Role of substitution on the photophysical properties of 5,5′-diaryl-2,2′-bipyridine (bpy*) in [Ir(ppy) 2(bpy*)]PF6 complexes: A combined experimental and theoretical study

Inorg. Chem. 2010, 49, 5625–5641 5625

DOI: 10.1021/ic100521t

Role of Substitution on the Photophysical Properties of 5,5⁰-Diaryl-2,2⁰-bipyridine

(bpy*) in [Ir(ppy)₂(bpy*)]PF₆Complexes: A Combined Experimental and

Theoretical Study

ꢀ

Sebastien Ladouceur, Daniel Fortin, and Eli Zysman-Colman*

ꢀ

Departement de chimie, Universite de Sherbrooke, 2500 Boul. Universite, Sherbooke, Quebec, Canada J1K 2R1

Received March 18, 2010

The synthesis of a family of 4⁰-functionalized 5,5⁰-diaryl-2,2⁰-bipyridines (bpy*; 6a-6g) is reported. These ligands

were reacted with the dimer [(ppy)₂IrCl]₂(ppyH = 2-phenylpyridine) and afforded, after subsequent counterion

exchange, a new series of luminescent cationic heteroleptic iridium(III) complexes, [(ppy)₂Ir(bpy*)]PF₆(8a-8g).

These complexes were characterized by electrochemical and spectroscopic methods. The crystal structures of two of

these complexes (8a and 8g) are reported. All of the complexes except for 8c and 8f exhibit intense and long-lived

emission in both 2-MeTHF and ACN at 77 K and room temperature. The origin of this emission has been assigned by

computational modeling to be an admixture of ligand-to-ligand charge-transfer [³LLCT; π(ppy) f π*(bpy*)] and

metal-to-ligand charge-transfer [³MLCT; dπ(Ir) f π*(bpy*)] excited states that are primarily composed of the former.

The luminescent properties for 8a-8c are dependent upon the functionalization at the 4⁰position of the aryl

substituents affixed to the diimine ligand, while those for 8d-8g are essentially independent because of an electronic

decoupling of the aryls and bpy due to the substitution of o,o-dimethyl groups on the aryls, causing a near 90ꢀ angle

between the aryl and bipyridyl moieties. A combined density functional theory (DFT)/time-dependent DFT study was

conducted in order to understand the origin of the transitions in the absorption and emission spectra and to predict

accurately emission energies for these complexes.

Introduction

coupling (ζ_Ir= 3909 cm^-1).¹³This, coupled with the use of

strong-field cyclometallating ligands (C^∧N) such as phenylpyri-

dine (ppyH), results in long-lived and highly luminescent¹⁴

complexes that usually emit from metal-to-ligand charge trans-

Since the turn of the millennium, there has been a near-

exponential interest in the study of the photophysical pro-

perties¹of iridium(III) complexes and their applications in

diverse fields such as organic light-emitting diodes (OLEDs)²

and light-emitting electrochemical cells (LEECs) for next-

generation visual displays,^2,3catalysts for photoinduced clea-

vage of water for hydrogen production,^3v,4biolabeling⁵and

chemosensors⁶for oxygen,⁷organic salts,⁸and ions such as

3

fer (³MLCT), ligand-centered (³LC), or mixed MLCT/³LC

states,^14b,15while rendering the thermal population of none-

3

missive metal-centered MC states all but impossible, unlike

Ru(N^∧N)₃complexes [N^∧N = dimine such as bipyridine

(bpy)].¹⁶

Heteroleptic iridium complexes of the form [(C^∧N)₂Ir-

(N^∧N)]^þare of particular interest because of their facile

capacity to finely modulate the emission energy through

independent modification of the cyclometallating and ancil-

lary ligands. Selective chemical substitution about the C^∧N

ligands modulatesthe ³LC state, while that of the N^∧N ligand

controls the mixed ³MLCT/³LC state.^{3g,j,x,15a,b,17}More

specifically, placement of electron-donating groups on the

ancillary ligand leads to an overall destabilization of the

lowest unoccupied molecular orbital (LUMO) and a result-

ing blue shift in the emission spectrum.¹⁸Furthermore, these

complexes possess reversible electrochemistry, are thermally

stable, and can be easily accessed in two steps from IrCl₃.^3b,19

One particular disadvantage to charged heteroleptic com-

plexes compared to their neutral homoleptic congeners of the

form Ir(C^∧N)₃is their relatively low quantum yields.

H^þ,⁸F^-,⁸Cl^-,⁹Ca^{2þ 10}Hg^{2þ 11}and Ag^þ.¹²The increased

,

interest results from the privileged photophysical and electro-

chemical properties of iridium(III) complexes resulting from

both the large ligand-field stabilization of iridium (color

tuning) and the efficient intersystem crossing (high phosphore-

scent quantum yields) between the singlet and triplet excited

states that is made possible by the high degree of spin-orbit

*To whom correspondence should be addressed. E-mail: Eli.Zysman-Colman@

usherbrooke.ca.

(1) (a) Dixon, I. M.; Collin, J.-P.; Sauvage, J.-P.; Flamigni, L.; Susana, E.;

Barigelletti, F. Chem. Soc. Rev. 2000, 29, 385. (b) Flamigni, L.; Barbieri, A.;

Sabatini, C.; Ventura, B.; Barigelletti, F. Top. Curr. Chem. 2007, 281, 143. (c)

Ulbricht, C.; Beyer, B.; Friebe, C.; Winter, A.; Schubert, U. S. Adv. Mater. 2009,

21, 4418.

(2) For a review of the science of OLEDs, see:Hung, L. S.; Chen, C. H.

Mater. Sci. Eng. Rep. 2002, 39, 143.

r

2010 American Chemical Society

Published on Web 05/25/2010

pubs.acs.org/IC

5626 Inorganic Chemistry, Vol. 49, No. 12, 2010

Ladouceur et al.

Whereas color tuning is reasonably well understood, it is far

less clear what strategies to use to augment emission quantum

yields. Recently, several examples^{3c,d,i,l,18,20}of highly lumi-

nescent [min 20% and up to 85% in acetonitrile (ACN)]

cationic heteroleptic complexes have been reported, but

these, unsurprisingly, mostly emit at high energy in the

green-to-blue region, adhering to the energy gap law that

describes the dependence of nonradiative decay processes

(and thus the quantum yield) on the emission energy of the

luminophore.²¹A key challenge in LEEC design is to develop

strategies that enable controlled color tuning, especially inthe

red, green, and blue regions, without adversely affecting

phosphorescent quantum yields. Architectures that can meet

this challenge would thus be highly desirable.

In this paper, we report the synthesis of a new family of

5,5⁰-diarylbipyridines (bpy*) and their incorporation as an-

cillary ligands into luminescent cationic iridium complexes of

the general formula [(ppy)₂Ir(bpy*)]PF₆. This family is sub-

divided into two series: the first series contains aryl groups

substituted with electron-releasing groups at the 4⁰position,

while the second series also possesses methyl groups at the 2⁰

and 6⁰positions. The latter series is designed to sequester the

arenes into an orthogonal orientation relative to the diimine.

The two series of isostructural complexes systematically

probe the interplay between the conjugative effects and

electron-releasing power of remote substituents, while simul-

taneously protecting the iridium center due to their steric

effects. These are compared to an archetypal complex,

[(ppy)₂Ir(bpy)]^þ. A detailed photophysical and electroche-

mical investigation is described in order to assess the afore-

mentioned structure-property relationship and its impact on

the quantum efficiency. We will show that the majority of

these complexes emit from an admixture of ligand-to-ligand

charge-transfer (³LLCT)/³MLCT states, while complexes

possessing potent electron-releasing substituents on the aryls

emit from an intraligand charge-transfer (³ILCT) state with a

marked decrease in the quantum yields.

(3) (a) Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Nature 2000, 403,

750. (b) Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest,

S. R. Appl. Phys. Lett. 1999, 75, 4. (c) Nazeeruddin, M. K.; Wegh, R. T.; Zhou,

€

Z.; Klein, C.; Wang, Q.; DeAngelis, F.; Fantacci, S.; Gratzel, M. Inorg. Chem.

2006, 45, 9245. (d) Bolink, H. J.; Cappelli, L.; Coronado, E.; Gratzel, M.; Orti, E.;

Costa, R. D.; Viruela, P. M.; Nazeeruddin, M. K. J. Am. Chem. Soc. 2006, 128,

14786. (e) Slinker, J. D.; Gorodetsky, A. A.; Lowry, M. S.; Wang, J.; Parker, S. T.;

Rohl, R.; Bernhard, S.; Malliaras, G. G. J. Am. Chem. Soc. 2004, 126, 2763. (f)

~

Slinker, J. D.; Bernards, D. A.; Houston, P. L.; Abruna, H. D.; Bernhard, S.;

Malliaras, G. G. Chem. Commun. 2003, 19, 2392. (g) Lowry, M. S.; Hudson,

W. R.; Pascal, R. A., Jr.; Bernhard, S. J. Am. Chem. Soc. 2004, 126, 14129. (h)

Slinker, J. D.; Koh, C. Y.; Malliaras, G. G.; Lowry, M. S.; Bernhard, S. Appl.

Phys. Lett. 2005, 86, 173506. (i) Tamayo, A. B.; Garon, S.; Sajoto, T.; Djurovich,

P. I.; Tsyba, I. M.; Bau, R.; Thompson, M. E. Inorg. Chem. 2005, 44, 8723. (j)

Lowry, M. S.; Bernhard, S. Chem.;Eur. J. 2006, 12, 7970. (k) Su, H.-C.; Wua,

C.-C.; Fang, F.-C.; Wong, K.-T. Appl. Phys. Lett. 2006, 89, 261118. (l) Su, H.-C.;

Fang, F.-C.; Hwu, T.-Y.; Hsieh, H.-H.; Chen, H.-F.; Lee, G.-H.; Peng, S.-M.;

Wong, K.-T.; Wu, C.-C. Adv. Funct. Mater. 2007, 17, 1019. (m) Slinker, J. D.;

A combined density functional theory (DFT) and time-

dependent DFT (TD-DFT) study provides insight into the

nature of the ground (S₀) and excited (S₁and T₁) states with

resulting detailed assignments of the orbitals involved in

absorption and emission transitions. The theoretical analysis

corroborates the experimental study. These will serve to

guide ligand design to develop complexes for better perform-

ing OLEDs for visual display and lighting applications.

~

Rivnay, J.; Moskowitz, J. S.; Parker, J. B.; Bernhard, S.; Abruna, H. D.; Malliaras,

G. G. J. Mater. Chem. 2007, 17, 2976. (n) Neve, F.; LaDeda, M.; Crispini, A.;

Bellusci, A.; Puntoriero, F.; Campagna, S. Organometallics 2004, 23, 5856. (o)

Lee, J. K.; Yoo, D. S.; Handy, E. S.; Rubner, M. F. Appl. Phys. Lett. 1996, 69,

1686. (p) Maness, K. M.; Masui, H.; Wightman, R. M.; Murray, R. W. J. Am.

Chem. Soc. 1997, 119, 3987. (q) Handy, E. S.; Pal, A. J.; Rubner, M. F. J. Am.

