Luminescent Iridium Complexes and Their Applications

Abstract

Considerable studies have been made on iridium complexes during the past 10 years, due to their high quantum efficiency, color tenability, and potential applications in various areas. In this chapter, we review the synthesis, structure, and photophysical properties of luminescent Ir complexes, as well as their applications in organic light-emitting diodes (OLEDs), biological labeling, sensitizers of luminescence, and chemosensors.

1 Introduction

There has been growing interest in luminescent iridium (Ir) complexes due to their high quantum efficiency and color tunability. The highly efficient emission is attributed to strong spin-orbit coupling caused by the presence of the 5d metal ion, which leads to efficient intersystem crossing from a singlet to a triplet excited state. Mixing of the singlet and triplet excited states via spin-orbit coupling relaxes the spin-forbidden nature of the radiative relaxation of the triplet state [1, 2]. Color diversity arises because excited states in Ir complexes are ligand-related. The energy of the lowest excited state can therefore be controlled by deliberately adjusting the energy of ligand orbitals through substituent effects [3, 4] or by entirely changing the cyclometalated ligand structure [5, 6].

Various cyclometalated Ir complexes have been reported, which can be classified into two main groups. The first group is neutral, containing Ir(C^N)₃ [2, 4, 7–18] (C is a cyclometalated carbon and N is a heterocyclic nitrogen) and Ir(C^N)₂(LX) [6, 19–28] (LX represents an ancillary ligand) type with tris-bidentate ligands, and Ir(N^C^N)(C^N^X) (X is an anionic ligand or cyclometalated carbon) [29–31] type with bi–tridentate ligands. The other group is ionic, including [Ir(C^N)₂(N^N)]⁺ [32–40] type with tris-bidentate ligands, [Ir(C^N)₂(L)₂]⁻ [41] (L denotes an anionic ancillary ligand) type with unidentate and bidentate ligands, and [IrN_6−n C _n ]⁽³⁻ⁿ⁾⁺ (n = 0, 1, 2) [31, 42–44] type with tridentate ligands. Because of their rich photophysical properties, Ir complexes have been widely studied in many applications, such as organic light-emitting diodes (OLEDs) [1, 2, 6, 8, 10, 13, 19, 20, 24, 26, 41, 45–51], light-emitting electrochemical cells (LECs) [52–60], biological labeling [33–36, 61–64], sensitizers of luminescence [65–67], and chemosensors [11, 68–71].

The synthesis, structure, and photophysical properties of luminescent Ir complexes, as well as their applications are introduced in this chapter.

2 Synthesis and Structure

2.1 Neutral Iridium Complex

2.1.1 Ir(C^N)₃ Type

In 1985, Watts et al. reported the first Ir(C^N)₃-type Ir complex facial-Ir(ppy)₃ (fac-1), which was formed in 10% yield as a side product in the reaction of Hppy (2-phenylpyridine) with hydrated IrCl₃ [16]. They subsequently attempted to extend this procedure with methyl-substituted ppy ligands, but only trace amounts of the desired complex were obtained [4]. In 1991, they developed a new procedure for the high-yield synthesis of fac-Ir(C^N)₃ with ppy and other substituted ppy ligands [4]. The procedure utilized the starting material Ir(acac)₃ (acac = 2,4-pentanedionate) instead of hydrated IrCl₃ (Fig. 1). The method typically produced fac-Ir(C^N)₃ in high yields of 40–75%, but did not prepare complexes for other structural cyclometalated ligands. To solve this problem, Güdel et al. reported another method in 1994 which involved treating a μ-dichloro bridged dimer complex [Ir(C^N)₂Cl]₂ with excess HC^N (free cyclometalated ligand). They found that fac-Ir(C^N)₃ complexes containing different cyclometalated ligands could be synthesized in high yield [17].

Fig. 1

Schematic representation of the reaction mechanism in the synthesis of Ir(C^N)₃ type complex fac-1

Since the initial report in 1999 using fac-1 as an emitter achieved highly efficient phosphorescent OLEDs, many Ir(C^N)₃-type Ir complexes, including small molecular complexes [8, 9, 15] and dendritic complexes [72–78], have been reported. The synthesis and structure of these complexes have been thoroughly investigated.

In general, there are three synthetic methods to prepare Ir(C^N)₃-type complexes (Fig. 2). Methods B and C have several advantages over method A. For example, Ir(C^N)₂(O^O) (O^O = 2,2,6,6-tetramethyl-3,5-heptanedionate) and [Ir(C^N)₂Cl]₂ compounds are easily prepared in high yield from a less expensive starting material; methods B and C give higher yields than method A in the last reaction step [15].

Fig. 2

Three synthetic methods to prepare Ir(C^N)₃ type Ir complex

Meridional (abbreviated as mer) isomers of Ir(C^N)₃ complexes can also be prepared as a different steric geometry configuration of the facial isomers (Fig. 3). Results show that fac-Ir(C^N)₃ tends to form under higher temperature (suggesting that fac isomers are thermodynamically favored products), whereas mer-Ir(C^N)₃ can be obtained at lower temperature (suggesting that mer isomers are kinetically favored products). This is consistent with the experimental phenomenon that the mer isomer can be converted to the fac isomer under treatment at high temperature [15].

Fig. 3

Steric configuration for fac-Ir(C^N)₃ and mer-Ir(C^N)₃

Meridional and fac isomers are different not only in steric geometry configuration, but also in Ir–N and Ir–C bond lengths. Key bond lengths of fac-Ir(tpy)₃ (fac-2, tpy is 4-methylphenylpyridine) and mer-Ir(tpy)₃ (mer-2) are compared in Fig. 4. In the fac isomer, all the Ir–C bonds are trans to a pyridyl group, and the Ir–N bonds are trans to a phenyl group, leading to identical Ir–C and Ir–N bond lengths. In the mer isomer, some of the bond lengths differ markedly from those of the fac isomer. This may be because Ir–C (or Ir–N) bonds in the mer isomer share an identical/different electronic environment to Ir–C (or Ir–N) bonds in the fac isomer.

