Iridium(III) complexes with phenyl-tetrazoles as cyclometalating ligands

Article abstract of DOI:10.1021/ic500999k

Ir(III) cationic complexes with cyclometalating tetrazolate ligands were prepared for the first time, following a two-step strategy based on (i) a silver-assisted cyclometalation reaction of a tetrazole derivative with IrCl₃ affording a bis-cyclometalated solvato-complex P ([Ir(ptrz)₂(CH₃CN)₂]⁺, Hptrz = 2-methyl-5-phenyl-2H-tetrazole); (ii) a substitution reaction with five neutral ancillary ligands to get [Ir(ptrz)₂L]⁺, with L = 2,2′-bypiridine (1), 4,4′-di-tert-butyl-2,2′-bipyridine (2), 1,10-phenanthroline (3), and 2-(1-phenyl-1H-1,2,3-triazol-4-yl)pyridine (4), and [Ir(ptrz)₂L₂]⁺, with L = tert-butyl isocyanide (5). X-ray crystal structures of P, 2, and 3 were solved. Electrochemical and photophysical studies, along with density functional theory calculations, allowed a comprehensive rationalization of the electronic properties of 1-5. In acetonitrile at 298 K, complexes equipped with bipyridine or phenanthroline ancillary ligands (1-3) exhibit intense and structureless emission bands centered at around 540 nm, with metal-to-ligand and ligand-to-ligand charge transfer (MLCT/LLCT) character; their photoluminescence quantum yields (PLQYs) are in the range of 55-70%. By contrast, the luminescence band of 5 is weak, structured, and blue-shifted and is attributed to a ligand-centered (LC) triplet state of the tetrazolate cyclometalated ligand. The PLQY of 4 is extremely low (<0.1%) since its lowest level is a nonemissive triplet metal-centered ( ³MC) state. In rigid matrix at 77 K, all of the complexes exhibit intense luminescence. Ligands 1-3 are also strong emitters in solid matrices at room temperature (1% poly(methyl methacrylate) matrix and neat films), with PLQYs in the range of 27-70%. Good quality films of 2 could be obtained to make light-emitting electrochemical cells that emit bright green light and exhibit a maximum luminance of 310 cd m^-2. Tetrazolate cyclometalated ligands push the emission of Ir(III) complexes to the blue, when compared to pyrazolate or triazolate analogues. More generally, among the cationic Ir(III) complexes without fluorine substituents on the cyclometalated ligands, 1-3 exhibit the highest-energy MLCT/LLCT emission bands ever reported.

Full text of DOI:10.1021/ic500999k

Article
pubs.acs.org/IC
Iridium(III) Complexes with Phenyl-tetrazoles as Cyclometalating
Ligands
†
‡
‡
ez,§
́ ́
M. Junquera-Hernandez, Domenica Tonelli, Andrea Mazzanti, Sara Muzzioli, Stefano Stagni,
§
Filippo Monti, Andrea Baschieri, Isacco Gualandi, Juan J. Serrano-Per
́
§
‡
‡
‡
‡
Jose
Cristina Roldan-Carmona, Antonio Pertegas
and Nicola Armaroli*
§
§
,§
,‡
́
, Henk J. Bolink, Enrique Ortí,* Letizia Sambri,*
,†
†
Istituto per la Sintesi Organica e la Fotoreattivita,
̀
Consiglio Nazionale delle Ricerche, Via P. Gobetti, 101, I-40129, Bologna, Italy
Dipartimento di Chimica Industriale “Toso Montanari”, Universita di Bologna, Viale Risorgimento 4, I-40136, Bologna, Italy
ico Jose Beltran 2, E-46980, Paterna, Spain
‡
̀
§
Instituto de Ciencia Molecular, Universidad de Valencia, Catedrat
́
́
́
*
S Supporting Information
ABSTRACT: Ir(III) cationic complexes with cyclometalating
tetrazolate ligands were prepared for the first time, following a
two-step strategy based on (i) a silver-assisted cyclometalation
reaction of a tetrazole derivative with IrCl affording a bis-
3
+
cyclometalated solvato-complex P ([Ir(ptrz) (CH CN) ] ,
2
3
2
Hptrz = 2-methyl-5-phenyl-2H-tetrazole); (ii) a substitution
+
reaction with five neutral ancillary ligands to get [Ir(ptrz) L] ,
2
with L = 2,2′-bypiridine (1), 4,4′-di-tert-butyl-2,2′-bipyridine
(
2), 1,10-phenanthroline (3), and 2-(1-phenyl-1H-1,2,3-
+
triazol-4-yl)pyridine (4), and [Ir(ptrz) L ] , with L = tert-
2
2
butyl isocyanide (5). X-ray crystal structures of P, 2, and 3 were solved. Electrochemical and photophysical studies, along with
density functional theory calculations, allowed a comprehensive rationalization of the electronic properties of 1−5. In acetonitrile
at 298 K, complexes equipped with bipyridine or phenanthroline ancillary ligands (1−3) exhibit intense and structureless
emission bands centered at around 540 nm, with metal-to-ligand and ligand-to-ligand charge transfer (MLCT/LLCT) character;
their photoluminescence quantum yields (PLQYs) are in the range of 55−70%. By contrast, the luminescence band of 5 is weak,
structured, and blue-shifted and is attributed to a ligand-centered (LC) triplet state of the tetrazolate cyclometalated ligand. The
3
PLQY of 4 is extremely low (<0.1%) since its lowest level is a nonemissive triplet metal-centered ( MC) state. In rigid matrix at
7
7 K, all of the complexes exhibit intense luminescence. Ligands 1−3 are also strong emitters in solid matrices at room
temperature (1% poly(methyl methacrylate) matrix and neat films), with PLQYs in the range of 27−70%. Good quality films of 2
could be obtained to make light-emitting electrochemical cells that emit bright green light and exhibit a maximum luminance of
−2
310 cd m . Tetrazolate cyclometalated ligands push the emission of Ir(III) complexes to the blue, when compared to pyrazolate
or triazolate analogues. More generally, among the cationic Ir(III) complexes without fluorine substituents on the cyclometalated
ligands, 1−3 exhibit the highest-energy MLCT/LLCT emission bands ever reported.
INTRODUCTION
Emission color tuning is conceptually simple and is
accomplished by modifying the energy gap separating the
highest-occupied (HOMO) and lowest-unoccupied molecular
orbital (LUMO): blue shifting (i.e., bandgap widening) is
obtained through stabilization of the HOMO and/or
destabilization of the LUMO, and red shifting (bandgap
■
Ir(III) cyclometalated (C^N) complexes continue to be the
most widely utilized class of triplet emitters in flat electro-
luminescent devices such as organic light-emitting diodes
1
−11
(
OLEDs) and light-emitting electrochemical cells (LECs).
Their high luminescence quantum yield and unsurpassed
versatility in tuning the emission color throughout the entire
visible spectral window by ligand design have prompted an
impressive amount of work on this class of metal complexes
worldwide. Over the past 20 years, a huge amount of blue,
green, and red Ir(III) emitters, made of both homoleptic
1
,5
narrowing) is accomplished with the opposite approach.
In the case of Ir(III) complexes, the HOMO orbital is
predominantly located on the cyclometalating ligand, whereas
the LUMO is usually centered on the ancillary diimino N^N
ligand, as long as it possesses low-lying π* orbitals such as 2,2′-
bipyridine (bpy) or 1,10-phenanthroline (phen), as is very
0
/+
[
Ir(C^N) ]- and heteroleptic [Ir(C^N) (L^L)] -type com-
3
2
1
1
,8,12
often the case. From the preparative standpoint, it is much
plexes, has been produced.
neutral and are targeted to OLED technology,
The former are electrically
7
,13
whereas the
latter are normally monocationic (L^L typically denotes a5
Received: May 2, 2014
neutral chelating ligand) and hence suitable for LEC devices.
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
easier to chemically modify the ancillary ligand, which is
prepared independently and eventually attached to the
Chart 1. Tetrazole-Based Ionic Ir(III) Complexes
Investigated in This Work
cyclometalated Ir(III) μ-dichloro-bridged dimer [(C^N) Ir(μ-
2
Cl)] that contains the four selected cyclometalated units.
2
Therefore, in the vast majority of cases, color tuning has been
1
4−17
pursued through modification of the ancillary ligand.
The less-frequently investigated modification of the C^N
ligand has been primarily addressed to the preparation of blue-
emitting complexes, which are presently the key target as active
1
5,18−26
materials for electroluminescent devices.
This is because
wide HOMO−LUMO gaps are more difficult to obtain and,
moreover, the high energy levels of blue emitters tend to be
1
reactive under the device operating conditions.
Two main modification strategies of the C^N ligand have
been pursued to stabilize the HOMO and shift the emission
band of Ir(III) complexes to the blue: (i) addition of electron-
27
withdrawing groups, typically fluorines, and (ii) reduction of
the ring size of the aromatic N-heterocycle. As far as the latter
approach is concerned, several examples of Ir(III) complexes
with cyclometalating ligands entailing 5-membered nitrogen-
containing rings have been reported, namely, 1-aryl-1,2-
1
8,28−30
31
pyrazoles,
aryl triazoles have been also proposed, such as 5-aryl-1,2,4-
and phenyl imidazoles. In addition, a few
3
2
33−38
triazoles and aryl-1,2,3-triazoles.
However, only cationic
Ir(III) complexes with 1-phenyl-1,2-pyrazoles led to a blue shift
of the emission compared to analogues with the prototypical 2-
+
phenylpyridine ligand  for instance: [Ir(ppy) (bpy)] (Hppy
2
21
+
=
=
2-phenylpyridine), λ = 602 nm; [Ir(ppz) (bpy)] (Hppz
max 2
28
1-phenyl-1,2-pyrazole), λmax = 563 nm. Surprisingly, the use
of phenyl-triazoles did not further blue shift the emission band
+
(
[Ir(phtl) (bpy)] (Hphtl = 1-phenyl-1,2,3-triazole), λ = 580
2 max
34
nm). Therefore, an increase of the number of nitrogen atoms
on the 5-membered heterocycle does not warrant an increase of
the emission energy of the complexes, although one might
expect that electron-rich triazoles in C^N ligands should
enhance the ligand field splitting of Ir(III) d orbitals.