Chem. Soc. 1999, 121, 3525. (r) Gao, F. G.; Bard, A. J. J. Am. Chem. Soc. 2000,

122, 7426. (s) Buda, M.; Kalyuzhny, G.; Bard, A. J. J. Am. Chem. Soc. 2002, 124,

~

6090. (t) Bernhard, S.; Barron, J. A.; Houston, P. L.; Abruna, H. D.; Ruglovsky,

J. L.; Gao, X.; Malliaras, G. G. J. Am. Chem. Soc. 2002, 124, 13624. (u)

Rudmann, H.; Shimada, S.; Rubner, M. F. J. Appl. Phys. 2003, 94, 115. (v)

Lowry, M. S.; Goldsmith, J. I.; Slinker, J. D.; Rohl, R.; Pascal, R. A.; Malliaras,

G. G.; Bernhard, S. Chem. Mater. 2005, 17, 5712. (w) Rudmann, H.; Shimida, S.;

Rubner, M. F. J. Am. Chem. Soc. 2002, 124, 4918. (x) Nazeeruddin, M. K.;

(10) Ho, M.-L.; Hwang, F.-M.; Chen, P.-N.; Hu, Y.-H.; Cheng, Y.-M.;

Chen, K.-S.; Lee, G.-H.; Chi, Y.; Chou, P.-T. Org. Biomol. Chem. 2006, 4, 98.

(11) Zhao, Q.; Cao, T.; Li, F.; Li, X.; Jing, H.; Yi, T.; Huang, C.

Organometallics 2007, 26, 2077.

€

Hunmphry-Baker, R.; Berner, D.; Rivier, B. S.; Zuppiroli, L.; Gratzel, M. J. Am.

(12) Schmittel, M.; Lin, H. Inorg. Chem. 2007, 46, 9139.

Chem. Soc. 2003, 125, 8790. (y) Zysman-Colman, E.; Slinker, J. D.; Parker, J. B.;

Malliaras, G. G.; Bernhard, S. Chem. Mater. 2008, 20, 388. (z) Adachi, C.;

Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Appl. Phys. 2001, 90, 5048. (aa)

Bolink, H. J.; Coronado, E.; Costa, R. D.; Ort, E.; Sessolo, M.; Graber, S.; Doyle,

K.; Neuburger, M.; Housecroft, C. E.; Constable, E. C. Adv. Mater. 2008, 20,

3910.

(13) Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, M. T. Handbook of

Photochemistry; Taylor and Francis: Boca Raton, FL, 2006.

(14) (a) Ohsawa, Y.; Sprouse, S.; King, K. A.; DeArmond, M. K.; Hanck,

K. W.; Watts, R. J. J. Phys. Chem. 1987, 91, 1047. (b) Garces, F. O.; King,

K. A.; Watts, R. J. Inorg. Chem. 1988, 27, 3464.

€

(15) (a) Colombo, M. G.; Hauser, A.; Gudel, H. U. Inorg. Chem. 1993, 32,

(4) (a) Tinker, L. L.; McDaniel, N. D.; Curtin, P. N.; Smith, C. K.;

Ireland, M. J.; Bernhard, S. Chem.;Eur. J. 2007, 13, 8726. (b) Goldsmith,

J. I.; Hudson, W. R.; Lowry, M. S.; Anderson, T. H.; Bernhard, S. J. Am. Chem.

Soc. 2005, 127, 7502.

3088. (b) Colombo, M. G.; G€udel, H. U. Inorg. Chem. 1993, 32, 3081. (c)

Colombo, M. G.; Brunold, T. C.; Riedener, T.; Guedel, H. U.; Fortsch, M.; Buergi,

H.-B. Inorg. Chem. 1994, 33, 545.

(16) Goze, C.; Chambron, J.-C.; Heitz, V.; Pomeranc, D.; Salom-Roig,

X. J.; Sauvage, J.-P.; Morales, A. F.; Barigelletti, F. Eur. J. Inorg. Chem.

2003, 2003, 3752.

(17) (a) Avilov, I.; Minoofar, P.; Cornil, J.; De Cola, L. J. Am. Chem. Soc.

2007, 129, 8247. (b) Hay, P. J. J. Phys. Chem. A 2002, 106, 1634. (c) 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. (d) 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. (e) 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. 2005, 44, 1713. (f) Kwon, T.-H.; Cho,

H. S.; Kim, M. K.; Kim, J.-W.; Kim, J.-J.; Lee, K. H.; Park, S. J.; Shin, I.-S.; Kim, H.;

Shin, D. M.; Chung, Y. K.; Hong, J.-I. Organometallics 2005, 24, 1578.

(18) De Angelis, F.; Fantacci, S.; Evans, N.; Klein, C.; Zakeeruddin,

S. M.; Moser, J.-E.; Kalyanasundaram, K.; Bolink, H. J.; Gratzel, M.;

Nazeeruddin, M. K. Inorg. Chem. 2007, 46, 5989.

(5) (a) Lo, K. K.-W.; Ng, D. C.-M.; Chung, C.-K. Organometallics 2001,

20, 4999. (b) Iridium complexes have been incorporated into dendritic structures.

See: Lo, S. C.; Namdas, E. B.; Burn, P. L.; Samuel, I. D. W. Macromolecules

2003, 36, 9721. (c) Lo, K. K.-W.; Chung, C.-K.; Zhu, N. Chem.;Eur. J. 2003, 9,

475. (d) Lo, K. K.-W.; Chung, C.-K.; Lee, T. K.-M.; Lui, L.-H.; Tsang, K. H.-K.;

Zhu, N. Inorg. Chem. 2003, 42, 6886. (e) Lo, K. K.-W.; Chan, J. S.-W.; Chung,

C.-K.; Tsang, V. W.-H.; Zhu, N. Inorg. Chim. Acta 2004, 357, 3109. (f) Lo,

K. K.-W.; Hui, W.-K.; Chung, C.-K.; Tsang, K. H.-K.; Ng, D. C.-M.; Zhu, N.;

Cheung, K.-K. Coord. Chem. Rev. 2005, 249, 1434. (g) Lo, K. K.-W.; Lau,

J. S.-Y. Inorg. Chem. 2007, 46, 700. (h) Lo, K. K.-W.; Zhang, K. Y.; Leung, S.-K.;

Tang, M.-C. Angew. Chem., Int. Ed. 2008, 47, 2213. (i) Lo, K. K.-W.; Chan,

J. S.-W.; Lui, L.-H.; Chung, C.-K. Organometallics 2004, 23, 3108.

(6) Kim, J. I.; Shin, I.-S.; Kim, H.; Lee, J.-K. J. Am. Chem. Soc. 2005, 127,

1614.

(7) Di Marco, G.; Lanza, M.; Mamo, A.; Stefio, I.; Di Pietro, C.; Romeo,

G.; Campagna, S. Anal. Chem. 1998, 70, 5019.

(8) Zhao, Q.; Liu, S.; Shi, M.; Li, F.; Jing, H.; Yi, T.; Huang, C.

Organometallics 2007, 26, 5922.

(19) King, K. A.; Watts, R. J. J. Am. Chem. Soc. 1987, 109, 1589.

(20) Lafolet, F.; Welter, S.; Popovic, Z.; Cola, L. D. J. Mater. Chem. 2005,

15, 2820.

(9) Goodall, W.; Williams, J. A. G. J. Chem. Soc., Dalton Trans. 2000,

2893.

(21) Caspar, J. V.; Kober, E. M.; Sullivan, B. P.; Meyer, T. J. J. Am.

Chem. Soc. 1982, 104, 630.

Article

Inorganic Chemistry, Vol. 49, No. 12, 2010 5627

Scheme 1^a

^a(a) (i) 1.1 equiv of n-BuLi/Et₂O, -78 ꢀC, 30 min; (ii) 1.1 equiv of I₂/Et₂O, -78 ꢀC to RT, 16 h. (b) 1 equiv of NBS/ACN, 0 ꢀC to RT, 16 h. (c) (i) 1.1 equiv of

n-BuLi/THF, -78 ꢀC, 30 min; (ii) 0.7 equiv of ZnCl₂/THF, -78 ꢀC to RT, 30 min; (iii) 1 equiv of 4, 1.6 mol % Pd(PPh₃)₄/THF, reflux, 18 h. (d) (i) 1.1 equiv of

n-BuLi/THF, -78 ꢀC, 30 min; (ii) 1.1 equiv of ZnCl₂/THF, -78 ꢀC to RT, 30 min; (iii) 1 equiv of 5a-5g, 5 mol % Pd(PPh₃)₄/THF, reflux, 18 h.

Table 1. Synthesis of 6a-6g

Results and Discussion

Synthesis. The targeted bpy* ligands were obtained

from the corresponding arene via a three-step protocol

(Scheme 1). Arenes 1b, 1c, and 1e-1g were brominated

regioselectively using N-bromosuccinimide (NBS) in

ACN to afford 2b, 2c, and 2e-2g in excellent yield,

97-99%; 2a and 2d were available commercially.²²These

bromoarenes were then subjected to a palladium-cata-

lyzed Negishi²³cross-coupling reaction with 2-bromo-5-

iodopyridine (4), itself obtained in 94% via iodine quench

of the lithiate formed by a metal-halogen exchange²⁴

of 2,5-dibromopyridine (3). 2-Bromo-5-arylpyridines

5a-5g were then homocoupled under Negishi conditions

to afford bpy* 6a-6g with overall yields ranging from 34

to 73% over the three-step linear sequence (Table 1)

under nonoptimized conditions. Nevertheless, the itera-

tive Negishi protocol proved to be an acceptable, straight-

forward, and general method for accessing functionalized

bpy*. Bipyridines 6a and 6d-6g were all found to be

readily soluble in a plethora of organic solvents; however,

6b and 6c were poorly soluble, most probably because of

increased conjugation.^23b

With 6a-6g obtained, scission of the μ-dichloro-bridged

iridium dimer²⁵7 occurred readily to afford heteroleptic

[(ppy)₂Ir(bpy*)]^þcomplexes (8a-8g) in good-to-excellent

-

yield. These complexes were isolated as their PF₆salts

after anion metathesis of the chloride (Scheme 2). The

archetypal [(ppy)₂Ir(bpy)]PF₆(9) was also synthesized to

serve as an external standard and reference complex. All

complexes were recrystallized (cf. the Supporting Informa-

tion, SI). Characterization of ligands and complexes by

melting point (mp) determination, ¹H NMR and ¹³C NMR

spectroscopy, and low- and high-resolution mass spectro-

metry provided satisfactory microanalysis.

triclinic P1 space group. The average Ir-C_ppy, Ir-N_ppy

,

Solid-State Structures from X-ray Crystallography.

X-ray-quality single-crystal structures, depicted in Figure 1,

for two (8a and 8g) of the seven complexes were obtained

by slow vapor diffusion of tert-butyl methyl ether into a

chlorobenzene solution. Complex 8a crystallizes in an

orthorhombic Pbca space group, while 8g crystallizes in a

and Ir-N_bpybond lengths are unremarkable when com-

pared with other [(ppy)₂Ir(bpy)]^þ-type complexes (1.994 (

˚

0.012, 2.047 ( 0.009, and 2.134 ( 0.017 A, respectively, for

˚

8a; 2.037 ( 0.013, 2.011 ( 0.023, and 2.148 ( 0.010 A,

respectively, for 8g).^3n,5c,e,26The longer Ir-N_bpybond is

(22) Zysman-Colman, E.; Arias, K.; Siegel, J. S. Can. J. Chem. 2009, 87, 440.

(23) (a) Negishi, E. Handbook of Organopalladium Chemistry for Organic

Synthesis; Wiley: New York, 2002. (b) Loren, J. C.; Siegel, J. S. Angew. Chem.,

Int. Ed. 2001, 40, 754.

(26) (a) Neve, F.; Crispini, A.; Campagna, S.; Serroni, S. Inorg. Chem.

