Effect of β-ketoiminato ancillary ligand modification on emissive properties of new iridium complexes

Article

pubs.acs.org/IC

Cite This: Inorg. Chem. XXXX, XXX, XXX −XXX

Eﬀect of β‑Ketoiminato Ancillary Ligand Modiﬁcation on Emissive

Properties of New Iridium Complexes

Ewelina Witkowska,^†Bartosz Orwat,^‡,§Myong Joon Oh,^‡,§Gabriela Wiosna-Salyga,*^,†

Ireneusz Glowacki,^†Ireneusz Kownacki,*^,‡,§Kamila Jankowska,^‡,§Maciej Kubicki,^‡Blazej Gierczyk,^‡

Michal Dutkiewicz,^§,∥Izabela Grzelak,^‡Marcin Hoﬀmann,^‡Julita Nawrocik,^‡,§Grzegorz Krajewski,^‡,§

Jacek Ulanski,^†Przemyslaw Ledwon,^⊥and Mieczyslaw Lapkowski^⊥,#

^†Department of Molecular Physics, Lodz University of Technology, Zeromskiego 116, 90-924 Lodz, Poland

^‡Faculty of Chemistry, Adam Mickiewicz University in Poznan, St. Uniwersytetu Poznanskiego 8, 61-614 Poznan, Poland

^§Center for Advanced Technology, Adam Mickiewicz University in Poznan, St. Uniwersytetu Poznanskiego 10, 61-614 Poznan,

Poland

^∥Poznan Science and Technology Park, Adam Mickiewicz University Foundation, ul. Rubiez 46, 61-612 Poznan, Poland

^⊥Faculty of Chemistry, Silesian University of Technology, St. Marcina Strzody 9, 44-100 Gliwice, Poland

^#Center of Polymer and Carbon Materials, Polish Academy of Sciences, Curie−Sklodowskiej 34, 41-819 Zabrze, Poland

S

* Supporting Information

ABSTRACT: A series of new bis(benzo[h]quinolinato) Ir(III)

complexes with modiﬁed β-ketoiminato ancillary ligands were

synthesized, and their electrochemical, photophysical proper-

ties were determined with the support of theoretical

calculations. Moreover, all the synthesized heteroleptic Ir(III)

complexes were examined as dopants of the host−guest type

emissive layers in solution-processed phosphorescent organic

light emitting diodes (PhOLEDs) of a simple structure. As

expected on the basis of voltammetry measurements as well as

DFT calculations, all the compounds appeared to be green

emitters. Their examination showed that alteration of β-

ketoiminato ligand structure causes frontier orbitals’ energy

levels to be slightly changed, while signiﬁcantly aﬀecting photoluminescence and electroluminescence eﬃciencies of iridium

phosphors containing these ligands. It was also found that modiﬁcation of ancillary ligands might enhance charge trapping on

the dopant, thus increasing its eﬃciency, especially in electroluminescence. From among the iridium complexes studied, the

compound bearing 1-naphthyl group bonded to the nitrogen atom of the ancillary ligand proved to be the most eﬃcient emitter.

The PhOLED fabricated on the basis of this dopant has reached a luminance level of 16000 cd/m², current eﬃciency close to

12 cd/A, and an external quantum eﬃciency around 3.2%.

Pd, or Os metals have been studied, but the most examined are

Ir(III) derivatives.⁴Popularity of the iridium compounds is

related to their intensive phosphorescence, even at room

temperature, that might be tuned within the whole visible

spectrum depending on the ligand structure modiﬁcation. It

should be emphasized that 2-phenylpyridine was one of the

ﬁrst ligands successfully used in this role and together with

benzo[h]quinoline gave rise to a group of iridium complexes

characterized by very intense green emission, for instance:

tris(2-phenylpyridinato)iridium(III) ([Ir(ppy)₃]) and bis(2-

phenylpyridinato-C²′,N) (acetylacetonato)iridium(III) ([Ir-

(ppy)₂(acac)]),⁵tris(benzo[h]quinolinato-C²,N′)iridium(III)

([Ir(bzq)₃]), and bis(2-benzo[h]quinolinato-C²,N′)(acetyl-

INTRODUCTION

■

Intensive research into new organic electroluminescent

materials has been ongoing since Tang and Van Slake

described the ﬁrst organic light emitting diodes (OLEDs),¹

and the researchers from the Cambridge University presented

the ﬁrst polymer light emitting diode (PLED).²In the ﬁrst

generation of electroluminescent devices, emission occurs

mainly from singlet states and is related to the phenomenon of

ﬂuorescence. Therefore, according to statistical rules, theoreti-

cal internal eﬃciency of such devices can reach maximally 25%.

This limit may be overcome by phosphorescent emitters, for

which quantum mechanics selection rules can be broken, and

theoretically their emission yield can increase up to 100%.³

Popular phosphorescent materials are heavy metal complexes

bearing organic ligands whose properties result from the

presence of strong spin−orbit coupling. The complexes of Pt,

Received: September 18, 2019

A

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

acetonato)iridium(III) ([Ir(bzq)₂(acac)]).⁵Although the ﬁrst

iridium(III) complexes for OLEDs were developed over 20

years ago, the synthesis and study of new eﬃcient iridium-

based emitters is still considered an attractive topic.^4b,6

In general, emissive materials show higher photolumines-

cence yield in solution than in solid state (i.e., thin layer)

which is related to the self−quenching of excited states. In

order to suppress this eﬀect, a host−guest system can be

implemented. In such a system, the emitter molecules are

dispersed in an active host, e.g. polymeric matrix, that increases

the mean distance between two emitter molecules.⁷On the

other hand, this approach emphasizes the importance of the

intermolecular energy transfer. The yield of energy transfer

nitrogen atom, were successfully synthesized. Some of the

desired derivatives (3a−f, h, i) were prepared by the classical

manner, i.e. the reﬂuxing of appropriate amines (2a−f, h, i)

with 2,4-pentanedione (acacH) in the presence of p-

toluenesulfonic acid in a benzene environment, using a

Dean−Stark apparatus. However, the conversion of more

hindered amines (2g, j−l) required an acceleration with

microwave irradiation as a non-classical energy source

(Scheme 1). Anyhow, these methods allowed us to obtain β-

Scheme 1. Synthesis of Ancillary Ligands

̈

from the excited states of the matrix to the dopant (by Forster

and/or Dexter mechanism) depends on the overlap of the

emission spectrum of the host and the absorption spectrum of

the guest.⁸Only when this condition is satisﬁed, a very eﬃcient

energy transfer from the host to the guest can take place.

Moreover, the emission originating from the guest molecules

can be also caused by the direct charge carrier trapping on the

dopant.⁹Thus, the host−guest systems can be successfully

applied in the form of an eﬃcient emission layer in

phosphorescent OLEDs (PhOLEDs).