Fig. 4

Crystal structures and key bond lengths of Ir(C^N)₃ type Ir complexes fac-2 and mer-2

2.1.2 Ir(C^N)₂(LX) Type

The universal synthetic method used to prepare an Ir(C^N)₂(LX)-type complex is shown in Fig. 5. Dimer complex [Ir(C^N)₂Cl]₂ is readily prepared from reaction of the ligand precursor and IrCl₃ · nH₂O [79], and chloro-ion ligands can be subsequently replaced with an LX chelate. The most studied ancillary ligand LX is acac, but it can be varied with other monoanionic bidentate ligands, such as picolinic acid, N-methylsalicylimine [23], 2-(5-phenyl-4H-[1,2,4]triazol-3-yl)-pyridine [28], 2,2,6,6-tetramethyl-3,5-heptanedionate, 1-phenyl-4,4-dimethyl-1,3-pentanedionate, 1,3-diphenyl-1,3-propanedionate, pyrazolyl, pyrazolyl-borate [80], and (2-pyridyl) pyrazolate derivatives [81].

Fig. 5

Synthetic route for Ir(C^N)₂(LX) type complex

To understand the structure of the Ir(C^N)₂(LX)-type Ir complex, the crystal structure of Ir(tpy)₂(acac) (3, Fig. 6) was compared with fac-2 and mer-2 mentioned above. The bis-cyclometalated fragment of 3 has the same disposition of tpy ligands as found in mer-2, and the mutually trans-disposed Ir–N bonds in both complexes have similar lengths (av = 2.032(5) Å). The weak trans influence of the acac ligand leads to shorter Ir–C bonds (av = 1.984(6) Å) for the complex 3 than those observed in complex fac-2 or mer-2.

Fig. 6

Chemical and crystal structures of Ir(C^N)₂(LX) type complex 3

There is also a polymerized Ir(C^N)₂(LX)-type Ir complex in which the bis-cyclometalated fragment Ir(C^N)₂ is incorporated on the side chain of a polymer. The design of the entire system could have better dissolution and film-forming ability as that of the polymer, as well as high quantum efficiency similar to that of luminescent Ir complexes. The chemical structure of a selected polymerized Ir(C^N)₂(LX)-type Ir complex is shown in Fig. 7 [82].

Fig. 7

Chemical structure of a polymerized Ir(C^N)₂(LX)-type Ir complex

2.1.3 Ir(N^C^N)(C^N^X) Type

Neutral Ir complexes mentioned above are all composed of bidentate ligands. Believing tridentate ligands can lead to different excited state, stereoisomer, and linear assembly, some groups focused their research on Ir complex with tridentate ligands. Figure 8 shows the synthetic methods of Ir(dpyx)(dppy) (4) and Ir(dpyx)(tppic) (5), two neutral Ir complexes with tridentate ligands. Complex 4 was obtained by heating the chloro-bridged dimer [Ir(dpyx)Cl(l-Cl)]₂ with AgOTf in molten 2,6-diphenylpyridine (dppyH), followed by rapid chromatographic purification [30], while complex 5 was synthesized upon reaction of the dimer with 4-hydroxybenzo[h]quinoline-2-carboxylic acid (tppicH) using molten benzoic acid as the solvent [29]. The N^C^N tridentate coordination mode of dpyx was confirmed in [Ir(dpyx)-(DMSO)Cl₂] (DMSO is dimethyl sulfoxide) by X-ray crystal structure [29].

Fig. 8

Synthetic methods of Ir(N^C^N)(C^N^X) type complexes 4 and 5

2.2 Ionic Iridium Complex

2.2.1 [Ir(C^N)₂(N^N)]⁺ Type

The cyclometalated Ir complexes investigated earlier with a formula of [Ir(C^N)₂(N^N)]⁺ are complexes [Ir(ppy)₂(bpy)]Cl (6) and [Ir(ppy)₂(bpy)]PF₆ (7) (bpy and PF₆ represent 2,2′-bipyridine and hexafluorophosphate, respectively) [39]. The two cationic Ir complexes were prepared by a modified method employed by Nonoyama [83]. The detailed synthetic route is shown in Fig. 9. The stereo structure of these [Ir(C^N)₂(N^N)]⁺ complexes is similar to those of Ir(C^N)₂(LX) complexes, with a trans-N,N configuration of the C^N ligands, whereas the diimine ligand is located opposite cis-oriented carbon atoms, completing an octahedral arrangement [84].

Fig. 9

Synthetic route to [Ir(C^N)₂(N^N)]⁺ type complexes 6 and 7

2.2.2 [Ir(C^N)₂(L)₂]⁻ Type

An [Ir(C^N)₂(L)₂]⁻ type anionic Ir complex can be conveniently synthesized in low-boiling solvent by reacting the corresponding [Ir(C^N)₂Cl]₂ dimer complex with a pseudohalogen ligand such as tetrabutylammonium cyanide, tetrabutylammonium thiocyanate, or tetrabutylammonium cyanate [41, 85]. Figure 10 shows the chemical and crystal structure of the complex (C₄H₉)₄N[Ir(ppy)₂(CN)₂] (8) [41, 85]. The Ir atom is octahedrally coordinated by four ligands, with the N atoms of the 2-phenylpyridine ligands in a trans disposition similar to that in [Ir(ppy)₂Cl]₂, whereas cyanide ligands coordinate through the carbon atom and adopt a cis configuration.

Fig. 10

Chemical and crystal structures of [Ir(C^N)₂(L)₂]⁻ type complex 8

These complexes are stable at room temperature as a solid and in solution containing noncoordinating solvents such as dichloromethane, chloroform, methanol, or ethanol. This type of Ir complex may undergo slow substitution of the pseudohalogen ligands upon standing for days in a strong coordinating solvent such as dimethyl sulfoxide [41].

2.2.3 [IrN_6−n C _n ]⁽³⁻ⁿ⁾⁺ (n ═ 0, 1, 2) Type

Figure 11 shows chemical structures for [IrN_6−n C _n ]⁽³⁻ⁿ⁾⁺ (n = 0, 1, 2) type Ir complexes with tridentate ligands. Complex 9 was first reported by Degraff and coworkers, with fusion reaction and arduous purification [86]. Subsequently, a milder and stepwise route was described by Sauvage, involving initial reaction of terpyridine (tpyH) with IrCl₃ · nH₂O in ethylene glycol to give [Ir(tpy)Cl₃] as an intermediate, followed by reaction with a second equivalent of tpy in refluxing ethylene glycol [87]. Purification of the tricationic complex is normally best achieved after ion exchange to the hexafluorophosphate salt, which offers the advantage of being soluble in polar organic solvents for column chromatography [31]. In order to obtain complexes 10 and 11, compound 1,3-di(2-pyridyl)-4,6-dimethylbenzene (dpyxH) was employed to react with IrCl₃ · nH₂O to give a chloro-bridged dimer [Ir(dpyx)Cl(μ-Cl)]₂, reactions of the dimer with 4′-tolylterpyridine (tppyH) and 6-phenyl-2,2′-bipyridine (phbpyH) gives complex 10 and 11, respectively [29, 30]. An X-ray diffraction study of a single crystal of the former confirms the mutually orthogonal orientation of the two ligands, each bound tridentately [30].

Fig. 11

Chemical structures of [IrN_6−n C _n ]⁽³⁻ⁿ⁾⁺ (n = 0, 1, 2) type complexes 9–11

3 Photophysical Properties

3.1 Neutral Iridium Complex

3.1.1 Ir(C^N)₃ Type

The photophysical properties of Ir(C^N)₃ have been examined by several research groups [8, 15]. Two principal transitions are observed in this type of Ir complex: (1) metal-to-ligand charge transfer (MLCT) in which an electron is promoted from a metal d orbital to a vacant π* orbital on one of the ligands and (2) ligand-centered (LC) transitions in which an electron is promoted between π orbitals on one of the coordinated ligands. Phosphorescence in Ir(C^N)₃-type complexes enabled by strong spin-orbit coupling mainly arise from a mixture of ³LC and ³MLCT excited states, whereas the emissive excited state is predominantly the excited state having the lowest energy.

The lowest excited state of a Ir(C^N)₃-type complex can be tuned by changing the substitution of electron-donating or electron-withdrawing groups on the cyclometalated ligand, or by entirely changing the cyclometalated ligand structure [15]. Table 1 lists the photophysical properties of complexes 1 (Fig. 1), 2 (Fig. 4), Ir(46dfppy)₃ (12), Ir(ppz)₃ (13), Ir(46dfppz)₃ (14), and Ir(tmfppz)₃ (15) (Fig. 12). Photophysical properties of complex 2 (14) are different from that of 12 (15), but both are similar to those of parent compound 1 (13). This is unsurprising because substitution of donor or acceptor groups tunes the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) levels of the metal complex in parallel, leading to marginal changes in maximal emission. Photophysical properties of complex 13 are entirely changed compared with complex 1, which is ascribed to skeletal change of the cyclometalated ligand.

Fig. 12

Chemical structures of complexes 12–15

Table 1 Photophysical properties of complexes 1, 2, 12–15 in 2-methyltetrahydrofuran

Photophysical properties of mer-Ir(C^N)₃-type complexes are very different to those of fac-Ir(C^N)₃-type complexes (Table 1). Meridional ones traditionally exhibit a broad, red-shifted emission, and lower quantum efficiencies, which can be attributed to the reason that the mer configuration usually arises from strongly trans influencing phenyl groups being opposite each other [88].

3.1.2 Ir(C^N)₂(LX) Type

The lowest triplet energy level of the ancillary ligand LX lie well above the energies of LC and MLCT excited states in most of the Ir(C^N)₂(LX)-type complexes, so luminescence of Ir(C^N)₂(LX) is dominated by ³LC and ³MLCT transitions. This leads to similar phosphorescence emission to the fac-Ir(C^N)₃ complexes with the same cyclometalated ligand [6, 23]. In such cases, density functional theory (DFT) calculations indicate that HOMOs are largely metal-centered, whereas LUMOs are primarily localized on the heterocyclic rings of the cyclometalated ligand. The ancillary is therefore not directly involved in the lowest excited state.

Though the ancillary ligand is not directly involved in the lowest energy excited state when it has a higher triplet energy level, it can alter the excited energy state by modifying electron density at the metal center [80]. Figure 13 shows the emission spectra of Ir(tpy)₂(pz₂Bpz₂) (16), Ir(tpy)₂(pz₂H) (17), Ir(tpy)₂(pzH)Cl (18), and Ir(tpy)₂(acac) (3) at room temperature. All have the same “Ir(tpy)₂” fragment, but there is a clear blue-shift in the maximum emission wavelength as the ancillary ligand is changed. This is attributed to a net electron-withdrawing ancillary ligand that pulls electron density away from the Ir atom, thereby stabilizing the metal orbitals and lowering the HOMO energy; whereas the LUMO of the Ir(tpy)₂(LX) complexes, localized on the pyridyl rings, is expected to be largely unaffected with respect to energy. The HOMO-LUMO gap should consequently increase for stronger electron-withdrawing ancillary ligands, leading to a blue-shift in emission.