1999, 38, 2250. (b) Cheung, K.-M.; Zhang, Q.-F.; Chan, K.-W.; Lam, M. H. W.;

Williams, I. D.; Leung, W.-H. J. Organomet. Chem. 2005, 690, 2913. (c) Neve,

F.; Crispini, A. Eur. J. Inorg. Chem. 2000, 2000, 1039. (d) Lepeltier, M.; Lee,

T. K.-M.; Lo, K. K.-W.; Toupet, L.; Le Bozec, H.; Guerchais, V. Eur. J. Inorg.

(24) (a) Parham, W. E.; Piccirilli, R. M. J. Org. Chem. 1977, 42, 257.

(b) Wang, X.; Rabbat, P.; O'Shea, P.; Tillyer, R.; Grabowski, E. J. J.; Reider, P. J.

Tetrahedron Lett. 2000, 41, 4335.

€

Chem. 2005, 2005, 110. (e) Urban, R.; Kramer, R.; Mihan, S.; Polborn, K.;

(25) Nonoyama, M. Bull. Chem. Soc. Jpn. 1974, 47, 767.

Wagner, B.; Beck, W. J. Organomet. Chem. 1996, 517, 191.

5628 Inorganic Chemistry, Vol. 49, No. 12, 2010

Ladouceur et al.

Figure 1. ORTEP perspective representations of complexes 8a and 8g (ellipsoids at 50% probability). The counterion PF₆^-has been removed for clarity.

Scheme 2^a

^a(a) (i) ethylene glycol, 150 ꢀC, 18 h; (ii) NH₄PF₆(aq).

due to the large trans influence imparted by the cyclo-

metalated carbon of the ppy ligands; the ppy ligands are

oriented in a mutually cis configuration, which is the

same as that in dimer 7.^1bThe overall geometry for each

is that of a distorted octahedron with (C-Ir-N)_avgand

N-Ir-N bite angles of 80.5 ( 0.3ꢀ and 89.9 ( 0.1ꢀ for 8a

and 79.7 ( 0.1ꢀ and 76.8 ( 0.1ꢀ for 8g. The latter

complex is slightly more distorted in order to accom-

modate the bulkier pentamethylphenyl groups attached

to the bpy fragment.

not necessarily translate into perturbations in the photo-

physical properties (absorption and emission). Bernhard

and co-workers had previously demonstrated that the

chloride and hexafluorophosphate salts of a large family

of heteroleptic iridium complexes exhibited nearly iden-

tical emission maxima.^3g

The ¹H NMR spectra of 8a-8c recorded in CDCl₃at

room temperature (RT) exhibit the expected C₂symmetry

(Figure 2).^19,27As the aryl substituent becomes more

electron-rich, the R-proton signal on the bpy is shifted

upfield from 9 to 8.5 ppm. The ¹H NMR spectra of 8d-8f

recorded in CDCl₃at RT exhibit the same trend, but it is

attenuated relative to that observed for 8a-8c because of

the loss of conjugation. Of particular interest is the

apparition of nonequivalent methyl peaks in the aliphatic

region in each of 8d-8g. For instance, all five methyl

groups about the aryl fragment are diastereotopic for 8g.

The presence of only five signals is indicative of a localized

σ_vmirror plane that renders the two aryls attached to the

bpy ligand enantiomorphic. The particular features of

this spectrum can be explained in terms of a ground-state

geometry that situates the arene perpendicular to the bpy

plane in concert with an elevated rotational barrier about

the C_aryl-C_bpybond. A similar set of observations were

made for each of 8d-8f. Here, two singlets were observed

in the aliphatic region, integrating with a ratio of 1:1.

Diastereotopic aryl protons were also clearly differentiated

for 8e, while those for 8f were unresolved but their signals

were broadened. Rotational barriers for 8a, 8d, and 8g were

calculated, and these results will be discussed below.

Solution-State Photophysical Properties. Absorption.

The photophysical properties of 8a-8g were investigated

and compared to those of [(ppy)₂Ir(bpy)]PF₆(9). The

absorption spectra are shown in Figure 3 and the data

The primary difference in structure between the two

complexes resides in the dihedral angles measured be-

tween the plane of the arene and that of the bpy. For 8a,

the average dihedral angle is 36 ( 3ꢀ, while for 8g, it is

60 ( 5ꢀ; the unexpectedly low dihedral angle for 8g is most

likely due to crystal-packing forces. Indeed, the computed

ground-state geometry for 8g predicts a dihedral angle of

ca. 90ꢀ (vide infra). The pseudo-orthogonal disposition in

8g with respect to the mean plane of the bpy is primarily

due to the presence of the o,o-dimethyl groups. The

methyl groups act to impede conjugation and to limit

rotation about the aryl-pyridyl bond. This elevated

barrier to rotation is also evident spectroscopically, and

its determination will be discussed below.

1

NMR Analysis of Solution Conformation. H and ¹³C

NMR spectra were obtained for all ligands and com-

plexes. The ¹H NMR spectra for the latter are shown in

Figure 2. For 8a-8g, each of the aromatic protons was

1

resolved by H NMR. We noted a large upfield shift of

several of these signals upon metathesis of the chloride

salt to the hexafluorophosphate salt, notably the R-pro-

ton designated H₆(bpy) (e.g., for 8b, see Figure S1 in the

SI). Thus, the counterion influences strongly the local

electronic environment around the iridium center, with

the bulkier PF₆^-being more loosely associated with the

iridium complex than the chloride. Such an influence does

(27) Garces, F. O.; Watts, R. J. Magn. Reson. Chem. 1993, 31, 529.

Article

Inorganic Chemistry, Vol. 49, No. 12, 2010 5629

Figure 2. ¹H NMR spectra for 8a-8g obtained in CDCl₃at RT.

Figure 3. Normalized absorption spectra of (A) 8a-8c and 9 and (B) 8d-8g and 9 in 2-MeTHF at 298 K.

summarized in Table 2. The absorption spectrum for model

complex 9 reproduces that found in the literature.^5i,14,28

Absorption spectra in 2-MeTHF for all complexes at RT

showed the presence of an intense (ε on the order

of 10⁴-10⁵M^-1cm^-1) band between 254 and 266 nm,

1

which was correlated to a spin-allowed LC(ppy) π-π*

transition.^1a,15a,b,29Molar extinction coefficient values for

1

8a-8g for the LC and MLCT transitions (see below)

were found to be intense when compared to those in the

literature. Classically, molar absorptivities are on the order

of (10-35) ꢀ 10³and (2-6) ꢀ 10³M^-1cm^-1for π-π* and

MLCT transitions, respectively.^14b,19,29a

(28) Plummer, E. A.; Hofstraat, J. W.; Cola, L. D. Dalton Trans. 2003,

2080.

(29) (a) Ho, C.-L.; Wong, W. Y.; Zhou, G. J.; Yao, B.; Xie, Z.; Wang, L.

Adv. Funct. Mater. 2007, 17, 2925. (b) Schmid, B.; Garces, F. O.; Watts, R. J.

Inorg. Chem. 1994, 33, 9. (c) Ichimura, K.; Kobayashi, T.; King, K. A.; Watts,

R. J. J. Phys. Chem. 1987, 91, 6104. (d) Colombo, M. G.; Hauser, A.; G€udel,

H. U. Top. Curr. Chem. 1994, 171, 143.

In the lower-energy region, well-defined spectra con-

sisting of two additional peaks were observed for 8a-8c

(Figure 3A). These strongly absorbing bands were

5630 Inorganic Chemistry, Vol. 49, No. 12, 2010

Ladouceur et al.

Table 2. Photophysical Properties of Complexes 8a-8g and 9^a

phosphorescence (nm)

Stokes shifts (cm^-1

)

lifetime^d

absorbance at 298 K (nm) [molar

absorptivities (ꢀ10⁴M^-1cm^-1)]

256 [2.4], 270 [2.3], 300 [1.7], 344 [0.8],

380 [0.6], 415 [0.5]

quantum

yield (%)^c

77 K 298 K

( μs)

k_r

k_nr

(ꢀ10⁵s^-1

77 K^b

532

298 K^b

595

77 K

5299

298 K

7290

(ns) (ꢀ10⁵s^-1

)

9

6.2

(5.00)^e(269) ^f

8a 254 [11.9], 266 [11.1], 297[10.6],

344 [10.3], 415 [0.9], 465 [0.2]

8b 254 [3.4], 315 [2.4], 375 [2.9]

8c 266 [6.0], 295 [2.2], 361 [2.2], 450 [4.6]

8d 258 [3.9], 316 [0.5], 379 [0.5], 410 [0.4],

471 [0.1]

8e 264 [4.0], 270 [4.0], 338 [2.0], 415 [0.3],

464 [0.02]

8f 266 [3.6], 304 [1.4], 418 [0.6], 442 [0.3],

488 [0.01]

8g 263 [5.4], 309 [2.8], 410 [0.4], 474 [0.04]

531

623

2673

5454

16.7

3.84

7.26

575

934

2.90

14.5

551

619

516

613

659

599

8518

6067

1852

10353

7048

4537

21.2

0.7

22.0

2.27

5.84

3.29

8.40

801

11.7

10.0

12.4

3.97

3.69

3.76

3.87

668

932

508

947

506

563

513

596

613

592

1789

2730

1604

4773

4179

4205

16.0

2.0

1.72

0.39

2.20

9.00

19.3

8.40

20.8

^aMeasured in 2-MeTHF. ^bλ_max. ^cIn ACN ( 5%, using Ru(bpy)₃(PF₆)₂as the standard (Φ = 6.2%). ^dMeasured at λ_max; relative error (5%. ^eFrom

ref 41b in EtOH/MeOH glass. ^fFrom ref 3g.

assigned primarily to different ¹ILCT(bpy) ¹π-π* transi-

tions (270-360 and 344-450 nm).^{3c,5d,g,14b,26a,29b,30}

These assignments are corroborated by DFT/TD-DFT

calculations (vide infra). As the donor availability of the

group (X) on the bpy ligand increases from H to OMe to

NMe₂, the energies of the second and third absorption

bands are red-shifted and adhere to a free-energy relation-

ship based on Hammett σ_pparameters (Figure S2 in the

Supporting Information). Thus, an increasing batho-

chromic shift is observed with enlargement of the effective

conjugative length of the ancillary ligand. De Cola and

co-workers had recently reported a similar observation in

the absorption spectra of a series of [(dFppy)₂Ir[4-(n-

oligo( p-phenylene)]bpy)]^þcomplexes [dFppyH = 2-(2,4-

difluorophenyl)pyridine; n = 0-3].²⁰The characteristic

¹MLCT and ³MLCT transitions^{14b,15a,b,29d}found in

[(C^∧N)₂Ir(N^∧N)]^þcomplexes is masked by the more

intense ¹ILCT transitions in these complexes.

Emission. Emission spectra at low (77 K) and room

(298 K) temperature were recorded for all of the com-

plexes in deaerated 2-MeTHF solutions and are compiled

in Table 2 and shown in Figures 4 and 5, respectively.

Spectra obtained from measurements at 77 K in the glass

state of 2-MeTHF frequently exhibited multiple emis-

sion maxima with emission energies (λ_max) ranging from

green (506 nm) to red-orange (619 nm). Complexes

8a-8c exhibited emission profiles different from those

of 8d-8g, with the latter resembling those determined for

9 (Figure 4). Note that, at this temperature, a single low-

energy conformation essentially exists during spectral

acquisition.