Very recently, we have reported the studies concerning new

heteroleptic complexes of the general formula [Ir-

(bzq)₂(O∧N)], for which we discussed the impact of the

number and distribution of ﬂuorine atoms directly bonded to

the N-aryl moiety (NR¹) in N,O-donating β-ketoiminato

ligand (RC(NR¹)CHC(−O⁻)R), on the photophysical

and emissive properties.¹⁰We have found that the introduction

of ﬂuorine atoms does not signiﬁcantly modify the optical and

electrochemical properties of the iridium complexes, while the

electroluminescence performance of PhOLEDs based on them

was strongly dependent on the structure of the modiﬁed β-

ketoiminato ligand. The most promising results were obtained

for the Ir complex with 4-ﬂuorophenyl substituent in the

ancillary ligand structure. Therefore, as a continuation of our

research, we decided to explore para substitution eﬀect by

investigation of Ir(III) heteroleptic complexes with diﬀerent

substituents in the mentioned position of the phenyl moiety in

4-phenylimino-2-pentanonate ancillary ligand. As follows from

the available reports on this subject, in particular those by

Teets’¹¹as well as our recent ﬁndings,¹⁰so far the subjects of

studies have been limited to phenyl-based N-substituents (

NR¹). Unfortunately, there are no other reports concerning the

inﬂuence of the higher condensed polycyclic aromatic

hydrocarbon used as −R¹substituents in N,O-donating β-

ketoiminato ligand. Therefore, we focused also on the

synthesis of 4-arylimino-2-pentanones (MeC(NR¹)CH₂−

C(O)Me) equipped with various polycyclic aromatic

moieties −R¹, which were further employed in the preparation

of new [Ir(bzq)₂(MeC(NR¹)CHC(−O)Me)] type com-

plexes. The modiﬁcation of β-ketoiminato ancillary ligand

allowed determination of the NR¹substitution eﬀect on the

emitter photochemical properties. The obtained iridium

complexes were applied as dopants in PhOLEDs, whose

work parameter analysis allowed identiﬁcation of the most

eﬃcient and promising iridium emitters.

ketoimines equipped with phenyl-based substituents bearing in

their structure groups characterized by strong electron

withdrawing properties (3c−f), including various regioisomers

of nitro−substituted derivatives (3d−f). Thus, in order to

explain the inﬂuence of the presence of a strong electron

withdrawing group and its position on photophysical proper-

ties of ﬁnal heteroleptic Ir(III) complexes, not only para-

substituted phenyl β-ketoimines, but also the compounds

bearing −NO₂group in meta and ortho positions were

synthesized. We assumed that diﬀerent position of the highly

electron withdrawing substituent in the phenyl group could

signiﬁcantly aﬀect the emissive parameters of the target iridium

complex, as we have previously described for the phenyl group

having one ﬂuorine substituent.¹⁰Additionally, on the basis of

the protocol involving the use of microwaves,¹⁰4-arylimino-2-

pentanones with selected bulky polycyclic aromatic systems

were synthesized (3g−l). The purity of all organic materials

1

was conﬁrmed by spectroscopic methods, such H, ¹³C NMR,

HRMS (high resolution mass spectrometry) as well as the

structures of three of them (3d, j, l) were solved using X-ray

methods.

In the next stage of our synthetic work, previously prepared

4-arylimino-2-pentanones were employed in the preparation of

new iridium(III) complexes, according to the earlier described

method.¹⁰However, this time instead of microwaves, a

classical source of heat was used. The protocol applied

consisted of two successive steps, namely, generation of 4-

arylimino-2-pentanonalate salt in the reaction of β-ketoimi-

RESULTS AND DISCUSSION

■

Synthesis. According to recently published protocols,¹⁰in

the initial step a series of 4-arylimino-2-pentanones, bearing

aryl substituents of variable stereoelectronic character 3a−l at

B

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

nates 3 with NaH or KH in THF solution and employment of

the former for cleavage of binuclear precursor 4 (Scheme 2).

Scheme 2. Synthesis of Heteroleptic β-Ketoiminato

Iridium(III) Complexes

However, due to encountered vulnerability of the initial

ketoimine to degradation during the deprotonation process,

the use of NaO^tBu as base in a one-pot protocol was required

when using 5b and 5f. The applied methodologies made it

possible to obtain a series of Ir(III)-based phosphors bearing in

their structures various types of ancillary ligands (5a−l).

Compounds 5a−l were obtained with moderate to good

1

yields and characterized by spectroscopic methods. In the H

NMR spectra, sets of resonance lines characteristic of protons

coming from speciﬁc parts of ligands were clearly visible,

namely, in the region typical of cyclometalated bzq ligand

(9.70−9.00 ppm), methine part of N,O-donating ancillary

ligand (c.a. 5.00 ppm), as well as magnetically and chemically

non-equivalent methyl groups of ketoiminato ligand (2.00−

1

1.60 ppm). However, the H NMR spectra of some of the

isolated materials, namely 5c, 5f, 5i, 5k, and 5l revealed the

presence of two isomeric forms. We suppose that the reaction

selectivity toward the formation of one isomer depends on

properties of aryl substituents bonded to imine nitrogen atom,

such as steric hindrance or its orientation relative to the C,N-

cyclometalated ligands. In our opinion, the presence of N-aryl

substituents in close proximity of the cyclometalated ligands

may induce their intramolecular π−π interaction (π-stacking),

which might be the driving force for the formation of one of

the two possible isomers, as it takes place for complexes 5b, 5d,

5g, 5h, and 5j. Heteroleptic octahedral iridium(III) complexes

with bidentate ligands may exist in the form of various

geometrical isomers, such as N,N-trans−mer, cis−fac, cis−mer

as well as C,C-trans−mer and their corresponding enantiomeric

forms, as reported for the FIrpic complex.¹²The authors of the

cited publication showed that the geometrical isomers do not

diﬀer in photophysical properties, thus we decided not to

isolate pure isomers but examine their mixtures. Nevertheless,

we wanted to determine exactly which of the possible isomers

is formed in predominance. Therefore, the structures of ﬁve

complexes (5f, g, i, j, l) were determined by X-ray analysis and

are discussed in the next paragraph.

X-Ray Analysis. X-ray diﬀraction structural analysis

conﬁrmed the obtainment of target molecules (Figures 1 and

2 and Figure 23S in the SI). Although the crystal structure of

3d has been published earlier,¹³for the sake of completeness,

we decided to add this structure determination as well. The

relevant geometrical parameters are listed in Table 2S (see the

SI). In all complexes the Ir atom is six-coordinated in quite a

regular octahedral fashion, surrounded by N and O atoms from

ketoiminato fragment, and N and C atoms from two

benzo[h]quinolinato ligands. Table 2S (see the SI) shows

that, upon complexation, the C4−N5−C6 angle becomes

signiﬁcantly smaller, while the conformation of the 4-imino-2-

Figure 1. Perspective views of ligands: (a) 3d, (b) 3h, (c) 3j, (d) 3l

(molecule A), together with the labeling schemes. Ellipsoids are

drawn at the 50% probability level, hydrogen atoms are shown as

spheres of arbitrary radii, and intramolecular hydrogen bonds are

drawn as dashed blue lines.

pentanone fragment remains stable, although less planar. This

observation can be connected with the presence of intra-

molecular N−H···O hydrogen bonds in free ligand molecules,

responsible for planarization (cf. Table 3S, see the SI). What is

more, the plane of the aromatic substituent at N5 (Table 2S,

the SI) is almost perpendicular to the OC−CC−N plane

(Table 2S, see the SI) in the coordinated ligands. The

conformations of all complexes, whose structural data are

collected, are quite similar. This can be shown by a comparison

of the dihedral angles between planar fragments: Ir−

ketoiminato cycle, two bzq planes, and substituent at N5.

Table 2S (see the SI) lists the appropriate values, and the data

can be summarized as follows: (i) the Ir−ketoiminato ring

plane is almost perpendicular to all of the other planes, (ii)

C

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

C,N-cyclometalated bzq ligand might induce the selective

formation of only one isomer as it takes place for compounds

5b−d and 5g−j (see ¹H NMR spectra in the SI). On the basis

of XRD analysis of single crystals obtained from isomerically

pure samples 5g and 5j, we assume that for the above-

mentioned complexes, N,N-trans−mer is the most preferred

geometric isomer, which is conﬁrmed by spectroscopic studies.

This might be supported by the fact that such a conﬁguration

of C,N-donating ligands around the metal atom is typical of

such a binuclear substrate.¹⁵However, in the case of any

disturbances of this type of interactions related to the mutual

arrangement of the above-mentioned elements of the complex

molecule, e.g. by introduction of a substituent at position 2

relative to the N−C(Aryl) bond (5f) or changing the

substitution site of the polyaromatic system (5i, 5k, 5l) (see

¹H NMR spectra in the SI), the formation of another isomer

was observed.

