Soc. 2005, 127,
15342–15343.
[54] A. Abbotto, C. Barolo, L. Bellotto, F. de Angelis, M. Grätzel, N. Manfredi,
C. Marinzi, S. Fantacci, J.-H. Yum, M. K. Nazeeruddin, Chem. Commun.
2008, 5318–5320.
[24] T. W. Rees, E. Baranoff, Polyhedron 2014, 82, 37–49.
[25] G. A. Crosby, Acc. Chem. Res. 1975, 8, 231–238.
[55] M. Ishida, S. W. Park, D. Hwang, Y. B. Koo, J. L. Sessler, D. Y. Kim, D. Kim,
J. Phys. Chem. C 2011, 115, 19343–19354.
[26] M. L. Stone, G. A. Crosby, Chem. Phys. Lett. 1981, 79, 169–173.
[27] J. R. Winkler, T. L. Netzel, C. Creutz, N. Sutin, J. Am. Chem. Soc. 1987,
109, 2381–2392.
[56] J.-H. Yum, S.-J. Moon, C. S. Karthikeyan, H. Wietasch, M. Thelakkat, S. M.
Zakeeruddin, M. Nazeeruddin, M. Grätzel, Nano Energy 2012, 1, 6–12.
[57] A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin,
E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin, M. Grätzel, Science 2011,
334, 629–634.
[28] M. K. Nazeeruddin, S. M. Zakeeruddin, R. Humphry-Baker, M. Jirousek,
P. Liska, N. Vlachopoulos, V. Shklover, C.-H. Fischer, M. Grätzel, Inorg.
Chem. 1999, 38, 6298–6305.
[29] M. K. Nazeeruddin, F. de Angelis, S. Fantacci, A. Selloni, G. Viscardi, P.
Liska, S. Ito, B. Takeru, M. Grätzel, J. Am. Chem. Soc. 2005, 127, 16835–
16847.
[58] K. C. D. Robson, B. D. Koivisto, T. J. Gordon, T. Baumgartner, C. P. Berlin-
guette, Inorg. Chem. 2010, 49, 5335–5337.
[59] K. C. D. Robson, B. Sporinova, B. D. Koivisto, E. Schott, D. G. Brown,
C. P. Berlinguette, Inorg. Chem. 2011, 50, 6019–6028.
[60] K. Hu, K. C. D. Robson, P. G. Johansson, C. P. Berlinguette, G. J. Meyer, J.
Am. Chem. Soc. 2012, 134, 8352–8355.
[30] M. Nazeeruddin, M. Grätzel, J. Photochem. Photobiol. A 2001, 145, 79–
86.
[31] H. T. Nguyen, H. M. Ta, T. Lund, Sol. Energy Mater. Sol. Cells 2007, 91,
1934–1942.
[61] M. B. Robin, P. Day, Adv. Inorg. Chem. 1968, 10, 247–422.
[62] C.-J. Yao, R.-H. Zheng, Q. Shi, Y.-W. Zhong, J. Yao, Chem. Commun.
2012, 48, 5680–5682.
[63] a) C.-J. Yao, H.-J. Nie, W.-W. Yang, J.-Y. Shao, J. Yao, Y.-W. Zhong, Chem.
Eur. J. 2014, 20, 17466–17477; b) B.-B. Cui, Y.-W. Zhong, J. Yao, J. Am.
Chem. Soc. 2015, 137, 4058–4061.
[32] a) T. P. Brewster, W. Ding, N. D. Schley, N. Hazari, V. S. Batista, R. H. Crab-
tree, Inorg. Chem. 2011, 50, 11938–11946; b) S. Kämper, A. Paretzki, J.
ˇ
Fiedler, S. Zµlis, W. Kaim, Inorg. Chem. 2012, 51, 2097–2104.
[33] a) A. Breivogel, S. Wooh, J. Dietrich, T. Y. Kim, Y. S. Kang, K. Char, K.
Heinze, Eur. J. Inorg. Chem. 2014, 2720–2734; b) A. Breivogel, C. Kreit-
ner, K. Heinze, Eur. J. Inorg. Chem. 2014, 5468–5490.
[64] Y.-W. Zhong, Z.-L. Gong, J.-Y. Shao, J. Yao, Coord. Chem. Rev. 2016, 312,
22–40.
[34] S. H. Wadman, J. M. Kroon, K. Bakker, M. Lutz, A. L. Spek, G. P. M. van
Klink, G. van Koten, Chem. Commun. 2007, 1907–1909.
[65] B. S. Brunschwig, C. Creutz, N. Sutin, Chem. Soc. Rev. 2002, 31, 168–
184.
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8927
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[66] R. K. Chitumalla, K. S. V. Gupta, C. Malapaka, R. Fallahpour, A. Islam, L.
Han, B. Kotamarthi, S. P. Singh, Phys. Chem. Chem. Phys. 2014, 16,
2630–2640.
[67] S. M. Feldt, P. W. Lohse, F. Kessler, M. K. Nazeeruddin, M. Grätzel, G.
Boschloo, A. Hagfeldt, Phys. Chem. Chem. Phys. 2013, 15, 7087–7097.
[68] J. W. Ondersma, T. W. Hamann, Coord. Chem. Rev. 2013, 257, 1533–
1543.
[69] M. Wang, C. Grätzel, S. M. Zakeeruddin, M. Grätzel, Energy Environ. Sci.
2012, 5, 9394–9405.
[70] E. Mosconi, J.-H. Yum, F. Kessler, C. J. G. García, C. Zuccaccia, A. Cinti,
M. K. Nazeeruddin, M. Grätzel, F. D. Angelis, J. Am. Chem. Soc. 2012,
134, 19438–19453.