The fine structure of the emission spectra for 9 at 77 K

(Figure 4B) has been attributed by Watts and co-workers

as an indication of dual emission resulting from thermally

nonequilibrated ³MLCT excited states.^14a,19,32Gudel and

€

co-workers would later assert that the structured lumi-

nescence of [(ppy)₂Ir(bpy)]^þoriginates from a mixed

³MLCT and ³LC(ppy) state.^29dThompson and co-workers^17e

more recently have advanced that the fine structure

observed results from emission from different vibronic

states of the excited species. The emission lifetimes for

8a-8g at RT and 77 K could all be fitted to a mono-

exponentially decaying function (vide infra), which would

seem to preclude a dual-emission pathway. Thus, for each

of 8a-8c, three emission bands were readily observed and

these were assigned to the different vibronic states of the

ground-state species (Figure 4A). As with the absorption

spectra, the emission energy (E_em) for 8a-8c correlated

well with the Hammett (σ_p) constants (Figure S2 in the

Supporting Information); the emission lifetimes at 77 K

were also found to increase from 3.84 to 10.0 μs with

increasing donor ability of the aryl substituents and also

were found to correlate linearly with the Hammett con-

stants. The vibronic coupling between the ground and

excited state, Δ [=(2S_M)^1/2, where S_Mis the Huang-

Rhys factor],^17e,33and the dominant vibrational mode

associated with the distortion between the ground and

In the case of complexes bearing o,o-dimethyl groups on

the arene (8d-8g), the respective spectra display a primary

¹LC absorption at ca. 260 nm, with one or two additional

moderate-to-weak absorption shoulders in a long tail

between 304 and 470 nm. Within this low-absorption

tailing region, the more intense peaks were assigned to an

1

admixture of π-π* and MLCT transitions, while less

intense peaks were reasonably assigned to both a mixed

3

³π-π* and MLCT transitions (Figure 3B)^3c,5g,31and to

1

geometrically forbidden ILCT(bpy) π-π* transitions

similar to those observed for, but exhibiting an order of

magnitude lower molar absorptivity than, 8a-8c

(evidenced by TD-DFT calculationsvide infra). The triplet

spin-forbidden transitions are accessible because of the

large spin-orbit coupling found in third-row heavy transi-

tion metals such as iridium.^15a,b,26a

(30) Maestri, M.; Balzani, V.; Deuschel-Cornioley, C.; von Zelewsky, A.

Adv. Photochem. 1992, 17, 1.

(31) (a) 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. (b) Polson, M.; Fracasso, S.; Bertolasi, V.; Ravaglia, M.; Scandola, F. Inorg.

Chem. 2004, 43, 1950. (c) Neve, F.; Deda, M. L.; Puntoriero, F.; Campagna, S.

Inorg. Chim. Acta 2006, 359, 1666. (d) Waern, J. B.; Desmarets, C.; Chamoreau,

L.-M.; Amouri, H.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletti, F. Inorg.

(32) Wilde, A. P.; King, K. A.; Watts, R. J. J. Phys. Chem. 1991, 95, 629.

(33) Rillema, D. P.; Blanton, C. B.; Shaver, R. J.; Jackman, D. C.; Boldaji,

M.; Bundy, S.; Worl, L. A.; Meyer, T. J. Inorg. Chem. 1992, 31, 1600.

ꢀ

Chem. 2008, 47, 3340. (e) Bolink, H. J.; Coronado, E.; Costa, R. n. D.; Lardies,

N.; Orti, E. Inorg. Chem. 2008, 47, 9149.

Article

Inorganic Chemistry, Vol. 49, No. 12, 2010 5631

Figure 4. Normalized emission spectra of (A) 8a-8c and (B) 8d-8g and 9 in 2-MeTHF at 77 K.

Figure 5. Normalized emission spectra of (A) 8a-8c and (B) 8d-8g and 9 in 2-MeTHF at 298 K.

Table 3. Calculated Excited-State Properties for 8a-8g at 77 K^a

excited states, ω, can be ascertained by modeling of the

low-temperature emission spectra using the time-depen-

dent theory of spectroscopy (Table 3 and eq S1 and

Figures S3-S11 in the Supporting Information). The

observed vibronic transitions (ω) correspond to the aro-

matic stretching modes of the aryl moiety^5i,17e,33attached

to the bpy but may also correspond to the bpy itself

because it is also present in 9.^5i,34These assignments imply

substantial contribution of the LLCT π-π* transition

during emission and a much less significant contribution

from the ³MLCT state.^29aIncreased rigidity of the com-

plex at low temperature might also contribute to the

increased vibronic coupling.³⁵

complex

Δ^b

ω^c(cm^-1

)

E_0-0^d(nm)

9

1.56

1.25

1.53

1.39

1.47

1.56

1.54

1320

1340

1250

1380

1350

505

530

552

622

514

506

562

510

18

8b

8c

8d

8e

8f

8g

^aThe vibronic structure of the spectra was fitted to a first approxima-

tion on the basis of a single mode. See Supporting Information for a

more complete description of these parameters. ^bCalculated horizontal

offset between the ground and excited state minima along the normal

coordinate at 77 K. ^cCalculated frequency of thevibrational modeat 77 K.

^dObtained from the highest energy band at 77 K.

Complex 8c, bearing the strongly electron-donating

NMe₂group, showed a fourth low-intensity, high-energy

band at 537 nm before the characteristic vibronic emis-

sion pattern. We have attributed this peak to a fluores-

cence mechanism, assigned primarily because of the small

observed Stokes shift. We were not able to confirm this

assignment by independent lifetime measurements at λ_max

for this band (537 nm) owing to the very low quantum

yield of 8c.³⁶A possible explanation for its presence is

through thermal repopulation of the singlet state in

(34) (a) Yam, V. W.-W.; Tang, R. P.-L.; Wong, K. M.-C.; Lu, X.-X.;

Cheung, K.-K.; Zhu, N. Chem.;Eur. J. 2002, 8, 4066. (b) Nakanishi, K.;

Solomon, P. H. Infrared Absorption Spectroscopy, 2nd ed.; Holden-Day Inc.:

San Francisco, 1977.

(35) Chen, H.-Y.; Yang, C.-H.; Chi, Y.; Cheng, Y.-M.; Yeh, Y.-S.; Chou,

P.-T.; Hsieh, H.-Y.; Liu, C.-S.; Peng, S.-M.; Lee, G.-H. Can. J. Chem. 2006,

84, 309.

(36) Attempts to verify and further confirm the origin of the 537 nm band

were not possible because of the weakly emissive nature of the compound

that led to the poor signal-to-noise ratio of the excitation spectra.

5632 Inorganic Chemistry, Vol. 49, No. 12, 2010

Ladouceur et al.

phosphorescent species possessing long lifetimes (10 μs

for 8c at 77 K; see Table 2), a property that is not

unprecedented.³⁷The observed bathochromic shifts in

the emission spectra within the series 8a-8c were attrib-

uted to the incorporation of successively stronger elec-

tron-donating groups (8a, H; 8b, OMe; 8c, NMe₂) that

upon initial analysis should effectively increase their

conjugation lengths and thus should, according to classi-

cal frontier molecular orbital (FMO) theory, increase the

energy of the HOMO and stabilize the LUMO.³⁸As will

be evidenced below, the arenes act, in fact, as electron-

donating groups, destabilizing both the HOMO and

LUMO, with an overall decrease in the HOMO-LUMO

gap. The effect of substitution at the 4⁰position of the

aryls in 8a-8c contrasts dramatically with those com-

plexes studied by Neve and co-workers, which have the

form [(ppy)₂Ir(2-phenyl-4-aryl-2,2⁰-bpy)]^þ.^26aThey ob-

served a hypsochromic shift with increased donating

ability of the remote substituents located at the 4⁰position

on the aryl ring. Thus, depending on the position of the

aryl about the bpy chelate, the aryl can act either to

decrease or increase the HOMO-LUMO gap.

The emission spectra at 77 K for 8d-8g, which qualita-

tively appear similar to 9, were hypsochromically shifted

when compared to the analogous complexes within the

series 8a-8c, owing to the drastically diminished con-

jugation resulting from the orthogonal conformation of

the arene relative to the bipyridine. The vibronic structure

was evident within the series 8d-8g, with two near-equal-

intensity principle bands clearly observed (Δ in Table 3)

as well as the presence of lower-energy shoulders. In

general, Δ is similar in magnitude for both series of

complexes, suggesting similar distortion profiles in the

complexes.^17e,33

Whereas the emission spectra at 77 K for 8a-8c

correlate well with the Hammett constants σ_p(Figure

S2 in the Supporting Information), no analogous rela-

tionship was observed for 8d-8g. All electronic commu-

nication is essentially cut because of the orthogonal

disposition of the aryls with respect to the bpy plane,

and there is essentially no difference in the emission

energies of 8d-8g, with the noticeable exception of 8f;

the lifetimes (τ_p) are also quite similar for this series,

ranging from 3.7 to 4.0 μs (vide infra). A reasonable

explanation for the observed exception in 8f is the im-

plication that the emissive transition occurs from distinct

MOs that become accessible due to the increased electron

density because of the inclusion of the NMe₂electron-

donating group. This hypothesis is corroborated by

combined DFT/TD-DFT calculations (vide infra).

All of the complexes luminesced in solution at 298 K

(Figure 5). Large Stokes shifts (4179-10353 cm^-1), mea-

sured from the lowest-energy absorption peak, reveal the

expected phosphorescence in each of the seven complexes

as well as in the reference, 9. The emission maximum

could be tuned marginally between 592 and 659 nm by

varying the nature of the aryl. By comparison, 9 emits at

599 nm at 298 K in ACN. With the exception of 8b, all

complexes exhibited a single broad structureless emission

3

band that was initially assigned to a MLCT [dπ(Ir) f

π*(bpy*)] excited state that is unremarkable in [(C^∧N)₂-

Ir(N^∧N)]^þ-type complexes such as 9.^{5g,15a,b,29d,39}Ana-

lysis of the DFT/TD-DFT calculations (vide infra) has

verified that the principle nature of the CT transitions is

that of a mixed ³LLCT/³MLCT transition for 8a and 8b,

while that of 8c is a ³ILCT transition. Surprisingly, dual

emission with strong bands at 577 and 613 nm was

observed from 8b. Given that the lifetimes at each of

these two emission wavelengths were found to be distinct

but invariant to changes in the concentration and given

that the excitation spectra obtained at each of these two

wavelengths were found to be superimposable, a reason-

able hypothesis is that the dual emission observed arises

from two distinct conformations in solution at RT.

Emission maxima (E_max) at RT for all complexes were

significantly bathochromically shifted compared to E_max

from spectra obtained at 77 K. At RT, rapid reorganiza-

tion of the solvent molecules can stabilize CT states prior

to emission, a process that is encumbered in viscous media

or a rigid matrix at low temperatures.^3n,8,40The particu-

larly large blue shift observed upon cooling is indicative

3

that emission occurs, in part, from a CT state.³ⁿThe

trends relating to the emission energies observed at 77 K

do not exist at RT, ostensibly because of the existence

of differing dynamic conformations that are certainly

operative for 8a-8c.

As with 9, the emission spectra for 8d-8g at 298 K were

generally broad and featureless, with emission maxima

at 599, 596, 613, and 592 nm for 8d-8g, respectively

(Figure 5B). Unlike 8a-8c, essentially no bathochromic

shift was observed, the explanation of which originates

from the electronic decoupling of the aryl groups from the

bpy. No trend could be ascertained between the character

of the arene and the emission energy for either series

(i.e., 8a-8c or 8d-8g). The small energy tuning effect for the

latter series (592-613 nm; cf. Table 2) can be explained by

the loss of conjugation between the aryl and the bpy

plane.^26dThus, all complexes in this series seem to exhibit

emission spectra essentially similar to that found for 9 and

are bathochromically shifted when compared with com-

plexes bearing inductivel_þy electron-donating groups such

as [(ppy)₂Ir(5,5⁰-dmbpy)] (λ_em= 570 nm, Φ = 18%, τ =

0.5 μs; 5,5⁰-dmbpy=5,5⁰-dimethyl-2,2⁰-bipyridine).^3gWith

the exception of 8f, the emission for 8d-8g was attributed

3

to originate from an admixture of LLCT and MLCT

excited states, similar to that found for 8a, 8b, and 9.^3d,15a

The small bathochromic shift (ca. 20 nm) observed for 8f

compared to those of the other complexes in the series can

be rationalized through an ³ILCT emissive transition,

similar to that found for 8c. These assignments were

corroborated by DFT/TD-DFT (vide infra).

Photoluminescent Quantum Yields and Lifetimes.

Quantum yields (Φ_p) for 8a-8g solutions in deaerated

ACN at RT ranged from 1 to 22%. With the exception of

(37) (a) Liu, Y.; Jiang, S.; Glusac, K.; Powell, D. H.; Anderson, D. F.;

Schanze, K. S. J. Am. Chem. Soc. 2002, 124, 12412. (b) Turro, N. J. Modern

Molecular Photochemistry; University Science Books: Sausalito, CA, 1991.

(38) (a) Mak, C. S. K.; Hayer, A.; Pascu, S. I.; Watkins, S. E.; Holmes,

A. B.; Kohler, A.; Friend, R. H. Chem. Commun. 2005, 4708. (b) Fleming, I.

Frontier Orbitals and Organic Chemical Reactions; John Wiley & Sons:

Toronto, Ontario, Canada, 1976.

(39) Dragonetti, C.; Falciola, L.; Mussini, P.; Righetto, S.; Roberto, D.;

Ugo, R.; Valore, A.; De Angelis, F.; Fantacci, S.; Sgamellotti, A.; Ramon,

M.; Muccini, M. Inorg. Chem. 2007, 46, 8533.

(40) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.;

von Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85.

Article

Inorganic Chemistry, Vol. 49, No. 12, 2010 5633

complexes bearing NMe₂groups (8c and 8f) that exhibit

particularly weak phosphorescence (Φ_p∼ 1%), these com-

plexes are remarkably bright (Φ_p= 16-22%; Table 2)

compared to similar complexes bearing extended π-conju-

gated frameworks found in the literature.^{3l,n,aa,26a,c,d,31c,41}

Even arguably, some of the best yellow-to-orange emitters

used in single-layer devices, [(ppy)₂Ir(dtb-bpy)]PF₆^3e,42and

3l

[(ppy)₂Ir(sb)]PF₆(dtb-bpy = 4,4⁰-di-tert-butylbipyridine;

sb = 4,5-diaza-9,9⁰-spirobifluorene), which luminesce at

581 and 605 nm in ACN with quantum yields of 23.5% and

22.6% and lifetimes of 557 and 330 ns, respectively, have

photophysical characteristics essentially similar to those

found in this study. Bolink and co-workers have, in fact,

recently shown that aryls appended onto the ancillary

diimine decrease nonradiative intermolecular charge

recombination through steric shielding of the iridium

center.^3dThus, it is gratifying to observe that similar

photophysical characteristics exist for most of the com-

plexes in this study.

Figure 6. ln(k_nr) vs emission energy (E_em) at 298 K in 2-MeTHF for

8a-8c (pink 9) and 8d-8g (blue [). The dashed lines represent linear

least-squares fits to the data.

By contrast, the lower-energy emission at RT coupled

with shorter observed lifetimes (τ_p), particularly for 8c

(τ_p= 12.4 ns), results in much lower quantum yields

(0.7% for 8c; 2.0% for 8f) and, accordingly, large calcu-

lated nonradiative decay rate constants (k_nr). The surpris-

ing marked decrease in the emission intensity implies a

new radiationless pathway operative in 8c and 8f, due to

the presence of distinctly electron-rich aniline substitu-

ents. The increased conjugation imparted by the presence

of NMe₂groups facilitates greatly nonradiative electron

transfer in the excited state for 8c. Owing to the ortho-

gonal orientation of the arene, 8f does not succumb so

readily to this deactivation pathway and so possesses rela-

tively longer-lived phosphorescence. Pendant alkylamines,^5c

alkylammonium salts,^3yor amines directly substituted onto

the ancillary ligand^3c,5dof heteroleptic iridium complexes do

not substantially impact the emission intensity; exception-

ally, [(ppy)₂Ir(5-NMe₂-1,10-phen)]^þexhibited a quantum

yield of only 0.3% in dichloromethane (DCM).³⁹

Observed phosphorescent lifetime (τ_p) values at RT in

2-MeTHF were modeled as single-exponential decays

and are on the nanosecond-to-submicrosecond time scale

(12.4-934 ns), values generally not dissimilar to other

cationic iridium(III) complexes^{3g,14a,31c,39}but generally

2-4 times longer than that found for 9.¹⁸The long-lived

emission is a further indication of a phosphorescence

radiative pathway.^29cWith measured τ_pand Φ_pdata in

hand, the radiative and nonradiative rate constants (k_r

and k_nr) were calculated with eqs 1 and 2, respectively.

The values for the radiative rate constant (k_r) are quite

similar for 8a-8g, are on the order of 10⁵s^-1, and are

similar to those previously reported by Mussini et al.³⁹

The limiting factor that seems to control the quantum

yield in 8a-8g is the nonradiative rate constant (k_nr), for

which we observe a significantly larger variation, with

values ranging from 10⁵to 10⁷s^-1. This range is much

larger than the one reported by the previous group (from

1.6 to 7.1 ꢀ 10⁵s^-1), and the difference may be attributed

to the loss of rigidity due to migration from a phenanthro-

line to a bipyridine ligand architecture. Surprisingly, the

relaxation dynamics of 8a and 8b, which possess freely

rotating aryl groups, are not adversely affected when

compared to their dimethylated congeners (8d-8e).

Φ_p

k_r¼

ð1Þ

τ_p

1 - Φ_p

τ_p

k_nr

¼

ð2Þ

Both series of isostructural complexes obey the

energy gap law²¹at RT when analyzed independently

(Figure 6) in accordance with other iridium(III) lumino-

phores.^17e,26a,43The fact that there is such a linear fit

would seem to indicate that the excited states for 8a-8c are

similar while those for 8d-8g are comparable, at least as far

as deactivation dynamics are concerned. The calculated

slope of -12.02 eV^-1for the series comprising 8d-8g is

similar to that found for other polypyridine complexes such

as tricarbonyl(bipyridine)rhenium(I) complexes whose va-

lues range from -9.0 to -11.9 eV^-1in ACN.⁴⁴By contrast,

the calculated slope for 8a-8c is significantly larger. This is

almost certainly an artifact due to a paucity of data points

coupled with the significant red shift observed for 8c. A

reasonable supposition, given that 8c emits via a pathway

distinct from 8a and 8b but similar to 8f, is that their

deactivation dynamics are also different.

Solvatochromic Study. For each of 8a-8g, absorption

and emission spectra obtained in either 2-MeTHF or

ACN were found to be superimposable. The apparent

lack of solvent dependence was surprising given the CT

nature of the observed transition and thus led us to

(41) (a) Cunningham, G. B.; Li, Y.; Liu, S.; Schanze, K. S. J. Phys. Chem.

B 2003, 107, 12569. (b) Glusac, K. D.; Jiang, S.; Schanze, K. S. Chem. Commun.

2002, 2504. (c) Kim, K.-Y.; Farley, R. T.; Schanze, K. S. J. Phys. Chem. B 2006,

110, 17302. (d) Zeng, X.; Tavasli, M.; Perepichka, I. F.; Batsanov, A. S.; Bryce,

M. R.; Chiang, C.-J.; Rothe, C.; Monkman, A. P. Chem.;Eur. J. 2008, 14, 933.

(e) Zhao, Q.; Liu, S.; Shi, M.; Wang, C.; Yu, M.; Li, L.; Li, F.; Yi, T.; Huang, C.

Inorg. Chem. 2006, 45, 6152. (f) Rothe, C.; Chiang, C.-J.; Jankus, V.; Abdullah,

K.; Zeng, X.; Jitchati, R.; Batsanov, A. S.; Bryce, M. R.; Monkman, A. P.

Adv. Funct. Mater. 2009, 19, 2038.

(43) Siebrand, W. J. Chem. Phys. 1967, 46, 440.

(44) (a) Caspar, J. V.; Westmoreland, T. D.; Allen, G. H.; Bradley, P. G.;

Meyer, T. J.; Woodruff, W. H. J. Am. Chem. Soc. 1984, 106, 3492. (b) Baiano,

J. A.; Murphy, W. R. Inorg. Chem. 1991, 30, 4594.

(42) Parker, S. T.; Slinker, J. D.; Lowry, M. S.; Cox, M. P.; Bernhard, S.;

Malliaras, G. G. Chem. Mater. 2005, 17, 3187.

5634 Inorganic Chemistry, Vol. 49, No. 12, 2010

Ladouceur et al.

Figure 7. Normalized emission spectra for (A) 8d and (B) 9 at 298 K in different solvents.

Figure 8. Correlation diagram for the emission energy with (A) Reichardt-Dimroth solvent E_T(30) parameters and (B) a Winnik-Dong solvent P_yscale

for 8d (blue [) and 9 (red 9).

investigate more closely solvatochromism in 8d, as a

prototypical complex of the family under investigation,

and compare its behavior to 9. The emission spectra for 8d

and 9 were obtained in six solvents of disparate polarity

and identity (Figures 7 and 8). Inspection of the emission

maximum as a function of the solvent polarity, as mea-

sured by the Reichardt-Dimroth⁴⁵solvent polarity func-

tion [E_T(30)] or the P_yscale developed by Winnik and

Dong,⁴⁶suggests that, in fact, both complexes are mod-

erately solvatochromic (Figure 8 and Table 4). In general,

polar solvents stabilized the dipole-dipole interaction of

the excited state, shifting the emission spectrum to the

red.⁴⁷No clear linear correlation could be obtained for 9;

however, there is an observed solvatochromic sensiti-

Electrochemical Characterization. Electrochemical

analysis via cyclic voltammetry (CV) was performed in

order to measure the redox potentials of 8a-8g and 9. The

electrochemical data are summarized in Table 5. The

reversible cyclic voltammogram for 9, acquired as an

ACN solution with 0.1 M of tetra-n-butylammonium

hexafluorophosphate (TBAPF₆) as the supporting elec-

trolyte at 298 K, exhibits an oxidation couple at 1.44 V

and a reduction couple at -1.32 V (ΔE_red-ox=2.76 V;

ox

red

Δ

_red-ox= E - E ). The CV accurately reproduces,

pa

pc

although is slightly cathodically shifted to, that reported

in the literature (E^o_p^x= 1.25 V; E^r_p^ed= -1.42 V; Δ_red-ox

=

2.67 V)^4band is in good agreement with the theoretical

value calculated for this complex (2.75 eV, vide infra).

The first oxidation wave for 9 has been mainly attributed

to a metal center Ir^IV/Ir^IIIcouple, with substantial con-

tributions from the cyclometalating ppy ligands. Gene-

rally, metal-centered oxidation is assumed to be a com-

pletely reversible process, with the degree of irreversibility

increasing with an increase in the contribution of the ppy ligand

to the electron density distibution of the HOMO.^{26a,39,41e,49}

The first reduction wave has been assigned to a localized

3

vity^5g,15bof the MLCT transition, in agreement with

that found by Watts and co-workers (Figure 8).^47b

similar monotonic correlation could be ascertained

A

between E_T(30) or P_yand λ_maxfor 8d.⁴⁸

(45) Reichardt, C. Chem. Rev. 1994, 94, 2319.

(46) Dong, D. C.; Winnik, M. A. Can. J. Chem. 1984, 62, 2560.