Thermal Analysis. Thermal decomposition process of all

discussed Ir complexes was essentially multimodal with two or

three well-pronounced steps observed. The ﬁrst decomposition

step occurred in the range of 250−450 °C, followed by the

second one, less pronounced, in the range of 450−650 °C and

the third one which could be observed between 650 and 950

°C. Comparing the results of TG analysis presented in Table 1

Table 1. Results of TG and DTG Analysis

mass loss temperature [°C]

a

sample

T_5%

T_10%

residue [%]

5a

5b

5c

5d

5e

5f

5g

5h

5i

272

300

238

302

278

217

349

233

339

311

313

238

333

340

313

343

338

313

376

318

367

358

351

329

23.9

24.3

26.2

24.3

24.4

23.5

25.6

21.7

23.2

22.6

21.1

5j

5k

5l

Figure 2. Perspective views of iridium complexes: (a) 5g, (b) 5j, (c)

5l. Only Ir and coordinated atoms are labeled, for clarity. Ellipsoids

are drawn at the 50% (a and c) and 33% (b) probability level, and

hydrogen atoms are shown as spheres of arbitrary radii.

a

Measured at 990 °C.

as well as TG curves presented in Figures 24S −33S, it can be

concluded that the most thermally stable are complexes 5g and

5b with the temperatures of 5% mass loss exceeding 230 °C.

Careful analysis of the obtained data leads to the observation

that not only the chemical nature of the functional group

present in the phenyl ring (see Figure 24S) inﬂuence the

thermal stability of examined compounds but also its

regiosubstitution. This phenomenon can be easily observed

in the example of TG curves of 4-, 3-, and 2-substituted

nitrophenyl 5d, 5e, and 5f derivatives presented in Figure 25S.

While only slight diﬀerences are observed between the

thermograms of 5d and 5g isomers, the thermal stability of

5f is signiﬁcantly lower, which was revealed by a decrease in

the temperatures T_5%and T_10%(see Table 1). The inﬂuence of

regioisomerism on thermal stability of the studied compounds

was also conﬁrmed by the results obtained for 1- and 2-

naphthyl (5h, 5i), as well as for 1-, 2-, and 9-anthracenyl (5j,

5k, and 5l) derivatives TG analysis, presented in Figure 26S

and 27S, respectively. Comparing the TG curves of the above-

mentioned naphthyl and anthracenyl derivatives with phenyl-

benzo[h]quinolinato planes are almost perpendicular to each

other, and (iii) the aromatic substituents at N5 assume

positions close to gauche with respect to one benzo[h]-

quinolinato ligand and almost parallel to the other. In all

complexes, the tendency toward such a parallel disposition is

accompanied by a relatively close interplanar separation (in all

cases ca. 3.6−3.7 Å; centroid−to−centroid distances are

around 3.8−3.9 Å) which can be connected with the weak

intramolecular π···π interactions.

The crystal structures are deﬁned mainly by weak dispersive

forces. Calculations of intermolecular (molecule-to-molecule)

potential using UNI force ﬁeld¹⁴indicate that the stabilizing

energy is the largest for pairs for which π···π (−57.5 kJ/mol for

5l, −67.0 kJ/mol for 5g), C−H···π (−65.8 kJ/mol for 5j), C−

H···S (−67.0 kJ/mol in 5g), or C−H···O (−48.1 kJ/mol in 5f)

intermolecular interactions can be found.

Considering the above, it seems that even weak intra-

molecular π···π interactions between the N-aryl group and the

D

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

derived complex (5a), it is easy to notice that also the number

of condensed aromatic rings present in the structure of R

substituent unambiguously inﬂuences the complex stability

(see Figures 30S −33S). Despite the above-mentioned diﬀer-

ences in the thermal stability of the investigated samples, the

relative residue at 990 °C was similar. The latter was expected

as the iridium should be the main component of the pyrolysis

residue, and its content in the examined compounds was

comparable.

Table 2. Electrochemical Properties of Studied Compounds

compound E_{red onset}[V] E_{ox onset}[V] E_g[eV] EA [eV] IP [eV]

5a

5b

5c

5d

5e

5f

5g

5h

5i

−2.32

−2.20

−2.15

−1.73

−1.70

−1.72

−2.35

−2.28

−2.31

−2.29

−2.26

−2.31

0.17

0.12

0.21

0.26

0.17

0.29

0.23

0.16

0.20

0.18

0.15

2.49

2.32

2.36

1.99

1.96

1.89

2.64

2.51

2.47

2.49

2.44

2.46

2.8

2.9

3.0

3.4

2.8

5.3

5.2

5.3

5.4

5.3

5.4

5.3

As a conclusion of this paragraph, it might be said that all the

complexes exhibited suﬃcient thermal stability to be applied as

potential phosphorescent emitters.¹⁶

5j

5k

5l

Electrochemical Properties. In order to determine

electrochemical properties of the studied iridium(III) com-

plexes, in particular their correlation with the structure of β-

ketoiminato ligand, cyclic voltammetry measurements were

performed. On the basis of the onsets of oxidation and

reduction potentials, the electron aﬃnity (EA) and ionization

potential (IP) were estimated. The measured curves are shown

in Figure 34S (SI). The oxidation onset potentials (E_{ox onset}) of

all studied complexes are in the range from 0.12 up to 0.29 V.

The oxidation is quasi-reversible and related to the oxidation of

Ir(III) to Ir(IV). These results are in good agreement with

those reported for the corresponding iridium complexes with

β-ketoiminato ligands.^10,11The incorporation of a methoxy

group to the phenyl moiety leads to a slight decrease in E_{ox onset}

with respect to that of 5a, while incorporation of nitrile or nitro

groups leads to E_{ox onset}increase (5d, 5e) or has no eﬀect (5f).

This is consistent with the electronic eﬀects of these groups,

since the methoxy group is electron donating, while nitrile and

nitro groups are electron withdrawing. For polycyclic aromatic

substituted complexes (5h−l), the most intensive inﬂuence on

E_{ox onset}was observed for α-naphthyl derived complex. As one

modiﬁcations of the studied complexes and, hence, mod-

iﬁcation of their HOMO and LUMO energy levels. Our

attention was drawn by the most outstanding EA change

observed for nitro-derived complexes, that must have come

from the presence of −NO₂groups. Thus, we decided to

implement computational chemistry methods in order to

explore the electronic structure of these three compounds

along with the other complexes, intending to explain the origin

of the diﬀerence.

Theoretical Considerations. As the basis for further

theoretical considerations, at ﬁrst the complexes’ geometries

were optimized at the B3LYP level without any symmetry

constrains. Cartesian coordinates of the ground states (S₀) of

the complexes are presented in Tables 10S −21S, while the

exemplary perspective view of 5f structure is shown in Figure

3. Table 3 illustrates the parameters of Ir−ligand bond lengths

can see, the measured reduction onset potentials (E_{red onset}

)

were signiﬁcantly more varied than E_{ox onset}. In general, most of

the complexes were characterized with E_{red onset}near −2.3 V,

while the values reported for methoxy- and nitrile-derived

complexes were slightly more positive. However, the most

outstanding values were measured for nitro-derived complexes

(5d−f). The three complexes showed quasi reversible

reduction in contrast to the other studied compounds whose

reduction was irreversible. These results are similar to those in

the other reports concerning iridium complexes equipped with

nitro group, in which a drastic change in the reduction

potential was also observed.¹⁷

Electrochemical EA and IP were estimated from E_{red onset}and

_{ox onset}, respectively. EA, which corresponds to the energy level

E

of LUMO (the lowest occupied molecular orbital), was found

to be more sensitive toward ancillary ligand chemical

modiﬁcation than IP, corresponding to the HOMO (the

highest occupied molecular orbital) energy level. According to

Table 2, EA values were within the 2.8−3.4 eV range, while the

IP spread was only 0.2 eV. Therefore, the chemical

modiﬁcation had noticeably greater impact on the LUMO

energy level. It is worth emphasizing that the complexes

equipped with naphthyl or anthracenyl substituents (5h−l)

were characterized by identical EA and IP values as the

reference complex 5a.