Fig. 13

Chemical structures of 16–18, and their photoluminescent spectra in 2-methyltetrahydrofuran compared with 3 at room temperature

If the lowest triplet energy level of the LX ligand is lower in energy than the ³LC or ³MLCT, it will be the lowest energy excited state, and thus a switch from “Ir(C^N)₂” to LX-based emission can be observed. Our recent work confirms this hypothesis [89]. Twelve Ir complexes (complexes 19–30, Fig. 14) with general formula Ir(C^N)₂(LX) were synthesized by changing the triplet energy level of the cyclometalated and the β-diketonate ligands, and photophysical properties were compared. Strong ³LC- or ³MLCT-based phosphorescence was observed if triplet-state density maps of the β-diketonate and the Ir(C^N)₂ fragment were not superimposed (triplet energy of the former is high). If the triplet-state density map of the two parts was superimposed, ³LC- or ³MLCT-based transition would be quenched at room temperature, and DFT calculations show that the lowest excited state is determined by the cyclometalated and ancillary ligand (Fig. 15).

Fig. 14

Chemical structures of complexes 19–30

Fig. 15

State density of triplet energy diagram for the β-diketonates and Ir(C^N)₂ fragments of complexes 19–30

Interligand electron transfer (ILET) from the cyclometalated ligand to the ancillary ligand is also possible in some excited Ir(C^N)₂(LX) complexes, particularly if the triplet energy of the ancillary ligand is much lower. Park et al. observed emission across the visible spectrum by changing the structure of the ancillary ligand in a series of complexes undergoing ILET before emission [90–92].

3.2 Ionic Iridium Complex

3.2.1 [Ir(C^N)₂(N^N)]⁺ Type

The photophysical properties of [Ir(C^N)₂(N^N)]⁺-type cationic Ir complexes are complicated. In some excited complexes, the neutral, diimine and cyclometalated ligands provide orbitals that participate in excited-state transitions. The cyclometalated ligand tends to be associated with the ³LC transition, and the diimine ligand with the ³MLCT transition [93]. In such cases, the excited state of the complex can be tuned directly through ligand modification because each ligand is linked to a different transition. Nazeeruddin et al. investigated three cationic Ir complexes 31–33 (Fig. 16) by modulating the electronic structure of the complex using selective ligand functionalization [94]. They developed a strategy to tune the phosphorescence wavelength for this class of compound [94]. The electron-withdrawing substituent on the C^N ligand decreased donation to the metal and therefore stabilized the metal-based HOMO, whereas the electron-releasing substituent on the N^N ligand led to destabilization of the N^N ligand-based LUMO, ultimately leading to increased HOMO-LUMO gaps and emission energies (Table 2).

Fig. 16

Chemical structures of 31–33

Table 2 Photophysical properties of complexes 31–33 in acetonitrile at 298 K

It was proved that photophysical properties of [Ir(C^N)₂(N^N)]⁺-type complexes can also be adjusted by modifying the conjugated lengths of the diimine ligand [84]. Theoretical calculations, photophysical studies and electrochemical studies of a series of cationic Ir complexes 34–39 (Fig. 17) showed that their excited states simultaneously contain ³MLCT, triplet ligand-to-ligand charge transfer (³LLCT), and ³LC transitions. Their emission wavelengths can therefore be tuned significantly (∼150 nm) by changing the conjugated length of N^N ligands.

Fig. 17

Chemical structures and emission spectra of 34–39

3.2.2 [Ir(C^N)₂(L)₂]⁻ Type

The structures of cyclometalated ligand C^N and spectator ligand L should be considered when evaluating the photophysical properties of [Ir(C^N)₂(L)₂]⁻-type Ir complexes. By keeping the cyclometalated ligand C^N as 2-phenylpyridine, Nazeeruddin et al. introduced three pseudohalogens, CN⁻, NCS⁻, and NCO⁻, as spectator ligand L [41]. They investigated the influence of the spectator ligand on the photoluminescent properties of this type of complex {(C₄H₉)₄N[Ir(ppy)₂(CN)₂], complex 8 in Fig. 10; (C₄H₉)₄N[Ir(ppy)₂(SCN)₂], complex 40, and (C₄H₉)₄N[Ir(ppy)₂(OCN)₂], complex 41, in Fig. 18)} in 2003 [41]. Introducing a strong ligand field strength spectator ligand such as CN⁻ increased the gap between LUMO of the phenyl pyridine ligand and metal t _2g orbitals, resulting in a blue-shift in the emission spectrum. The gap between the LUMO of the phenyl pyidine ligand and the vacant metal e _g orbitals effectively increased because of the inhibition of nonradiative pathways by the cyanide ligands, which also led to these complexes displaying unusually high phosphorescence quantum yields in solution at room temperature.

Fig. 18

Chemical structures of complexes 40–45

While maintaining the spectator ligand L as CN⁻, Nazeeruddin et al. traced the effect of cyclometalated ligands (8 in Fig. 10; 42–45 in Fig. 18) in 2008 [85]. They found that introduction of 4-dimethylaminopyridine can destabilize LUMO and HOMO orbitals in parallel, leading to marginal changes in emission spectra, but it increased nonradiative rate constants, which in turn led to a reduction of overall quantum yield. They also found that introducing fluorine atoms on the phenyl groups led to stabilization of HOMO orbitals. This led to an increase in the HOMO-LUMO gap and was accompanied by a blue-shift in emission spectra, but it did not substantially influence quantum yields because of a slight increase in radiative rate constants or a decrease in overall nonradiative rate constants. Based on these findings, they provided an interesting approach for tuning the phosphorescence wavelength from 470 to 450 nm of anionic Ir complexes, maintaining a high phosphorescence quantum yield by modification with donor and acceptor substituents on the pyridine and phenyl moieties of 2-phenylpyridine (detailed photophysical data are summarized in Table 3).