(47) (a) Lo, K. K.-W.; Lau, J. S.-Y.; Lo, D. K.-K.; Lo, L. T.-L. Eur. J.

Inorg. Chem. 2006, 2006, 4054. (b) Wilde, A. P.; Watts, R. J. J. Phys. Chem.

1991, 95, 622.

(48) It is frequently difficult to ascertain the magnitude and origin of the

solvatochromism in complexes of the type [(C^∧N)₂Ir(N^∧N)]^þ. For instance,

(49) (a) Didier, P.; Ortmans, I.; Kirsch-De Mesmaeker, A.; Watts, R. J.

Inorg. Chem. 1993, 32, 5239. (b) Calogero, G.; Giuffrida, G.; Serroni, S.;

Ricevuto, V.; Campagna, S. Inorg. Chem. 1995, 34, 541. (c) Neve, F.; Crispini,

A.; Serroni, S.; Loiseau, F.; Campagna, S. Inorg. Chem. 2001, 40, 1093.

λ

_maxfor emission in ACN for 9 has been measured to be 580,^31b583,^3g585,^3y

and 588 nm;^3sadditionally, λ_maxfor emission for the chloride salt of 9 has

been measured to be 616 nm.^23b

Article

Inorganic Chemistry, Vol. 49, No. 12, 2010 5635

Table 4. Solvent-Induced Shifts in Emission Maxima for 8d and 9

emission energy [cm^-1(nm)]

8d

9

solvent

E_T(30) (kcal/mol)

P_y

this study

lit.

ref

2-MeTHF

THF

CHCI₃

DCM

acetone

DMSO

ACN

36.5

37.4

39.1

40.7

42.2

45.1

45.6

55.4

16 695 (599)

16 807 (595)

a

1.35

1.25

1.35

1.64

1.95

1.79

1.35

16 949 (590)

16 667 (600)

16 750 (597)

16 556 (604)

17 007 (588)

16 502 (606)

41b

47b^b

5i

16 863 (593)

17 094 (585)

16 667 (600)

16 420 (609)

16 639 (601)

16 722 (598)

16 779 (596)

16 556 (604)

16 695 (599)

MeOH

5i

^aNo data reported in the literature. ^bThe measurement was obtained from the chloride salt of 9.

Table 5. Redox Properties for 8a-8g and 9^a,b

oxidation

reduction

ppy

Ir^IV/Ir^III

bpy

E_pa

E_pc

ΔE

E_pa

E_pc

ΔE

E_pa

E_pc

ΔE

ΔE_red-ox

E_pa

E_pc

ΔE

E_pa

E_pc

ΔE

E_pa

E_pc

ΔE

9

2.20 2.04 0.15^c1.44 1.27 0.17^c

2.76

2.61

2.66

2.44

2.76

2.77

2.39

2.67

-1.30 -1.32 0.02^c-2.01 -2.04 0.03^c-2.17 -2.29 0.12^c

-1.19 -1.27 0.08^c-1.69 -1.79 0.10^c

-1.22 -1.29 0.06^c-1.76 -1.85 0.09^c

-1.27 -1.41 0.14^c-1.85 -1.93 0.08^c

-1.26 -1.33 0.08^c-1.91 -1.99 0.08^c

-1.27 -1.40 0.13^d

e

8a 2.13

8b 1.94

8c

1.34 1.31 0.03^c

1.38 1.27 0.11^c

e

1.44 1.33 0.11^c1.03 0.88 0.15^d

1.42 1.26 0.16^c

e

8d 2.27

8e 1.90

8f

e

1.37

e

1.46

1.39 1.32 0.08^c

0.97

-1.34 -1.42 0.08^c-2.10 -1.99 0.11^d

-1.30 -1.36 0.06^c-1.97 -2.07 0.10^c

e

8g 2.05

^aCV was carried out at 100 mV/s in 0.1 M TBAPF₆/CH₃CN at a platinum working electrode with glassy carbon as the counter electrode and an

Ag/Ag^þelectrode as a pseudoreference. Ferrocene was used as an internal standard, and potentials are reported with respect to a saturated calomel

electrode (SCE). ^bE_paand E_pccorrespond to anodic and cathodic peak potentials (V), and ΔE is the difference calculated between these two potentials.

ΔE_red-oxis the difference between the smaller E_paand the biggest E_pcvalues (V). Values are evaluated with a 10% relative error. ^cRefers to a reversible

process. ^dRefers to a quasi-reversible process. ^eRefers to an irreversible process.

reduction of the bpy ligand, while the second reduction

wave is assigned to the localized reduction of the ppy ligand.

These assignments match those previously reported and are

also consistent with DFT calculations (vide infra).^{3c,n,14,39,50}

The following observations were made with respect to

the CVs of 8a-8g (cf. Figures S12-S13 in the Supporting

Information) regardless of the scan rate (50-200 mV/s).

Compared to 9, the first oxidation potentials for each of

8a, 8d, and 8g are anodically shifted by 100, 20, and

50 mV, respectively, and were found to be reversible.

Similar to 9, this first oxidation wave was attributed to a

one-electron oxidation eminating from a combined metal

phenylpyridine MO. A second irreversible oxidation

wave at 2.20, 2.27, and 2.05 V for 8a, 8d, and 8g,

respectively, was also detected. The irreversibility of this

second oxidation peak distinguishes these three com-

plexes from 9, which was found to be reversible.

Their respective first reduction peak potentials are also

reversible, with 8a being cathodically shifted by 50 mV

and 8d and 8g being anodically shifted by 10 and 40 mV,

respectively, relative to 9. The first reduction potential

corresponds to the energy of the LUMO, which is localized

on the diimine as ascertained by computations (vide infra).

The larger cathodic shift for 8a is attributed to the greater

conjugative influence compared to the more orthogonally

oriented arenes found in 8d and 8g, which should result

in a lowering of the energy of the LUMO for the former.

Accordingly, there is an overall decrease in the HOMO-

LUMO energy gap (ΔE_red-ox) for 8a to 2.61 V, while the

energy gap remains essentially unchanged at 2.76 and

2.67 V for 8d and 8g, respectively. This decrease in the

HOMO-LUMO energy gap is not generally reproduced

computationally, the results of which indicate a tiny blue

shift. Additionally, each of these complexes exhibit two

additional reversible or quasi-reversible reduction waves

between -1.79 and -2.05 V, which have been assigned to

successive reductions of the cyclometalating ppyH.^5c,d,26a,d

The CV traces for 8b and 8c exhibit reversible oxidation

peaks for metal-centered orbitals with contributions from

the cyclometalating ppyH at 1.38 and 1.44 V, respectively,

that are only slightly anodically shifted compared to that

of 9. Additionally, 8c possesses a second oxidation peak

at 1.03 V that has been attributed to oxidation of the

aniline fragment, a conclusion consistent with DFT cal-

culations (vide infra). The quasi-reversible oxidation

peak for 8b observed at 1.94 V is attributed to oxidation

of the anisole fragment based on analysis of the MOs

from DFT calculations, below. Each displays two reduc-

tion peak potentials that are generally slightly cathodi-

cally shifted with respect to 9; the first reduction peak for

8c is anodically shifted by 90 mV. As such, ΔE_red-oxfor 8c

is significantly smaller at 2.44 V than that found for 8b

(2.66 V). Overall, the HOMO-LUMO energy gap for

(50) (a) Bandini, M.; Bianchi, M.; Valenti, G.; Piccinelli, F.; Paolucci, F.;

Monari, M.; Umani-Ronchi, A.; Marcaccio, M. Inorg. Chem. 2010, 49, 1439.

(b) Stagni, S.; Colella, S.; Palazzi, A.; Valenti, G.; Zacchini, S.; Paolucci, F.;

Marcaccio, M.; Albuquerque, R. Q.; De Cola, L. Inorg. Chem. 2008, 47, 10509.

5636 Inorganic Chemistry, Vol. 49, No. 12, 2010

Ladouceur et al.

Table 6. Comparison of Measured and Calculated Ground-State Geometries for 8a

experimental structure

structural

calculated structure

relative

error (%)^b

relative

error (%)^b

parameter

crystal structure

uncertainty (%)

method 1^a

relative error (%)^b

method 2^c

method 3^d

A1 (deg)

A2 (deg)

32 ( 3.0

9

5

35

36

9

3

27

23

15

37

15

0

37 ( 1.7

˚

D1 (A)

2.146 ( 0.009

2.121 ( 0.010

1.985 ( 0.012

2.002 ( 0.012

2.041 ( 0.010

2.053 ( 0.011

0.4

0.5

0.6

0.5

2.214

2.211

2.022

2.081

3.2

4.2

1.9

1.0

2.0

1.4

2.180

2.179

2.035

2.039

2.078

2.082

15.8

2.7

2.5

1.8

1.4

2.165

2.039

2.074

0.9

2.1

2.7

1.8

1.6

1.0

˚

D2 (A)

˚

D3 (A)

˚

D4 (A)

˚

D5 (A)

˚

D6 (A)

^aGaussian 03: B3LYP/6-31þG* for C, H, N, and O; B3LYP/LAN2DZ for Ir. ^bRelative error between the optimized and X-ray structures. ^cGamess:

B3LYP/SBKJC-31G for C, H, N, O, and Ir. ^dGaussian 03: B3LYP/3-21G* for C, H, N, and O; B3LYP/SBKJC-VDZ for Ir.

8a-8c decreases from 8b>8a>8c, a trend that mirrors

that observed for emission at 298 K (cf. Figure 5).

observed spectroscopically (vide supra). The radical cation

on the aniline results in a large net destabilization of the

HOMO in these two complexes. The remainder of the

complexes exhibited a destabilization of the HOMO asso-

ciated with the Ir^IV/Ir^IIIcouple. With the exception of 8a

and 8b, their corresponding LUMO energies, as measured

as E_pcof the first reduction peak, were found to be

anodically shifted and their destabilization is of a magni-

tude comparable to that found for the destabilization of

the HOMO, resulting in an overall slight decrease in the

HOMO-LUMO gap compared to 9. A more detailed

investigation intothe electrochemical properties isongoing

and will be reported elsewhere.

Computational Modeling Approaches. Recently, several

groups have employed computations as a tool for the predic-

tion and modeling of the photophysical properties of lumi-

nescent organometallic complexes.^{3g,17a,b,e,f,18,31b,39,50a,52}In

order to obtain accurate photophysical property predic-

tions, a rigorous methodology must be established. First,

the ground- and excited-state geometries must be modeled

accurately⁵³because these will directly impact accurate

absorption and emission energy predictions. In our current

work, theoretical predictions of the photophysical proper-

ties of 8a-8g and 9 were obtained through a combined

TD-DFT/DFT study (cf. Tables S4-S11 in the Supporting

Information). We employed DFT methods using the

B3LYP functional with low-cost basis sets (3-21G*) used

The behavior of the CV traces for 8e and 8f is different.

For 8f, two quasi-reversible oxidation peaks at 0.97 and

1.46 V were observed, similar to that found for 8c. DFT

calculations place the HOMOs for 8c and 8f on the aniline

fragment of the diimine (vide infra). Unfortunately,

although 8e was found to be highly phosphorescent,

because of exhibition of an irreversible oxidation wave,

this complex does not seem to be a good candidate for

device fabrication. The lack of reversibility observed for

the oxidation waves has, in other heteroleptic iridium

complex systems, been attributed to an increased con-

tribution of the cyclometalating ligand via augmented

covalent σ(Ir-C) character in the HOMO, resulting in its

destabilization.^{3n,26d,41e,49a,51}This is not the case here

because we observe a net yet small stabilization of the

HOMO by CV. The quasi-reversible nature of the first

reduction wave for 8e suggests that the anisole fragment is

somewhat involved in the reduction process. The first

reduction peak for 8f, on the other hand, was found to be

reversible. This first reduction peak is anodically shifted

by 80 and 100 mV for 8e and 8f, respectively. For the

moment, we have no satisfactory explanation for the

difference of behavior between 8e and 8f.