Results of electrochemical measurement show that the

incorporation of diﬀerent electron withdrawing or electron

donating groups to β-ketoiminato ligand might lead to a

change in their electrochemical properties. Obviously, it is a

consequence of electron density shift caused by chemical

Figure 3. Optimized structure of 5f in the ground state.

and bond angles in vacuo and in the C₆H₅Cl media, together

with the X-ray crystal structure data of 5f. Due to the d⁶

conﬁguration of the Ir(III) ion, all complexes have the

expected pseudo-octahedral coordination geometry around

the iridium center. As presented in Figure 3 and Table 3, two

coordinating nitrogen atoms in the bzq ligands (N(1) and

N(2)) are in trans positions and the valence angles N(1)−Ir−

N(2) are nearly 180°, while the coordinating carbon atoms

E

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

In general, the DFT (density functional theory) calculation

results are in reasonable agreement with the X-ray crystal

structure data. The deviations measured by the mean unsigned

error are 0.08 Å and 2° for bond lengths and valence angles,

respectively. The slight discrepancy between the calculated and

the measured values is reasonable, because the former results

were obtained adopting the complex molecules in vacuo,

whereas the latter ones were examined in the crystalline state.

The above diﬀerences of geometrical parameters were indeed

expected as B3LYP method overestimates the bond lengths in

transition metal complexes.¹⁸

Table 3. Comparison of Selected Bond Lengths and Valence

Angles from the Optimized Geometries with the

Experimental Values for 5f

gas

C₆H₅Cl

X-ray

Bond Lengths (Å)

Ir−N1

Ir−N2

Ir−N3

Ir−C13

Ir−C25

Ir−O1

2.102

2.076

2.265

2.034

2.028

2.180

2.105

2.080

2.271

2.036

2.029

2.191

1.946

1.899

2.195

2.034

2.009

2.125

Spectral properties are strongly dependent on the frontier

molecular orbitals’ (FMOs) properties. Therefore, DFT

calculations were employed to investigate the HOMO and

LUMO of the complexes. The molecular orbitals were

calculated assuming the optimized geometry of the ground

state. For all complexes, the energy gaps between the HOMO

and LUMO were correlated with their experimental values

obtained by cyclic voltammetry measurements. In this study,

apart from B3LYP, other commonly used functionals, namely

M06 and WB97XD with basis set composed of 6-311++G(d,p)

for H, C, N, and O and SDD for Ir atom, were tested to obtain

the best possible correlation with experimental values. Because

the geometries optimized with B3LYP were comparable to

those obtained from the experimental XRD results, we initially

adopted this approach in our studies. The linear regression

coeﬃcient determined, characterizing the strength of the

correlation between the theoretical energies and electro-

chemical data, was 0.86 (Figure 35S SI). From among the

other tested functionals, B3LYP was characterized by the best

correlation between experimental results and allowed us to

explain the unusually low bandgap for 5d−f; thus, it was

chosen for further considerations. This choice was also

supported by the fact that B3LYP functional is applicable for

prediction with good accuracy of the trend of structural

changes eﬀect on the electronic structure of the modiﬁed

compound with respect to that of the unmodiﬁed one.¹⁹The

exemplary HOMO and LUMO contours and HOMO−LUMO

energy gaps of the studied complexes are presented in Figure 4,

while the full set of the numerical values of orbital distributions

are compiled in Table 9S (see the SI).

Valence Angles (deg)

C13−Ir−C25

C13−Ir−N2

C25−Ir−N2

C13−Ir−N1

C25−Ir−N1

N2−Ir−N1

C13−Ir−N3

C25−Ir−N3

N2−Ir−N3

N1−Ir−N3

C13−Ir−O1

C25−Ir−O1

N2−Ir−O1

N1−Ir−O1

N3−Ir−O1

89.2

93.4

80.9

80.5

98.1

173.8

174.4

95.8

89.9

96.2

89.0

174.8

94.3

86.5

86.2

88.4

93.6

92.5

91.2

81.6

82.7

98.3

173.9

173.8

93.4

91.6

94.5

85.9

177.0

96.0

84.0

88.4

80.8

80.4

98.0

173.9

174.7

96.2

89.8

96.2

89.5

175.1

94.9

86.0

86.1

(C(13) and C(25)) are in cis positions and the valence angle

C(13)−Ir−C(25) is close to 90°. Furthermore, two valence

angles between the coordinating atoms from the bzq ligand

and the central iridium atom, namely C(13)−Ir−N(1) and

C(25)−Ir−N(2) (Table 3 and Figure 3) are nearly identical,

ca. 80°. The comparison of the structural parameters calculated

in the chlorobenzene environment and in vacuo showed that

the solvent eﬀects have minor inﬂuence on the optimized

geometries of these complexes. The calculated Ir−N, Ir−C,

and Ir−O bond lengths are slightly longer (up to 0.01 Å) in

C₆H₅Cl media, while changes in the bond angles are less than

1.0°.

Figure 4. Frontier molecular orbital diagrams of complexes 5a and 5f computed at the B3LYP/SDD/6-311++G(d,p) level of theory.

F

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

As usually observed for cyclometalated cationic iridium(III)

complexes,^18,19the iridium atom brings a signiﬁcant con-

tribution to HOMO, while being much less involved in

LUMO. This fact indicates that HOMO−LUMO transition

should be characterized by strong MLCT (metal to ligand

charge transfer) character. Therefore, the HOMOs are

localized on both types of ligands, but their distribution is

limited to the N,O-donating atoms of the β-ketoiminato ligand

and is also spread over the whole bzq ligands, with slight shift

of electron density toward the C-donating part of bzq. On the

contrary, the LUMOs are almost totally present on cyclo-

metalating ligands (except 5d−f) with less than 5%

contribution of the Ir atom. It is worth noting that due to

the lack of symmetry constraint in the calculated structures, the

two bzq ligands are inequivalent, so their LUMOs are split into

two subsequent levels (e.g., LUMO and LUMO + 1) that are

very close in energy. However, for complexes 5d−f equipped

with a nitro group, the LUMOs are mainly located on the

ancillary ligand, in particular its nitrophenyl moiety, while the

successive unoccupied molecular orbitals LUMO + 1 and

LUMO + 2 primarily originate from the cyclometalating

ligands (see Table 9S in the SI). According to Table 6S, the

bandgap values calculated for 5a−c and 5g−l are very close,

the spread is 0.21 eV, while the respective electrochemical

bandgap spread is 0.32 eV. The calculated bandgap spread for

5d−f is 0.11 eV, while electrochemically determined bandgaps

are within 0.1 eV range. This indicates that the electrochemical

data and theoretical predictions are well correlated and show

that the mentioned complexes should exhibit similar properties

related to the bandgap values within the above−speciﬁed

groups, despite the fact that their LUMO levels are

overestimated. Additionally, the calculated 5d−f bandgaps

are lower than the value obtained for the reference 5a by about

0.8 eV for B3LYP functional (Table 6S). The respective

electrochemically determined diﬀerence is in the 0.5−0.6 eV

range, which indicates good reproducibility of the experimental

data by theory. Interestingly, if one calculates HOMO and

LUMO + 1 energy level diﬀerences for 5d−f, they will obtain a

value close to the calculated HOMO−LUMO bandgap of 5a.