[71] T. W. Hamann, Dalton Trans. 2012, 41, 3111–3115.
[72] J.-H. Yum, E. Baranoff, F. Kessler, T. Moehl, S. Ahmad, T. Bessho, A.
Marchioro, E. Ghadiri, J.-E. Moser, C. Yi, M. K. Nazeeruddin, M. Grätzel,
Nat. Commun. 2012, 3, 631; DOI: 10.1038/ncomms1655.
[73] S. M. Feldt, G. Wang, G. Boschloo, A. Hagfeldt, J. Phys. Chem. C 2011,
115, 21500–21507.
[87] C. Kreitner, M. Grabolle, U. Resch-Genger, K. Heinze, Inorg. Chem. 2014,
53, 12947–12961.
[88] K. Heinze, K. Hempel, Chem. Eur. J. 2009, 15, 1346–1358.
[89] G. Boschloo, A. Hagfeldt, Acc. Chem. Res. 2009, 42, 1819–1826.
[90] X. Wang, D. M. Stanbury, Inorg. Chem. 2006, 45, 3415–3423.
[91] T. L. J. Huang, D. G. Brewer, Can. J. Chem. 1981, 59, 1689–1700.
[92] Y. Liu, J. R. Jennings, Y. Huang, Q. Wang, S. M. Zakeeruddin, M. Grätzel,
J. Phys. Chem. C 2011, 115, 18847–18855.
[93] D. K. Lee, K.-S. Ahn, S. Thogiti, J. H. Kim, Dyes Pigm. 2015, 117, 83–91.
[94] V. V. Pavlishchuk, A. W. Addison, Inorg. Chim. Acta 2000, 298, 97–102.
[95] I. Krivokapic, M. Zerara, M. L. Dakua, A. Vargas, C. Enachescu, C. Am-
brusc, P. Tregenna-Piggott, N. Amstutz, E. Krausz, A. Hauser, Coord.
Chem. Rev. 2007, 251, 364–378.
[96] M. Chavarot, Z. Pikramenou, Tetrahedron Lett. 1999, 40, 6865–6868.
[97] K. Suzuki, A. Kobayashi, S. Kaneko, K. Takehira, T. Yoshihara, H. Ishida, Y.
Shiina, S. Oishi, S. Tobita, Phys. Chem. Chem. Phys. 2009, 11, 9850–
9860.
[98] G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman,
B. M. Stoltz, J. E. Bercaw, K. I. Goldberg, Organometallics 2010, 29,
2176–2179.
[74] S. M. Feldt, E. A. Gibson, E. Gabrielsson, L. Sun, G. Boschloo, A. Hag-
feldt, J. Am. Chem. Soc. 2010, 132, 16714–16724.
[75] S. Caramori, J. Husson, M. Beley, C. A. Bignozzi, R. Argazzi, P. C. Gros,
Chem. Eur. J. 2010, 16, 2611–2618.
[76] B. M. Klahr, T. W. Hamann, J. Phys. Chem. C 2009, 113, 14040–14045.
[77] J. J. Nelson, T. J. Amick, C. M. Elliott, J. Phys. Chem. C 2008, 112, 18255–
18263.
[78] H. Nusbaumer, S. M. Zakeeruddin, J.-E. Moser, M. Grätzel, Chem. Eur. J.
2003, 9, 3756–3763.
[79] S. A. Sapp, C. Elliott, C. Contado, S. Caramori, C. A. Bignozzi, J. Am.
Chem. Soc. 2002, 124, 11215–11222.
[80] H. Nusbaumer, J.-E. Moser, S. M. Zakeeruddin, M. K. Nazeeruddin, M.
Grätzel, J. Phys. Chem. B 2001, 105, 10461–10464.
[81] A. Breivogel, C. Fçrster, K. Heinze, Inorg. Chem. 2010, 49, 7052–7056.
[82] C. Fçrster, K. Mack, L. M. Carrella, V. Ksenofontov, E. Rentschler, K.
Heinze, Polyhedron 2013, 52, 576–581.
[99] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910.
[100] F. Neese, WIREs Comput. Mol. Sci. 2012, 2, 73–78.
[101] F. Neese, F. Wennmohs, A. Hansen, U. Becker, Chem. Phys. 2009, 356,
98–109.
[102] R. Izsµk, F. Neese, J. Chem. Phys. 2011, 135, 144105.
[103] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[104] A. Schäfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97, 2571–2577.
[105] A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829–5835.
[106] E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem. Phys. 1993, 99,
4597–4610.
[107] D. A. Pantazis, X.-Y. Chen, C. R. Landis, F. Neese, J. Chem. Theory
Comput. 2008, 4, 908–919.
[108] S. Sinnecker, A. Rajendran, A. Klamt, M. Diedenhofen, F. Neese, J. Phys.
Chem. A 2006, 110, 2235–2245.
[83] A. K. C. Mengel, W. Cho, A. Breivogel, K. Char, Y. S. Kang, K. Heinze, Eur.
J. Inorg. Chem. 2015, 3299–3306.
[84] N. C. Bruno, M. T. Tudge, S. L. Buchwald, Chem. Sci. 2013, 4, 916–920.
[85] M. D. Charles, P. Schultz, S. L. Buchwald, Org. Lett. 2005, 7, 3965–3968.
[86] A. Breivogel, M. Meister, C. Fçrster, F. Laquai, K. Heinze, Chem. Eur. J.
2013, 19, 13745–13760.
Received: March 2, 2016
Published online on May 19, 2016
Chem. Eur. J. 2016, 22, 8915 – 8928
www.chemeurj.org
8928
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim