The role of substituents on functionalized 1,10-phenanthroline in controlling the emission properties of cationic iridium(III) complexes of interest for electroluminescent devices

Article abstract of DOI:10.1021/ic700414z

The photophysical and electrochemical properties of the novel complexes [Ir(ppy)₂(5-X-1,10-phen)][PF₆] (ppy = 2-phenylpyridine, phen = phenanthroline, X = NMe₂, NO₂), [Ir(pq) ₂(5-X-1,10-phen)][PF

Full text of DOI:10.1021/ic700414z

Inorg. Chem. 2007, 46, 8533−8547
The Role of Substituents on Functionalized 1,10-Phenanthroline in
Controlling the Emission Properties of Cationic Iridium(III) Complexes of
Interest for Electroluminescent Devices
Claudia Dragonetti,†,§, Luigi Falciola,‡,§ Patrizia Mussini,*,‡,§ Stefania Righetto,†,§,
|
,
|,
Dominique Roberto,*,†,§, Renato Ugo,†,§, and Adriana Valore†,§,
Dipartimento di Chimica Inorganica, Metallorganica e Analitica, Dipartimento di Chimica Fisica
ed Elettrochimica, Centro di Eccellenza CIMAINA dell’UniVersita` di Milano,
Istituto di Scienze e Tecnologie Molecolari, CNR, and UdR dell’INSTM di Milano,
Via Venezian 21, 20133 Milano, Italy
Filippo De Angelis, Simona Fantacci,* and Antonio Sgamellotti
Istituto di Scienze e Tecnologie Molecolari, CNR, c/o Dipartimento di Chimica,
UniVersita` di Perugia, Via Elce di Sotto 8, I-06123 Perugia, Italy
Miguel Ramon and Michele Muccini
Istituto per lo Studio dei Materiali Nanostrutturati, CNR, Via P. Gobetti 101,
I-40129 Bologna, Italy
Received March 3, 2007
The photophysical and electrochemical properties of the novel complexes [Ir(ppy)2(5-X-1,10-phen)][PF6] (ppy
2-phenylpyridine, phen phenanthroline, X NMe2, NO2), [Ir(pq)2(5-X-1,10-phen)][PF6] (pq 2-phenylquinoline,
H, Me, NMe2, NO2), [Ir(ppy)2(4-Me,7-Me-1,10-phen)][PF6], [Ir(ppy)2(5-Me,6-Me-1,10-phen)][PF6], [Ir(ppy)2(2-
)
)
)
)
X
)
Me,9-Me-1,10-phen)][PF6], and [Ir(pq)2(4-Ph,7-Ph-1,10-phen)][PF6] have been investigated and compared with those
of the known reference complexes [Ir(ppy)2(4-Me or 5-H or 5-Me-1,10-phen)][PF6] and [Ir(ppy)2(4-Ph,7-Ph-1,10-
phen)][PF6], showing how the nature and number of the phenanthroline substituents tune the color of the emission,
its quantum yield, and the emission lifetime. It turns out that the quantum yield is strongly dependent on the
nonradiative decay. The geometry, ground state, electronic structure, and excited electronic states of the investigated
complexes have been calculated on the basis of density functional theory (DFT) and time-dependent DFT approaches,
thus substantiating the electrochemical measurements and providing insight into the electronic origin of the absorption
spectra and of the lowest excited states involved in the light emission process. These results provide useful guidelines
for further tailoring of the photophysical properties of ionic Ir(III) complexes.
Introduction
organic light-emitting diodes (OLED) with improved ef-
ficiency. These complexes are able to harvest both triplet
and singlet excitons due to increased spin-orbit coupling,
which in turn increases the efficiency of intercrossing into
the emitting triplet state.1 Neutral iridium(III) cyclometalated
complexes are of particular interest because (i) they typically
exhibit high phosphorescence quantum yields (φ ) 10-90%)
and (ii) their emission spectrum can be effectively tuned over
Recently, many neutral heavy-transition-metal complexes
have emerged as promising phosphorescent emitters for
* To whom correspondence should be addressed. E-mail: simona@
thch.unipg.it (S.F.), patrizia.mussini@unimi.it (P.M.), dominique.roberto@
unimi.it (D.R.).
† Dipartimento di Chimica Inorganica, Metallorganica e Analitica.
‡ Dipartimento di Chimica Fisica ed Elettrochimica.
§ Centro di Eccellenza CIMAINA dell’Universita` di Milano.
| Istituto di Scienze e Tecnologie Molecolari, CNR.
UdR dell’INSTM di Milano.
(1) Evans, R. C.; Douglas, P.; Winscom, C. J. Coord. Chem. ReV. 2006,
250, 2093.
10.1021/ic700414z CCC: $37.00
Published on Web 09/21/2007
© 2007 American Chemical Society
Inorganic Chemistry, Vol. 46, No. 21, 2007 8533

Dragonetti et al.
a large spectral range by an appropriate choice and combina-
tion of ligands and ligand substituents.1,2 These characteristics
suggested their use as emitting materials in host-guest
Almost all investigated ionic iridium complexes contain
two cyclometalated 2-phenylpyridines and one bipyridine.
The spectral tuning of the emission is accomplished by
functionalizing the former4,8,11 or the latter9 ligand. In
particular, it was reported that the emission color of
complexes such as [Ir(pic)2(L)][PF6] (pic ) cyclometalated
1-phenylisoquinoline; L ) bidentate nitrogen donor ligand
such as 1,2-bipyridine, 1,10-phenanthroline, 2-pyridyl-quino-
line, 2,2′-biquinoline, 1,1′-biisoquinoline, or 2-(2-quinolinyl)-
quinoxaline)) can be significantly tuned by changing the
nature of the ligand L.13 Recently, a combinatorial approach
highlighted the potential emission color tunability of [Ir-
(ppy)2(substituted 1,10-phen)]+ (phen ) phenanthroline)
complexes by changing the nature and position of substit-
uents.10,12 Basically, the quantum yield of [Ir(ppy)2(1,10-
phen)]+ is reported to be 2 times higher than that of
[Ir(ppy)2(2,2′-bipyridine)]+, but phenanthrolines substituted
with alkyl groups such as 3,4,7,8-tetramethyl-1,10-phenan-
throline and 4- or 5-methyl-1,10-phenanthroline were pre-
dicted to lead to even higher quantum yields,10 whereas a
slightly lower quantum yield was predicted for the complex
with 4,7-diphenyl-1,10-phenanthroline.12 However, contrary
to what was predicted by the combinatorial approach, for
[Ir(ppy)2(4-Ph,7-Ph-1,10-phen)][PF6], a quantum yield 1.5
times higher than that of [Ir(ppy)2(1,10-phen)][PF6] was
reported, suggesting that predictions need to be verified
experimentally.12 Finally, the potential application of these
cationic Ir(III) complexes as electroluminescent materials was
confirmed by the preparation of a stable single-layer light-
emitting electrochemical cell using [Ir(ppy)2(4-Ph,7-Ph-1,-
10-phen)][PF6].15b
architectures typically employed in OLED devices.1,3
A
recent concept in solid-state electroluminescence proposes
the use of “ionic transition-metal complexes” (iTMCs)4 thin
films as active layers in electroluminescent devices. Unlike
neutral metal complexes, iTMCs contain mobile counterions
that facilitate charge transport across the film and eliminate
the need for electron- and hole-injection layers. The iTMC-
based single-layer electroluminescent devices may also
operate at low voltage with high workfunction air-stable
electrodes (e.g., gold), due to the high electric fields produced
by ions accumulating at the anode and cathode interfaces,
which favor charge injection.4 The first reports on single-
layer devices involving ionic transition-metal complexes were
based on ruthenium(II) (e.g., [Ru(2,2′-bipyridine)3]2+),5 os-
mium(II),6 and Re(I)7 complexes, characterized primarily by
an orange-red emission and by a limited possibility of tuning
the emission wavelength. It was only 3 years ago that an
ionic cyclometalated iridium(III) complex, [Ir(ppy)2(4,4′-di-
tert-butyl-2,2′-bipyridine)][PF6] (ppy ) cyclometalated 2-phe-
nylpyridine), was first used in electroluminescent devices
affording yellow emission.8 Since then, ionic iridium(III)
complexes have attracted wide interest mainly due to the
possibility of inducing broad emission color tuning by
changing the ligand.4,8-15
(2) See, for example: (a) Nazeeruddin, M. K.; Humphrey-Baker, R.;
Berner, D.; Rivier, S.; Zuppiroli, L.; Graetzel, M. J. Am. Chem. Soc.
2003, 125, 8790 and references therein. (b) Lepeltier, M.; Le Bozec,
H.; Guerchais, V.; Lee, T. K.-M.; Lo, K. K.-W. Organometallics 2005,
24, 6069 and references therein. (c) Hwang, F.-M.; Chen, H. Y.; Chen,
P.-S.; Liu, C.-S.; Chi, Y.; Shu, C. F.; Wu, F.-L.; Chou, P. T.; Peng,
S.-M.; Lee, G.-H. Inorg. Chem. 2005, 44, 1344.
These promising results prompted us to perform a sys-
tematic investigation of the photophysical and electrochemi-
cal properties of the new class of ionic Ir(III) complexes
with substituted 1,10-phenanthrolines in order to experimen-
tally assess the role of substituents located on various
positions of the phenanthroline ligand. Thus, a large number
of Ir(III) complexes was investigated (1-7, Figure 1).
The electronic origin of the experimental photophysical
and electrochemical data was also substantiated by a density
functional theory (DFT) investigation, with particular refer-
ence to the nature of the highest-occupied molecular orbital
(HOMO) and lowest-unoccupied molecular orbital (LUMO)
(3) See, for example: (a) Baldo, M. A.; Lamansky, S.; Burrows, P. E.;
Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 1999, 75, 4. (b)
Adachi, C.; Baldo, M. A.; Forrest, S. R.; Lamansky, S.; Thompson,
M. E.; Kwong, R. C. Appl. Phys. Lett. 2001, 78, 1622. (c) Adachi,
C.; Kwong, R. C.; Djurovich, P.; Adamovich, V.; Baldo, M. A.;
Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 2001, 79, 2082. (d)
Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. J. Appl.
Phys. 2001, 90, 5048. (e) 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. (f) Xie, H.
Z.; Liu, M. W.; Wang, O. Y.; Zhang, X. H.; Lee, C. S.; Hung, L. S.;
Lee, S. T.; Teng, P. F.; Kwong, H. L.; Zhen, H.; Che, C. M. AdV.
Mater. 2001, 13, 1245. (g) Duan, J.-P.; Sun, P.-P.; Cheng, C.-H. AdV.
Mater. 2003, 15, 224. (h) Tsuzuki, T.; Shirasawa, N.; Suzuki, T.;
Tokito, S. AdV. Mater. 2003, 15, 1455. (i) Laskar, I. R.; Chen, T.-M.
Chem. Mater. 2004, 16, 111. (j) Laskar, I. R.; Hsu, S.-F.; Chen, T.-
M. Polyhedron 2005, 24, 189. (k) Tokito, S.; Tsuzuki, T.; Sato, F.;
Iijima, T. Curr. Appl. Phys. 2005, 331.
(10) Lowry, M. S.; Hudson, W. R.; Pascal, R. A., Jr.; Bernhard, S. J. Am.
Chem. Soc. 2004, 126, 14129.
(11) Slinker, J. D.; Koh, C. Y.; Malliaras, G. G.; Lowry, M. S.; Bernhard,
S. Appl. Phys. Lett. 2005, 86, 173506.
(12) Goldsmith, J. I.; Hudson, W. R.; Lowry, M. S.; Anderson, T. H.;
Bernhard, S. J. Am. Chem. Soc. 2005, 127, 7502.
(4) Lowry, M. S.; Goldsmith, J. I.; Slinker, J. D.; Rohl, R.; Pascal, R. A.,
Jr.; Malliaras G. G.; Bernhard, S. Chem. Mater. 2005, 17, 5712 and
references therein.
(5) (a) Handy, E. S.; Pal, A. J.; Rubner, M. F. J. Am. Chem. Soc. 1999,
121, 3525. (b) Gao, F. G.; Bard, A. J. J. Am. Chem. Soc. 2000, 122,
7426.
(13) Zhao, Q.; Liu, S.; Shi, M.; Wang, C.; Yu, M.; Li, L.; Li, F.; Yi, T.;
Huang, C. Inorg. Chem. 2006, 45, 6152.
(14) Neve, F.; La Deda, M.; Puntoriero, F.; Campagna, S. Inorg. Chim.
Acta 2006, 359, 1666.
(15) See, for example: (a) Nazeruddin, M. K.; Wehg, R. T.; Zhou, Z.;
Klein, C.; Wang, Q.; De Angelis, F.; Fantacci, S.; Gra¨tzel, M. Inorg.
Chem. 2006, 45, 9245. (b) Bolink, H. J.; Cappelli, L.; Coronado, E.;
Gra¨tzel, M.; Ort`ı, E.; Costa, R. D.; Viruela, P. M.; Nazeeruddin, M.
K. J. Am. Chem. Soc. 2006, 128, 14787. (c) Wilkinson, A. J.; Goeta,
A. E.; Foster, C. E.; Williams, J. A. G. Inorg. Chem. 2004, 43, 6513.
(d) De Angelis, F.; Fantacci, S.; Evans, N.; Klein, C.; Zakeeruddin,
S. M.; Moser, J.-E.; Kalyanasundaram, K.; Bolink, H. J.; Gra¨tzel, M.;
Nazeeruddin, M. K. Inorg. Chem. 2007, 46, 5989.
(6) Bernhard, S.; Gao, X.; Abruna, H. D.; Malliaras, G. G. AdV. Mater.
2002, 14, 433.
(7) Gong, X.; Ng, P. K.; Chan, W. K. AdV. Mater. 1998, 10, 1337.
(8) Slinker, J. D.; Gorodetsky, A. A.; Lowry, M. S.; Wang, J.; Parker, S.;
Rohl, R.; Bernhard, S.; Malliaras, G. G. J. Am. Chem. Soc. 2004, 126,
2763.
(9) (a) Neve, F.; La Deda, M.; Crespini, A.; Bellusci, A.; Puntoriero, F.;
Campagna, S. Organometallics 2004, 23, 5856. (b) Lepeltier, M.; Lee,
T. K.-M.; Lo, K. K.-W.; Toupet, L.; Le Bozec, H.; Guerchais, V. Eur.
J. Inorg. Chem. 2005, 110.
8534 Inorganic Chemistry, Vol. 46, No. 21, 2007

Cationic Iridium(III) Complexes
Figure 1. Complexes investigated corresponding to [Ir(ppy)2(substituted-1,10-phen)][PF6] and [Ir(pq)2(substituted-1,10-phen)][PF6] (ppy ) 2-phenylpyridine
and pq ) 2-phenylquinoline, both cyclometalated).
states and also of the lowest excited states involved in the
absorption and emission processes, by means of time-
dependent DFT calculations (TDDFT). While ground-state
DFT calculations have become increasingly popular together
with experimental electrochemical measurements to define
the nature of the redox processes involving iridium(III)
compounds,4,10,13,16 only a few TDDFT excited-state calcula-
tions on iridium(III) complexes have been reported to date,17
a pioneering paper in the field being the work by Hay17a on
the neutral [Ir(ppy)3] complex.
Experimental Section
General Comments. All reagents and solvents were purchased
from Sigma-Aldrich, while IrCl3 hydrate was purchased from
Engelhard. The cyclometalated Ir(III) chloro-bridge dimers [Ir-
(ppy)2Cl]2 and [Ir(pq)2Cl]2 (pq) 2-phenylquinoline) were prepared
according to the literature.18 The 5-NMe2-1,10-phen compound was
synthesized as previously reported.19 All reactions were carried out
(17) See, for example: (a) Hay, P. J. J. Phys. Chem. A 2002, 106, 1634.
(b) 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. (c) Polson, M.; Ravaglia, M.; Fracasso, S.; Garavelli, M.;
Scandola, F. Inorg. Chem. 2005, 44, 1282. (d) Yang, C. H.; Su, W.
L.; Fang, K. H.; Wang, S. P.; Sun, I. W. Organometallics 2006, 25,
4514. (e) Polson, M.; Fracasso, S.; Bertolasi, V.; Ravaglia, M.;
Scandola, F. Inorg. Chem. 2004, 43, 1950. (f) Obara, S.; Itabashi,
M.; Okuda, F.; Tamaki, S.; Tanabe, Y.; Ishii, Y.; Nozaki, K.; Haga,
M.-A. Inorg. Chem. 2006, 45, 8907.
(16) See, for example: (a) Tamayo, A. B.; Garon, S.; Sajoto, T.; Djurovich,
P. I.; Tsyba, I. M.; Bau, R.; Thompson, M. E. Inorg. Chem. 2005, 44,
8723. (b) You, Y.; Park, S. Y. J. Am. Chem. Soc. 2005, 127, 12438.
(c) Lo, S. C.; Shipley, C. P.; Bera, R. N.; Harding, R. E.; Cowley, A.
R.; Burn, P. L.; Samuel, I. D. W. Chem. Mater. 2006, 18, 5119. (d)
Dedeian, K.; Shi, J.; Shepherd, N.; Forsythe, E.; Morton, D. C. Inorg.
Chem. 2005, 44, 4445.
(18) Sprouse, S.; King, K. A.; Spellane, P. J.; Watts, R. J. J. Am. Chem.
Soc. 1984, 106, 6647.
(19) Shan, Y.; Sullivan, B. P. Inorg. Chem. 1995, 34, 6235.
Inorganic Chemistry, Vol. 46, No. 21, 2007 8535

Dragonetti et al.
under nitrogen. 1H NMR spectra were obtained on a Bruker Avance
400 MHz instrument. Elemental analyses were carried out in the
Dipartimento di Chimica Inorganica, Metallorganica e Analitica
of Milan University.
Synthesis of Complexes. All [Ir(ppy)2L][PF6] and [Ir(pq)2L]-
[PF6] (L ) 1,10-phen) complexes were prepared with the same
procedure, described below in the case of 1a, slightly modified
with respect to the literature.9,13 Complexes 1d and 3a were
previously prepared and isolated;10,12 1a was previously obtained
by a combinatorial approach10 but never isolated whereas all the
other complexes (1b, 1c, 2a, 2b-2d, 3b, and 4-6) are new, to our
knowledge.
[Ir(ppy)2(5-Me-1,10-phen)][PF6] (1a). A solution of [Ir-
(ppy)2Cl]2 (0.083 g, 0.077 mmol) and 5-Me-1,10-phen (0.030 g,
0.154 mmol) in CH2Cl2-MeOH (15 mL, 2:1 v/v) was heated under
reflux. After 5-6 h, the orange solution was cooled to room
temperature, and then a 10-fold excess of ammonium hexafluoro-
phosphate was added. The suspension was stirred for 15 min and
then filtered to remove insoluble inorganic salts. The solution was
evaporated to dryness under reduced pressure to obtain a crude
orange solid. The solid was dissolved in CH2Cl2 and filtered to
remove the residual traces of inorganic salts. Diethyl ether was
layered onto the orange filtrate, and the mixture was cooled to
∼0 °C. Orange plates of the desired product formed overnight.
Yield: 0.093 g (72%). 1H NMR (400 MHz, CD2Cl2) δ: 8.74 (d, J
) 8.46 Hz, 1H), 8.55 (d, J ) 8.26 Hz, 1H), 8.38 (d, J ) 5.01 Hz,
1H), 8.31 (d, J ) 5.03 Hz, 1H), 8.04 (s, 1H), 8.00 (d, J ) 8.30 Hz,
2H), 7.83 (m, 6H), 7.35 (t, J ) 5.36 Hz, 2H), 7.16 (t, J ) 7.56 Hz,
2H), 7.03 (t, J ) 8.08 Hz, 2H), 6.88 (t, J ) 6.63 Hz, 2H), 6.46 (d,
J ) 7.16 Hz, 2H), 2.95 (s, 3H). Anal. Calcd for Ir C35H26N4PF6:
C, 50.06; H, 3.12; N, 6.67. Found: C, 49.89; H, 3.08; N, 6.78.
[Ir(ppy)2(5-NMe2-1,10-phen)][PF6] (1b). Yield: 0.082 g (61%,
starting from 0.0825 g, 0.077 mmol of [Ir(ppy)2Cl]2). 1H NMR (400
MHz, CD2Cl2) δ: 8.86 (d, J ) 8.52 Hz, 1H), 8.44 (d, J ) 8.18
Hz, 1H), 8.34 (d, J ) 3.49 Hz, 1H), 8.14 (d, J ) 3.39 Hz, 1H),
7.99 (d, J ) 8.00 Hz, 1H), 7.76 (m, 6H), 7.59 (m, 1H), 7.52 (s,
1H), 7.38 (s, 2H), 7.15 (t, J ) 7.37 Hz, 2H), 7.02 (d, J ) 6.44 Hz,
2H), 6.89 (t, J ) 6.38 Hz, 2H), 6.44 (t, J ) 8.35 Hz, 2H), 3.11(s,
6H). Anal. Calcd for Ir C36H29N5PF6: C, 49.77; H, 3.36; N, 8.06.
Found: C, 49.69; H, 3.28; N, 7.98.
2H), 6.81 (t, J ) 7.94 Hz, 2H), 6.7 (m, 2H), 2.72 (s, 3H). Anal.
Calcd for Ir C43H30N4PF6: C, 54.95; H, 3.22; N, 5.96. Found: C,
54,78; H, 3.16; N, 6.00.
[Ir(pq)2(5-NMe2-1,10-phen)][PF6] (2b). Yield: 0.0731 g (60%,
starting from 0.080 g, 0.0629 mmol of [Ir(pq)2Cl]2). 1H NMR (400
MHz, CD2Cl2) δ: 8.72 (d, J ) 8.5 Hz, 1H), 8.58 (d, J ) 4.5 Hz,
1H), 8.39 (d, J ) 4.80 Hz, 1H), 8.27 (m, 5H), 8.16 (dd, J ) 10.65
Hz, 2H), 7.82 (dd, J ) 8.45 Hz, 1H), 7.69 (m, 3 H), 7.27 (m, 6H),
7.18 (d, J ) 8.95 Hz, 1H), 6.87 (m, 4H), 6.69 (dd, J ) 18 Hz,
2H), 2.92 (s, 2H). Anal. Calcd for Ir C44H33N5PF6: C, 54.54; H,
3.43; N, 7.23. Found: C, 54.46; H, 3.35; N, 7.12.
[Ir(pq)2(5-nitro-1,10-phen)][PF6] (2c). Yield: 0.060 g (80%,
starting from 0.0491 g, 0.0386 mmol of [Ir(pq)2Cl]2). 1H NMR (400
MHz, CD2Cl2) δ: 9.16 (d, J ) 8.72 Hz, 1H), 8.83 (s, 1H), 8.78 (d,
J ) 5.08 Hz, 1H), 8.71 (d, J ) 8.20 Hz, 1H), 8.29 (m, 4H), 8.19
(d, J ) 7.88 Hz, 2H), 8,02 (m, 2H), 7.69 (d, J ) 8.09 Hz, 2 H),
7.28 (m, 4H), 7.14 (m, 2H), 6.93 (t, J ) 7.48 Hz, 2H), 6.85 (t, J
) 8.58 Hz, 2H), 6.70 (d, J ) 7.25 Hz, 2H). Anal. Calcd for Ir
C42H27N5O2PF6: C, 51.96; H, 2.80; N, 7.21. Found: C, 51.78; H,
2.78; N, 7.16.
[Ir(pq)2(1,10-phen)][PF6] (2d). Yield: 0.077 g (68%, starting
1
from 0.080 g, 0.0629 mmol of [Ir(pq)2Cl]2). H NMR (400 MHz,
CD2Cl2) δ: 8.61 (dd, J ) 5.14 Hz, 1H), 8.47 (dd, J ) 8.20 Hz,
1H), 8.26 (q, J ) 8.82, 2H), 8.16 (d, J ) 7.07 Hz, 1H), 7.92 (s,
1H), 7.84 (dd, J ) 8.20 Hz, 1H), 7.66 (dd, J ) 8.06 Hz, 1H), 7.26
(m, 3H), 6.92 (t, J ) 7.59 Hz, 1H), 6.79 (t, J ) 6.94 Hz, 1H), 6.71
(dd, J ) 7.66 Hz, 1H). Anal. Calcd for IrC42H28N4PF6: C, 54.49;
H, 3.05; N, 6.05. Found: C, 54.56; H, 3.10; N, 5.96.
[Ir(ppy)2(4-Ph,7-Ph-1,10-phen)][PF6] (3a). Yield: 0.081 g
1
(61%, starting from 0.0728 g, 0.0689 mmol of [Ir(ppy)2Cl]2). H
NMR (400 MHz, CD2Cl2) δ: 8.43 (d, J ) 5.26 Hz, 2H), 8.23 (s,
2H), 8.04 (d, J ) 8.12 Hz, 2H), 7.85 (m, 4H), 7.78 (d, J ) 5.27
Hz, 2H), 7.63 (m, 10H), 7,54 (d, J ) 5.24 Hz, 2H), 7.16 (t, J )
7.65 Hz, 2 H), 7.06 (t, J ) 7.50 Hz, 2H), 6.97 (t, J ) 7.70 Hz,
2H), 6.50 (d, J ) 6.91 Hz, 2H). Anal. Calcd for Ir C46H32N4PF6:
C, 56.49; H, 3.29; N, 5.73. Found: C, 56.38; H, 3.28; N, 5.68.
[Ir(ppy)2(4-Me,7-Me-1,10-phen)][PF6] (3b). Yield: 0.071 g
1
(64%, starting from 0.0696 g, 0.0649 mmol of [Ir(ppy)2Cl]2). H
NMR (400 MHz, CD2Cl2) δ: 8.37 (s, 2H), 8.22 (d, J ) 5.00 Hz,
2H), 7.99 (d, J ) 5.00 Hz, 2H), 7.81 (d, J ) 10.00 Hz, 2H), 7.76
(t, J ) 7.5 Hz, 2H), 7.64 (d, J ) 5.00 Hz, 2H), 7.37 (d, J ) 5.0
Hz, 2H), 7.15 (t, J ) 7.5 Hz, 2H), 7.02 (t, J ) 5 Hz, 2H), 6.88 (t,
J ) 5.8 Hz, 2H), 6.45 (d, J ) 6.6 Hz, 2H), 2.99 (s, 6H). Anal.
Calcd for Ir C36H28N4PF6: C, 50.64; H, 3.30; N, 6.56. Found: C,
50.70; H, 3.26; N, 6.60.
[Ir(pq)2(4-Ph,7-Ph-1,10-phen)][PF6] (4). Yield: 0.070 g (60%
starting from 0.0689 g, 0.0541 mmol of [Ir(pq)2Cl]2). 1H NMR (400
MHz, CD2Cl2) δ: 8.72 (d, J ) 5.37 Hz, 2H), 8.31 (m, 4H), 8.19
(d, J ) 7.87 Hz, 2H), 7.93 (s, 2H), 7.79 (d, J ) 5.38 Hz, 2H), 7.72
(d, J ) 7.16 Hz, 2H), 7.59 (m, 6H), 7.48 (m, 4H), 7.32 (m, 6H),
6.91 (m, 4H), 6.72 (d, J ) 7.65 Hz, 2H). Anal. Calcd for Ir
C54H36N4PF6: C, 60.16; H, 3.34; N, 5.20. Found: C, 60.18; H,
3.38; N, 5.30.
[Ir(ppy)2(5-nitro-1,10-phen)][PF6] (1c). Yield: 0.058 g (87%,
1
starting from 0.0410 g, 0.0383 mmol of [Ir(ppy)2Cl]2). H NMR
(400 MHz, CD2Cl2) δ: 9.32 (d, J ) 8.74 Hz, 1H), 9.16 (s, 1H),
8.89 (d, J ) 8.18 Hz, 1H), 8.56 (d, J ) 4.49 Hz, 1H), 8.52 (d, J )
4.99 Hz, 1H), 8.02 (m, 4H), 7.80 (m, 4H), 7.38 (t, J ) 4.51 Hz, 2
H), 7.17 (t, J ) 7.48 Hz, 2H), 7.05 (t, J ) 7.40 Hz, 2H), 6.93 (t,
J ) 7.06 Hz, 2H), 6.44 (d, J ) 7.51 Hz, 2H) Anal. Calcd for Ir
C34H23N5O2PF6: C, 46.90; H, 2.66; N, 8.04. Found: C, 46.92; H,
2.70; N, 8.10.
[Ir(ppy)2(1,10-phen)][PF6] (1d). Yield: 0.103 g (80%, starting
from 0.0489 g, 0.0385 mmol of [Ir(ppy)2Cl]2). 1H NMR (400 MHz,
CD2Cl2) δ: 8.66 (d, J ) 8.00 Hz, 2H), 8.38 (d, J ) 4.00 Hz, 2H),
8.24 (s, 2H), 8.00 (d, J ) 8.00 Hz, 2H), 7.83 (m, 6H), 7.35 (d, J
) 5.2 Hz, 2H), 7.16 (t, J ) 7.3 Hz, 2H), 7.04 (t, J ) 7.3 Hz, 2H),
6.88 (t, J ) 5.8 Hz, 2H), 6.46 (d, J ) 6.6 Hz, 2H). Anal. Calcd for
Ir C34H24N4PF6: C, 49.46; H, 2.93; N, 6.78. Found: C, 49.38; H,
2.88; N, 6.72.
[Ir(pq)2 5-Me-1,10-phen)][PF6] (2a). Yield: 0.0772 g (67%,
starting from 0.0780 g, 0.0613 mmol of [Ir(pq)2Cl]2). 1H NMR (400
MHz, CD2Cl2) δ: 8.62 (d, J ) 5.05 Hz, 1H), 8.55 (m, 2H), 8.37
(d, J ) 8.26 Hz, 1H), 8.24 (m, 4H), 8.17 (d, J ) 7.4 Hz, 2H), 7.87
(d, J ) 8.41 Hz, 1H), 7.80 (d, J ) 8.19 Hz, 1H), 7.72 (s, 1 H),
7.67 (d, J ) 8.06 Hz, 2H), 7.25 (m, 6H), 6.91 (t, J ) 7.20 Hz,
[Ir(ppy)2(5-Me,6-Me-1,10-phen)][PF6] (5). Yield: 0.060 g
1
(65%, starting from 0.0579 g, 0.0541 mmol of [Ir(ppy)2Cl]2). H
NMR (400 MHz, CD2Cl2) δ: 8.78 (d, J ) 5.00 Hz, 2H), 8.32 (d,
J ) 5.00 Hz, 2H), 8.00 (d, J ) 10.00 Hz, 2H), 7.81 (m, 6H), 7.34
(d, J ) 5.0 Hz, 2H), 7.16 (t, J ) 7.5 Hz, 2H), 7.03 (t, J ) 7.5 Hz,
2H), 6.87 (t, J ) 8.33 Hz, 2H), 6.46 (d, J ) 5.0 Hz, 2H), 5.36 (d,
J ) 5.0 Hz, 2H), 2.90 (s, 6H). Anal. Calcd for Ir C36H28N4PF6: C,
50.64; H, 3.30; N, 6.56. Found: C, 50.70; H, 3.26; N, 6.68.
[Ir(ppy)2(2-Me,9-Me-1,10-phen)][PF6] (6). Yield: 0.110 g
1
(83%, starting from 0.0832 g, 0.0776 mmol of [Ir(ppy)2Cl]2). H
8536 Inorganic Chemistry, Vol. 46, No. 21, 2007

Cationic Iridium(III) Complexes
Table 1. Photophysical Data Obtained in CH2Cl2 Solution of Complexes 1-6: Absorption and Emission Maxima, Quantum Yield (Φ), Lifetime (τ),
Radiative Rate Constant (kr), and Nonradiative Rate Constant (knr)
absorption maxima
emission maxima
(nm; eV)
Φa
(%)
τb
(µs)
kr
knr
(nm) [ꢀ (M-1 cm-1)]
(105 s-1
)
(105 s-1
)
1a
255 (sh), 268 [60300]
333 (sh), 377 [8070]
410 (sh)
559; 2.218
38
3.17
1.20
1.96
1b
1c
1d
2a
252 (sh), 264 [81400]
334 (sh)
254 (sh), 264 [86900]
378 [12400]
252 (sh), 264 [58500]
377 [9130]
273 [70400]
568; 2.183
0.3
<0.1
7
n.d.
n.d.
1.31
4.03
n.d.
n.d.
0.53
0.84
n.d.
n.d.
7.10
1.64
575; 2.156
556; 2.230
34
329 [20700], 347 (sh)
431 [4750]
2b
2c
2d
260 [97000]
332 [44100], 349 (sh)
434 [10900]
268 [81100]
324 [27600]
557; 2.226
6
<0.1
6
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
430 [6930]
269 [95000]
552; 2.246
330 [41300]
433 [14700]
269 [51300], 282 (sh)
385 [9020]
255 (sh), 265 [73000]
375 (sh)
267 (sh), 282 [80800]
329 [34400], 389 (sh)
433 [6060]
3a
3b
4
592; 2.094
552; 2.246
562; 2.206
20c
26
1.31
1.43
2.39
1.53
1.82
0.79
6.11
5.17
3.39
19
5
6
242 [61800], 250 (sh)
269 [58200]
377 [8940]
252 (sh), 269 [106000]
372 (sh)
568; 2.183
534; 2.322
22
19
1.27
2.31
1.73
0.82
6.14
3.51
a By using coumarin 540 (Φ is 56% in EtOH; λmax emission ) 540 nm) as the reference;21a it is worth pointing out that lower quantum yields were
obtained by working in CH3CN or toluene instead of CH2Cl2 solution: Φ in CH3CN is 7.7, 8, 4, and 5.7% for 1a, 2a, 3a, and 4, respectively; Φ in toluene
is 2.7% for 3a. b Values of τ significantly decrease (1a, 0.69; 1d, 0.19; 2a, 0.82; 3a, 0.33; 3b, 0.19; 4, 0.67; 5, 0.25; 6, 0.26) when working with non-
deaerated solutions, probably due to oxygen-triplet state interaction. c For this complex, various quantum yields were reported (Φ in CH3CN ) 18.18%,
using [Ir(ppy)2(bpy)]PF6 as the standard);12 Φ in CH3CN ) 53%, using [Ru(4,7-diphenyl-1,10-phen)3]Cl2 as the standard;15b for the same cationic Ir complex
but with Cl- instead of PF6- as the counterion, Φ in CH3CN ) 10.25%, obtained by a combinatorial method and using [Ir(ppy)2(bpy)]Cl as the standard12).
NMR (400 MHz, CD2Cl2) δ: 8.51 (d, J ) 10.00 Hz, 2H), 8.08 (s,
2H), 8.00 (d, J ) 10.00 Hz, 2H), 7.81 (t, J ) 7.5 Hz, 2H), 7.72 (d,
J ) 10.0 Hz, 2H), 7.65 (d, J ) 10 Hz, 2H), 7.49 (d, J ) 5 Hz,
2H), 7.03 (t, J ) 7.5 Hz, 2H), 6.91 (t, J ) 6.67 Hz, 2H), 6.85 (t,
J ) 7.5 Hz, 2H), 6.20 (d, J ) 10.0 Hz, 2H), 2.19 (s, 6H). Anal.
Calcd for Ir C36H28N4PF6: C, 50.64; H, 3.30; N, 6.56. Found: C,
50.70; H, 3.31; N, 6.70.
Photophysical Measurements. UV spectra were recorded at
room temperature in a 1 cm quartz cuvette using a Jasco V-570
spectrometer. Emission spectra were recorded using a Jobin-Yvon
Fluorolog-3 spectrometer equipped with double monochromators
and a Hamamatsu-928 photomultiplier tube as the detector. All
complexes were excited at 260 nm. Emission quantum yields have
been determined, using the method of Demas and Crosby,20 by
comparison with the emission of coumarin 540, employed as a
standard.21a The use of coumarin 540 as a reference standard seems
more suitable than the use of [Ru(bpy)3]Cl2,21b which is very air
sensitive. All solutions were deaerated by nitrogen bubbling for
30 min before measurements.
harmonic (355 nm) of a Nd:YAG laser with a 25 ns pulse duration
and a 20 Hz repetition rate. The photoluminescence was detected
by a fast photomultiplier and then fed into a fast digital oscilloscope
for synchronous time-resolved data acquisition. The time resolution
of the setup is about 50 ns, well below all measured lifetimes. All
solutions were in a 10-5 M concentration range and were deaerated
by argon bubbling for 10 min before measurement. Lifetimes
significantly decrease (values in the range 0.19-0.82 µs) in
solutions that were not deaerated, probably due to oxygen-triplet
state interaction (see note in Table 1).
Electrochemical Measurements. The cyclovoltammetric (CV)
measurements were carried out using an Autolab PGSTAT 30
potentiostat/galvanostat (EcoChemie, The Netherlands) run by a
PC with GPES software, at scan rates in the range of 0.02-2 V
s-1, working in acetonitrile (CH3CN, Merck, HPLC grade), with
0.1 M tetrabutylammonium hexafluorophosphate [NBu4]PF6 (Fluka,
puriss electrochemical grade) as the supporting electrolyte. The
ohmic drop was corrected by means of the positive feedback
technique.22 A glassy-carbon (GC) electrode (Amel, surface area
) 0.071 cm2) was used as the working electrode, a platinum wire
was used as the counter electrode, and an aqueous saturated calomel
electrode (SCE) was used as the operating reference electrode. The
Lifetimes of the electronic emitting states were measured by
exciting photoluminescence of CH2Cl2 solutions by the third
(20) Demas, J. N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991.
(21) (a) Kubin, R. F.; Fletcher, A. N. Chem. Phys. Lett. 1983, 99, 49. (b)
Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583.
(22) Bard, A. J.; Faulkner, L.R. Electrochemical Methods. Fundamentals
and Applications; Wiley: New York, 2002; pp 648-650.
Inorganic Chemistry, Vol. 46, No. 21, 2007 8537

Dragonetti et al.
data have been subsequently referred to the Me10Fc+|Me10Fc
(decamethylferricinium|decamethylferrocene) reference redox couple,
currently proposed as an improved alternative23-25 to ferrocene,
recommended by IUPAC,26,27 and whose formal redox potential in
our operative solvent is -0.110 V against our aqueous SCE
reference. The cell was thermostated at 298 K, and the solutions
were carefully deaerated by nitrogen bubbling before the scans.
The optimized polishing procedure for the working GC electrode
consisted of surface treatment with diamond powder (Aldrich,
diameter ) 1 µm) on a wet cloth (DP-Nap, Struers). The chemical
and electrochemical reversibility of each CV peak were checked
by means of classical tests,28 including analyses of: (a) half-peak
width (Ep - Ep/2); (b) Ep vs log V characteristics; (c) the distance
between the forward and backward peak potential (Ep,f - Ep,b), and
(d) the “stationary” steplike waves obtained by means of convolu-
tion analysis of the original CV characteristics,29 for the evaluation
of the half-wave potentials (E1/2).
singlet excitations at the ground-state-optimized geometries allowed
us to simulate a large (up to 250 nm) portion of the absorption
spectrum. The simulation of the absorption spectra has been
performed by a Gaussian convolution with fwhm ) 0.4 eV; the
experimental spectra have been rescaled so that the intensity of
the experimental and theoretical main features in the absorption
spectra match.
At the triplet-optimized geometries, we calculated the three
lowest singlet-singlet (S) and singlet-triplet (T) TDDFT excita-
tions in dichloromethane at the B3LYP/LANL2DZ level, the triplet
excited state being related to the emission process. It has to be
noticed that the singlet-triplet excitation oscillator strengths are
set to zero due to the neglect of spin-orbit coupling in the TDDTF
calculations as implemented in G03.
Results and Discussion
Synthesis. Complexes [Ir(ppy)2(5-X-1,10-phen)][PF6] (1)
and [Ir(pq)2(5-X-1,10-phen)][PF6] (2) were synthesized with
phenanthrolines carrying a weak electron-donor substituent
(1a and 2a, X ) Me), a strong electron-donor substituent
(1b and 2b, X ) NMe2), and a strong electron-withdrawing
substituent (1c and 2c, X ) NO2). With the aim of
investigating the effect of a larger substitution by methyl
groups on the phenanthroline, we also synthesized the
complexes [Ir(ppy)2(4-Me,7-Me-1,10-phen)][PF6] (3b), [Ir-
(ppy)2(5-Me,6-Me-1,10-phen)][PF6] (5), and [Ir(ppy)2(2-
Me,9-Me-1,10-phen)][PF6] (6). The complex [Ir(ppy)2(4-Me-
1,10-phen)][PF6] (7) was only investigated theoretically by
DFT calculations. The complexes [Ir(ppy)2(1,10-phen)][PF6]
(1d) and [Ir(pq)2(1,10-phen)][PF6] (2d) were taken as refer-
ences. Finally, to verify the difference between a methyl and
a π-delocalized substituent such as a phenyl ring, [Ir(ppy)2-
(4-Ph,7-Ph-1,10-phen)][PF6] (3a) was synthesized and stud-
ied together with [Ir(pq)2(4-Ph,7-Ph-1,10-phen)][PF6] (4). All
complexes, except 7, were prepared by a bridge-splitting
reaction of the dinuclear precursor [Ir(ppy)2(µ-Cl)]218 or [Ir-
(pq)2(µ-Cl)]218 with the appropriate 1,10-phen ligand added
in a stoichiometric amount (see the Experimental Section).
Photophysical Properties. Photophysical data for com-
plexes 1-6 are reported in Table 1. All complexes display
strong absorption bands between 250 and 290 nm that can
be attributed to π-π* ligand-centered (LC) transitions
typically involving the excitation of electrons from filled π
to vacant π* orbitals of the cyclometalated ppy or pq and
phen ligands. Weaker absorption bands at longer wavelengths
(in the range of 300-450 nm) have been usually attributed
to metal-to-ligand charge transfer (MLCT, arising from the
excitation from a filled t2g orbital of the Ir(III) ion to a vacant
π* orbital of the phenanthroline ligand acting as acceptor)
Computational Details. The geometries of all complexes were
optimized by DFT calculations using the BP86 exchange-correlation
function,30,31 together with a TZP (DZP) basis set for Ir (N, C, O,
H), including scalar-relativistic corrections as implemented in the
ADF program.32 The symmetric complexes 1d, 3a, 3b, 4, 5, and 6
were optimized within C2 symmetry constraints, while no symmetry
constraints were considered for all the others. For complexes 1a,
1d, 2a, 3a, 3b, and 4 we optimized the geometry of the lowest
excited state by means of self-consistent field calculations imposing
a triplet spin multiplicity.
On the ground-state-optimized geometries we performed single-
point calculations at the B3LYP/LANL2DZ level of theory33,34 in
dichloromethane and acetonitrile solutions by means of the PCM
solvation model,35 as implemented in the Gaussian 03 (G03)
program package.36 For complexes 1a,1d, 3a, and 3b we performed
TDDFT calculations at the B3LYP/LANL2DZ level of theory in
dichloromethane solution. Calculation of the lowest 70 singlet-
(23) Falciola, L.; Gennaro, A.; Isse, A. A.; Mussini, P. R.; Rossi, M. J.
Electroanal. Chem. 2006, 593, 47.
(24) Ruiz, J.; Astruc, D. C. R. 'Acad. Sci., Se´rie IIc: Chim. 1998, 1 (1),
21.
(25) Noviandri, I.; Brown, K. N.; Fleming, D. S.; Gulyas, P. T.; Lay, P.
A.; Masters, A. F.; Phillips, L. J. Phys. Chem. B 1999, 103, 6713.
(26) Gritzner, G.; Kuta, J. Pure Appl. Chem. 1984, 56, 461.
(27) Gritzner, G. Pure Appl. Chem. 1990, 62, 1839.
(28) Ref 22, pp 228-232.
(29) Ref 22, pp 247-252.
(30) Becke, A. D. Phys. ReV. A 1988, 38, 3098.
(31) Perdew, J. P. Phys. ReV. B 1986, 33, 8822.
(32) Velde, G. te; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca-Guerra,
C.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput.
Chem. 2001, 22, 931.
(33) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(34) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
(35) Miertus, S.; Scrocco, S.; Tomasi, T. Chem. Phys. 1981, 55, 117.
(36) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. J. A.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G.
A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J.
W.; Ayala, P. Y.; Morokuma, K.; Vot, G. A.; Salvador, P.; Dannen-
berg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.;
Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian 03, revision B.05; Gaussian, Inc.: Wallingford, CT, 2004.
and to ligand-to-ligand charge transfer transitions.9,10,13
A
detailed assignment by our DFT investigation of the bands
of the absorption spectra of the investigated complexes is
given below. The emission spectra of some representative
complexes (1a, 3a, and 6) are shown in Figure 2. The
emission maximum can be slightly tuned by varying the
nature and number of the phenanthroline’s substituents: in
CH2Cl2 solution it is in the range of 534-592 nm for the
investigated complexes (Table 1 and Figure 2). As can be
noticed from Figure 2, the emission spectra are broad and
8538 Inorganic Chemistry, Vol. 46, No. 21, 2007

Cationic Iridium(III) Complexes
Figure 2. Emission spectra of compounds 1a (‚‚‚‚), 3a (—), and 6 (-) in CH2Cl2 at room temperature.
structureless, evidence of the substantial MLCT character
of the emitting state. Indeed, the increasing role of the π-π*
LC transitions in the emission process is usually associated
to a more pronounced vibronic substructure in the emission
spectra, even at room temperature.37 Our TDDFT investiga-
tion (see later) has confirmed for some selected complexes
the substantial MLCT character of the light-emitting states.
As evidenced in Table 1, the quantum yields of 1,
measured in CH2Cl2 solution, strongly depend on the
electronic nature of the substituent X. The quantum yield of
1a, where X is the weak electron-donor methyl group, is
quite high (Φ ) 38% in CH2Cl2) whereas that of complex
1b, where X is a strong electron-donor NMe2 substituent, is
very low (Φ ) 0.3% in CH2Cl2). With a strong electron-
withdrawing group such as NO2 (1c) the photoluminescence
is below the detection limit. The reference complex (1d) has
only a rather small quantum yield (Φ ) 7% in CH2Cl2), as
previously reported.10,12 A similar trend is observed for the
related 2 complexes (Table 1). Besides, it appears that the
presence of two methyl groups on the phenanthroline ligand
(e.g., 4-Me,7-Me-1,10-phen, 5-Me,6-Me-1,10-phen, and
2-Me,9-Me-1,10-phen) leads to a continuous decrease of the
quantum yield (Φ ) 26, 22, and 19% in CH2Cl2 solution
for complexes 3b, 5, and 6, respectively) when compared
with that of the complex bearing 1a. Complex 3a, bearing
the 4-Ph,7-Ph-1,10-phen ligand, shows a quantum yield (Φ
) 20% in CH2Cl2 solution) slightly lower than the related
complex of 3b with methyl instead of phenyl substituents.
Therefore, we confirmed, with an experimental investiga-
tion on a large number of cationic cyclometalated Ir(III)
complexes carrying substituted 1,10-phenanthrolines, that the
higher quantum yields are shown when the substituents on
the phenanthroline ring are alkyl groups, although the
increase of the substitution to two methyl substituents does
not produce an increase of the quantum yields. The position
of the substitution is relevant since complexes 3b, 5, and 6
show quite different quantum yields (Table 1).
The emissive lifetimes, measured in deaerated solutions,
are in the range of 1.27-4.03 µs (Table 1), in agreement
with the values reported for other cationic Ir(III) complexes.14
From the quantum yields Φ and the lifetimes τ values, the
overall radiative and nonradiative rate constants kr and knr
were calculated for a series of complexes through the
equations kr ) Φ/τ and knr ) (1 - Φ)/τ38 (see Table 1). It
appears that the kr values show a smaller variability compared
with that of the knr values; experimental quantum yields of
the various complexes can be analyzed in terms of the
different contributions of kr and knr for 1a and 1d and for 3b
and 2a, which show the highest quantum yields within the
complexes of the series 1-3. According to Einstein’s law39
2
of spontaneous emission, kr should be proportional to ν3|µ| ,
where µ is the transition dipole moment of the emitting
transition and ν3 is the cube of the emission frequency, while,
according to the so-called energy-gap law, knr should be
proportional to e-ν. Therefore knr is expected to decrease
significantly by increasing the emission frequency.40 For the
investigated complexes (see Table 1), all emitting in a rather
similar range of the emission frequency, comparable kr values
2
are found; this in turn relates to similar |µ| values,
confirming a common nature of the emitting excited states.
Complexes 1a and 2a show comparably higher quantum
yields than all the other complexes, mainly as a result of the
relatively low knr values compared with, e.g., 1d and 3b
(Table 1). However, while for 3b the rather high knr value is
partially compensated by the highest kr value among all the
investigated complexes, providing a quantum yield of 26%,
for 1d, the low kr value leads to a quantum yield of 7%.
The energy gap law is effectively obeyed by 1d, 1a, and 2a
(Table 1 and Supporting Information) while the higher knr
value of 3b is not easily correlated with the knr values of
(38) Straughan, B. P.; Walker, S. Spectroscopy; Science Paperbacks;
Chapman & Hall, London, 1976; Vol. 3.
(39) (a) Strickler, S. J.; Berg, R. A. J. Chem. Phys. 1962, 37, 814. (b)
Einstein, A. Phys. Z. 1917, 18, 121.
(40) (a) Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J. J. Phys.
Chem. 1986, 90, 3722. (b) Henry, B. R.; Siebrand, W. In Organic
Molecular Photophysics; Birks, J. B., Ed.; Wiley, New York, 1973;
Vol. 1, Chapter 4. (c) Freed, K. F.; Jortner, J. J. Chem. Phys. 1970,
52, 6272. (d) Bixon, M.; Jortner, J. J. Chem. Phys. 1968, 48, 715.
(37) (a) Colombo, M. G.; Hauser, A.; Gu¨del, H. U. Inorg. Chem. 1993,
32, 3088. (b) Colombo, M. G.; Brunold, C. T.; Riedener, T.; Gu¨del,
H. U.; Fo¨rtsch, M.; Bu¨rgi, H.-B. Inorg. Chem. 1994, 33, 545.
Inorganic Chemistry, Vol. 46, No. 21, 2007 8539

Dragonetti et al.
convolution analysis of the original CV characteristics are
also reported. Moreover, in Table 2 the HOMO-LUMO
energy gaps, calculated from the difference between the
anodic and cathodic peak potentials, are also reported and
compared with those calculated by a DFT theoretical
approach (see below).
All compounds show a chemically and electrochemically
reversible anodic peak, with the exception of compounds 1b
and 2b, carrying a NMe2 group, which show a complex
oxidation signal composed by three subsequent multielec-
tronic and very close peaks at 1.33, 1.42, and 1.58 V (vs
Me10Fc, where Fc is ferrocene), suggesting that the NMe2
group is probably also involved in the oxidation process. At
low scan rates (<0.1 V s-1), a loss of chemical reversibility
occurs, suggesting a chemical reaction following the oxida-
tion processes. Moreover, all the compounds show in the
reduction part a chemically and electrochemically reversible
cathodic peak generally followed by other peaks (see Figure
3 and the Supporting Information), different in intensity,
shape, and number, depending on the compound.
As reported for these types of Ir(III) complexes,2c,4,9,41 the
oxidation processes are linked to metal-centered orbitals with
a contribution from the phenyl part of cyclometalated
fragments (ppy or pq), which usually lead to an increased
loss of electrochemical reversibility. On the other hand,
reduction processes are assigned to orbitals centered mainly
on the Ir-phen fragment.2c,4,9,24 Both these assignments are
confirmed by our DFT calculations (see below). The presence
of one or two weak electron-donor groups (CH3) on the
phenanthroline ring shifts, as expected, the reduction peak
potentials to lower values with the stronger effect (130 mV)
for compound 3b, bearing two methyl groups in positions 4
and 7. The effect of the methyl groups is remarkably lower
on the oxidation peak, where a slight anticipation of the peak
potential (20 mV) can be perceived only in the presence of
two methyl groups. However, in compound 6, we have a
significant perturbing effect (see Table 2), probably caused
by the steric influence of the two methyl groups in positions
2 and 9 on the availability of orbitals of the Ir site, as
confirmed by DFT calculations (see below).
Figure 3. CV patterns of compounds 1a-d and 3a and 3b, working at
(1.8-2.8) × 10-4 M concentrations in CH3CN + 0.1 M [NBu4]PF6 and
with a GC electrode. The CV pattern of the reference model compound
ferrocene (Fc), as a standard for a monoelectronic chemically and
electrochemically reversible reaction, is also reported.
1d, 1a, and 2a. Overall, the relative quantum yields of the
investigated complexes are mainly controlled by nonradiative
deactivation processes, since, in contrast with the rather
limited variations of kr (in the range of (0.5-1.8) × 105 s-1),
significant variations of knr values (in the range of (1.64-
7.1) × 105 s-1) are found. The wider variation range of knr
values probably originates in the significant effect of the
nature and position of substituents on the phenanthroline ring
deformation motions, which might control the nonradiative
pathway of decay.
Electrochemical Properties. Figure 3 shows the CV
pattern of the investigated Ir(III) complexes working in a
1.8-2.8 10-4 M range of concentration in CH3CN solutions
added with 0.1 M [NBu4]PF6 and working with a GC
electrode (see Experimental Section for details). The CV
pattern of the reference model compound ferrocene, used as
the standard for a monoelectronic chemically and electro-
chemically reversible reaction, is also reported in Figure 3.
In Table 2, the oxidation and reduction potentials are given,
together with the reversibility parameters (the half-peak width
(Ep - Ep/2) and the distance between the forward and
backward peak potential (Ep,f - Ep,b)22). Half-wave potentials
from the “stationary” steplike waves obtained by means of
The presence of a strong electron-donor (NMe2) or
electron-acceptor (NO2) group on the phenanthroline ligand
significantly shifts the oxidation peak potentials to lower or
higher values (60-120 mV), respectively, according to the
electron enrichment/depletion of the metal-centered redox
site (in Table 2, compare 1b, 2b and 1c, 2c with compound
1d) due to the quite different donor properties of the
phenanthroline ligand. Symmetrically, while the presence of
NMe2 shifts the reduction peak potential to more negative
values, due to the higher electron density on the phenan-
throline ligand, the opposite effect is observed by the
presence of the NO2 electron-withdrawing group, resulting
in a dramatic anticipation of the reduction peak potentials
(41) Calogero, G.; Giuffrida, G.; Serroni, S.; Ricevuto, V.; Campagna, S.
Inorg. Chem. 1995, 34, 541.
8540 Inorganic Chemistry, Vol. 46, No. 21, 2007

Cationic Iridium(III) Complexes
Table 2. Oxidation EA and Reduction EC Potentials (against Me10Fc), Reversibility Parameters, Half-Wave Potentials (as Volts) of the Investigated
Ir(III) Complexes on a GC Electrode, in a 1.8-2.8 10-4 M Range of Concentrations in CH3CN Solution Added with 0.1 M [NBu4]PF6 as Supporting
Electrolyte. HOMO-LUMO Energy Gaps (Band Gap), Obtained from the Difference between the Anodic and Cathodic Peak Potentials, Compared with
Those Calculated by DFT Calculations (∆E H-L), Are Also Reported
oxidation
reduction
compound
MW
EA/V
Ep - Ep/2/V
Ep,f - Ep,b/V
E1/2/V
Ec/V
Ep - Ep/2/V
Ep,f - Ep,b/V
E1/2/V
band gap
∆E H-L
1a
1b
1c
1d
2a
2b
2c
3a
3b
4
839.74
868.82
870.75
825.4
939.86
968.94
970.87
977.93
853.2
1.42
1.33
1.48
1.39
1.42
1.33
1.51
1.43
1.37
1.45
1.39
1.48
0.053
0.058
0.061
0.061
0.054
0.059
1.39
-1.32
-1.36
-0.49
-1.30
-1.32
-1.37
-0.51
-1.22
-1.43
-1.22
-1.36
-1.41
0.059
0.074
0.041
0.046
0.058
0.062
0.041
0.065
0.053
0.059
0.060
0.055
-0.063
-0.068
-0.071
-0.067
-0.064
-0.069
-0.047
-0.050
-0.058
-0.060
-0.059
-0.086
-1.28
-1.31
-0.48
-1.28
-1.28
-1.34
-0.49
-1.20
-1.40
-1.20
-1.32
-1.37
2.74
2.69
1.97
2.69
2.74
2.70
2.02
2.65
2.80
2.67
2.75
2.89
3.16
3.24
2.08
3.18
3.16
3.24
2.07
3.15
3.27
3.14
3.18
3.34
0.070
0.063
0.060
1.44
1.36
1.40
0.061
0.056
0.055
0.060
0.060
0.060
0.068
0.068
0.069
0.052
0.062
0.094
1.46
1.39
1.35
1.42
1.35
1.43
1080.12
853.2
853.2
5
6
Table 3. Selected Optimized Bond Lengths (Å) and Angles (deg) for Complexes 1-7 (Optimized Parameters for the Triplet-State Geometries Are
Reported in Parentheses for Selected Complexes)
Ir-Nppy
2.059
(2.057)
2.058
2.060
2.060
(2.056)
2.058
(2.058)
2.058
(2.119)
2.059
2.058
Ir-Cppy
2.009
(1.980)
2.010
2.009
2.009
(1.989)
2.009
(1.980)
2.009
(1.973)
2.009
2.011
Cppy-Cppy
1.451
(1.447)
1.451
1.451
1.451
(1.449)
1.451
(1.446)
1.451
(1.441)
1.451
1.449
1.451
Cppy-Cppy
-
Ir-Nphen
2.165
(2.177)
2.157
2.158
2.164
(2.170)
2.155
(2.170)
2.162
(2.208)
2.150
2.247
Cphen-Cphen
1.428
(1.406)
1.425
1.431
1.428
(1.407)
1.429
(1.408)
1.431
Cppy-Ir-Nppy
80.2
Nphen-Ir-Nphen
76.7
1a
(81.1)
80.2
80.2
80.3
(81.0)
80.2
(81.05)
80.2
(80.7)
80.3
80.2
80.2
(76.3)
76.4
76.4
76.7
(76.5)
76.5
(76.1)
76.3
(74.5)
76.7
76.0
76.6
1b
1c
1d
3a
3b
(1.416)
1.425
1.436
5
6
7
2.060
2.010
2.163
1.429
Ir-Nppy
2.057(6)
2.057(6)
2.056(5)
Ir-Npq
2.1275
(2.117)
2.117
Ir-Cppy
2.015(7)
1.997(6)
2.014(6)
Ir-Cpq
1.9995
(2.119)
2.000
Ir-Nbpy
2.151(6)
2.262(5)
2.217(5)
Ir-Nphen
2.180
(2.208)
2.177
2.181
Cbpy-Cbpy
Cppy-Ir-Nppy
80.1(3)
84.6(2)
Nbpy-Ir-Nbpy
76.3(3)
75.44(18)
X-raya
X-rayb
-
-
-
Cpq-Cpq
1.446
(1.442)
1.446
1.446
1.445
Cphen-Cphen
1.426
(1.412)
1.424
Cpq-Ir-Npq
79.65
Nphen-Ir-Nphen
75.9
2a
(80.7)
79.75
79.7
(74.6)
75.5
75.4
2b
2c
4
2.119
2.128
2.001
2.000
1.429
1.428
2.178
79.8
75.9
(2.119)
(1.973)
(1.441)
(2.208)
(1.416)
(80.7)
(74.5)
b
aX-ray data refer to those reported for [Ir(ppy)2(4,4′-C9H19-2,2′-bipyridine)]+ in ref 9a. X-ray data refer to those reported for [Ir(ppy)2(6,6′-diphenyl-
tBu-bipyridine)]+ in ref 9b.
(810 mV). In this latter case, however, the reduction process
might be centered on the NO2 group.42
ergy gap was obtained (see Table 2 and the Supporting
Information).
Furthermore, the substitution of ppy with pq (compare
1a-c and 3a with 2a-c and 4, respectively) more signifi-
cantly shifts, as expected, the oxidation potentials to positive
values (shift of 20-30 mV), while the reduction potentials
are shifted to slightly more negative values (shift of only 10
mV), i.e., in both cases toward the higher energy required
for the oxidation or the reduction, respectively. These effects
could be ascribed to a screening effect of the bulkier pq
ligand (with respect to the ppy one) on the electron transfer,
which must be more relevant on the oxidation process that
involves both the cyclometalated ligands and the Ir center.
Finally, a good linear correlation (r2 ) 0.98) between the
experimental and the DFT-calculated HOMO-LUMO en-
DFT and TDDFT Calculations. To provide more detailed
insight into the electronic energy levels and nature of the
electronic transitions controlling the electrochemical and
optical properties, we performed DFT calculations on all the
cationic iridium complexes investigated experimentally and
also on complex 7. We optimized the singlet ground-state
geometries of all the various cationic complexes and the
lowest triplet excited-state geometries of 1a, 1d, 2a, 3a, 3b,
and 4 complexes. Selected optimized geometrical parameters
of the ground and some triplet states are collected in Table
3 and compared to the available structural data obtained from
an X-ray investigation of the related complex [Ir(ppy)2(4,4′-
C9H19-2,2′-bipyridine)]+ 9a and Ir(ppy)2(6,6′-diphenyl-tBu-
bipyridine)]+.9b Even though these complexes carry the
bipyridine ligand instead of the phenanthroline ligand, our
optimized geometrical parameters are in excellent agreement
(42) (a) Annoni, E.; Pizzotti, M.; Ugo, R.; Quici, S.; Morotti, T.; Bruschi,
M.; Mussini, P. Eur. J. Inorg. Chem. 2005, 3857. (b) Morotti, T.;
Pizzotti, M.; Ugo, R.; Quici, S.; Bruschi, M.; Mussini, P.; Righetto,
S. Eur. J. Inorg. Chem. 2006, 1743.
Inorganic Chemistry, Vol. 46, No. 21, 2007 8541

Dragonetti et al.
Figure 4. Schematic representation of the frontier molecular orbitals of complexes 1a-d and 3a in acetonitrile solution
with the X-ray data. For instance, for the ppy complexes,
we calculated Ir-N(ppy), Ir-C(ppy), and Ir-N(phen) bond
distances of 2.057-2.060, 2.008-2.011, and 2.150-2.166
Å compared to X-ray experimental values9 of 2.057(6),
2.015(7), and 2.151(6) Å, respectively. Interestingly, for
complex 6, we evidenced a significantly lengthened Ir-
N(phen) bond distance (2.25 vs 2.16 Å for 6 and 1d/3b,
respectively). This effect is due to the presence of two methyl
substituents in the 2 and 9 positions, which sterically perturbs
the metal coordination sphere, as also shown by our
electrochemical investigation, thus producing a strain on the
Ir-N(phen) bond. A similar effect was found for the X-ray
structure of Ir(ppy)2(6,6′-diphenyl-tBu-bipyridine)]+,9b due
to the presence in this case of bulky phenyl substituents in
the 2 and 9 positions. Bond angles show minimal variations
among the complexes investigated and are in good agreement
with X-ray experimental values.9 Compared with the ppy
complexes, the corresponding pq complexes show longer Ir-
N(pq and phen) and shorter Ir-C(pq) bond distances. This
effect is magnified for the Ir-N bonds, which are on average
ca. 0.05 Å longer than those of the corresponding ppy
complex as a result of the hindrance exerted by the more
bulky pq ligand, as clearly shown in the case of the sterically
crowded complex 4 (see Table 3). Interestingly, for some
complexes, we have evidenced a correlation between the
steric hindrance around the metal center due to the ppy ligand
and the ratio of the deaerated/not deaerated excited-state
lifetimes (see notes of Table 1). Indeed, the more sterically
crowded complex 4 shows a decreased deaerated/not deaer-
ated lifetimes ratio (3.57) compared with 3a (3.97), which
in turn has a lower ratio than, e.g., 1d (6.89), which is the
less sterically hindered complex among those investigated.
This behavior might be due to the shielding of the metal
coordination sphere by more bulky ligands, which in turn
protect the metal center of the octahedral complex in the
excited state from interacting with triplet molecular oxygen.
On the optimized geometries, we performed single-point
calculations in dichloromethane and acetonitrile solutions.
The results, in terms of frontier orbitals, are reported in
Figures 4-6 and in the Supporting Information. The two
solvents were chosen because they are those experimentally
employed for the photophysical and electrochemical mea-
surements, respectively. In Figure 4, for the complexes
bearing the ppy ligands, the effect of a substituent in position
5 of the phenanthroline (1a-c) and of two phenyl substit-
uents in positions 4 and 7 (3a) is reported together with the
frontier orbitals of the reference complex 1d.
For complexes 1a-d and 3a, the HOMO is, as found for
the related Ir(III) complexes, an antibonding combination
of Ir(t2g) and ppy(π) orbitals.4,10,13,15,16 The HOMO is
followed, in order of decreasing energy, by two additional
combinations of Ir(t2g) and ppy(π) orbitals, with decreasing
ppy character. However, for complex 1b, the HOMO-1 is a
π-bonding orbital of the NMe2-substituted phenanthroline,
which lies close to the HOMO, as a result of the increased
electron donation of the NMe2 group on the phenanthroline
ring. This low-energy orbital, also obtained for the related
pq complex 2b, might be responsible for the complex
8542 Inorganic Chemistry, Vol. 46, No. 21, 2007