Considering the above, the DFT calculations reproduced the

trends observed in electrochemical measurements, conﬁrming

that incorporation of −NO₂groups into the ancillary ligand

structure is associated with the insertion of its own lowest

unoccupied level between the original HOMO and LUMO

levels of the unmodiﬁed reference complex 5a (see Figure 4).

Therefore, it supports our supposition that E_{red onset}decreases

for 5d−f complexes relative to the other complexes which

originates from the reduction of the nitro group itself. This

conclusion is very important since it indicates that the

electrochemically determined bandgaps for these three

complexes are ﬂawed. The correlation of electronic structure

with photophysical properties of the complexes is further

discussed in the following sections.

Figure 5. Normalized absorption spectra of the investigated

complexes in chlorobenzene: 5a−5g (a) and 5h−5l compared with

5a (b).

centered (¹LC) transitions. However, the exact assignment of

the lowest energy states may be inaccurate, because it is

3

generally known that they can be a mixture of LC, MLCT,

4b

1

and MLCT. The fact that for tris(benzo[h]quinolinato)-

iridium(III) complex ([Ir(bzq)₃]) the absorption bands in the

1

370 nm region were assigned to MLCT states might support

our interpretation.²⁰

As one can see, the performed modiﬁcations of β-

ketoiminato ancillary ligand structure did not aﬀect signiﬁ-

cantly the absorption spectra of the studied complexes.

Nonetheless, the absorption spectra of the complexes equipped

with polycyclic aromatic substituents, especially anthracenyl,

diﬀer mostly in the range of ¹MLCT/LC transitions.

Interestingly, the change in anthracenyl regio-substitution

from position 2 (5k) to 1 or 9 (5j, 5l respectively) resulted in

alteration in the probability of these transitions. The bands

located around 400 nm are more pronounced for 5j and 5l

complexes, and the most intense bands (350−380 nm) are

slightly red-shifted in comparison to the corresponding signals

in the spectra of 5a, 5h, 5i, and 5k. Interestingly, the

absorption onset for 5d−5e is similar to those of the other

complexes and no additional low−energy absorption bands

were observed for them. This fact is in contrast to the

electrochemically determined energy levels, which suggests a

signiﬁcant decrease in the bandgaps for the nitro-substituted

complexes. This indicates that the excitation from HOMO to

LUMO levels are not preferable for them. Therefore, it can be

supposed that the observed lowest in energy transitions involve

the central atom and bzq ligands (corresponding to HOMO →

LUMO+1 transition). The reason for that might be the

Photophysical Properties. Absorption spectra of the

studied complexes 5a−5l in chlorobenzene under ambient

conditions are presented in Figure 5. The strong spin−orbit

coupling of iridium atom makes the formally forbidden triplet

metal−ligand charge transfer transitions (³MLCT) possible,

and for the examined complexes, they are observed as long-

wavelength absorption bands in the range of 440−550 nm.

Generally, the bands located around 400 nm can be assigned to

1

the higher extinction singlet MLCT, and the bands falling in

the short-wavelength range (below 400 nm), to π−π* ligand

G

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

location of LUMO on the nitrophenyl moiety of ancillary

ligand, which hampers the electron transition from HOMO

mostly located on bzq ligands, due to very poor overlapping of

the orbitals.

Complexes from the ﬁrst group (5b−5g) exhibited the

structured emission which was a mirror image of the excitation

spectra (compare Figures 6a and 7), as relative intensities of

Normalized photoluminescence (PL) spectra of the studied

complexes in chlorobenzene are shown in Figure 6. Excitation

Figure 7. (a) Normalized photoluminescence excitation and

absorption spectra of 5e in chlorobenzene. (b) Normalized

photoluminescence spectra of 5e in chlorobenzene, measured at

diﬀerent excitation wavelengths.

Figure 6. Normalized photoluminescence spectra of the investigated

complexes in chlorobenzene (λ_ex= 485 nm): 5a−5g (a) and 5h−5l

compared with 5a (b).

the vibrational peaks in the emission correlated with the

excitation spectra (see Figure 37S in the SI for a full data set).

Considering the spectra recorded for the nitro-derived

complexes 5e and 5f, the emission structure is the most

visible and the probability of higher energy transition relative

to the other bands is higher. For many common ﬂuorophores,

the vibrational energy levels spacing is similar for the ground

and excited states, which results in the emission spectrum that

strongly resembles the mirror image of the absorption

spectrum. However, for our complexes, the vibrational mirror

image is only visible in the excitation spectra and no structure

is observed in the absorption spectra. It may suggest that these

two spectra have diﬀerent excited-state origins. It is supposed

that the ﬁrst absorption band reﬂects a combination of

diﬀerent transitions to MLCT states, whereas the excitation

spectrum comes from the speciﬁc state responsible for the

emission. In general, MLCT transitions are coupled to low

energy vibrations and are more inﬂuenced by the molecule’s

environment, thus the absorption bands should reveal more

inhomogeneous broadening with increasing degree of MLCT

character. As shown for 5e (Figure 7), when its molecule was

excited at the lowest energy absorption bands observed in the

excitation spectra, a structured emission spectrum in photo-

luminescence was recorded. Nevertheless, when higher energy

3

at 485 nm, corresponding to the MLCT transition of the

complexes, resulted in a broad emission band in the range of

500−750 nm with a maximum at ∼570 nm for the complexes

with phenyl, naphthyl, and anthracenyl substituents (5a and

5h−5l). All other complexes (5b−5g) showed a structured

emission with narrow bands located at ∼538, ∼577, and ∼628

nm. The most structured emission spectrum was observed for

complexes 5e and 5f, equipped with −NO₂groups. Nonethe-

less, the emission spectra of the complex with −NO₂groups in

the para position (5d) revealed its instability and thus, it was

not further examined (see Figure 36S in the SI). As one can

see, the most pronounced structure is observed for the

complexes bearing electron withdrawing substituents (e.g.,

−NO₂, −CN) and less distinct bands are noticed when the

electron donating groups (e.g., −OCH₃, −Ph) are present in

the ancillary ligand structure. The structured emission spectra

of transition metal complexes often indicate a large

contribution of LC in the emissive excited states.²¹Addition-

ally, for the complexes with electron withdrawing group, the

most intense is the 0−0 transition that changes a little bit the

color of the emitted light and suggests small geometry changes

during excitation and relaxation.

H

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

was supplied to the molecule, a broad and structureless

emission spectrum was found. Usually, the structureless

spectrum is assigned to the emission from MLCT states,

whereas the structured emission indicates that LC emissive

excited states play a crucial role in generation of visible light.

Therefore, the probable explanation of this behavior could be

that upon tuning the excitation energy, the emission from

diﬀerent close lying excited states is observed. However, in

order to examine this phenomenon, additional experiments are

needed. The strong excitation wavelength dependency of

emission have been reported by Hsu et al. for osmium(II) and

silver(I) complexes²²in which the phosphorescence/ﬂuores-

cence intensity ratio is enhanced when the excitation energy is

increased.

of poly(N-vinylcarbazole) (PVK) and 2-(4-tert-butylphenyl)-5-

(4-biphenylyl)-1,3,4-oxadiazole (PBD), providing hole and

electron transport, respectively. The PVK/PBD mixture was

selected as a host material due to its wide energy gap in

accordance with the general rule that the emitter energy gap

should be localized between HOMO and LUMO levels of the

matrix to ensure good energy transfer of the excitons from the

matrix to the emitter molecules.

The PL spectra of thin PVK/PBD ﬁlms doped with 1 wt %

of the studied emitters upon excitation with the wavelength

corresponding to the lowest absorption band of the PVK/PBD

matrix (λ_ex= 340 nm) are presented in Figure 8. All tested

Additionally, LC transitions are characterized with higher

energy than MLCT transitions.²³However, our results

obtained for the studied iridium complexes suggests almost

the same energy for both transitions and a strong mixing of LC

and MLCT. Since the structured emission was observed only

for the complexes bearing phenyl-based ketoimine ancillary

ligands, we expected that this phenomenon would be

connected with the presence of this moiety. Surprisingly,

with the higher energy applied for excitation of 5e and 5f, their

emission spectra were almost the same as those of the other

compounds. Once again, it proved that the electron transitions

involving nitrophenyl moiety are not preferable and rather Ir-

bzq are engaged, which correspond to HOMO−LUMO+1 and

HOMO−LUMO+2 transitions. This explanation was further

supported by the results of electroluminescence studies.

[Ir(bzq)₃] is classiﬁed as a complex whose emission

originates from the ³MLCT state.²⁰A comparison of its

emissive properties with the results reported here suggest that

β-ketoiminato ancillary ligand plays rather a marginal role in

the emission processes that originates mainly from the states in

which bzq and iridium are mostly involved. As a consequence,

the observed structured photoluminescence spectrum can

correspond to the vibronic−sublevel, especially for the electron

transition between the orbitals localized mostly on the ligand,

when the electronic−vibronic coupling occurs and/or for low

energy metal−ligand vibrations, when the metal participation

in the electronic state increases. Since electron transitions are

much faster than nuclear oscillations, the structured spectra

should be observed when a small molecular geometry change

takes place in the excited state. The latter was conﬁrmed

previously on the basis of the intensity of 0−0 transition.

Therefore, taking into account the above considerations, it can

be concluded that substitution of N-donating atom in the β-

ketoiminato with various aryl groups aﬀects only the LC/

MLCT excited states contribution to the total emission, while

all the studied [Ir(bzq)₂(N∧O)] complexes exhibit green

emission in the same range. Therefore, it should be expected

that the modiﬁcation of ancillary ligand should inﬂuence the

luminescence performance, since the emission from the

³MLCT state is expected to be more eﬃcient, but this will

not tune the emitted wavelength. Driven by this conclusion, we

wanted to examine if the structured emission would occur in

photo- and electroluminescence in the solid state.

Figure 8. Normalized photoluminescence spectra of thin layers of the

PVK/PBD blend doped with 1 wt % of emitter molecules: 5a−5g (a),

and 5h−5l compared with 5a (b). Excitation wavelength was 340 nm,

that corresponds to ﬁrst absorption band of PVK/PBD matrix.

layers showed the emission related to the phosphorescent

dopants as well as host component (λ_max∼ 430 nm),

originating from singlet exciplexes formed between carbazole

moieties and oxadiazole molecules. A contribution of the

matrix emission indicates incomplete energy transfer from the

matrix to the emitters. The best energy transfer was observed

for naphthyl−derived complex (5i), whereas the worst

eﬃciency of this process was noted for the complexes

characterized by structured emission spectra in solution

(primarily 5e, 5f). The main photoluminescence bands are

generally similar to those detected for the emitters in solution

upon high energy excitation. For the neat ﬁlms, the emission

distribution is broadened and less structured, thus indicating

that mostly MLCT energy states are responsible for the

emission in the examined host−guest systems. Eﬃciency of

such an energy transfer inﬂuences the photoluminescence

Photoluminescence of PhOLED Active Layers. In order

to avoid the concentration induced emission quenching, for

which Ir(III) complexes are known,²⁴the host−guest systems

are commonly used as an active layer, which means that the

emitter is homogeneously dispersed in a polymer matrix. In

this work, we used a well-known ambipolar matrix, composed

I

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Table 4. Parameters of PhOLEDs Based on PVK/PBD Layers Doped with 1 wt % of All Tested Ir Complexes

compound

λ_EL[nm]

559

L_max[cd/m²]

η_max[cd/A]

EQE [%]

λ_PLfilm[nm]

552

QY_film[%]

5a

5b

5c

5e

5f

5g

5h

5i

9 500

1400

600

180

580

4200

15700

5500

50

100

70

9.1

2.6

1.5

0.2

0.3

3.8

12.3

4.2

0.02

0.06

0.02

2.71

0.83

0.48

0.07

0.10

1.13

3.20

1.45

0.01

0.03

0.02

15

8

9

2

12

18

4

<1

565

559

567

563

559

548

545

534

540

548

556

569

549

563

5j

5k

5l

587, 715

625, 697, 765

575, 740

554, 703, 775

561, 685, 753

559, 723

quantum yield (QY), which was up to 18% for the investigated

layers (Table 4).

its emission originated from the matrix is visible in the ﬁlm PL

spectrum. However, the photoluminescence QY of this

compound in thin ﬁlm was very low (4%, Table 4). The

opposite relationship was observed for 5g, for which the energy

transfer was much less eﬀective, but the compound showed

higher QY. Furthermore, 5h was characterized with the best

quantum yield (18%), despite showing noticeable contribution

of matrix emission in the photoluminescence spectrum. These

observations indicate that aryl-substitution of β-ketoiminato

ancillary ligand can tune photoluminescence eﬃciency of the

complexes in a way not fully understood. Nevertheless, the

conﬁrmed spectral overlap of the complex absorption and the

matrix emission was a good premise to examine the eﬃciency

of energy transfer in electroluminescence studies.

Electroluminescent Properties. The electrolumines-

cence properties of the iridium complexes were tested in one

of the simplest PhOLED structure. The devices with a hole

injection layer (HIL) and an emissive layer were produced by a

spin-coating technique. Such a simple construction of the

device is preferable for manufacturing by economical wet

printing techniques. Additionally, this minimalistic approach is

an easy screening procedure for emitters’ electroluminescent

properties. The emissive layers were based on PVK/PBD

matrix systems as used in photoluminescence studies. It should

be noted that a PVK/PBD blend with a 7:3 weight ratio is

characterized by balanced charge transport properties, which is

a basic requirement for eﬃcient OLEDs.

Electroluminescence (EL) spectra recorded for the devices

based on the investigated emitters are presented in Figure 10.

The EL spectra obtained for PhOLEDs doped with emitters

5a−5i were very similar to the PL spectra in thin ﬁlms. The

maxima of emission were in the range of 555−570 nm, with

one exception, and were only slightly red-shifted in relation to

the PL spectra (compare Figures 8 and 10). It should be

emphasized that the matrix emission disappeared in the EL

spectra, even when it was predominant in PL of thin ﬁlms (for

complexes 5e and 5f). The lack of emission from the matrix in

the EL spectra suggests a strong contribution of the charge

trapping process in EL phenomenon, which can promote

formation of excitons on the dopant molecules.

The emission spectra of thin layers doped with anthracenyl

substituted complexes (5j−5l) showed additional low energy

bands (680−800 nm) that were not visible in the spectra

recorded in solution. The presence of these bands is most

likely related to the anthracenyl degradation products. These

complexes exhibited poor emissive properties (QY < 1%, Table

4), as well as low stability of emissive states that was also

observed in further electroluminescence studies. Additionally,

incorporation of 5j−5l complexes resulted in a hypsochromic

shift of the matrix emission as presented in Figure 8b. It can be

attributed to the disruption of exciplex formation between

carbazole moieties from PVK and oxadiazole molecules.

Therefore, 5j−5l complexes are not good candidates for

eﬃcient emitters in the PVK/PBD matrix, even though their

energy transfer was acceptable.

Dominant contribution to the emission of dopant

̈

introduced to the layers may originate from long-range Forster

and/or from short-range Dexter energy transfer. The Dexter

mechanism is less likely because of low emitter concentration.