Table 3 Photophysical properties of complexes 8 and 40–45 in acetonitrile solution at 298 K

The photophysical properties of both neutral Ir(N^C^N)(C^N^X) type and ionic [IrN_6−n C _n ]⁽³⁻ⁿ⁾⁺(n = 0, 1, 2) type Ir complexes with tridentate ligands, in particular how the number of cyclometalating carbon atoms in the coordination sphere of the metal ion influences the luminescence, were comprehensively reviewed by Williams in 2008 [31].

4 Applications of Luminescent Iridium Complexes

4.1 Applications in OLEDs

OLEDs are electroluminescent devices in which electrical energy is converted to luminous energy. A typical configuration is depicted in Fig. 19. Two electrodes sandwich one or more layers of the organic functional films. A voltage of 2–10 V is typically applied between the electrodes, and electrons are injected into the LUMO of the organic material from a low work function metallic cathode such as Mg–Ag or Li–Al. Holes are injected into the HOMO of the organic material from a high work function anode such as indium–tin–oxide (ITO). Electrons and holes move towards the middle region of the emitting layer under the influence of the applied field. Energetic electrons can drop into the holes to form singlet and triplet excitons which release their energy as photons escaping through the transparent electrode.

Fig. 19

Typical configuration of OLEDs, ETL: electron-transporting layer; EML: emitting layer; HTL: hole-transporting layer

4.1.1 Emitters for Doping OLEDs

Ir complexes have been doped into a charge-transporting host material to optimize device efficiency up to a theoretical limit. The remarkable enhancement in efficiency upon doping has been ascribed to a favorable triplet energy level alignment between host material and Ir complex dopant. Triplet–triplet annihilation, which is a key adverse factor for phosphorescence-based emitters, can also be greatly inhibited by dispersing emitter molecules into the host matrix. Highly efficient red, green, blue (RGB) devices (which are necessary for full-color displays) can be explored by doping different color emission Ir complexes into an appropriate host.

Green-phosphorescent OLED is the best developed RGB device. Cyclometalated fac-1 and its heteroleptic analog such as Ir(ppy)₂(acac) (20 in Fig. 14) have been extensively applied in fabricating green-emitting electroluminescent devices. In 1999, Forrest et al. described high-efficiency OLEDs employing fac-1 as the dopant and CBP (Fig. 20) as the host [2]. The combination of a short triplet lifetime and reasonable photoluminescent efficiency allows fac-1 based OLEDs to achieve peak quantum efficiencies of 8.0%. In 2000 they demonstrated much higher efficiency OLEDs by doping fac-1 into an electron-transport layer host TAZ (Fig. 20) [10]. A maximum external quantum efficiency of 15.4 ± 0.2% was achieved. In 2001, they again demonstrated very high efficiency OLEDs employing 15 doped into TAZ host [20]. A maximum external quantum efficiency of 19.0 ± 1.0% was achieved. The calculated internal quantum efficiency of 87 ± 7% is supported by the observed absence of thermally activated nonradiative loss in the photoluminescent efficiency of 20. Very high external quantum efficiencies are therefore due to the nearly 100% internal phosphorescence efficiency of the Ir complex coupled with balanced hole and electron injection, as well as triplet exciton confinement within the light-emitting layer.

Fig. 20

Chemical structures of CBP and TAZ

Compared to green OLEDs, development of red OLEDs has been slower. This is because red emission originated from a smaller energy gap transition, which increased the difficulty of material searching. Many nonradiative pathways originating from strong π–π interaction or charge transfer between ligands may be present, leading to a reduction in efficiency in the OLEDs. The earliest report concerning a red emitting Ir complex as an emitter was by Forrest et al. in 2001 [19]. The device achieved a maximum external quantum efficiency of 7.0 ± 0.5% using a red phosphor Btp₂Ir(acac) (46, Fig. 21) as the dopant and CBP as the host. In 2003, Tsuboyama et al. fabricated an efficient red OLED device using a novel Ir emitter Ir(piq)₃ (47, Fig. 21) as a dopant [8]. The maximum external quantum efficiency was 10.3%. Duan et al. synthesized two new orange–red Ir complexes, Ir(DBQ)₂(acac) (21, Fig. 14) and Ir(MDQ)₂(acac) (48, Fig. 21) [24]. The devices based on these two complexes emit orange–red light, and the maximum external quantum efficiency is 12%.

Fig. 21

Chemical structure of red emitters 46–48

Highly efficient blue OLEDs are necessary to realize RGB full-color displays. One of the best known phosphorescence blue emitters is FIrpic (49, Fig. 22) [95]. A maximum external electroluminescent quantum efficiency of 7.5 ± 0.8% was obtained by doping 49 into mCP (Fig. 23). The result represents a significant increase in efficiency over a similar device structure based on the host CBP, where the maximum external quantum efficiency was 6.1 ± 0.6%. Data mentioned above indicate that the energy differences in the triplet energies of host and guest materials are very important for confinement of electro-generated triplet excitons on dopant molecules, thus the key factor to achieve highly efficient blue OLEDs is to find appropriate guest and host materials. Based on this guidance, Thompson et al. demonstrated efficient, deep-blue OLEDs using a charge-trapping phosphorescent guest FIr6 (50, Fig. 22) doped in the wide-energy-gap hosts UGH1 and UGH2 (Fig. 23) [25]. Peak quantum efficiencies of 8.8 ± 0.9% in UGH1 and 11.6 ± 1.2% in UGH2 were obtained. In 2005, Yeh et al. reported a series of OLEDs based on FIrN4 (51, Fig. 22) or 49 dopant emitters coevaporated with mCP or SimCP (Fig. 23) [48]. The device based on 51 and mCP was found to be one of the bluest OLEDs. The device based on 49 and SimCP achieved maximum external electroluminescent quantum efficiency of 14.4%. Considering UGH1 and UGH2 are less conductive compared with the carbazole derivatives, and that the glass transition temperatures of these two compounds are low for practical application, Lin et al. synthesized two hosts, BSB and BST (Fig. 23), as well as a new blue Ir complex FIrpytz (52, Fig. 22) [96]. By using BSB as the host for the blue emitter 52, highly efficient blue OLEDs with external quantum efficiency of 19.3% was achieved. Su et al. recently reported a unique molecular design strategy of combining a carbazole electron donor with a high triplet energy and a pyridine electron acceptor with high electron affinity to give a novel bipolar host material of 26DCzPPy (Fig. 23) [97]. By using the host for 49-based blue phosphorescent OLEDs, an external quantum efficiency of 24% was achieved at the practical brightness of 100 cd m⁻². Even at a brighter emission of 1,000 cd m⁻², an efficiency of 22% was obtained.