Cationic Iridium(III) Complexes
Figure 5. Schematic representation of the frontier molecular orbitals of complexes 2a-c and 4 in acetonitrile solution
oxidation peaks observed experimentally (Table 2). The
LUMO of all the complexes 1a-d and 3a is a π* orbital
localized on the phenanthroline ligand acting as an accep-
tor.4,10,13,15,16 For these complexes, with the exception of 1c
(see below), the LUMO is followed by a second phenan-
throline π* orbital (LUMO+1), whose energy relative to the
LUMO depends upon the substituent, and at higher energies
by a couple of ppy π* orbitals (LUMO+2 and LUMO+3)
almost degenerate, irrespective of the substituent on the
phenanthroline ligand. This electronic structure is not affected
by a change of solvent, even though a general energy down-
shift of both occupied and unoccupied orbitals is calculated
for dichloromethane compared with that of acetonitrile (see
the Supporting Information).
The presence of a methyl group in position 5 does not
substantially affect the electronic structure with respect to
the reference complex 1d, while the presence in position 5
of a stronger donor such as the NMe2 group (1b) leads to a
destabilization of the HOMO and to major effects on the
LUMO/LUMO+1 and LUMO+1/LUMO+2 separations.
The pattern of the unoccupied orbitals of complex 3a with
phenyl groups in positions 4 and 7 of the phenanthroline
ligand is similar to that of the reference 1d complex, but
with the smallest LUMO/LUMO+1 and largest LUMO+1/
LUMO+2 gaps, as a result of the phenanthroline π* LUMOs
stabilization due to some π conjugation with the phenyl
ligands. For 1c, the strongly electron-withdrawing NO2 group
leads to a dramatic stabilization of all the phenanthroline
π* orbitals, which ultimately leads to a drastic reduction by
more than 1 eV of the HOMO-LUMO gap. The LUMO,
LUMO+1, and LUMO+2 orbitals become phenanthroline
π* orbitals strongly mixed with the NO2 orbitals.
In Figure 5, the electronic structure of the phenanthroline
complexes bearing two pq ligands (2a-c and 4) is reported,
to be compared to the electronic structure of the correspond-
ing complexes with two ppy ligands (1a-c and 3a). The
pattern of the occupied orbitals is essentially the same in
the two series of complexes, with the HOMO being a
combination of Ir(t2g) and pq(π) orbitals. Complex 2b, due
to the presence of a NMe2 group, shows a HOMO-1 of
π-bonding phenanthroline character inserted just below the
HOMO. Also, the pattern of unoccupied orbitals is similar
to that of the corresponding ppy complexes, even though
sizable differences in their relative energies are calculated.
In particular, the LUMO and LUMO+1 phenanthroline π*
orbitals are now almost degenerate, reflecting a more-
symmetrical coordination sphere. In addition, the pq-based
π* LUMO+2/LUMO+3 orbitals are considerably stabilized
Inorganic Chemistry, Vol. 46, No. 21, 2007 8543

Dragonetti et al.
Figure 6. Schematic representation of the frontier molecular orbitals of complexes 3b and 5-7 compared to those of complex 1a in acetonitrile solution.
with respect to the corresponding ppy-π* LUMO+2/
LUMO+3 orbitals, lying about 0.4-0.5 eV below, as an
effect of the more-extended π delocalization of the pq ligand.
Such an effect is particularly pronounced in the case of the
NO2-substituted complex 2c, for which extensive mixing of
π* phenanthroline and π* pq orbitals is calculated.
Information), even though a quantitative agreement between
the HOMO-LUMO gap and the energy difference between
redox potentials is not to be expected. In fact, we are not
taking into account the relaxation of the electronic structure
upon the population/depopulation of the unoccupied/occupied
orbitals (see the Supporting Information).
The effect of the number of methyl substituents and of
their position on the phenanthroline ligand is reported in
Figure 6. Moving the methyl substituent from the 5 (1a) to
the 4 (7) position does not significantly affect the electronic
structure. Also, for complex 5, with two methyl groups in
positions 5 and 6, we notice that, compared to 1a, no
substantial change of the electronic structure occurs, apart
from a slightly destabilization of the LUMO. For complex
3b, dimethylated in positions 4 and 7, a slight LUMO
destabilization is calculated, which brings the LUMO and
LUMO+1 orbitals to become almost perfectly degenerate,
and leads to a HOMO-LUMO gap increase of ca. 0.1 eV.
Complex 6 shows a sizable stabilization of the HOMO,
reflecting the effect on the t2g metal orbitals accomplished
by the presence of two methyl substituents in positions 2
and 9, which reduces the donation from the phenanthroline
ligand, leading to substantial elongated Ir-N bond. The final
result is an overall increase of the HOMO-LUMO gap and
an increase of the LUMO/LUMO+1 splitting.
For complexes 1a, 1d, 3a, and 3b, with the higher quantum
yields (Table 1), we simulated the absorption spectra by
computing the lowest 70 excitation energies by TDDFT in
dichloromethane solution, thus allowing us the simulation
of a large portion of the spectrum, below 250 nm. Given
the similarity of the absorption spectra for these complexes,
we discuss in detail only the absorption spectrum of 1a (see
the Supporting Information for complexes 1d, 3a, and 3b).
It has to be stressed here that, even though singlet-triplet
excitations can be calculated, their oscillator strengths are
set to zero due to the neglect of spin-orbit coupling in the
TDDTF calculations, so that these transitions do not con-
tribute to the overall spectral profile. This should represent
an acceptable approximation in the high-energy region
dominated by the more intense singlet-singlet transitions,
while the contribution of singlet-triplet excitation might
become relevant in the low-energy portion of the spectrum.
A comparison of the calculated and experimental spectra
for complex 1a in dichloromethane solution is reported in
Figure 7. As can be observed, the agreement between theory
and experiment is good, both in terms of absolute band
positions and relative intensities, especially considering the
Comparison of the calculated HOMO-LUMO gap with
the electrochemical energy difference between redox poten-
tials shows good correlation (Table 2 and Supporting
8544 Inorganic Chemistry, Vol. 46, No. 21, 2007

Cationic Iridium(III) Complexes
Figure 7. Comparison between the experimental (red line) and computed (blue line) absorption spectra of the 1a complex in dichloromethane solution.
Inset: detail of the spectra in the 300-550 nm range. The experimental spectrum has been rescaled so that the intensity of the 268 nm band of the experimental
spectrum matches the theoretical main feature.
rather limited dimensions of the basis set and the neglect of
spin-orbit coupling in the excitation energies. The experi-
mental absorption spectrum shows an intense feature at 268
nm with a shoulder at 333 nm. At longer wavelengths, a
less-intense absorption band is found at 377 nm, with a
shoulder at 410 nm and a tail extending above 450 nm. The
calculated absorption spectrum nicely reproduces the main
features of the experimental one, with an overall deviation
within 0.15 eV between the main calculated and experimental
values. In particular, we calculate an intense band at 260
nm with a shoulder at 322 nm, in excellent agreement with
the 268 and 333 nm experimental values. At longer wave-
lengths, we calculate two overlapping absorptions at 366 and
399 nm, which can be related to the 377 nm band of the
experimental absorption spectrum and to its 410 nm shoulder,
respectively.
Inspection of the TDDFT eigenvectors allows the assign-
ment of the main spectral features in terms of their singly
occupied to virtual molecular orbital contributions. The
lowest excitation energies of low intensity for 1a are
calculated to be of singlet-triplet character and are located
at 524 and 466 nm, with the corresponding lowest singlet-
singlet excitations, of negligible intensity, found at 519 and
465 nm. The 524 and 519 nm excitations correspond to
HOMO-LUMO transitions, followed at higher energy by
the 466 and 465 nm excitations of HOMO-1-LUMO
character. These two couples of low-intensity transitions
therefore have MLCT character, with arriving states localized
on the phenanthroline ligand. The lowest-energy transitions
of sizable intensity are calculated at 404 and 395 nm (f )
0.064 and 0.064, respectively); the 404 nm excitation is of
HOMO-LUMO+2 character and is therefore still a MLCT
transition, with arriving states localized on the phenylpyridine
ligands, while the 395 nm feature has mainly HOMO-1-
LUMO character, therefore being a MLCT transition with
arriving states on the phenanthroline ligand. These latter two
transitions give rise to the feature of the simulated spectrum
at 399 nm, which can be related to the 410 nm shoulder
found experimentally. The band experimentally found at 377
nm appears to be related to an intense transition calculated
at 361 nm (f ) 0.101) composed by almost equal contribu-
tions of HOMO-2-LUMO+1 and HOMO-3-LUMO ex-
citations, therefore showing substantial mixing of MLCT and
π-π* character, with arriving states localized on the
phenanthroline ligand. Moving to higher energies, the MLCT
contribution decreases, so the band measured at 268 nm
corresponds to a series of intense, almost overlapping π-π*
excitations of both the phenanthroline and phenylpyridine
ligands, giving rise to an overall feature at 260 nm.
To gain insight into the nature of the emitting excited states
of the investigated complexes, we performed TDDFT
calculations in a dichloromethane solution on the optimized
lowest triplet excited-state geometry of some selected
compounds; in particular, we investigated complexes 1a, 1d,
2a, 3a, 3b, and 4. The triplet-optimized geometries are not
too different for all the above complexes from those of the
ground singlet state (see Table 3). Taking 1a as an example,
the largest difference is observed for the C-C(phen) bond
distances connecting the two pyridine units of the phen
moiety, which decreases from 1.428 to 1.406 Å. This is due
to the bonding interaction between the p orbitals of the two
connecting carbon atoms characterizing the LUMO orbital.
Slight differences are computed also for the Ir-N(phen) and
Ir-C(ppy) bond distances which, respectively, increase and
decrease compared to their value in the ground-state geom-
etry. The reinforced Ir-C interaction in the excited state is
due to the depopulation of the HOMO orbital based on Ir
and ppy contributions, which in turn weakens the Ir-N(phen)
Inorganic Chemistry, Vol. 46, No. 21, 2007 8545