This unexpected result prompted us to explore the possibility
of using the 5-phenyltetrazole cyclometalating ligand that,
according to a preliminary appraisal by quantum chemical
calculations, should have afforded complexes with blue-shifted
remarkably high emission quantum yields (∼70%) that are
comparable only to some polyfluorinated Ir-iTMCs (e.g.,
A2).
34
emissions compared to the pyrazole-based analogue [Ir-
+
There are at least three main advantages in using the
cyclometalating ligands herein presented for the synthesis of
cationic Ir(III) complexes: (i) the high-field nature of phenyl-
tetrazole that can lead to blue-shifted emissions without the
need to use electron-deficient polyfluorinated ligands while, at
the same time, preserving the relatively short-lived and highly
emissive metal-to-ligand charge transfer (MLCT) nature of the
(
ppz) (bpy)] . However, to the best of our knowledge, 5-
2
aryltetrazoles had never been used as cyclometalating ligands
for Ir(III) complexes, probably because the cyclometalation of
the iridium metal center by the 5-aryltetrazole ligand cannot be
directly accomplished (vide infra).
Herein, we present not only the first series of iridium(III)
phenyltetrazolate complexes, but also the first example of a
silver-assisted cyclometalation reaction of a tetrazole derivative
2
emitting states; (ii) the absence of reactive C(sp )−F bonds
that normally afford instability when the complex is used in
starting directly from IrCl . In this way we obtained the bis-
3
2
0,39,40
LECs;
(iii) the higher radiative rate constants compared
cyclometalated phenyl-tetrazole Ir(III) solvato complex [Ir-
+
to standard blue-emitting fluorine-free Ir-iTMCs (with, e.g.,
(
ptrz) (CH CN) ] (P, Hptrz = 2-methyl-5-phenyl-2H-tetra-
2
3
2
22,23,26
isocyanide ancillary ligands),
which allows optimized
22,26
zole), which was used as a precursor for the preparation of five
new Ir-iTMCs (1−5; iTMCs = ionic transition-metal
complexes) by reaction with different neutral ancillary ligands.
The complete series of complexes based on the novel Hptrz
cyclometalating ligand is reported in Chart 1, along with the
precursor P.
performance in devices preventing exciton quenching.
EXPERIMENTAL SECTION
Synthesis of 2-Methyl-5-phenyl-2H-tetrazole (Ia, Scheme 1).
A solution of 5-phenyl-2H-tetrazole (I) (1.0 g, 6.8 mmol) in CH CN
■
3
(
70 mL) was treated with K CO (3.8 g, 27.4 mmol) followed by
2 3
As correctly anticipated by our preliminary theoretical
assessment, this series of compounds displays markedly blue-
shifted emission bands (λmax ≈ 530 nm), if compared to other
nonfluorinated Ir(III) complexes (A1, B1, and C1, Chart
addition of methyl iodide (0.86 mL, 13.6 mmol), according to a
42
described procedure. The reaction was stirred at reflux for 72 h, then
it was cooled to room temperature, the residual solid was filtered off,
and the solution was concentrated in vacuo to give a dark oil. The oil
was then purified by column chromatography on silica eluting with
2
1,28,34
2).
In addition to that, some of them (1−3) show
B
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
CH), 129.8 (CH), 130.0 (C), 133.1 (C), 133.6 (CH), 174.3 (C)].
Chart 2. Selected Archetypal Examples of Ir-iTMCs Having
2
1,28,34,41
Anal. Calcd for C H BF IrN (721.18) (X-ray crystals): C, 36.61; H,
2
,2′-Bipyridine as Neutral Ancillary Ligand
22 23
4
11
3
.21; N, 21.36%. Found: C, 36.50; H, 3.08; N, 21.29%. Electrospray
+
+
ionization mass spectrometry (ESI-MS ): 593 (M − BF ) ; 552 (M −
BF − ACN) ; 511 (M − BF − 2ACN) .
4
+
+
4
4
Synthesis of Complexes 1−4. The solvato complex P (50 mg,
0
.074 mmol) was dissolved in a 3:1 mixture of CH Cl /EtOH (4 mL),
2
2
and then the appropriate bidentate diimino ligand (0.22 mmol) was
added, that is, 2,2′-bypiridine (bpy), 4,4′-di-tert-butyl-2,2′-bipyridine
t
(
Bu-bpy), 1,10-phenanthroline (phen), and 2-(1-phenyl-1H-1,2,3-
triazol-4-yl)pyridine (taz) for complexes 1, 2, 3, and 4, respectively.
The mixture was stirred overnight at room temperature (RT), and
then the solvent was evaporated. The solid was dissolved in acetone,
precipitated with Et O, and the obtained solid was purified by column
2
chromatography (CH Cl /MeOH = 98/2) to give pure complexes 1−
2
2
4
.
1
Complex 1 (42 mg, 75% yield). H NMR (CD Cl ) δ: 4.27 (s,
2
2
6
2
H), 6.24 (d, 2H, JHH = 8.8 Hz), 6.98 (t, 2H, JHH = 8.8 Hz), 7.05 (t,
H, JHH = 8.8 Hz), 7.36 (t, 2H, JHH = 7.0 Hz), 7.70 (d, 2H, J = 7.0
HH
Hz), 8.00−8.10 (m, 4H), 8.51 (d, 2H, J = 7.0); 13C NMR (CD Cl )
HH
2
2
δ: 42.1 (CH ), 123.6 (CH), 124.8 (CH), 125.1 (CH), 127.7 (CH),
3
1
29.2 (C), 131.5 (CH), 132.6 (CH), 140.2 (CH), 145.6 (C), 151.7
CH), 156.7 (C), 174.6 (C). Anal. Calcd for C H BF IrN
10
(
(
2
6
22
4
754.17): C, 41.37; H, 2.94; N, 18.57%. Found: C, 41.44; H, 2.85;
N, 18.49%. ESI-MS : 667 (M − BF ) .
+
+
4
1
Complex 2 (44.3 mg, 70% yield). H NMR (CD Cl ) δ: 1.45 (s,
2
2
1
8H), 4.37 (s, 6H), 6.33 (d, 2H, JHH = 7.4 Hz), 7.06 (dt, 2H, JHH = 7.4
Hz, JHH = 1.7 Hz), 7.14 (dt, 2H, JHH = 7.4 Hz, JHH = 1.3 Hz), 7.43 (dd,
H, JHH = 5.8 Hz, JHH = 1.9 Hz), 7.80 (dd, 2H, JHH = 7.4 Hz, J = 1.7
2
HH
1
3
Scheme 1. Reactivity of 5-Phenyltetrazoles 1a−c under
Hz), 8.02 (d, 2H, JHH = 5.8 Hz), 8.30 (d, 2H, JHH = 1.9 Hz); C NMR
(CD Cl ) δ: 30.0 (CH ), 35.6 (C), 41.6 (CH ), 120.4 (CH), 123.5
Various Conditions
2
2
3
3
(
1
CH), 124.7 (CH), 124.8 (CH), 129.2 (C), 131.4 (CH), 132.5 (CH),
45.0 (C), 151.3 (CH), 156.3 (C), 164.2 (C), 174.7 (C). Anal. Calcd
for C H BF IrN Cl (950.25) (X-ray crystals): C, 44.20; H, 4.24; N,
35
40
4
10
2
+
1
4.74%. Found: C, 44.00; H, 4.15; N, 14.94%. ESI-MS : 779 (M −
+
BF ) .
4
1
Complex 3 (31 mg, 54% yield). H NMR (CD Cl ) δ: 4.17 (s,
2
2
6
H), 6.35 (d, 2H, JHH = 7.6 Hz), 7.02 (dt, 2H, JHH = 7.5 Hz, JHH = 1.6
Hz), 7.08 (dt, 2H, JHH = 8.8 Hz, JHH = 1.9 Hz), 7.36 (t, 2H, JHH = 7.0
Hz), 7.70−7.75 (m, 4H), 8.02 (s, 2H), 8.33 (dd, 2H, J = 5.5 Hz, J
HH
HH
1
3
=
1.9 Hz), 8.57 (dd, 2H, J = 8.5 Hz, J = 1.6 Hz); C NMR
HH HH
(
CD Cl ) δ: 41.5 (CH ), 123.7 (CH), 124.8 (CH), 126.0 (CH), 128.2
2 2 3
CH Cl /petroleum ether (4/6) and then CH Cl /EtOAc (98/2).
(CH), 129.4 (C), 131.1 (C), 131.4 (CH), 132.7 (CH), 138.7 (CH),
144.9 (C), 147.7 (C), 152.0 (CH), 174.6 (C). Anal. Calcd for
C H BF IrN Cl (862.12) (X-ray crystals): C, 40.36; H, 2.80; N,
16.24%. Found: C, 40.50; H, 2.89; N, 16.50%. ESI-MS : 691 (M −
BF ) .
2
2
2
2
Homogeneous fractions by thin-layer chromatography were combined
and concentrated in vacuum to give 2-methyl-5-phenyl-2H-tetrazole
29
24
4
10
2
1
+
(
(
0.67 g, 59.1% yield, less polar spot) as a pale yellow solid [ H NMR
13
+
CDCl ) δ: 4.33 (s, 3H), 7.35−7.50 (m, 3H), 8.05−8.15 (m, 2H);
NMR (CDCl ) δ: 39.5 (CH ), 126.8 (CH), 127.3 (C), 128.9 (CH),
C
3
4
1
Complex 4 (25 mg, 42% yield). H NMR (CD Cl ) δ: 4.29 (s,
3
3
2
2
1
2
7
30.3 (CH), 165.2 (C)] and 1-methyl-5-phenyl-2H-tetrazole (0.26 g,
3H), 4.30 (s, 3H), 6.27 (d, 2H, JHH = 7.5 Hz), 6.90−7.10 (m, 4H),
1
3.7% yields) as a white solid [ H NMR (CDCl ) δ: 4.18 (s, 3H),
7.22 (bt, 1H, JHH = 6.7 Hz), 7.40−7.50 (m, 3H), 7.69 (dd, 2H, J
=
3
HH
13
.55−7.60 (m, 3H), 7.70−7.75 (m, 2H); C NMR (CDCl ) δ: 35.0
7.1 Hz, JHH = 11.3 Hz), 7.78 (bd, 1H, J = 7.1 Hz), 7.89 (d, 1H, J
3
HH HH
13
(
CH ), 123.8 (C), 128.6 (CH), 129.3 (CH), 131.3 (CH), 165.9 (C)].
= 5.2 Hz), 8.0 (bs, 1H), 8.5 (bs, 2H), 9.5 (s, 2H); C NMR (CD Cl )
2 2
3
Synthesis of Ir(III) Solvato Complex (P). The reactions were
δ: 41.6 (CH ), 41.7 (CH ), 120.8 (CH), 123.2 (CH), 123.5 (CH),
3 3
performed under N and in the absence of light. The solvents were not
123.6 (CH), 124.1 (C), 124.3 (CH), 124.7 (CH), 125.6 (CH), 129.3
(C), 130.0 (CH), 130.2 (CH), 130.6 (CH), 131.3 (CH), 132.5 (CH),
133.0 (CH), 136.1 (CH), 140.0 (CH), 141.8 (C), 145.1 (C), 150.2
(C), 150.3 (C), 151.0 (CH), 159.3 (C), 174.4 (C), 174.8 (C). Anal.
Calcd for C H BF IrN (820.19): C, 42.43; H, 2.95; N, 20.49%.
2
dried but were deoxygenated by bubbling with N . The ligand 2-
2
methyl-5-phenyl-2H-tetrazole (96 mg, 0.6 mmol) was dissolved in a
3
:1 mixture of ethoxyethanol/water (4 mL), and then IrCl ·xH O (60
3 2
mg) was added. The mixture was stirred overnight at 140 °C under N2.
After cooling, the obtained yellow solid was filtered, washed with
water, dried, and dissolved in CH Cl (15 mL). AgBF (80 mg) was
29
24
4
12
+
+
Found: C, 42.71; H, 2.87; N, 20.37%. ESI-MS :733 (M − BF ) .
4
2
2
4
Synthesis of Complex 5. Caution! tert-Butyl isocyanide is a foul-
smelling volatile liquid; therefore, ensure adequate ventilation! The solvato
complex P (60 mg, 0.088 mmol) was dissolved in a 3:1 mixture of
added, and the solution was vigorously stirred and then refluxed for 6
h. After cooling, the solution was filtered through a paper filter to
remove AgCl and was evaporated to dryness. The resulting crude
t
CH Cl /EtOH (4 mL), and tert-butyl isocyanide (CN Bu, 0.88 mmol)
2
2
Ir(III) solvato complex was dissolved in CH CN, precipitated twice
was added. The mixture was left to stir overnight at 30 °C, and then
the solvent was evaporated. The solid was dissolved in acetone, and
3
with Et O, and then crystallized from slow diffusion of Et O vapors in
2
2
a CH CN solution to give colorless crystals of P (40 mg, 60% yield).
the product was precipitated with Et O to give pure complex 5 (27
3
2
1
1
H NMR (crystals, CD OD) δ: 2.03 (s, 6H), 4.65 (s, 6H), 6.14 (d, 2H,
mg, 40% yield). H NMR (CD Cl ) δ: 1.44 (s, 18H), 4.70 (s, 6H),
3
2
2
JHH = 7.5 Hz), 6.80 (dt, 2H, JHH = 7.5 Hz, JHH = 1.4 Hz), 6.92 (dt, 2H,
6.15 (d, 2H, JHH = 7.5 Hz), 7.01 (dt, 1H, JHH = 7.5 Hz, JHH = 1.6 Hz),
J
= 7.5 Hz, JHH = 1.2 Hz), 7.64 (d, 2H, JHH = 7.5 Hz, JHH = 1.4 Hz);
7.09 (dt, 2H, J = 7.5 Hz, JHH = 1.4 Hz), 7.76 (d, 2H, JHH = 7.5 Hz);
HH
HH
1
3
13
C NMR (precipitate, CD OD) δ: 40.8 (CH ), 122.8 (CH), 123.6
C NMR (CD Cl ) δ: 30.0 (CH ), 42.2 (CH ), 65.6 (C), 124.2
3
3
2
2
3
3
C
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
1
4
CH), 124.8 (CH), 128.3 (C), 131.2 (CH), 131.3 (CH), 149.1 (C),
Scheme 2. Synthesis of Complexes 1−5 from P
57.6 (C), 174.8 (C). Anal. Calcd for C H BF IrN (764.25): C,
26 32 4 10
0.82; H, 4.22; N, 18.32%. Found: C, 41.02; H, 4.33; N, 18.27%. ESI-
+
+
MS :677 (M − BF ) .
4
Experimental details on the electrochemistry, photophysics, and
LEC devices and computational details on density functional
calculations are reported in the Supporting Information.
RESULTS AND DISCUSSION
■
Synthesis. Direct cyclometalation of phenylpyridines by
Ir(III) derivatives is an established reaction that is widely
exploited for the synthesis of hundreds of complexes.
43−45
The
classical cyclometalating methods work well also with cyclo-
metalating ligands containing 5-membered rings with two or
1
3,23−26,28−30
three nitrogen atoms.
Repeated attempts to obtain the classical dimer (C^N) Ir(μ-
rings are mutually in a trans position. The two acetonitrile
molecules of P fill the remaining equatorial and axial positions.
During the final reaction step, the two acetonitrile molecules
are displaced by the ancillary ligand (Figure 1).
2
Cl) Ir(C^N) by direct reaction of IrCl ·xH O with a N-
substituted 5-phenyltetrazole as cyclometalating ligand follow-
ing literature procedures
2
2
3
2
4
4,46
failed. On the other hand, the use
The Ir−N and Ir−C distances are summarized in Table 1
of high-boiling solvents, such as glycerol or ethylene glycol, or
the addition of a base did not bring any significant results. We
always observed the formation of an unknown product,
presumably a salt involving the ligand and IrCl3, without
any cyclometalation (Scheme 1, route A).
Therefore, in an attempt to induce cyclometalation, an
intermediate synthetic step consisting of the Ag(I)-mediated
halide extraction was introduced. Following this approach, the
yellow solid obtained by refluxing Ia with IrCl ·xH O in an
and are very similar to other iridium complexes reported in the
3
4,35,38,50−59
literature.
Therefore, the different geometry of the
tetrazole with respect to heterocycles used previously does not
alter the bonding distances.
47
Electrochemical Properties. The electrochemical behav-
ior of 1−5 was investigated by cyclic voltammetry (CV) in
acetonitrile (Figure 2). The nature of the ancillary ligands
affects both the potentials and the reversibility of the redox
processes. All compounds display at least one oxidation and
two reduction processes. The formal half-wave potentials of the
reversible processes and the peak potentials of the irreversible
ones are reported in Table 2.
The CVs of 1−3 display one reversible wave in the cathodic
side with a formal potential of approximately −1.8 V, along
with another irreversible reduction process at around −2.6 V.
The former is a monoelectronic process, as evidenced from a
47
3
2
1
ethoxyethanol/water mixture  whose H NMR spectrum
exactly matched that of the free ligand 1a  was reacted with
AgBF in CH Cl (Scheme 1, route B). The species obtained
4
2
2
was treated with CH CN and Et O vapors, affording the
cyclometalated solvato complex P, whose structure was
resolved by X-ray diffraction.
A thorough choice of the nature and position of the tetrazole
substituent was essential for the success of the reaction. Among
the ligands tested, only 2-methyl-5-phenyltetrazole Ia gave
satisfactory results. Indeed, when the methyl substituent is
shifted to the 1-position (Ib, Scheme 1) the reaction does not
occur. This is likely because 1-alkyl-5-phenyltetrazoles, either as
3
2
ΔE value of ∼70 mV. As far as the anodic side is concerned,
p
1−3 exhibit only one oxidation process around +1.2 V, which
−1
appears irreversible at a scan rate of 0.05 V s . As the scan rate
is increased, the backward peak indicates a complex kinetics of
this process (see Figure S1 in the Supporting Information for
complex 1). From the graphs of normalized peak currents
4
8
49
“
free” or as coordinated ligands, do not exhibit the
preferentially coplanar arrangement of the aromatic rings that
would be required for any ligand capable of coordinating a
metal center through a bis-chelate geometry. Moreover, if the
substituent is changed from methyl to tert-butyl, a C−H
versus the square root of the scan rate, an E CE-type kinetics
rev
is determined, that is, a reversible electrochemical process
followed by a chemical reaction that in turn is followed by
another electrochemical process. At a scan rate of 10 V s− the
shapes of the voltammograms are typical of a quasi-reversible
process; therefore, the calculated formal potentials can be
considered as thermodynamic data. Moreover, the appearance
of a backward wave for the process on the anodic side for 2
1
3
activation of an sp C−H bond can compete with the
cyclometalation of the arylic C−H bond, affording a complex
mixture of products.
The reaction of the solvato complex P with the bidentate
t
−1
ligands bpy, Bu-bpy, phen, and taz, under mild reaction
appears at a lower scan rate (0.1 V s ) than that needed for 1
−1
conditions, gave complexes 1−4 (Scheme 2) in satisfactory
and 3 (0.5 V s ). This observation suggests that, under the
conditions of the electrochemical experiments, the species
formed upon oxidation of 2 is more stable than those formed
yields. Similarly, complex 5 was obtained by treatment of P
t
with an excess of CN Bu.
−1
X-ray Characterization. Good-quality single crystals of P,
with 1 and 3. (0.5 V s ).
2
, and 3 were obtained by slow diffusion of Et O in CH CN
The CV of 4 displays two cathodic waves at −2.226 and
2
3
−1
and CH Cl solutions. The X-ray structures of the solvato
−2.68 V and one anodic wave at +1.256 V. At 0.05 V s only
the forward peaks are present for all the processes, but by
increasing the scan rate, the backward peaks of the less cathodic
process and of the anodic one appear. The mechanisms of the
electrochemical processes are E C and E CE, respectively
2
2
complex P and of complexes 2 and 3 showed the same
arrangement of the two phenyl-tetrazole ligands around a
pseudo-octahedral iridium ion. One of the phenyl-tetrazoles is
bound to two equatorial positions of the octahedral Ir center,
whereas the second one occupies an equatorial position with
nitrogen and an axial position with carbon (Figure 1). This
arrangement implies that the two nitrogens of the tetrazole
rev
rev
(see footnotes in Table 2).
The CV of 5 exhibits two cathodic waves at −2.504 and
−2.71 V and an anodic feature at +1.569 V that do not display
D
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Selected Data from X-ray Analysis
a
b
bite angle (deg) Ir−N (Å) Ir−C (Å) Ir−N (Å)
C^N
N^N
87.90
P
3
80.11
2.024
2.041
2.025
2.017
2.044
2.011
1.997
2.000
2.034
2.046
2.034
2.030
1.998
2.016
2.074
2.084
2.117
2.108
2.139
2.131
2.135
2.145
2.120
8
0.01
79.49
9.62
80.30
77.52
75.57
77.37
7
c
2
8
7
7
0.83
7.06
7.91
2.112
a
b
c
Tetrazole nitrogen. Nitrogens of the ancillary ligand. Two
independent molecules are present in the unit cell.
Figure 2. Cyclic voltammograms of 1−5 recorded at a scan rate of
−1
0
.05 V s in acetonitrile.
Table 2. Electrochemical Properties of 1−5
a
a
a
Ered1°′ (V)
Ered2/3°′ (V)
Eox°′ (V)
b
b
1
2
3
4
5
−1.785 (rev.)
−1.866 (rev.)
−1.769 (rev.)
−2.226 (ErevC)
−2.49 /−2.71
+1.156 (ErevCE)
+1.139 (ErevCE)
+1.179 (ErevCE)
+1.256 (ErevCE)
b
−2.64
−2.69
−2.68
b
b
b
b
b
−2.504 (irrev.)
−2.71
+1.569 (irrev.)
a
+
b
Potential values vs Fc /Fc. Peak potential value. rev. = reversible;
irrev. = irreversible; E C = reversible process followed by chemical
rev
reaction; E CE = reversible process followed by chemical reaction
rev
and a second electron transfer process. E °′ refers to the oxidation
ox
process; E °′ and E
refer to the first and second/third reduction
waves, respectively. The first electrochemical cathodic and anodic
red1
red2/3
processes were monoelectronic for all the investigated compounds.
The redox gap of the whole series (calculated as the difference
between the formal potentials of the cathodic and anodic
systems) spans over a wide range, namely, ∼2.9−4.1 V. As far
as the electrochemical behavior of 1−3 is concerned, we can
make the following observations: (i) the first reduction process
occurs at less cathodic potentials than that of 4 because the
triazole moiety is more electron-rich than bipyridine and
Figure 1. Experimental X-ray structures of the solvato complex P and
of complexes 2 and 3. ORTEP representations are at the 50%
probability. Counter anion, solvents, and hydrogen atoms are omitted
for the sake of clarity.
the corresponding backward wave either at 0.05 V s− or at
higher scan rates, suggesting irreversibility. In such a case, the
peak potentials, which can be considered only a rough
estimation of the thermodynamic values, are reported in
Table 2.
1
60
phenanthroline; (ii) the electrochemistry of 5 is highly
affected by the tert-butyl isocyanide ligands that have a strong
electron-withdrawing character, which lowers the energy of the
26
Ir-aryl centered HOMO. Notably, 5 exhibits by far the highest
Overall, the electrochemical data suggest that the nature of
the ancillary ligand affects the first reduction potentials more
remarkably than the oxidation potentials. Indeed, the first
reduction of the Ir complexes usually occurs on the ancillary
ligands, while the oxidation is centered on the Ir-aryl moiety.
redox gap of the series.
61
The excited-state electrochemical potentials were also
estimated for 1, taken as a model for 1−3 that are characterized
by similar spectroscopic and electrochemical features. These
quantities are easily calculated from the ground-state electro-
E
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
chemical potentials and the spectroscopic energy (E°°, in eV
by ∼0.3 eV. This stabilization supports the higher oxidation
potentials measured for 1−4 (around +1.15 V, Table 2)
62
units) related to the involved transition. The resulting formal
+
64
potentials for the MLCT excited-state couples of 1 (vs Fc/Fc )
compared with A1 (+0.84 V). For complex 5, the HOMO,
+
−
are +0.965 ([Ir(ptrz) (bpy)] * + e ⇌ [Ir(ptrz) (bpy)]) and
which is also localized on the Ir-phenyl environment, is further
stabilized (∼0.4 eV) by the electron-withdrawing effect of the
tert-butyl isocyanide ligands. This additional stabilization
explains the anodic shift (∼0.4 V) measured for the oxidation
potential in passing from 1−4 to 5 (Table 2).
2
2
2
+
−
+
−
1.594 V ([Ir(ptrz) (bpy)] + e ⇌ [Ir(ptrz) (bpy)] *),
2
2
respectively. Thus, the excited state of complex 1 ([Ir-
+
(
ptrz) (bpy)] *) is a moderately strong oxidant as well as a
2
very strong reductant. The latter finding is in line with many
other Ir(III) complexes having similar triplet energy, whereas
the excited state oxidative power of 1 is particularly high when
systematically compared to Ir(III) complexes previously
−
The cyclometalating ptrz ligands have no special effect on
the LUMO of complexes 1−4 because this orbital is located on
the ancillary ligand (see Figure 3). The LUMO appears at very
similar energies for 1 (−2.43 eV), 3 (−2.40 eV), and A1 (−2.42
eV), justifying the almost identical values recorded for the first
reduction potential of these complexes (−1.79, −1.77, and
1
6
reported.
Theoretical Calculations: Electronic Ground State. The
molecular and electronic structures of complexes 1−5 were
investigated by density functional theory (DFT) calculations at
the B3LYP/(6-31G**+LANL2DZ) level considering solvent
effects (see the Supporting Information for full computational
details). The geometries of the complexes in the electronic
63
−1.77 V, respectively). For 2, the LUMO (−2.25 eV) is
slightly destabilized by the electron-donating effect of the tert-
butyl substituents in accord with the more cathodic reduction
potential measured for this complex (−1.87 V). Substitution of
a pyridine ring by a triazole ring in complex 4 causes a LUMO
destabilization of 0.40 eV with respect to 1 (Figure 3), in
accord with the cathodic shift of 0.44 V measured
experimentally. The lack of low-energy π* virtual orbitals in
the isocyanide ligands determines that the LUMO of complex 5
relies on the tetrazole rings of the cyclometalating ligands and
appears at high energies (−1.62 eV) compared to 1 (−2.43
eV). This justifies the cathodic shift of 0.72 V observed for the
first reduction potential in passing from 1 to 5 (Table 2).
Theoretical calculations therefore predict that the HOMO−
LUMO energy gap increases along the series A1 (3.22 eV) < 1
ground state (S ) were fully optimized, and the values obtained
0
for the bond lengths and bond angles defining the coordination
sphere of the iridium center are collected in Table S2 in the
Supporting Information. Both the ancillary N^N ligand and the
−
cyclometaling ptrz ligands remain mainly planar and define a
near-octahedral coordination for the Ir metal. The maximum
t
deviation from planarity (1.32°) is found for the Bu-bpy ligand
−
in complex 2. The bite angles formed by the ptrz ligands with
the Ir center are always around 79.3°, in good agreement with
the X-ray values collected in Table 1, and are similar to those
previously reported for iridium(III) complexes bearing 1-
63
phenylpyrazole (Hppz) ligands.
(
3.50 eV) ≈ 3 (3.53 eV) < 2 (3.65 eV) < 4 (3.90 eV) < 5 (4.75
Figure 3 displays the energy and the atomic orbital
composition calculated for the HOMO and LUMO of
eV). This series is in perfect agreement with the electro-
chemical gaps inferred from CV measurements (2.61, 2.94,
2
.95, 3.00, 3.48, and 4.07 V, respectively). A noticeable blue
shift is therefore expected in both absorption and emission
properties in passing from 1−3 to 4 and especially to 5.
Photophysical Properties in Solution. Absorption
Spectra. The RT absorption spectra of complexes 1−5 in
acetonitrile are reported in Figure 4; spectra in dichloro-
methane are depicted in Figure S2 in the Supporting
Information. All complexes display high stability in both
solvents for several weeks, under standard laboratory
conditions.
The 200−275 nm spectral window is dominated by intense
4
−1
−1
absorption bands (ε ≈ (2−7) × 10 M cm ) that are
Figure 3. Energy diagram showing the isovalue contours (±0.03 au)
and the energy values calculated for the HOMO and LUMO of
complexes A1, 1, 4, and 5. Hydrogen atoms are omitted.
complexes 1, 4, and 5, together with those computed for the
+
archetypical complex [Ir(ppy) (bpy)] (A1) for comparison
2
purposes. Complexes 2 and 3 present identical molecular
orbital topologies to those depicted in Figure 3 for 1. As found
for A1, the HOMO of 1−4 is mainly localized on the Ir center
and the phenyl rings of the cyclometalating ligands, the main
Figure 4. Absorption spectra of complexes 1−5 in acetonitrile at 298
K: 1 (black), 2 (orange), 3 (red), 4 (green), and 5 (blue). (inset) The
direct transition S → T is magnified.
−
effect of the ptrz ligands being the stabilization of the HOMO
0
1
F
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
assigned to spin-allowed ligand-centered (LC) π−π* transitions
involving both the cyclometalating and the ancillary ligands. On
the other hand, the weaker and broader bands at longer
3
−1
−1
wavelengths (275−375 nm; ε ≈ (1−15) × 10 M cm ) are
attributed to charge transfer transitions with mixed metal-to-
ligand and ligand-to-ligand charge transfer (MLCT/LLCT)
1
character. In the case of 5, the absorption band at 325−350
nm is absent because isocyanide ancillary ligands do not have
low-energy π* acceptor orbitals allowing MLCT transitions
from the iridium metal center; this is a common feature in
2
2,23,26
cationic Ir(III) complexes with isocyanides.
The weak and long tail observed in the spectra above 375 nm
inset, Figure 4) is due to direct spin-forbidden absorption
(
from the singlet ground state (S ) to the first triplet excited
0
state (T ) of the complexes, enabled by the high spin−orbit
1
−1
Figure 6. Normalized emission spectra of 1, 2, 3, and 5 in acetonitrile
solution at 298 K. 1−3 exhibit a broad and unstructured emission
band underpinning MLCT character, while 5 shows an LC-type
emission profile, typical of isocyanide-based iridium(III) complexes.
coupling constant of the iridium metal core (ζ = 3909 cm ).
Ir
2
−1
−1
This band is stronger for 1−3 (ε ≈ (5−10) × 10 M cm )
where T is expected to have a marked MLCT character, but is
1
1
much less pronounced for 4. On the other hand, it is
completely absent in 5, in agreement with the fact that
isocyanide-based Ir(III) complexes lack MLCT transitions and,
accordingly, display only strong ligand-centered (LC) lumines-
All the complexes of the tetrazole-based series display blue-
green emission with maxima between 440 and 540 nm, that is, a
range similar to polyfluorinated A2, B2, and C2 (see Chart 2).
2
2,23,26
cence.
It is noteworthy that the phenyl-tetrazole cyclometalating
ligands are capable of blue shifting not only the MLCT
−
Indeed, the strong-field cyclometalating ptrz ligand is able to
blue shift the T emission even more strongly than the pristine
1
absorption transitions, but also the direct S → T band. This
−
0
1
ppz ligand is able to.
can be observed in Figure 5, where the absorption spectra of
In line with the absorption spectra, 1−3 show similar
emission properties with broad and structureless bands typical
of a triplet T state with strong MLCT character, while 5
1
exhibits a much bluer and more vibronically structured
emission, clearly indicating an emitting triplet state of
predominant LC character. Peculiarly, 4 is a very weak emitter,
and the reasons for this behavior will be rationalized in the next
section, with the support of quantum-chemical calculations.
The different nature of the emitting states is also crucially
influencing the photoluminescence quantum yield (PLQY) of
this family of complexes. Complexes 1−3 show very high
PLQYs (up to 70%), whereas 5 is almost nonemissive (Φ
=
em
0
.5%). To the best of our knowledge, the PLQYs of 1−3 in RT
solutions are the highest ever reported for fluorine-free cationic
iridium(III) complexes having a bipyridine-type ancillary ligand.
In fact, these PLQY values are only comparable to some
65
polyfluorinated complexes (e.g., A2 and C2 that exhibit
Figure 5. Absorption spectra in dichloromethane at 298 K of some
archetypal Ir(III) complexes having the same ancillary ligand (2,2′-
bypiridine) but different cyclometalating ligands: 2-methyl-5-phenyl-
4
1
34
PLQYs of 55% and 73%, respectively). Complexes 1−3
exhibit radiative constants similar to many others Ir-iTMCs (k
r
5
−1
2H-tetrazole (1, full), 2-phenylpyridine (A1, dashed), and 2-(2,4-
≈ 4 × 10 s ), so the high PLQY values are primarily
difluorophenyl)pyridine (A2, dotted). (inset) The S → T transition
0
1
attributable to the lack of effective nonradiative pathways. This
is magnified to allow easy comparison.
picture drastically changes if the T state does not have a
1
pronounced MLCT character, as in the case of 5 (vide infra).
For this complex, in which the lowest electronic state has an LC
character, the radiative constant drastically drops by a factor of
two archetypal Ir-iTMCs widely used as emitters in LECs,
+
+
namely, [Ir(ppy) (bpy)] (A1) and [Ir(dfppy) (bpy)] (A2),
2
2
3
−1
are compared to 1; all complexes have the same ancillary ligand,
100 (k ≈ 6 × 10 s ) and makes 5 barely emissive. The
r
but different cyclometalating units. The effect on the S → T1
photophysical properties of 5 are very similar to those of
0
t
+
26
transition is particularly remarkable with respect to A2, because
no electron-withdrawing fluorine substituents are present on
the tetrazolate complex 1.
Emission Properties. The emission spectra of 1, 2, 3, and 5
at 298 K in acetonitrile are reported in Figure 6; spectra in
dichloromethane are depicted in Figure S3 in the Supporting
Information. The luminescence properties and photophysical
parameters of 1−5 are summarized in Table 3; emission
maxima can be compared to those of some selected archetypal
complexes (A−C) gathered in Chart 2.
[Ir(ppz) (CN Bu) ] that have been reported recently.
2 2
Therefore, it can be concluded that 5-phenyltetrazole (ptrz),
1-phenylpyrazole (ppz), and phenyl-[1,2,3]triazole (phtl)
cyclometalating ligands act in a very similar way in cationic
iridium(III) complexes bearing high-field ancillary ligands, that
1
9,26,33
is, they exhibit faint luminescence.
A totally different
t
+
behavior is observed for [Ir(ppy) (CN Bu) ] and analogous 2-
2
2
phenylpyridine complexes, where, in spite of rather small k
r
3
−1
values ((2−18) × 10 s ), very high PLQYs (∼30−70%) were
2
2,23
found.
G
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Luminescence Properties and Photophysical Parameters of 1−5 in Solution at 298 and 77 K
7
7 K rigid matrix,
CH CN oxygen-free solution, 298 K
CH Cl oxygen-free solution, 298 K
CH CN
3
2
2
3
a
a
b
(ns) kr (10 s−1) knr (10 s−1) λem (nm) Φem (%)
c
5
d
5
e
e
b
(ns) kr (10 s−1) knr (10 s−1)
c
5
d
5
a
b
λem (nm) Φem (%)
τ
τ
λem (nm)
τ
(μs)
3.7
1
2
3
545
530
540
54.8
67.6
70.3
1220
1181
1665
4.49
5.72
4.22
3.70
2.74
1.78
532
514
520
65.2
60.9
68.9
1267
1077
1547
5.15
5.65
4.45
2.75
3.63
2.01
450, 482, 505
442, 466, 500
462, 495, 530,
3.6
11.2
5
70
4
5
524
442
<0.1
0.5
f
f
f
524
442
∼0.1
0.5
f
f
f
410, 436, 462
396, 422, 446
7.5
873
0.06
11.4
787
0.06
12.6
106
a
b
c
d
e
f
λexc = 360 nm. λ = 407 nm. Radiative constant: k = Φ /τ. Nonradiative constant: k = 1/τ − k . λ = 375 nm. Data not reported due to
exc
r
em
nr
r
exc
very low emission.
The most puzzling case is 4, which bears a pyridine-triazole
ancillary ligand. This complex exhibits a broad emission band
complex 4, for which T1 is predicted to have the same
electronic nature as 1−3. To investigate this discrepancy, the
lowest triplet excited states of complexes 1−5 were further
examined by optimizing their geometry using the spin-
unrestricted UB3LYP approach.
similar to 1−3, indicative of a T level with a largely
1
predominant MLCT character. However, the emission
quantum yields in solution are extremely low (Φ < 0.1%,
em
Table 3), that is, even smaller than 5.
The relative energy position and unpaired-electron spin
density distributions calculated for the lowest-energy triplets of
complexes 1 and 4 are depicted in Figure 7. After full-geometry
relaxation, the T state is computed to lie 2.63 eV above S for
Theoretical Calculations: Triplet Excited States. In an
attempt to rationalize the absorption and emission properties,
the lowest triplet excited states of complexes 1−5 were first
1
0
calculated at the optimized geometry of the ground state (S )
1 and 3.03 eV above S for 4 (adiabatic energy differences,
0
0
using the time-dependent DFT (TD-DFT) approach. Table S3
in the Supporting Information lists the vertical excitation
energies and the electronic descriptions computed for the three
ΔE(T − S ) in Figure 7). The spin-density distribution
1
0
calculated for both 1 (Ir: 0.49e, C^N ligands: 0.48e, N^N
ligand: 1.03e) and 4 (Ir: 0.47e, C^N ligands: 0.50e, N^N
ligand: 1.03e) perfectly matches the topology of the HOMO →
LUMO excitation (Figure 3) and confirms the electron transfer
from the Ir-ptrz environment to the ancillary ligand, that is, the
first triplet excited states. The lowest-energy triplet state (T ) of
1
complexes 1, 2, and 4 mainly results from the HOMO →
LUMO excitation that implies an electron transfer from the Ir-
ptrz environment, where the HOMO is localized, to the bpy
3
3
MLCT/ LLCT character of T . The T state corresponds in
1
2
3
ligand, where the LUMO resides (see Figure 3). The T state
both cases to a LC state. For 1, the T state is localized over
̂
the ancillary NN ligand and is calculated 0.21 eV above T . By
1
1
2
3
3
therefore shows a mixed MLCT/ LLCT character. For
complex 3, the T state is more multiconfigurational, but it
also shows a MLCT/ LLCT character, and the HOMO →
contrast, in 4, T is centered on one of the C^N ligands and is
1
2
3
3
mostly degenerate with T (ΔE(T − T ) = 0.01 eV).
1
2
1
LUMO excitation gives rise to the T state that is very close in
The most important difference between 1 and 4 is that for
2
3
energy to T . TD-DFT calculations predict the T state at 2.79
the latter a metal-centered ( MC) state, in which the Ir center
1
1
eV (444 nm) for 1 and 3 and at 2.95 eV (421 nm) for 2, in
accumulates a large part of the unpaired electrons (Ir: 1.23e,
good agreement with the very weak bands observed in the
C^N ligands: 0.65e, N^N ligand: 0.13e), appears 0.21 eV below
3
4
00−450 nm range of the absorption spectra (Figure 4, inset).
T (Figure 7b). This MC state implies the decoordination of
1
For complex 4, the T state is calculated at higher energies
the pyridine ring that twists around the inter-ring bond and
1
(
3.19 eV, 389 nm) due to the destabilization of the LUMO
moves away from the iridium center (Ir−N distance = 3.4 Å,
pyr
caused by the substitution of a pyridine ring by a triazole ring.
This shift to lower wavelengths implies that the associated weak
absorption band is hidden under the tail of the singlet
absorption centered around 350 nm and is not observed in the
experimental spectrum (Figure 4). Finally, for 5, there is no
Table S2 in the Supporting Information). A similar behavior
was found for complexes incorporating pyridine-carbene
6
6,67
ancillary ligands,
with formation of a pentacoordinated
iridium center. For complex 1, all of the attempts made to
3
localize the MC state invariably led to a related description
low-energy MLCT triplet and, as expected, the T state has a
LC character involving the cyclometalating ligands.