Fig. 22

Chemical structures of blue emitters 49–52

Fig. 23

Chemical structures of some host materials for blue emission iridium complex

Despite the investigation of RGB OLEDs for display applications, Ir complexes have also been used to exploit near-infrared (NIR) OLEDs for optical communication and biomedical application. Jabbour et al. demonstrated the first example of NIR OLEDs fabricated by the cyclometalated Ir complex NIR1 (53, Fig. 24) by increasing the size of the cyclometalated ligand π system [98]. The devices exhibited exclusive emission with a peak value at 720 nm, and the external quantum efficiency was nearly 0.1%.

Fig. 24

Chemical structures of 53–55

Ir complexes were also used to develop white OLEDs (WOLEDs) for large-scale production of solid-state light sources and backlights in liquid-crystal displays. Several device architectures have been introduced to achieve high brightness and efficiency in WOLEDs. By controlling the recombination current within individual organic layers, emission from red 46, yellow 54 (Fig. 24), and blue 49 was balanced to obtain white of the desired color purity [99]. The most significant disadvantage of this structure is its relatively high operating voltage due to the combined thicknesses of the many layers used in the emission region. To solve this problem, Forrest et al. mixed red 55 (Fig. 24), green fac-1, and blue 50 into a wide energy gap UGH2 host to ensure that all emission originates from a single thin layer. The process of direct triplet exciton formation of the 50-UGH2 system leads to a reduction in operating voltage, and hence an increase in power efficiency up to 42 ± 4 lm W⁻¹ was obtained [100].

Doping OLEDs mentioned above are all using fluorescent materials as the host material. Phosphorescent materials can also be used as host material and their performances appear to be significant [50, 101]. Our results proved that highly efficient phosphorescent OLEDs can be pursued using Ir complexes as the host materials [102]. In detail, six devices using 20, 23 (Fig. 14), 56, 57 and 58 (Fig. 25) or CBP as the host for an orange–red Ir complex 21 (Fig. 14) were fabricated with an identical configuration. Results show that the devices using Ir complex hosts had better performances than that of the device based on a CBP host. In addition, steric hindrance and exciton-transporting ability of the host were found to be the most important factors for this type of doped system.

Fig. 25

Chemical structures of 56–58

4.1.2 Emitters for Non-doped OLEDs

Up to now, most positive results based on Ir complex emitters have been obtained by using a host–guest doped emitter system to improve energy transfer and to avoid triplet–triplet annihilation which occurs when phosphorescent materials are used in OLEDs. Considering that reproducibility of the optimum doping level requires careful manufacturing control, and that long-playing operation of devices may lead to phase separation of guest and host materials, doped OLEDs are relatively more difficult to adapt to practical application than their nondoped counterparts [103]. High-performance phosphorescent OLEDs fabricated by a much simplified, nondoped method are rarely studied, and their typical performances with respect to brightness and efficiency are far from satisfactory [13, 25, 104, 105]. It is therefore desirable to design nondoped electrophosphorescent devices with high brightness and efficiency.

A novel red phosphorescent Ir complex containing carbazole-functionalized β-diketonate 24 (Fig. 14) was designed, synthesized, and characterized in our earlier report [106]. Electrophosphorescent properties of a nondoped device using complex 24 as emitter were examined. The nondoped device achieved maximum lumen efficiency of 3.49 lm W⁻¹. This excellent performance can be attributed mainly to the improved hole-transporting property that benefits exciton transportation. Encouraged by the fact that functionalized β-diketonate can have such an important role in the Ir complex, and with a wish to design high-efficiency nondoped green and blue OLEDs, we synthesized two complexes both containing carbazole-functionalized β-diketonate, 22 and 23, and subsequently fabricated nondoped devices using these two complexes [107]. Lumen efficiency of the green emission device using 23 as emitter was 4.54 lm W⁻¹, whereas the device based on a blue–green 22 achieved maximum lumen efficiency of 0.51 lm W⁻¹. A very simple device and two double-layer devices 23 were also fabricated, and the designed Ir complex was proved to be with good hole-transporting ability and electron-transporting ability. This study provided a special type of doping technology that should have a more general use in emitter design.

As opposed to neutral Ir complexes, charged Ir complexes are mainly synthesized to develop LECs (another type of OLED) [55, 59, 108–110]. In these devices, electrons and holes injected from two air-stable electrodes into a single layer of organic semiconductor recombine, giving rise to light emission. LECs have advantages over traditional OLEDs: (1) LECs require only a single layer of organic semiconductor, whereas traditional OLEDs require a multilayered structure for charge injection, transport, and light emission, and (2) charge injection in an LEC is insensitive to the work function of the electrode material, thereby permitting use with a wide variety of metals as cathode materials. These suggest that LECs may be a promising alternative for solid-state lighting technologies.