According to the CVs, the overall impact for the

inclusion of aryl groups on the bpy ligand is to destabilize

the HOMO relative to 9. There did not seem to be a

general trend in terms of the magnitude of this destabiliza-

tion, but this behavior is corroborated by the computational

predictions. For 8c and 8f, there is a substantial reduction

in the HOMO-LUMO gap. This red shift was also

(52) (a) Bolink, H. J.; Cappelli, L.; Cheylan, S.; Coronado, E.; Costa,

R. D.; Lardies, N.; Nazeeruddin, M. K.; Orti, E. J. Mater. Chem. 2007, 17,

5032. (b) Di Censo, D.; Fantacci, S.; De Angelis, F.; Klein, C.; Evans, N.;

Kalyanasundaram, K.; Bolink, H. J.; Gratzel, M.; Nazeeruddin, M. K. Inorg.

Chem. 2008, 47, 980. (c) Velusamy, M.; Thomas, K. R. J.; Chen, C.-H.; Lin, J. T.;

Wen, Y. S.; Hsieh, W.-T.; Lai, C.-H.; Chou, P.-T. Dalton Trans. 2007, 3025.

(53) Predicted ground-state geometries are deemed accurate if the struc-

tural parameters fall within 1 standard deviation of the experiment.

(51) Serroni, S.; Juris, A.; Campagna, S.; Venturi, M.; Denti, G.; Balzani,

V. J. Am. Chem. Soc. 1994, 116, 9086.

Article

Inorganic Chemistry, Vol. 49, No. 12, 2010 5637

to model carbon, hydrogen, nitrogen, and oxygen centers,

while the SBKJC-VDZ basis set was used to model the

iridium center (method 3, Table 6). We then compared our

predicted geometries to two other methods commonly used

in the literature.

As an example of the effectiveness of method 3, selected

measured and calculated ground-state structural para-

meters for 8a are compared within Table 6. Ir-N(bpy),

Ir-N(ppy), and Ir-C(ppy) distances for 8a were calcu-

˚

lated to be 2.17, 2.04, and 2.07 A, respectively, which

˚

slightly overestimate distances of 2.13, 1.99, and 2.04 A,

respectively, found in the X-ray structure; other bond

lengths and bond and dihedral angles were, in general,

also accurately reproduced. This combined protocol

proved to be a more accurate predictor of the geometry

for complexes such as 8a at far smaller computational

cost than the most commonly used computational meth-

od found in the literature for these types of complexes,

namely, the B3LYP⁵⁴functional with the 6-31þG*

Pople⁵⁵double-ζ split valence basis set for carbon, hydro-

gen, oxygen, and nitrogen and LANL2DZ⁵⁶for iridium

(method 1, Table 6). This latter method overestimated the

bond lengths by 1-4%; however, it did slightly more

accurately predict the dihedral angles found in the X-ray

structure than method 3 did. The use of the SBKJC-31G⁵⁷

basis set for all atoms within Gamess⁵⁸(method 2,

Table 6) proved even less effective, severely underestimat-

ing the dihedral angles formed by the intersection of the

arene and bpy planes. Calculated excited state (T₁) bond

lengths using method 3 mirror those found in the ground

state with only ca. 1% contraction observed (cf. Table S2

in the Supporting Information). Generally, there is a

slight planarization observed in the triplet state for both

ligands: the two pyridines in bpy and the phenyl and

pyridine in ppyH become more coplanar. Dihedral angles

between the arene and the bpy plane for 8a-8g diminish

from 1 to 43%. Complexes bearing methoxy groups on

the arene (8b and 8e) were found to exhibit the greatest

conformational change within each series.

Figure 9. Calculated rotational barriers about the C_aryl-C_bpybond.

1

methyl signals in the H NMR spectra for the latter two

complexes. We performed a relaxed dihedral potential

energy scan using our validated method 3. The rotational

barriers for 8a, 8d, and 8g were calculated to be 2.5, 14.6,

and 21.4 kcal/mol, respectively (Figure 9). The barrier for

8a is similar to that found for biphenyl (exptl,⁵⁹1.4 kcal/mol;

calcd at MP2/cc-pV¥Z,⁶⁰2.2 kcal/mol; calcd at CC-cf,⁶¹

1.9 kcal/mol), while that for 8d is significantly smaller than

that determined for 2,6-dimethylbiphenyl (exptl,⁶²>23.9

kcal/mol). Rotation about the C_aryl-C_bpybond in 8g is

calculated to be further restricted (21.4 kcal/mol) because of

steric interactions between the methyl groups found meta to

this bond and the remainder of the ligand framework about

the complex. The elevated calculated rotational barriers for

8d and 8g plausibly explain the observed atropisomerism

evident in the ¹H NMR spectra (vide supra).

Following accurate geometry optimization, we next

modeled the electronic structure of each of the complexes.

TD-DFT calculations are particularly amenable to pro-

viding insight into the different singlet and triplet excited

states, which can be related to absorption and emission

processes, respectively. TD-DFT has of late become

increasingly popular for describing the photophysical

properties of luminescent complexes, but to date, only

a few studies of iridium(III) complexes have been

reported.^{3e,17a,b,18,31b,39,52b,63}Thus, absorption spectra

were predicted by means of TD-DFT calculations based

on S₀-optimized geometries (cf. the Supporting In-

formation). The calculated absorption transitions repro-

duce well the principle spectral features and the energies

(λ_max) of the absorption spectra for 9 as well as 8a-8g

(e.g., for 8a and 8c, parts A and C of Figure 10; see

Figures S14-S29 in the Supporting Information for other

complexes).

With ground-state structure prediction sound, we turned

our attention toward an examination of the torsional

potentials for the C_aryl-C_bpybond for each of 8a, 8d, and

8g in an effort to explain the observed anisochronous

(54) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Becke, A. D. Int. J.

Quantum Chem. 1994, S28, 625. (c) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B

1988, 37, 785. (d) Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett.

1989, 157, 200.

(55) (a) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54,

724. (b) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257.

(c) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213. (d) Hariharan,

P. C.; Pople, J. A. Mol. Phys. 1974, 27, 209. (e) Gordon, M. S. Chem. Phys. Lett.

1980, 76, 163.

(56) (a) Dunning, J. T. H. In Methods of Electronic Structure Theory;

Schaefer, I., H. F., Ed.; Plenum Press: New York, 1977; Vol. 3, p 1. (b) Hay, P. J.;

Wadt, W. R. J. Chem. Phys. 1985, 82, 299. (c) Hay, P. J.; Wadt, W. R. J. Chem.

Phys. 1985, 82, 270. (d) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 284.

(57) (a) Stevens, W. J.; Basch, W. J.; Krauss, M. J. Chem. Phys. 1984, 81,

6026. (b) Stevens, W. J.; Krauss, M.; Basch, H.; Jasien, P. G. Can. J. Chem. 1992,

70, 612. (c) Cundari, T. R.; Stevens, W. J. J. Chem. Phys. 1993, 98, 5555.

(58) (a) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.;

Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su,

S.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem. 1993,

14, 1347. (b) Gordon, M. S.; Schmidt, M. W. In Theory and Applications of

Computational Chemistry: the First Forty Years; Dykstra, C. E., Frenking, G.,

Kim, K. S., Scuseria, G. E., Eds.; Elsevier: Amsterdam, The Netherlands, 2005;

p 1167.

(59) Bastiansen, O.; Samdal, S. J. Mol. Struct. 1985, 128, 115.

(60) Sancho-Garcia, J. C.; Cornil, J. J. Chem. Theory Comput. 2005, 1,

581.

(61) Johansson, M. P.; Olsen, J. J. Chem. Theory Comput. 2008, 4, 1460.

(62) Grumadas, A. J.; Poshkus, D. P.; Kiselev, A. V. J. Chem. Soc.,

Faraday Trans. 2 1982, 78, 2013.

(63) (a) 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. (b)

Polson, M.; Ravaglia, M.; Fracasso, S.; Garavelli, M.; Scandola, F. Inorg. Chem.

2005, 44, 1282. (c) Yang, C.-H.; Su, W.-L.; Fang, K.-H.; Wang, S.-P.; Sun, I. W.

Organometallics 2006, 25, 4514. (d) Obara, S.; Itabashi, M.; Okuda, F.; Tamaki,

S.; Tanabe, Y.; Ishii, Y.; Nozaki, K.; Haga, M.-a. Inorg. Chem. 2006, 45, 8907. (e)

Liu, T.; Zhang, H.-X.; Zhou, X.; Zheng, Q.-C.; Xia, B.-H.; Pan, Q.-J. J. Phys.

Chem. A 2008, 112, 8254.

5638 Inorganic Chemistry, Vol. 49, No. 12, 2010

Ladouceur et al.

Figure 10. (A) Measured (violet) and simulated (blue) absorption spectra for 8a. The oscillator strengths are represented by unbroadened vertical lines of

calculated singlet-singlet transitions (red). (B) Calculated energies (for method 1, see the text) and isodensity surface plots for selected orbitals of 8a. (C)

Measured (violet) and simulated (blue) absorption spectra for 8c. The oscillator strengths are represented by unbroadened vertical lines of calculated

singlet-singlet transitions (red). (D) Calculated (for method 1, see the text) energies and isodensity surface plots for selected orbitals of 8c.

The results indicate that, unlike 9, 8a-8g possess an

additional and important allowed transition in the form

of an ILCT that occurs on the ancillary ligand, wherein

the arenes act as electron-releasing groups and the bpy

ligand acts as the electron acceptor. The importance of

this transition is evident in, for instance, 8c and explains

the dominant absorption band observed at ca. 450 nm

(Figure 10C,D). What distinguishes series 1 (8a-8c) from

series 2 (8d-8g) is that in the former this transition is

dominant, in agreement with that observed in their

absorption spectra. The lowest-energy permitted transi-

tion can be modulated by increasing the electron density

of the arene. This has the effect of increasing the energy

in both the π and π* orbitals of the bpy* ligand, which

correlates well with a free-energy relationship, as exhi-

bited in Figure 11. Thus, whereas the lowest-lying per-

mitted absorption transition with a significant oscil-

lator strength is primarily comprised of a mixture of the

π(ppy) f π*(bpy*) transition (LLCT) with some MLCT

character for 8a, 8d, 8g, and 9, the dominant low-energy

absorption is that of an ILCT transition for the remainder

of the complexes; 8g also exhibits a low-energy transition

with some ¹LC_ppycharacter (HOMO f LUMOþ1). The

orthogonal orientation of the aryl groups in 8d and 8g

impedes electronic communication, effectively attenuat-

ing the intensity of the ILCT transition. Thus, the gross

structure of their absorption spectra is essentially distilled

to that of 9. This is not the case for 8a, where the aryl

groups are somewhat implicated in the LUMO and the

ILCT remains the dominant feature in the absorption

spectrum. The first permitted transition for 8a occurs for

HOMO-1 f LUMO ( f ∼ 0.01). Additionally, there is a

second important transition, HOMO-3 f LUMO ( f ∼

0.04), wherein the occupied MO is nearly isosymmetric

with the first and that of the HOMO ( f ∼ 0.0004 for

HOMO f LUMO) (cf. Figure 10B). Complex 8d behaves

similarly, with the first permitted transition occurring for

HOMO-1 f LUMO ( f ∼ 0.01), with the presence of a

second important transition, HOMO-3 f LUMO (f ∼

0.03). For 8g, absorption occurs primarily from a mixed

Article

Inorganic Chemistry, Vol. 49, No. 12, 2010 5639

Figure 11. Calculated electronic structures for 9 and 8a-8g by DFT methods at their S₀-optimized geometries (see the text). Black indicates that the

electrondensityismostlylocalizedonπ(bpy) andπ(ppy) with contributions fromd(Ir). Orange and turquoiseindicate the mostenergeticπ(bpy) and π(ppy),

respectively. In each of these cases, minor d(Ir) contributions may exist. Red indicates the LUMO, which is localized on π*(bpy).