Calculations also rationalize the blue shift observed for the
implying the nitrogen atoms of the bpy ligand moving away
from the Ir center (2.5 Å) but still coordinated. The resulting
MC state (Ir: 1.17e, C^N ligands: 0.64e, N^N ligand: 0.18e) is
1
3
3
lowest-energy absorption band of 1 when compared with the
calculated 0.33 eV above T (Figure 7a). The larger structural
distortion found in the MC state of 4 is due to the asymmetric
nature of the taz N^N ligand that favors the preferred
1
+
3
archetype complexes [Ir(ppy) (bpy)] (A1) and [Ir-
2
+
(
(
dfppy) (bpy)] (A2), which bear the same ancillary ligand
Figure 5). The ptrz ligand brings about a shift of the S → T
2
−
decoordination of the pyridine ring. The Ir−N distances
0
1
transition to higher energies for 1 (2.79 eV, 444 nm) compared
calculated for the S state of 4 (Ir−N = 2.235 Å, Ir−N
=
0
pyr
triazol
to A1 (2.50 eV, 491 nm), which is even larger than that caused
2.175 Å, Supporting Information, Table S2) actually indicate
that the pyridine ring is less strongly attached to the iridium
center than the triazol ring. In contrast, in complex 1 the two
nitrogen atoms of the N^N ligand (bpy) are equally
coordinated to the iridium center, and they prefer to move
−
by the difluorinated dfppy ligand in A2 (2.74 eV, 451 nm).
Concerning the emission properties, TD-DFT calculations
support the efficient and unstructured emission observed for
complexes 1−3 based on the MLCT/LLCT nature of the T1
state and the blue-shifted structured emission recorded for
3
away from Ir in a synchronous way in the MC state. Therefore,
complex 5 due to the LC nature of T . However, they do not
provide an explanation for the poor emission exhibited by
emission in 1 originates from the lowest-energy HOMO →
1
3
3
LUMO T state of the MLCT/ LLCT character. The emission
1
H
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
unstructured emission band of 4 likely stems from the
3
3
MLCT/ LLCT state T . To further support the different
1
energy ordering found for the excited states in complexes 1 and
, the geometry of the triplet states was reoptimized at the
4
more accurate TD-DFT level. TD-DFT calculations predict
that the MC state in complex 4 lies lower in energy than T by
3
1
0
.37 eV, while for complex 1 it resides 0.38 eV above the
emitting T state, thus confirming the ordering shown in Figure
1
7
.
Low-Temperature and Thin-Film Photophysical Prop-
erties. To get a wider understanding of the photophysical
properties of complexes 1−5, we performed luminescence
experiments at 77 K. Spectral data and lifetimes are summarized
in Table 3, and emission bands are collected in Figure 8. In
Figure 8. Normalized 77 K emission spectra of complexes 1−5
recorded in frozen acetonitrile matrix.
rigid matrix at 77 K, all complexes exhibit bright emissions of
comparable intensity when excited at the same experimental
conditions. Therefore, also 4 and 5, which are very poor
emitters at 298 K, show intense luminescence at low
temperature.
2
1,28
In line with previous findings on Ir(III) complexes,
1 and
Figure 7. Schematic energy diagram showing the adiabatic energy
2
(T state with a strong MLCT character) exhibit more
differences (ΔE) between S and the lowest triplet states and the
1
0
emission energy (E ) from T calculated for (a) 1 and (b) 4. The
structured emission profiles and undergo a considerable blue
shift (around 50 nm, passing from 298 to 77 K) due to
rigidochromic effects. On the other hand, 5 displays an
emission spectrum that is not temperature-dependent, further
em
1
unpaired-electron spin-density contours (0.003 au) calculated for the
fully relaxed triplet states are depicted together with the electronic
nature of the states.
corroborating the LC nature of its T state, with emission
1
energy of T (528 nm), calculated as the vertical energy
mainly centered on the cyclometalating ligand (Figure 9a).
A dramatic change in the spectral shape (i.e., electronic
nature) and emission intensity from 298 to 77 K is observed for
1
difference between T and S at the optimized minimum-energy
1
0
geometry of T (E in Figure 7a), is in good agreement with
1
em
3
experiment (545 nm in CH CN and 532 nm in CH Cl ). For 2
ligand 4. At 77 K, population of the MC state is prevented
3
2
2
and 3 the triplet state distribution is similar to that obtained for
because the frozen medium inhibits the rotation of the pyridine
ring required for the relaxation of this state (see above).
3
3
1
, and emission also takes place from the MLCT/ LLCT T
1
state.
Notably, the energy distribution of the triplet states
calculated for 4 is completely different. The T
1
3
3
3
(
MLCT/ LLCT) and T ( LC centered on the C^N ligands)
2
3
states are almost degenerate, and a lower-energy MC state is
present, which opens a deactivation pathway to S . At its
0
3
minimum-energy geometry, the MC state is indeed calculated
to be only 0.33 eV above S , which facilitates the nonemissive
0
3
deactivation of 4 once the MC state is populated. Theoretical
3
calculations therefore suggest that, at RT, the low-energy MC
state largely quenches the emission of complex 4, justifying the
low PLQYs experimentally measured. The observed weak and
Figure 9. Unpaired-electron spin-density contours (0.003 au)
calculated for (a) the T state of 5 and (b) the T state of 3.
1
2
I
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Therefore, at low temperature, emission from the almost
degenerate T and T states (Figure 7b) is not quenched by the
1
2
3
MC state. In addition, T and T split at low temperature, that
1
2
is, the polar MLCT/LLCT state moves to higher energies
similarly to 1 and 2) due the blocking of the solvent
(
reorganization at 77 K, whereas the apolar LC level (centered
on the cyclometalating ligands as in 5) remains substantially
unaffected. As a result, the 77 K emission band of 4 shows the
same complex vibronic progressions of 5, though it is located at
lower energy due to the smaller d−d orbital splitting of
pyridine-triazole with respect to isocyanide ligands (Supporting
Information, Figure S4). The different nature of the emitting
level of 4 versus 1 and 2 at 77 K (LC vs MLCT) is likely at the
origin of its longer excited-state lifetime (Table 3).
The longer lifetime of 3 at 77 K with respect to 1, 2, and 4
can be rationalized through the triplet state ordering predicted
for this phenanthroline-based complex. As discussed above, 3
shows the same triplet distribution depicted in Figure 7a for 1.
Figure 10. Normalized emission spectra of complexes 1−3 in the solid
state at 298 K. Full lines: in 1% PMMA matrix; dashed lines: in neat
film.
That is, T corresponds to the HOMO → LUMO
1
3
3
3
MLCT/ LLCT state, and T is a LC state centered on the
2
with respect to 1 and 2 in both media. The k value of 3 is
r
5
−1
phenanthroline ancillary ligand (Figure 9b); they are separated
by only 0.10 eV. In analogy with 4 (see above), at 77 K we can
assume that the LC state becomes the lowest-lying because of
the energy increase of the MLCT/LLCT state, due to
rigidochromic effects. These theoretical findings are corrobo-
rated by the 77 K emission profile of 3, which is practically
superimposable onto that of pristine phen under the same
conditions. Indeed, the spin density distribution depicted in
Figure 9b fully corresponds to that calculated for the first triplet
state of phen.
The new series of Ir(III) tetrazolate complexes was
ultimately designed and prepared to open a new route for
green/blue highly luminescent Ir emitters for light-emitting
electrochemical cells (LECs) without fluorine-containing
ligands. Therefore, we also investigated the emission properties
of 1−5 in the solid state, that is, dispersed in a poly(methyl
methacrylate) (PMMA) matrix at a concentration of 1% by
weight and as neat films. Table 4 and Figure 10 collect the
particularly low in PMMA (2.2 × 10 s ), as a consequence of
a longer lifetime, which might reflect a slightly different
character of the triplet level, with some LC character. This is in
line with 77 K phosphorescence data that show an unusually
long phosphorescence lifetime of 11.2 μs in CH CN at 77 K for
3
3 (Table 3). On the other hand, a comparison of highly dilute
samples with neat films is not appropriate due to the
intermolecular interactions occurring in the latter. However,
notice that 3 exhibits the smallest radiative rate constant also in
6
8
5
−1
5 −1
neat films, that is, 4.2 × 10 s , compared to ∼9 × 10 s for 1
and 2.
LEC Devices. As mentioned in the Introduction, it is very
challenging to obtain solid-state electrophosphorescent devices
at high energy, that is, emitting in the blue region. This is
especially the case for LECs, where the transition-metal
complex is responsible not only for the emission of light but
also for the transport of electrons and holes throughout the
film. We have selected complex 2, due to its superior PLQY in
neat film, to prepare LECs following the procedure described in
detail in the Supporting Information.
Table 4. Luminescence Properties of 1−3 at 298 K in the
Solid State
The electroluminescence spectra of such devices are very
similar to those obtained by photoexcitation in neat films with
an emission maximum at 552 nm. The devices were driven by a
pulsed current using a block wave at a frequency of 1 kHz with
a duty cycle of 50% and an average current density of 100 A
1% PMMA matrix
neat film
a
a
b
a
a
b
λem (nm) Φem (%)
τ
(μs) λem (nm) Φem (%)
τ
(ns)
537
1
2
3
490
485
497
69
51
57
1.1
1.0
2.6
520
527
535
45
70
27
−2
774
650
m . These operative conditions were previously shown to
69
enhance the performances of LECs. LECs built up with 2
−2
a
b
reach a maximum luminance of 310 cd m that drops within 2
h to 20 cd m− (Figure 11). The luminance observed is among
the best observed for blue-emitting LECs. The efficiency is low,
λexc = 310 nm. λ = 407 nm.
exc
2
1
experimental data obtained for complexes 1−3 only, which
exhibited PLQYs in the range of ∼30−70%, unlike 4 and 5,
which have PLQYs below 5%. The similar emission behavior of
and the stability is similar to many blue-emitting LECs.