4.2 Biological Labeling Reagents

To investigate the potential of luminescent Ir complexes as biological labeling reagents, Lo et al. [33–36, 61–63] reported a series of [Ir(C^N)₂(N^N–R)](PF₆) (R = CHO, NCS and NHCOCH₂I) complexes incorporating aldehyde, isothiocyanate, and iodoacetamide groups in the ligands. These Ir complexes can be functionalized as biological labeling reagents due to their ability – covalently or noncovalently – to bind biomolecules. Crosslinked products showed different emission colors than their free analogs due to the more hydrophobic environment associated with the protein molecule. A comprehensive review of luminescent Ir complexes used as biological labeling reagents was reported by Lo et al. in 2005 [64].

We recently reported two cationic Ir complexes, 59 (Fig. 26) with bright green emission, and 60 (Fig. 26) with red emission, as phosphorescent dyes for imaging of living cells [111]. These two Ir complexes have advantages when used as bio-imaging agents: exclusive staining in cytoplasm, low cytotoxicity, reduced photobleaching, permeability to cell membranes, and moderate luminescence efficiencies in buffer solution.

Fig. 26

Chemical structures of 59, 60, and their bio-imaging application

As shown in Fig. 27, after continuous excitation at 405 nm for 480 s, luminescence intensity of 4′,6′-diamidino-2-phenylindole dihydrochloride (DAPI) (460 ± 20 nm, region 1, channel 1) decreased to 1% of its initial value (owing to photobleaching). Luminescence intensity of 60 (620 ± 20 nm, region 2, channel 2) stayed at essentially one-eighth of the original value during the same period of excitation. This result establishes that the Ir complex show reduced photobleaching and higher photostability than the organic dye. These findings open-up interesting possibilities for using luminescent Ir complexes for imaging of living cells. Investigation of other Ir complexes used in this field will be published in a forthcoming article.

Fig. 27

Comparison of 60 and DAPI for resistance to photobleaching. (a) Confocal luminescence images of fixed HeLa cells stained with 60 and DAPI under continuous excitation at 405 nm with different laser scan times (0, 200, 480 s). (b) Luminescence decay curves of 60 and DAPI during the same period. The signals of DAPI and 60 were collected from region 1 of channel 1 (460 ± 20 nm) and region 2 of channel 2 (620 ± 20 nm), respectively

4.3 Sensitizer of Lanthanide Luminescence

Lanthanide ions such as europium (Eu^III), neodymium (Nd^III), ytterbium (Yb^III), and erbium (Er^III) are becoming increasingly important in applications such as OLEDs [112–115], optical communication [116, 117], medical imaging, and biological labeling [118, 119] due to their excellent luminescence properties originating from f–f transitions. These ions show little or no absorption in the visible region of the spectrum, and often require application of strongly absorbing “antennae” for light harvesting to obtain efficient photoluminescence [120]. Many organic chromophores having high absorbing ability were therefore introduced into lanthanide complexes. The use of strongly-absorbing d-block chromophores (e.g., Ir complexes) as sensitizers has attracted increasing attention [65–67].

The first example of sensitization of Eu^III emission through an Ir complex was reported by De Cola et al. in 2005 (Fig. 28) [67]. Energy transfer from the Ir fragments to the Eu^III center is incomplete, leading to emission of a composite white light instead of Eu^III-based emission of pure red light.

Fig. 28

The first example of sensitization of Eu^III emission through an iridium complex

To make energy transfer more efficient and obtain pure-red emission from Eu^III, a novel ligand with four coordination sites was designed as a bridge to link the Ir^III center and the Eu^III center; and a new Ir^III complex Ir(dfppy)₂(phen5f) [dfppy represents 2-(4′,6′-difluorophenyl)-pyridinato-N,C²′, phen5f denotes 4,4,5,5,5-pentafluoro-1-(1′,10′-phenanthrolin-2′-yl)-pentane-1,3-dionate] was obtained [65]. By using the “complexes as ligands” approach, the novel bimetallic complex {[(dfppy)₂Ir(μ-phen5f)]₃EuCl}Cl₂ was synthesized, and efficient pure-red luminescence from Eu^III was sensitized by ³MLCT energy from the Ir^III moiety (Fig. 29). The excitation window can extend up to 530 nm due to the introduction of the d-block metal moiety, so this bimetallic complex can readily emit red light under sunlight irradiation.

Fig. 29

(a) The crystal structure of “Ir₃Eu” assembly. (b) Pure-red emission of the “Ir₃Eu” assembly in EtOH (1 × 10⁻³ M) upon excitation of a 532-nm laser

As important phosphorescent materials, many Ir^III complexes with different energies of the lowest excited states have been extensively investigated by modifying cyclometalated ligands. This is a good strategy to seek more suitable Ir^III complexes to sensitize NIR lanthanide ions. Our research group has changed the cyclometalated ligand from dfppy to ppy to reduce the triplet energy level of the whole d-block complex-ligand and make it more suitable for NIR Ln^III. Another Ir^III complex Ir(ppy)₂(phen5f) was introduced to form Ir₂Ln (Ln = Nd, Yb, Er) arrays. NIR emission upon photoexcitation of the Ir^III-centered antenna chromophore was successfully obtained (Fig. 30) [121]. An article involving sensitized NIR emission from Yb via direct energy transfer from Ir in a heterometallic neutral complex was published at virtually the same time by De Cola et al. [66].