Dragonetti et al.
Table 4. Lowest Singlet-Singlet (S) and Singlet-Triplet (T) Excitation Energies (nm; eV) and Composition in Terms of Molecular Orbital Excitations
for 1a, 1d, 2a, 3a, 3b, and 4 Calculated at the Triplet-State-Optimized Geometriesa
Em. exp
(nm; eV)
T1
T2
T3
478; 2.593
S1
592; 2.093 (0.0003) 486; 2.551 (0.0001) 429; 2.893 (0.0785)
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
585; 2.121 (0.0003) 485; 2.554 (0.0000) 422; 2.937 (0.0765)
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
572; 2.167 (0.0008) 512; 2.421 (0.0010) 505; 2.455 (0.0762)
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
S2
S3
1a 559; 2.218 600; 2.067
488; 2.543
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
1d 575; 2.156 594; 2.086 488; 2.539 478; 2.592
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
2a 556; 2.230 582; 2.130 555; 2.235 544; 2.279
HOMO-LUMO HOMO-LUMO+2 HOMO-LUMO+1/
HOMO-LUMO+3
479; 2.587
3a 592; 2.094 600; 2.068
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
3b 552; 2.246 565; 2.195 501; 2.475 483; 2.565
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
562; 2.206 580; 2.138 560; 2.215 558; 2.224
501; 2.472
590; 2.102 (0.0004) 498; 2.489 (0.0001) 427; 2.904 (0.0748)
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
557; 2.228 (0.0004) 498; 2.488 (0.0000) 432; 2.868 (0.0782)
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
567; 2.186 (0.0015) 522; 2.373 (0.0004) 508; 2.440 (0.0711)
HOMO-LUMO HOMO-LUMO+1 HOMO-LUMO+2
4
HOMO-LUMO HOMO-LUMO+2 HOMO-LUMO+1/
HOMO-LUMO+3
a The oscillator strengths of the singlet-singlet transitions are reported in parentheses. Experimental values (Em. exp) of the emission maxima are reported
for comparison.
bonds trans to the Ir-C bonds. The singlet ground-state
electronic structure calculated at the triplet geometry shows
the same orbital pattern calculated for the ground-state
geometry, even though changes in the orbital energies take
place.
compare complex 8, bearing a doubly substituted bipyridine
ligand with alkyl groups, with 3b, which shows a phenan-
throline with two alkyl substituents in the same positions.
Complex 8 in acetonitrile shows a slightly red-shifted
emission wavelength compared with 3b in dichloromethane,
581 vs 552 nm, respectively, in line with DFT/TDDFT
calculations performed in acetonitrile on 8,15a,d showing a
lowest S0-T1 transition at 583 nm, as compared with the
565 nm value calculated here for 3b in dichloromethane.
The slightly red-shifted emission observed for 8 is entirely
due to the slight LUMO stabilization (by 0.11 eV) compared
to 3b, the HOMO lying at the same energy in both species
in acetonitrile solution. Relevant differences between 3b and
8 are found, on the other hand, when looking at the emission
lifetimes, even though the two complexes show similar
quantum yields (0.238 vs 0.26 for 8 and 3b, respectively)
and the two data sets have been obtained in different solvents;
for 3b, we measure an emission lifetime of 1.43 µs, while a
value of 0.557 µs was reported for 8.8 Given the similar
quantum yields, this difference translates into a comparable
difference in the radiative and nonradiative rate constants
(1.82 and 5.17 vs 4.31 and 13.6 105 s-1 for 3b and 8,
respectively). Since in both complexes the lowest excited
state is of 3MLCT character, originating from the same Ir-
ppy HOMO, the observed differences might reside in the
different solvents used in the experiments and/or in the
different Franck-Condon factors exhibited by metallorganic
complexes bearing phenanthroline and bipyridine ligands.40a
The energy and character of the three computed lowest
excited singlet states S1, S2, and S3 and triplet states T1, T2,
and T3 for 1a, 1d, 2a, 3a, 3b, and 4 are reported in Table 4.
At the triplet-optimized geometries, the lowest excited state
is of 3MLCT character (S0-T1 in Table 4), involving a single
orbital excitation from the Ir(t2g)/ppy(π)- or Ir(t2g)/pq(π)-
based HOMO orbital to the phen π*-based LUMO orbital.
The lowest singlet excited state (S0-S1) has the same MLCT
3
character and lies close to the MLCT state, showing in all
cases negligible oscillator strength. The comparison of
calculated lowest excitation energies with experimental
emission maxima is not straightforward, since the latter are
broad and structureless and it is therefore quite difficult to
localize the E0-0 transition, which can be estimated by the
lowest excitation energy calculated by TDDFT. Nevertheless,
the calculated T1 excitation energies are in good agreement
with experimental emission maxima: the largest discrepancy
is calculated for 1a, for which a slight red-shifted value
(0.151 eV) is calculated (see Table 4) and small discrepancies
(between 0.02 and 0.06 eV) in the energy trends between
calculated and measured values are found. These deviations,
which are well below the accuracy of the theoretical setup,
might also be related to an inhomogeneous comparison
between theory and experiment, as discussed above. We
notice, in general, that a substantial stabilization of the
excited states takes place at the triplet-optimized geometry,
compared with the singlet ground state ones. For the lowest
singlet and triplet excited states, this stabilization energy
amounts to ca. 0.3 eV while more limited variations, within
ca. 0.1 eV, are calculated for the second lowest excitation
energies.
Conclusions
This work shows how the nature and number of substit-
uents on the phenanthroline can significantly influence some
of the photochemical properties of [Ir(ppy)2(substituted 1,-
10-phen)][PF6] (ppy ) 2-phenylpyridine) and [Ir(pq)2-
(substituted 1,10-phen)][PF6] (pq ) 2-phenylquinoline)
complexes, a promising class of cationic Ir(III) complexes
for electroluminescent devices. While the absorption and
emission energies are slightly affected by substitution on the
phenanthroline ligand, quantum yields and emission lifetimes
are affected by significant changes. The highest quantum
yields (Φ ) 34-38%) are obtained with 5-methyl-1,10-
Finally, it is interesting to compare our results, obtained
in the context of the present investigation, with the experi-
mental8 and theoretical results15a,d reported for the prototype
[Ir(2-phenylpyridine)2(4,4′-tert-butyl-2,2′-bipyridine)]+ com-
plex, hereafter 8: in particular, it seems worthwhile to
8546 Inorganic Chemistry, Vol. 46, No. 21, 2007

Cationic Iridium(III) Complexes
phenanthroline, while negligible values are obtained by
substitution in position 5 of the methyl group with an
electron-donor group such as NMe2 (Φ ) 0.3-6%) or an
electron-withdrawing substituent such as NO2 (Φ < 0.1%).
The presence of two methyl groups or of two π-delocalized
phenyl substituents on the phenanthroline ligand leads to
lower quantum yields (Φ ) 20-26%) but still higher than
that of the reference complex bearing 1,10-phenanthroline
(Φ ) 7%). The investigated complexes display only a slight
color tunability, with the emissions ranging from 534 to 592
nm, but an emissive lifetime ranging from 1.27 to 4.03 µs
and from 0.19 to 0.82 µs under deaerated and non-deaerated
conditions, respectively. Interestingly, the more sterically
crowded complex 4 shows the lowest deaerated/not-deaerated
lifetimes ratio, due to steric hindrance toward the interaction
with triplet oxygen. Detailed theoretical calculations and
electrochemical investigations show a nice correlation of the
trend of the HOMOs and LUMOs energy levels of the
complexes investigated. TDDFT calculations of the lowest
singlet and triplet excited states at the singlet ground-state
and triplet excited-state geometries have allowed us to assign
the main bands of the absorption spectra to π-π* transitions
centered on the π-delocalized ppy, pq, and phen ligands
(higher energy) and to the MLCT with arriving states on
the phenanthroline ligand (lower energy), with the emission
(iii) Replacing the ancillary phenylpyridine ligand by
phenylquinoline does not affect the nature and composition
of the lowest excited states; the only effect is a stabilization
of the phenylquinoline-based LUMO+2/LUMO+3, which
eventually become almost isoenergetic with the phenanthro-
line-based LUMO/LUMO+1.
(iv) The overall quantum yields of complexes carrying
methyl or phenyl substituents on the phenanthroline ligand
are mainly controlled by nonradiative processes of decay,
since rather limited variations of kr are measured, as opposed
to a wider knr variation range. The limited kr variation is in
agreement with the modest electronic-structure change
observed for these complexes. The knr variability seems
therefore not to be related to major changes of the electronic
structure but rather to significant perturbation of the phenan-
throline ring deformation motions either by the different
nature of the substituents or by their positions on the
phenanthroline ligand. However, the presence of a strong
electron donor like NMe2 or of a strong electron-withdrawing
substituent produces a quenching of the quantum yield, which
could be related partially to the significant perturbation of
the electronic levels.
Overall, our work has shown that substitution on the
phenanthroline ligand in cationic Ir(III) complexes of type
[Ir(ppy)2(phen)]+ and [Ir(pq)2(phen)]+ does not influence the
emission color very much but strongly influences quantum
yields. Our work also provides evidence that the dominant
role of the nonradiative pathway of decay is strongly
dependent upon the nature and type of substitution.
3
band originating from a MLCT excited state localized on
the phenanthroline ligand.
Some general observations can be finally made, based
upon our combined experimental and theoretical investiga-
tions:
(i) Weak electron-donating substituents like the methyl
group do not significantly alter the geometry and the
electronic structure of the Ir(III) cationic complexes inves-
tigated in this work; an exception is represented by complex
6 with two methyl groups in positions 2 and 9 of the
phenanthroline ligand, for which steric hindrance to the metal
coordination alters the metal-to-phenanthroline bond dis-
tances and consequently perturbs the electronic structure, as
clearly evidenced also by the electrochemical investigation.
The substitution of methyl groups by phenyl groups in
positions 4 and 7 (e.g., complexes 3a and 4) does not perturb
the electronic structure, in particular the pattern of the LUMO
orbitals, although the LUMO/LUMO+1 and LUMO+1/
LUMO+2 gaps are slightly affected.
(ii) Strong electron-withdrawing substituents, such as the
NO2 group, lead to extensive perturbation of the phenan-
throline π* levels, which is reflected in a strong LUMO
stabilization. Strong electron-donor substituents, such as
NMe2, destabilize the π-bonding levels of the phenanthroline
ligand, which shift quite close to the HOMO. This is clearly
reflected by the more complex oxidation behavior observed
in the electrochemical investigation of complexes 1b and
2b. In addition, the presence of a NMe2 group leads not only
to the destabilization of the HOMO but also to major effects
on the LUMO/LUMO+1 and LUMO+1/LUMO+2 gaps as
expected for a significant perturbation of both the π and π*
levels of the phenanthroline ligand.
Acknowledgment. We deeply thank Prof. Camilla Fer-
rante (Universita` di Padova) for fruitful discussions and Drs.
Martina Bonacina, Li Cong, Francesco Morosini, and Matteo
Parravicini for experimental help. This work was supported
by the Ministero dell’Istruzione, dell’Universita` e della
Ricerca (Programma di ricerca FIRB, year 2003; research
title, Molecular compounds and hybrid nanostructured ma-
terials with resonant and non-resonant optical properties for
photonic devices) and by CNR (PROMO 2006; research title,
Sistemi molecolari e nanodimensionali con proprieta` funzi-
onali: Nanostrutture organiche, organometalliche, polimer-
iche ed ibride; ingegnerizzazione supramolecolare delle
proprieta` fotoniche e dispositivistica innovativa per opto-
elettronica).
Supporting Information Available: Energies of the frontier
molecular orbitals and energy differences between some frontier
molecular orbitals for complexes 1-7 in CH3CN solution and in
CH2Cl2; isodensity plots of selected molecular orbitals of 1b-d
and 2b complexes; comparison between the experimental and
computed UV-vis absorption spectra of 1a, 1d, 3a, and 3b
complexes; cyclovoltammetric patterns of compounds 2a-c and
4-6; graph of Eem vs ln(knr) for 1a, 1d, and 2a; graph of the
experimental vs the calculated HOMO-LUMO energy gaps. This
materialisavailablefreeofchargeviatheInternetathttp://pubs.acs.org.
IC700414Z
Inorganic Chemistry, Vol. 46, No. 21, 2007 8547