However, the exciton diﬀusion can enhance both paths of

energy transfer. Both mechanisms can operate eﬃciently only

when the emission spectrum of the donor and the absorption

spectrum of the acceptor overlap considerably. All the studied

complexes are characterized by a good spectral overlap of their

MLCT absorption bands with the matrix emission spectrum, as

shown in Figure 9, for exemplary 5a and 5h−5i complexes. In

the whole group of tested compounds, emitter 5i stands out in

the energy transfer eﬃciency since only a small contribution of

Similarly to the PL spectra, the EL spectra of PhOLEDs

doped with emitters 5j−5l showed long-wavelength bands in

addition to the desired ³MLCT bands. For the PhOLED

created on the basis of 5k, this red emission even dominated.

During prolonged measurements of these OLED emissions, we

observed that the contribution of long-wavelength bands

changed irreversibly along with the device operation time (see

Figure 38S in the SI). Therefore, it can be concluded that the

Figure 9. Normalized absorption spectra of four selected compounds

(5a, 5g, 5h, 5i) in chlorobenzene solutions and normalized

photoluminescence spectra of neat PVK/PBD blend.

J

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

V. The obtained luminance values were ∼9500 and ∼16000

cd/m²at 16 V for 5a and 5h based PhOLEDs, respectively.

The turn-on voltage (the voltage value at which the luminance

reaches 1 cd/m²) was around 7 V. Although the 5h-based

PhOLED was characterized by a greater leakage of current;

nevertheless, the maximum luminance was higher when

compared to the system made with the use of 5a. In Figure

12, the current eﬃciencies are plotted versus luminance for the

Figure 12. Current eﬃciency−luminance dependencies of PhOLEDs

with emitting layer made of PVK/PBD doped with 1 wt % of 5a and

5h complexes.

tested devices. The highest value (over 12 cd/A) was obtained

for PhOLED with 5h emitter, derived with a 1-naphthyl

substituent. This result correlates well with the fact that the

highest photoluminescence QY was measured for thin ﬁlms

with this dopant (Table 4). Moreover, measured current

eﬃciency was relatively stable in the wide luminance range

(1500−4000 cd/m²).

Figure 10. Normalized electroluminescence spectra of PhOLEDs

based on PVK/PBD doped with 1 wt % of the investigated emitter

molecules: 5a−5g (a) and 5h−5l compared with 5a (b).

A comparison of the parameters determined for all tested

devices and their emissive layers is shown in Table 4. It should

be emphasized that the best devices exhibited good

parameters, considering that the diodes had such a simple

structure and were produced using solution methods. OLEDs

of similar structure, based on [Ir(bzq)₂(acac)], exhibited the

maximum current eﬃciency of 4.2 cd/A.²⁵Therefore, the

obtained current eﬃciencies of 9.5 and 12.0 cd/A for 5a and

5h can be considered to be quite high. Of course, further

optimization of the devices’ structure by incorporation of

additional layers should lead to PhOLEDs of better perform-

ance.²⁶

The devices that exhibited the best EQE were those with 5a,

b and 5g−i phosphorescent dopants (Table 4). As one can see

from a comparison of QY_filmand EQE, they correlate well for

all complexes with the exception of 5i. This deviation might be

related to a noticeable impact of the exciton energy transfer on

the device performance. Considering the above, it seems that

the best external eﬃciencies were obtained for the emitters

equipped with unmodiﬁed aromatic moieties (5a and 5g−i)

and electron-donating methoxy group (5b), all rich in

electrons. The very low performance of anthracenyl-derived

complexes (5j−k) might deny it at a ﬁrst glance, but their poor

stability in electroluminescence should be kept in mind. Taking

into account the above-mentioned details, it is clear that their

eﬃciency might be suppressed by their unexpected instability.

In this way, we proved that β-ketoiminato ligands aﬀect the

photophysical properties of the examined complexes, in

particular luminescence eﬃciency, although they do not tune

the maximum emission wavelength.

complexes with the anthracenyl substituent (5j−5l) are

unstable. This indicates that the degradation process of

anthracenyl groups may occur during the radiative excitation

and can be reinforced during operation of the device. Such a

scenario is possible because the degradation might be caused,

e.g., by trace amounts of oxygen molecules embedded in the

bulk of the emissive layer, despite their production in a

nitrogen atmosphere.

Current density−voltage (J−V) and luminance−voltage (L−

V) characteristics of the best performing PhOLEDs, based on

5a and 5h emitters, are shown in Figure 11. These plots

represent typical PhOLED characteristics in the range up to 16

Figure 11. Current density−luminance−voltage (J−L−V) character-

istics of PhOLEDs with emitting layer made of PVK/PBD doped with

1 wt % of emitter molecules: 5a (empty symbols) and 5h (full

symbols).

K

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

It is known that the performance of PhOLEDs strongly

depends on the concentration of emitter in the active layer.

After screening all the emitters as dopants for OLEDs at 1 wt

%, we decided to optimize the dopant concentration for the

complex with the best−performance, i.e. 5h. This emitter was

chosen inter alia because of its highest quantum photo-

luminescence eﬃciency (18%). Furthermore, the devices based

on it exhibited the highest luminance and current eﬃciency

from among the tested PhOLEDs. To determine the optimal

dopant concentration, a series of devices with diﬀerent emitter

content in the PVK/PBD matrix were prepared (from 0.5 to

7.0 wt %). The observed maximum luminance values (at 17 V)

were in the range from ∼4000 cd/m²(for 7 wt %) to ∼16000

cd/m²(for 1 wt %). The maximum value of external quantum

eﬃciency (EQE ∼ 3.2%) was obtained for the devices with 1

wt % emitter additive. A little bit lower value, but with slower

reduction of the eﬃciency with the increase in current density,

was observed for the diodes with 2 wt % emitter content

(Figure 13). Therefore, it can be assumed that the optimal

Figure 14. Normalized photoluminescence spectra of emissive layers

based on PVK/PBD doped with 5h in diﬀerent concentrations (from

0.5 to 7 wt %). The excitation wavelength, corresponding to ﬁrst

absorption band of PVK/PBD matrix, was 340 nm.

nm (0.5 wt %) to 556 nm (7 wt %) was observed, suggesting

stronger intermolecular interactions. This is a normal

consequence of increasing concentration since it should also

reduce the mean distance between emitter’s molecules in the

emissive layer. It is worth noting that this phenomenon is more

visible in photoluminescence and less in electroluminescence.

Nevertheless, it is undoubtedly clear that the lower dopant

concentration gave better results in OLEDs.

As mentioned above, the charge carrier trapping might play

an important role in electroluminescence, therefore this

phenomenon was further explored by spectrally resolved

thermoluminescence (SRTL) studies in the range of 20−300

K. In recent years, it has been proved that guest and matrix

molecules compete in charge carriers trapping in emissive

layers based on host−guest systems.^9b,27To maximize the

emitter eﬃciency, this balance should be shifted toward

eﬃcient formation of excitons on guest molecules and

quenching of host emission, even for the systems with minor

content of the emitter molecules in the matrix.

The SRTL results for the neat PVK/PBD and for the PVK/

PBD matrix doped with 1 wt % of 5h are shown in Figure 15.

A comparison of the TL maps showed that doping of 5h into

the PVK/PBD matrix results in a change in the emitted light

range. The spectrally resolved luminescence (SRL) recorded

close to the TL maximum (115 K) revealed that the emission

occurs in the range of 490−730 nm with one very broad band

with a maximum at 565 nm (see Figure 15c, red dashed line)

which does not change its position with temperature (see

Figure 15a). For the neat PVK/PBD matrix (see Figure 15c,

blue solid line), the emission occurred in the broader range of

390 to 750 nm. One can see two separated emission bands: a

dominant one with a maximum at 550 nm and the second one

with a maximum at 450 nm. Both emission bands from the

PVK/PBD blend are attributed to the exciplexes formed

between carbazolyl group and PBD molecule. The dominant

band originates from the triplet exciplexes, and the minor one

can be related to the singlet exciplexes and/or monomeric

triplet states of carbazolyl groups.²⁷However, the identiﬁcation

of the radiative recombination active center in the doped

matrix is problematic because the dominant emission band of

the matrix (550 nm) is close to the emission originating from

the iridium complex molecules (565 nm). Nevertheless,

comparing both spectra, one can see that the spectrum of

the matrix doped with 5h is more intensive and has no short-

Figure 13. External quantum eﬃciency−current density dependencies

of PhOLEDs with emitting layer made of PVK/PBD doped with

diﬀerent concentrations (from 0.5 to 7 wt %) of 5h emitter.

emitter content for such PhOLEDs ranges between 1 and 2 wt

%. Higher concentration may result in the emitter molecules

aggregation process resulting in quenching of emissive states

that is a common phenomenon observed in guest−host

systems. This might be the reason for the decrease in EQE

with increasing emitter concentration from 4 to 7 wt % in the

examined case.