Therefore, the devices operate with reasonable efficiency, but
the performance is rapidly decreasing, probably due to
undesired quenching of the excited state caused by a
recombination at one of the extremes of the light-emitting
layer. This can be inferred from the average voltage versus time
curve, which decreases quickly over the first minutes and
continues decreasing at longer time scales (Figure 11). It was
demonstrated that the initial decrease is caused by the
reduction of the injection barriers as ions accumulate at the
4
and 5 under these conditions is not surprising because in rigid
3
PMMA matrix the MC state of 4 cannot fully relax and is no
longer the lowest one. Therefore, in both cases, the lowest
3
excited state is of LC nature, which is typically poorly emissive
in the presence of a 5-membered heteroatomic ring on the
26
cyclometalated ligand. As far as the photophysical data in
Table 4 are concerned, notice that 1 and 2 exhibit similar values
5
70
of the radiative constant (k = Φ/τ) in PMMA (5 ÷ 6 × 10
interface with the electrodes.
r
−1
s ), which are also comparable to those in CH CN and
Once electrons and holes are injected in the light-emitting
material in the presence of uncompensated ions, some of the
3
CH Cl (Table 3). By contrast, the rate constant of 3 is lower
2
2
J
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 11. Luminance, average driving voltage, and efficacy for a LEC using 2 as phosphorescent material, which is driven by a pulsed current of 100
−2
A m (1000 Hz, 50% duty cycle, block wave). (inset) The electroluminescence spectrum of the device obtained under these driving conditions.
electronic charges get compensated by the ionic charges. The
compensation of excess electrons on the iridium complexes by
the uncompensated positive ions leads to the formation of
electrochemically doped layers. These doped layers are
beneficial for charge injection and transport, as the conductivity
of these zones is significantly higher than the intrinsic
this problem, we have designed a novel two-step synthetic
strategy based on a silver-assisted cyclometalation reaction with
IrCl that afforded tetrazolate complexes 1−5 with medium-to-
3
high yields (40−75%).
Our primary goal to blue shift the emission of cationic Ir(III)
complexes through the modification of the C^N ligand, while
keeping an electronic transition with predominant charge
transfer character, was fully accomplished with compounds 1−
3, which exhibit the highest-energy MLCT/LLCT emission
bands ever reported for fluorine-free cyclometalating ligands.
They display green-bluish photoluminescence, with quantum
yields in the range of 55−70% in acetonitrile at RT. Similar
PLQY values are found in solid matrices, making 1−3 suitable
for LEC device testing. The strongest emitter in neat films (2)
was tested and afforded good quality films with remarkable
(nondoped) layers. Hence, the remaining potential drops
over the nondoped region that is thinner than the original light-
emitting layer. This process is reflected in the continued slow
decrease of the average voltage versus time, as less voltage is
needed to maintain a current density over a thinning intrinsic
layer. An additional, less desirable effect of the doped layers is
their capability to quench excited states. Hence, with rapidly
expanding doped zones, the number of quenched excited states
is increased, leading to the reduction of the luminance and,
consequently, of the efficacy. Hence, to summarize the
performance in electroluminescent devices, complex 2 is able
to simultaneously transport electrons and holes and leads to
efficient luminescence with an emission spectrum similar to
that observed under photoexcitation. The observed low stability
is most likely related to unbalanced charge transport within the
−
2
maximum luminance (310 cd m ) but limited stability.
This paper discloses a novel synthetic strategy to an
unprecedented class of cationic iridium cyclometalated
complexes. Chemical modifications of the tetrazole ligands
will now offer the possibility to further push the emission band
toward the blue, presently one of the most important goals in
this area of investigation. Moreover, based on the deep
understanding of the electronic and photophysical properties
of 1−5 acquired through experimental and theoretical methods,
we now aim to design cationic Ir-tetrazolate complexes with
tailored features for easy device preparation and enhanced
electroluminescence performance.
1
device.
CONCLUSION
■
The first cationic Ir(III) complexes with tetrazole units on the
C^N cyclometallating ligand (1−5, [Ir(ptrz) (L^L)] , Hptrz =
-methyl-5-phenyl-2H-tetrazole) have been synthesized and
+
2
2
fully characterized. The presence of four nitrogen atoms on the
five-membered ring makes tetrazole a stronger electron donor
ASSOCIATED CONTENT
■
compared to its analogues pyrazole and triazole. Accordingly,
*
S
Supporting Information
+
stabilization of the HOMO of the [Ir(ptrz) (L^L)] complexes
Full experimental details of the electrochemical, theoretical,
2
1
13
is attained, enhancing the ligand field and pushing the emission
to the blue. This strategy had proved effective in the past, with
blue shift of the luminescence by passing from complexes
photophysical and device part; H NMR, C NMR, and mass
spectra (ESI) of all complexes; crystal data for P, 2 and 3; cyclic
voltammograms of 1 recorded at different scan rates; DFT and
TD-DFT data regarding bond distances and angles (S , T , and
28
containing 2-phenylpyridine to 1-phenylpyrazole. Indeed,
0
1
0/+
3
[
Ir(C^N) (L^L)] complexes with triazole units on the C^N
MC) and triplet vertical excitations; absorption and emission
2
ligands were also made but, surprisingly, they proved
unsuccessful in promoting further emission to blue shift
relative to pyrazole counterparts.
spectra in dichloromethane solution. This material is available
free of charge via the Internet at http://pubs.acs.org.
34
In this context, the ultimate step to increase the nitrogen
atoms on the C^N ligand, that is, the use of 5-phenyltetrazoles,
had remained unexplored, because the classic protocol to
prepare cyclometalated Ir(III) complexes, based on the
preliminary formation of the μ-dichloro-bridged dimer
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: enrique.orti@uv.es. (E.O.)
*E-mail: letizia.sambri@unibo.it. (L.S.)
*E-mail: nicola.armaroli@isof.cnr.it. (N.A.)
[
(C^N) Ir(μ-Cl)] , is ineffective with tetrazoles. To address
2 2
K
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Notes
Slinker, J. D.; De Cola, L.; Zysman-Colman, E. J. Mater. Chem. C 2013,
1
, 7440.
25) Yang, C.-H.; Mauro, M.; Polo, F.; Watanabe, S.; Muenster, I.;
Frohlich, R.; De Cola, L. Chem. Mater. 2012, 24, 3684.
26) Shavaleev, N. M.; Monti, F.; Scopelliti, R.; Baschieri, A.; Sambri,
L.; Armaroli, N.; Gratzel, M.; Nazeeruddin, M. K. Organometallics
013, 32, 460.
27) Grushin, V. V.; Herron, N.; LeCloux, D. D.; Marshall, W. J.;
Petrov, V. A.; Wang, Y. Chem. Commun. 2001, 1494.
28) Tamayo, A. B.; Garon, S.; Sajoto, T.; Djurovich, P. I.; Tsyba, I.
M.; Bau, R.; Thompson, M. E. Inorg. Chem. 2005, 44, 8723.
29) Ho, M.-L.; Hwang, F.-M.; Chen, P.-N.; Hu, Y.-H.; Cheng, Y.-M.;
The authors declare no competing financial interest.
(
̈
ACKNOWLEDGMENTS
(
■
(
̈
This work was supported by the Italian Ministry of Research
MIUR) (PRIN 2010 INFOCHEM, Contract No. CX2TLM;
2
(
FIRB Futuro in Ricerca SUPRACARBON, Contract No.
RBFR10DAK6), UNIBO-RFO, the Consiglio Nazionale delle
Ricerche (CNR) (MACOL PM. P04. 010, and Progetto
Bandiera N−CHEM), the Spanish Ministry of Economy and
Competitiveness (MINECO) (MAT2011-24594 and
CTQ2012-31914), the Generalitat Valenciana (PROMETEO/
(
(
Chen, K.-S.; Lee, G.-H.; Chi, Y.; Chou, P.-T. Org. Biomol. Chem. 2006,
4, 98.
(30) He, L.; Duan, L.; Qiao, J.; Dong, G.; Wang, L.; Qiu, Y. Chem.
Mater. 2010, 22, 3535.
2012/053), and European FEDER funds (CTQ2012-31914).
A.P. acknowledges the MINECO for an FPI grant.
(31) Baranoff, E.; Fantacci, S.; De Angelis, F.; Zhang, X.; Scopelliti,
R.; Gratzel, M.; Nazeeruddin, M. K. Inorg. Chem. 2011, 50, 451.
(32) Lai, W.-Y.; Levell, J. W.; Jackson, A. C.; Lo, S.-C.; Bernhardt, P.
̈
REFERENCES
■
(
1) Costa, R. D.; Ortí, E.; Bolink, H. J.; Monti, F.; Accorsi, G.;
V.; Samuel, I. D. W.; Burn, P. L. Macromolecules 2010, 43, 6986.
(33) Beyer, B.; Ulbricht, C.; Escudero, D.; Friebe, C.; Winter, A.;
Armaroli, N. Angew. Chem., Int. Ed. 2012, 51, 8178.
(
(
2) Chou, P. T.; Chi, Y. Chem.Eur. J. 2007, 13, 380.
Gonzalez, L.; Schubert, U. S. Organometallics 2009, 28, 5478.
(34) Ladouceur, S.; Fortin, D.; Zysman-Colman, E. Inorg. Chem.
2011, 50, 11514.
3) Evans, R. C.; Douglas, P.; Winscom, C. J. Coord. Chem. Rev. 2006,
2
(
2
50, 2093.
4) Gather, M. C.; Ko
33.
hnen, A.; Meerholz, K. Adv. Mater. 2011, 23,
(35) Fernan
V.; Polo, F.; Fro
2011, 133, 10543.
́
dez-Hernan
́ ́
dez, J. M.; Yang, C.-H.; Beltran, J. I.; Lemaur,
̈
̈
hlich, R.; Cornil, J.; De Cola, L. J. Am. Chem. Soc.
(
(
(
5) Hu, T.; He, L.; Duan, L.; Qiu, Y. J. Mater. Chem. 2012, 4206.
6) Lowry, M. S.; Bernhard, S. Chem.Eur. J. 2006, 12, 7970.
7) Wang, J.; Zhang, F.; Zhang, J.; Tang, W.; Tang, A.; Peng, H.; Xu,
(36) de Barros e Silva Botelho, M.; Fernandez-Hernandez, J. M.; de
Queiroz, T. B.; Eckert, H.; De Cola, L.; de Camargo, A. S. S. J. Mater.
Chem. 2011, 21, 8829.
Z.; Teng, F.; Wang, Y. J. Photochem. Photobiol. C 2013, 17, 69.
8) Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletti, F.
Top. Curr. Chem. 2007, 281, 143.
9) Slinker, J.; Bernards, D.; Houston, P. L.; Abrun
S.; Malliaras, G. G. Chem. Commun. 2003, 2392.
10) Slinker, J. D.; Koh, C. Y.; Malliaras, G. G.; Lowry, M. S.;
Bernhard, S. Appl. Phys. Lett. 2005, 86.