Table 7. Comparison of Experimental and Calculated Emission Energies for 9

and 8a-8g

HOMO f LUMOþ1 and HOMO f LUMOþ2 transi-

tion ( f = 0.027; 74:21), which is characterized as an admix-

ture of ¹LC and ¹LLCT. There is an additional important

experimental emission (nm)

calculated emission (nm)

¹LLCT/¹MLCT transition, HOMO-7 f LUMO ( f ∼

0.04), which is close in energy to the first (see Figure S20

in the Supporting Information).

method method relative method

1^b

complex

77 K^a

298 K^a

2^c

error (%)^d

3^e

Although DFT/TD-DFT calculations predict the gen-

eral form of the absorption spectra for 8b and 8c and for

8e, and 8f, they poorly estimate absorption energies; the

predictions are bathochromically shifted by ca. 70 and

150 nm for 8b and 8c and for 8e and 8f, respectively (cf.

Figures S22-S29 in the Supporting Information). The

bathochromic shift for 8b and 8c (e.g., Figure 10C for 8c)

can be rationalized by the difference in conformation

between the gas-phase and solution-state structures ex-

isting at RT. More likely is that there is an overestima-

tion of the planarization computed in the triplet state.

For each of these four complexes, the transition from

the ground state to the lowest-energy singlet (S₁) state

[HOMO-1fLUMO for 8b (f ∼ 0.57) and 8e ( f ∼ 0.09);

HOMO f LUMO for 8c (f ∼ 1.03) and 8f ( f ∼ 0.43)]

involves primarily a π(bpy*)-π*(bpy*) ¹ILCT transition.

The oscillator strength for this transition is particularly large

( f >0.6) for 8b and 8c but hypochromically shifted by over

60% for 8e and 8f because of the orthogonal orientation of

the aryl groups in these complexes, a characteristic also

observed experimentally. Complexes 8b and 8e also exhibit

a low-energy admixture of LLCT and MLCT transitions

(HOMO-4 f LUMO for 8b; HOMO-5 f LUMO for

8e), similar to those described for the other complexes in this

study. In the particular case of 8e, this transition is, in fact,

9

532

531

551

619

516

506

563

513

595

623

613

659

599

596

613

592

549

642

690

747

563

608

735

570

561

551

576

659

541

564

633

530

5.5

3.7

619

592

566

669

580

567

676

566

8a

8b

8c

8d

8e

8f

8g

4.5

6.5

4.8

11.4

12.3

3.3

^aλ_max

.

^bΔE = E^T_H¹_SOMO- E^S_H⁰_OMO. ^cΔE = E^T_to¹_tal- E_t^S_o⁰_tal

.

^dCalculated

between method 2 and experimental emission at 77 K. ^eΔE = E_t^T_o¹_tal

E^S_to⁰_talby TD-DFT.

-

each, which has the effect of raising both the HOMO and

LUMO energy levels. In fact, the next-lowest-lying transi-

tion after the characteristic ILCT transition was identified

as ¹LMCT.

Emission energies were calculated using three distinct

methods found in the literature. For the first two meth-

ods, we imposed a triplet spin multiplicity by selectively

populating the LUMO on complexes whose geometries

had been previously optimized at the S₀state. The

geometry of the newly formed HSOMO triplet (T₁) state

was then reoptimized at the same level of theory (using

UB3LYP) to account for any geometry relaxation in the

triplet excited state. Method 1^17b,31a,64calculates the

energy difference between HSOMO(T₁) and HOMO(S₀),

while method 2^3gcalculates the energy difference between

1

more dominant than the lower-energy ILCT transition.

The absence of evidence for this contribution to low-energy

transitions for 8c and 8f is a result of the increased electron

density contribution of the NMe₂groups incorporated onto

(64) Brooks, J.; Babayan, Y.; Lamansky, S.; Djurovich, P. I.; Tsyba, I.;

Bau, R.; Thompson, M. E. Inorg. Chem. 2002, 41, 3055.

5640 Inorganic Chemistry, Vol. 49, No. 12, 2010

Ladouceur et al.

Figure 12. (A) Comparison between experimental (77 and 298 K) and calculated emission energies (λ_max) using methods 1-3 for 8a-8g (see the text).

(B) Comparison between calculated (method 2) and experimental emission energies at 77 K (λ_max).

the T₁and S₀states (ΔSCF approach). Method 3 calculates

the energy difference between the T₁and S₀states as

determined directly from TD-DFT methods using restricted

B3LYP.⁶⁵The calculated emission energies using these three

methods are summarized in Table 7, and a comparison with

those measured is shown in Figure 12.

studies indicate that intense emission in ACN occurs primar-

3

ily from a mixed MLCT and LLCT transition except for

those complexes bearing dimethylamino substituents at the 4⁰

position of the aryl groups. These latter complexes emit with

much diminished intensity from a ³ILCT state. Modest

emission wavelength tuning of (ca. 30 nm) at RT was achieve

by increasing the effective conjugative length of the ancillary

ligand. Complexes with aryl substituents bearing o,o-dimeth-

yl groups do not impact the emission energy at RT because of

their relative orthogonal orientation with respect to the bpy

plane, yet they emit brightly in the orange, with quantum

yields ranging from 16 to 22% in ACN, except for 8f. The

appending of peripheral aryl groups about the bpy ancillary

ligand to promote supramolecular π-stacking interactions

has recently been shown to increase the device stability.⁶⁷The

use of ancillary ligands such as ours, which can effectively

shield the iridium(III) center and its bonding to the surround-

ing ligands from adventitious attack by small molecules,

would be amenable to exploiting this same interaction.^3d

In fact, with the exception of 8c and 8f, the complexes within

this study luminesce in the orange-red as brightly as the

best cationic iridium-based luminophores found in the litera-

ture. The construction and physical and photophysical

characterization of LEECs using our complexes will be

reported elsewhere.

The electronic structures, as illustrated by the orbital

patterns, in the T₁state are similar to those found for

the S₁state (e.g., Figures 10B,D and S14-S21 in the Sup-

porting Information). The electronic structures obtained

from DFT calculations indicate that emission (HOMO r

LUMO) for all complexes, except 8c and 8f, can be described

as mainly an admixture of LLCT/MLCT transitions. In these

latter two cases, emission is best described as occurring

through an ILCT transition. The predicted emission energies

not surprisingly better reflect those measured at 77 K because

at this temperature the frozen conformation is better mir-

rored to that of the predicted geometry. Overall, all three

methods systematically underestimate the emission energy;⁶⁶

however, method 2 correlates quite accurately with experi-

ment, as demonstrated by the overall linear fit found in

Figure 12B. Charge-density analysis for 8d-8g indicates that

functionalization at the 4⁰position has essentially no impact

up the electronics of the bpy ligand, and thus there is

negligible electronic communication through the σ network.

These results suggest that the aryl groups in 8d,8e, and 8g

essentially act to shield the iridium center without affecting

much the luminescence properties of the complexes.

Our calculations demonstrate that method 2 is the most

accurate for reproducing experimental emission energies.

The TD-DFT study provides insight into the origin of the

transitions. Combined DFT/TD-DFT calculations indicate

that frequently the S₀-state HOMO orbitals are localized

over the ppy cyclometalating ligand and the iridium center,

whereas the LUMO is mainly localized on the bpy fragment.

Conclusion

We have synthesized, studied computationally, and mea-

sured the photophysical and electrochemical properties of a

series of new cationic iridium(III) complexes based on 5,5⁰-

diaryl-2,2⁰-bipyridines as ancillary ligands. These combined

Acknowledgment. We acknowledge Professor Pierre

D. Harvey for helpful discussions and for access to the fluo-

rimeter and UV spectrophotometer found in his laboratory.

We thank M. Shawkat Aly for aide during fluorimetry

measurement acquisition. We thank Professor Christian

Reber for modeling the low-temperature emission spectra.

(65) The primary difference between methods 1 and 3 is the following. For

method 1, the energies of the S₀and T₁states are computed separately to obtain

optimized MOs foreach ofthesestates, withthe latter defined asanm_l= 1 state.

In the case of method 3 (using TD-DFT), the excited states are expanded as

determinants built from the ground-state (S₀) orbitals and the triplets are

formally assigned an m_l= 0 state and are coupled as two determinants: single

excitation HOMO/LUMO R with HOMO/LUMO β with a positive sign.

(66) (a) Grimme, S.; Parac, M. ChemPhysChem 2003, 4, 292. (b) Jacquemin,

D.; Perpete, E. A.; Scuseria, G. E.; Ciofini, I.; Adamo, C. J. Chem. Theory Comput.

2008, 4, 123.

(67) Graber, S.; Doyle, K.; Neuburger, M.; Housecroft, C. E.; Constable,

E. C.; Costa, R. n. D.; Ortiz, E.; Repetto, D.; Bolink, H. J. J. Am. Chem. Soc.

2008, 130, 14944.

Article

Inorganic Chemistry, Vol. 49, No. 12, 2010 5641

We also acknowledge Professor Jean Lessard for access to

CV instrumentation and Dr. Jean Marc Chapuzet for aide

in voltammogram interpretation. Computational time on

procedures for all new compounds, methodologies used for

photophysical and electrochemical characterization, details of

computational methodology, ¹H NMR comparison of the

chloride and PF₆salts for 8d, calculated low-temperature emis-

sion spectra, cyclic voltammograms for 8a-8g and 9, table of

calculated S₁and T₁geometries for all complexes, table of

population analysis for four HOMOs and four LUMOs for all

complexes, calculated absorption and emission energies and

isodensity surface plots for selected orbitals of 8a-8g, tables

of the 50 lowest-energy transitions calculated by TD-DFT for

8a-8g and 9, simulated (red) absorption spectra for 8a-8g [the

oscillator strengths are represented by unbroadened vertical

lines of calculated singlet-singlet transitions (green)]. This

material is available free of charge via the Internet at http://

pubs.acs.org.

ꢀ

the Mammouth supercomputer at the Universite de Sher-

ꢀ

ꢀ ꢀ

brooke was underwritten by RQCHP (reseau quebecois

de calculs de haute performance). E.Z.-C. acknowledges

CFI (Canadian Foundation for Innovation), NSERC (the

National Sciences and Engineering Research Council of

ꢀ ꢀ

Canada), FQRNT (Le Fonds quebecois de la recherche

ꢀ

sur la nature et les technologies), and the Universite de

Sherbrooke for financial support.

Supporting Information Available: X-ray crystallographic

data for 8a and 8g in CIF format, complete experimental

Article Doi

Role of substitution on the photophysical properties of 5,5′-diaryl-2,2′-bipyridine (bpy*) in [Ir(ppy) 2(bpy*)]PF6 complexes: A combined experimental and theoretical study

DOI: 10.1021/ic100521t

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ic100521t

Full text of DOI:10.1021/ic100521t

Products guided by the article

R&D Labs maybe for 1232567-91-7

Relevant to this article

Hot Product

Role of substitution on the photophysical properties of 5,5′-diaryl-2,2′-bipyridine (bpy) in [Ir(ppy) 2(bpy)]PF6 complexes: A combined experimental and theoretical study