Fig. 30

Structures of Ir₂Ln (Ln = Nd, Yb, Er) arrays and corresponding near-infrared emissions. C^N denotes cyclometalated ligand ppy, and Y denotes bidentate nitrate anion

4.4 Sensor Applications

Photoluminescent materials in which luminescence output can be modified by interaction with a substrate are being extensively investigated for use as sensors. Luminescent Ir complexes have been used as oxygen sensors, homocysteine sensors, metal cation sensors, and volatile organic compound sensors because of their rich photophysical properties.

4.4.1 Oxygen Sensor

Luminescence-based oxygen sensors work on the principle of luminescence quenching by oxygen; the excited luminophore enables efficient energy transfer to the triplet ground state of molecular oxygen, resulting in a nonradiative luminophore and formation of singlet oxygen [122]. Ir complexes are attractive candidates as novel luminescence-based oxygen-sensing materials (primarily because Ir complexes are the best known luminescent metal complexes). They can be excited with visible light, and their excited state is mainly a mixed ³MLCT and ³LC level, which is prone to quenching by molecular oxygen.

The first example of an Ir complex-based oxygen sensor was reported by Donckt et al. in 1994 [123]. They embedded fac-1 in polystyrene and studied the luminescence properties of the system for use as an oxygen sensor to avoid self-quenching of fac-1 at higher concentration. In 1996, DiMarco et al. [124] reported another Ir complex [Ir(ppy)₂(dpt-NH₂)](PF₆) (where dpt-NH₂ = 4-amino-3,5-di-2-pyridyl-4H-l,2,4-triazole), immobilized in a polymerized poly-(ethyleneglycol) ethyl ether methacrylate (pPEGMA) matrix. The system exhibited excellent properties as a quenchometric oxygen sensor (e.g., linear Stern–Volmer plot, photostability, thermal stability, and reproducibility). They extended their study in 1998 and reported the oxygen sensor ability of a series of mononuclear and dinuclear cyclometalated Ir complexes immobilized in pPEGMA matrixes [125]. The sensitivity and response of the sensor system was affected not only by the lifetime of the Ir complex and the permeability of matrices, but also by the size and charge of the Ir complex. This is important for the design of new solid-state luminescent sensors with improved performance. Efforts were subsequently focused on covalently binding the luminophore to different polymeric hosts, as well as adjusting the structure of the Ir complex [11, 68, 70, 122, 126].

4.4.2 Homocysteine Sensor

Homocysteine has a unique role within physiologic matrices because it is an important amino acid containing a free thiol moiety. Detection of homocysteine from other amino acids is therefore important. A selective phosphorescence chemosensor for this aim was developed based on the reaction shown in Fig. 31 [127]. Upon addition of homocysteine to a semiaqueous solution of 61, a color change from orange to yellow and a luminescent variation from deep red to green were evident to the naked eye. This can be attributed to formation of a thiazinane group by selective reaction of the aldehyde group of 61 with homocysteine.

Fig. 31

Luminescent iridium complex-based sensor for homocysteine

4.4.3 Metal Cation Sensor

Binding an appropriate metal ion to the ligand may lead to sufficient changes in luminescence because the photophysical properties of the Ir complex are influenced by the structure or surrounding environment of the organic ligands. In 2006, Ho et al. [128] presented a novel system in which an azacrown receptor was attached to the pyridyl pyrazolate chelate of a heteroleptic Ir complex. Photophysical study showed that phosphorescence was gradually blue-shifted from 560 to 520 nm, and was accompanied by an increase of emission intensity upon addition of Ca²⁺, which made the complex a highly sensitive phosphorescence probe. In 2007, Schmittel et al. [129] synthesized 62 by introducing a crown group. The complex exhibited selective binding properties toward Ag⁺ and Hg²⁺ in aqueous media, accompanied by characteristic luminescence responses. As shown in Fig. 32, luminescence was enhanced by >10 times in the presence of excess Ag⁺; this is because the diimine ligand becomes an increasingly better acceptor by binding Ag⁺. The emission quenched about 80% upon addition of Hg²⁺, which may be explained by competition between emission-enhancing effects (decrease of the electron-donating ability of nitrogen upon Hg²⁺ binding) and the electron- or energy transfer-quenching effects of unbound Hg²⁺ ions. Ho et al. [130] demonstrated the concept of Pb²⁺ cation-sensing using the emissive Ir complex, which is based on the associated reduction of phosphorescence at room temperature upon chelate interaction between the Ir complex and metal analyte.

Fig. 32

Iridium-based chemosensors for Ag⁺ and Hg²⁺ in aqueous media

4.4.4 Acetonitrile or Propiononitrile Vapor Sensor

Sensor applications involving significant color and/or luminescence efficiency tuning mentioned above are usually based on structural or coordinated environment changes in the cyclometalated or ancillary ligand. We recently found the first example of an Ir complex PIrqnx (63) having a unique, fast vapochromic and vapoluminescent behavior towards acetonitrile or propiononitrile vapor based on molecular packing transformation (Fig. 33) [131]. Complex 63 exists as black and red forms (visible to the naked eye) in the solid state. The black form can be transformed to the red form upon exposure to acetonitrile or propiononitrile vapor, but no response was observed when it was exposed to other volatile organic compounds. Crystallographic and DFT studies indicated that the black form adsorbed acetonitrile vapor first because of a porous packing structure, then the red form was formed with different color and luminescence properties induced by weak intermolecular interactions (e.g., hydrogen bonding, π–π interactions).

Fig. 33

An iridium complex that can detect acetonitrile or propiononitrile vapor

5 Summaries and Outlook

As can be seen from the evolution of Ir complexes discussed in this chapter, chemical syntheses are carried out to coordinate different ligand structures to the metal center to control the excited state. Ir complexes are therefore tailored to express specific luminescent properties, and are being explored for many applications (e.g., OLEDs, biological labeling reagents, chemosensors). And with the efforts of research groups investigating luminescent Ir complexes, we believe the future is very bright indeed.