11) Slinker, J. D.; Gorodetsky, A. A.; Lowry, M. S.; Wang, J. J.;
Parker, S.; Rohl, R.; Bernhard, S.; Malliaras, G. G. J. Am. Chem. Soc.
004, 126, 2763.
12) Xiao, L.; Chen, Z.; Qu, B.; Luo, J.; Kong, S.; Gong, Q.; Kido, J.
Adv. Mater. 2011, 23, 926.
13) Organic Light-Emitting Diodes (OLEDs). Materials, Devices and
Applications; Buckley, A., Ed.; Woodhead Publishing: Cambridge, U.K.,
013.
14) Li, J.; Djurovich, P. I.; Alleyne, B. D.; Yousufuddin, M.; Ho, N.
N.; Thomas, J. C.; Peters, J. C.; Bau, R.; Thompson, M. E. Inorg. Chem.
005, 44, 1713.
15) Rausch, A. F.; Thompson, M. E.; Yersin, H. J. Phys. Chem. A
009, 113, 5927.
16) Ladouceur, S.; Zysman-Colman, E. Eur. J. Inorg. Chem. 2013,
985.
17) Stagni, S.; Colella, S.; Palazzi, A.; Valenti, G.; Zacchini, S.;
(
(37) Fernan
Lopez, M.-D.; Chien, C.-H.; Polo, F.; Orselli, E.; Fro
J.; De Cola, L. Inorg. Chem. 2013, 52, 1812.
(38) Donato, L.; Abel, P.; Zysman-Colman, E. Dalton Trans. 2013,
42, 8402.
(39) Kiplinger, J. L.; Richmond, T. G.; Osterberg, C. E. Chem. Rev.
1994, 94, 373.
(40) Sivasubramaniam, V.; Brodkorb, F.; Hanning, S.; Loebl, H. P.;
van Elsbergen, V.; Boerner, H.; Scherf, U.; Kreyenschmidt, M. J.
Fluorine Chem. 2009, 130, 640.
(41) Constable, E. C.; Neuburger, M.; Rosel, P.; Schneider, G. E.;
Zampese, J. A.; Housecroft, C. E.; Monti, F.; Armaroli, N.; Costa, R.
D.; Ortí, E. Inorg. Chem. 2013, 52, 885.
(42) Thomas, E. W.; Cudahy, M. M. J. Org. Chem. 1993, 58, 1623.
(43) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.;
Kwong, R.; Tsyba, I.; Bortz, M.; Mui, B.; Bau, R.; Thompson, M. E.
Inorg. Chem. 2001, 40, 1704.
(44) Tamayo, A. B.; Alleyne, B. D.; Djurovich, P. I.; Lamansky, S.;
Tsyba, I.; Ho, N. N.; Bau, R.; Thompson, M. E. J. Am. Chem. Soc.
2003, 125, 7377.
(45) Nonoyama, M. Bull. Chem. Soc. Jpn. 1974, 47, 767.
(46) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.; Lee,
H. E.; Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J.
Am. Chem. Soc. 2001, 123, 4304.
́
dez-Hernan
́
dez, J. M.; Beltran
́
, J. I.; Lemaur, V.; Gal
́
vez-
́
̈
hlich, R.; Cornil,
(
̃
a, H. D.; Bernhard,
(
(
2
(
(
2
(
2
(
2
(
2
(
Paolucci, F.; Marcaccio, M.; Albuquerque, R. Q.; De Cola, L. Inorg.
Chem. 2008, 47, 10509.
(
18) Yang, C.-H.; Li, S.-W.; Chi, Y.; Cheng, Y.-M.; Yeh, Y.-S.; Chou,
P.-T.; Lee, G.-H.; Wang, C.-H.; Shu, C.-F. Inorg. Chem. 2005, 44, 7770.
19) Sajoto, T.; Djurovich, P. I.; Tamayo, A. B.; Oxgaard, J.;
Goddard, W. A.; Thompson, M. E. J. Am. Chem. Soc. 2009, 131, 9813.
20) Kozhevnikov, V. N.; Dahms, K.; Bryce, M. R. J. Org. Chem.
011, 76, 5143.
21) Costa, R. D.; Monti, F.; Accorsi, G.; Barbieri, A.; Bolink, H. J.;
Ortí, E.; Armaroli, N. Inorg. Chem. 2011, 50, 7229.
22) Shavaleev, N. M.; Monti, F.; Costa, R. D.; Scopelliti, R.; Bolink,
H. J.; Ortí, E.; Accorsi, G.; Armaroli, N.; Baranoff, E.; Gratzel, M.;
Nazeeruddin, M. K. Inorg. Chem. 2012, 51, 2263.
23) Shavaleev, N. M.; Monti, F.; Scopelliti, R.; Armaroli, N.; Grat
M.; Nazeeruddin, M. K. Organometallics 2012, 31, 6288.
24) Fernandez-Hernandez, J. M.; Ladouceur, S.; Shen, Y.; Iordache,
A.; Wang, X.; Donato, L.; Gallagher-Duval, S.; de Anda Villa, M.;
(47) Henderson, W. Adv. Organomet. Chem. 2006, 54, 207.
(48) Butler, R. N. Comprehensive Heterocyclic Chemistry II; Storr, R.
C., Ed.; Pergamon Press: Oxford, U.K., 1996; Vol. 4, p 621.
(49) Stagni, S.; Palazzi, A.; Zacchini, S.; Ballarin, B.; Bruno, C.;
Marcaccio, M.; Paolucci, F.; Monari, M.; Carano, M.; Bard, A. J. Inorg.
Chem. 2006, 45, 695.
(
(
2
(
(50) Zhao, N.; Wu, Y.-H.; Shi, L.-X.; Lin, Q.-P.; Chen, Z.-N. Dalton
Trans. 2010, 39, 8288.
(
(51) Lee, P.-K.; Liu, H.-W.; Yiu, S.-M.; Louie, M.-W.; Lo, K. K.-W.
Dalton Trans. 2011, 40, 2180.
(52) Bandini, M.; Bianchi, M.; Valenti, G.; Piccinelli, F.; Paolucci, F.;
Monari, M.; Umani-Ronchi, A.; Marcaccio, M. Inorg. Chem. 2010, 49,
1439.
̈
(
̈
zel,
(
(53) Ladouceur, S.; Fortin, D.; Zysman-Colman, E. Inorg. Chem.
2010, 49, 5625.
L
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
54) Shavaleev, N. M.; Scopelliti, R.; Gratzel, M.; Nazeeruddin, M.
K.; Pertegas, A.; Roldan-Carmona, C.; Tordera, D.; Bolink, H. J. J.
Mater. Chem. C 2013, 1, 2241.
(
55) Routledge, J. D.; Hallett, A. J.; Platts, J. A.; Horton, P. N.; Coles,
S. J.; Pope, S. J. A. Eur. J. Inorg. Chem. 2012, 2012, 4065.
56) Costa, R. D.; Ortí, E.; Bolink, H. J.; Graber, S.; Housecroft, C.
E.; Neuburger, M.; Schaffner, S.; Constable, E. C. Chem. Commun.
009, 2029.
57) Hanss, D.; Freys, J. C.; Bernardinelli, G.; Wenger, O. S. Eur. J.
Inorg. Chem. 2009, 2009, 4850.
58) Helms, M.; Lin, Z.; Gong, L.; Harms, K.; Meggers, E. Eur. J.
Inorg. Chem. 2013, 2013, 4164.
59) Su, H.-C.; Chen, H.-F.; Fang, F.-C.; Liu, C.-C.; Wu, C.-C.;
Wong, K.-T.; Liu, Y.-H.; Peng, S.-M. J. Am. Chem. Soc. 2008, 130,
413.
60) Felici, M.; Contreras-Carballada, P.; Vida, Y.; Smits, J. M. M.;
(
2
(
(
(
3
(
Nolte, R. J. M.; De Cola, L.; Williams, R. M.; Feiters, M. C. Chem.
Eur. J. 2009, 15, 13124.
(
61) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.;
von Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85.
(
(
62) Armaroli, N. Chem. Soc. Rev. 2001, 30, 113.
́
63) Tordera, D.; Bunzli, A. M.; Pertegas, A.; Junquera-Hernandez, J.
̈
́
M.; Constable, E. C.; Zampese, J. A.; Housecroft, C. E.; Ortí, E.;
Bolink, H. J. Chem.Eur. J. 2013, 19, 8597.
(
́
64) Costa, R. D.; Ortí, E.; Tordera, D.; Pertegas, A.; Bolink, H. J.;
Graber, S.; Housecroft, C. E.; Sachno, L.; Neuburger, M.; Constable,
E. C. Adv. Energy Mater. 2011, 1, 282.
(
65) Bolink, H. J.; Coronado, E.; Costa, R. D.; Lardies, N.; Ortí, E.
Inorg. Chem. 2008, 47, 9149.
66) Kessler, F.; Costa, R. D.; Di Censo, D.; Scopelliti, R.; Ortí, E.;
(
Bolink, H. J.; Meier, S.; Sarfert, W.; Gratzel, M.; Nazeeruddin, M. K.;
Baranoff, E. Dalton Trans. 2012, 41, 180.
(
67) Monti, F.; Kessler, F.; Delgado, M.; Frey, J.; Bazzanini, F.;
Accorsi, G.; Armaroli, N.; Bolink, H. J.; Ortí, E.; Scopelliti, R.;
Nazeeruddin, M. K.; Baranoff, E. Inorg. Chem. 2013, 52, 10292.
(
68) Armaroli, N.; De Cola, L.; Balzani, V.; Sauvage, J. P.; Dietrich-
Buchecker, C. O.; Kern, J. M. J. Chem. Soc., Faraday Trans. 1992, 88,
53.
69) Tordera, D.; Meier, S.; Lenes, M.; Costa, R. D.; Ortí, E.; Sarfert,
W.; Bolink, H. J. Adv. Mater. 2012, 24, 854.
70) Lenes, M.; Garcia-Belmonte, G.; Tordera, D.; Pertegas
Bisquert, J.; Bolink, H. J. Adv. Funct. Mater. 2011, 21, 1581.
5
(
(
́
, A.;
M
dx.doi.org/10.1021/ic500999k | Inorg. Chem. XXXX, XXX, XXX−XXX