In order to compare the energy transfer eﬃciency in electro-

and photoluminescence, the same amounts of 5h dopant were

used in both experiments. The obtained normalized PL spectra

are presented in Figure 14. As one can see, the contribution of

matrix emission decreases as the dopant concentration

increases. However, even for 7 wt % emitter additive, a small

emission from the matrix exciplexes was observed, so the

energy transfer was still incomplete. It strongly contrasts with

the electroluminescence spectra, in which almost complete

quenching of PVK/PBD emission was observed for the lowest

examined concentration (0.5 wt %) of emitter 5h (see Figure

39S in the SI). As mentioned above, these results suggest that

charge carrier trapping processes play an important role in the

electroluminescence phenomenon. EL spectra do not change

noticeably with the concentration increase and only a slight

shift of the band maximum was observed from 557 nm (0.5 wt

%) to 560 (7 wt %) nm. In the dopant concentration−

photoluminescence studies, a slightly bigger redshift from 547

L

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

are concerned (cf. Figure 10b). This indicates that in both

cases, after optical excitation at low temperatures (SRTL) and

during the excitation induced by the current ﬂow at room

temperature (EL), the generation mechanism of triplet exciton

for the emitter molecules added to PVK/PBD is similar.

CONCLUSIONS

■

In summary, we reported a synthetic protocol for the

obtainment of heteroleptic [Ir(bzq)₂(MeC(NR)CHC-

(−O)Me)] complexes series and studied their electrochemical,

photoluminescent, and electroluminescent properties in

reference to the structure of β-ketoiminato ligand. The

spectrum of investigated R substituents included phenyl with

methoxy, cyano, nitro, and 5-methyl-2-benzothiazyl groups in

para position, as well as all possible regioisomers of naphthyl

and anthracenyl moieties. In the case of nitro-substituted

complexes, also ortho and meta isomers were investigated in

order to explore in detail their outstanding electrochemical

properties. Namely, it was found that the presence of nitro

group drastically increases electron aﬃnity. With the aid of

theoretical calculations, we proved that this phenomenon was

related to the reduction of nitro group and should have rather

minor eﬀect on other properties related to the bandgap

between the orbitals involved in excitation and emission. Apart

from the three nitro-derived complexes, the variation of

ancillary ligand structure had only slight impact on the electron

aﬃnity and ionization potential.

Figure 15. SRTL spectra for PVK/PBD layer doped with 1 wt % of

emitter 5h and for neat PVK/PBD matrix layer: (a) the TL maps; (b)

normalized monochromatic TL curves recorded at λ = 565 and 550

nm, respectively; (c) non-normalized isothermal spectra of emitted

light recorded at 105 and 115 K, respectively. The red dashed and

blue solid curves show results for the doped and neat PVK/PBD

matrixes, respectively. Straight lines on the TL maps indicate selected

wavelength for monochromatic TL curves (horizontal lines) and

selected temperature for isothermal spectra of emitted light (vertical

lines).

The photoluminescence studies conﬁrmed the results

obtained from electrochemical measurements and theoretical

considerations, showing that the ancillary ligand structure

modiﬁcation does not aﬀect the bandgap. As a result, the

complexes were found to be green emitters with emission

maximum within the range of 30 nm. Surprisingly, it evidenced

that the electrochemically determined bandgap for nitro-

derived complexes was ﬂawed because of the incorporation of

lowest unoccupied orbital of nitro moiety. Therefore, together

with theoretical calculations, it proved that HOMO localized

mostly on iridium and cyclometalating ligands, and LUMOs

present mostly on bzq, are predominantly involved in radiative

transitions. It also correlates with the measured photo-

luminescence spectra, which resemble the typical broad

MLCT character of the transition. Additionally, the com-

pounds equipped with polyaromatic substituents exhibited

broad and structureless emission, whereas the complexes

bearing phenyl-based substituents were characterized by

structured emission. The observed structure could correspond

to ligand vibrations or low energy metal−ligand vibrations.

Additionally, the structured emission spectra were not

observed in solid state for thin ﬁlm layers of PVK/PBD

doped with 1 wt % of studied complexes, suggesting that

mainly MLCT states were ﬁlled by exciton transfer from the

host to the guest.

We found that although the structure modiﬁcation of β-

ketoiminato ligand did not allow tuning of the maximum

emission wavelength, the other properties related to

luminescence eﬃciency were strongly aﬀected. A particular

impact was observed in the photoluminescence quantum yield

of PVK/PBD layers doped with 1 wt % of the studied Ir(III)

phosphors, which ranged between <1 and 18%. A similar, but

less spectacular change, was observed for external quantum

eﬃciencies of the devices based on the same emissive layer,

namely from 0.01 to 3.20%. Interestingly, N,O-donating

ligands bearing unsubstituted aromatic moiety seemed to be

wavelength band; the main emission band is slightly red-shifted

by 15 nm and has a smaller full width at half-maximum

(FWHM). Considering the diﬀerences, it can be assumed that

in doped ﬁlm the emission from 5h dominates over PVK/PBD

emission. It is clear that the TL intensity of PVK/PBD with 1

wt % of 5h is signiﬁcantly higher than the TL intensity of the

matrix alone (about 30%). In addition, two normalized

monochromatic TL curves shown in Figure 15b diﬀer as

well. The monochromatic TL curve of the pure matrix

(recorded at λ = 550 nm) has a main maximum at 105 K.

Introduction of 1 wt % of 5h causes an increase in the relative

intensity of the TL signal in the high temperature range (above

110 K), as well as TL signal broadening in comparison to that

of its pure matrix analogue. In this case, high-temperature TL

signal contribution was greater in comparison to the previous

TL results obtained for the system doped with the iridium

complex containing one ﬂuorine atom in the para position of

phenyl ring of the β-ketoiminato ligand.¹⁰This comparison

points out the impact of deep traps located on the emitter

molecules. The depth of the traps might be approximated to

0.5 eV for holes and 0.4 eV for electrons, on the basis of

diﬀerences between HOMO and LUMO levels of the emitter

(5.3 and 2.8 eV, see Table 1) and the matrix (PVK (5.8, 2.2

eV) and PBD (6.2, 2.4 eV)). However, it should be taken into

account that HOMO and LUMO levels obtained from CV

experiments might be slightly varied in thin ﬁlms due to

diﬀerent environment of the molecules (THF as solvent in CV,

PVK/PBD in thin ﬁlms).

It can be concluded that introduction of the iridium complex

molecules to PVK/PBD matrix causes formation of new,

deeper trapping states that promote radiative recombination

on the dopant molecules. In addition, it must be emphasized

that the spectrum observed in the TL experiment for PVK/

PBD doped with 5h shown in Figure 15c is similar to the EL

spectrum, as far as their shapes and location of the maximum

M

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

intensity, chosen from the whole experiment. The structures were

solved with SHELXT³⁰and reﬁned with the full-matrix least-squares

procedure on F²by SHELXL-2013.³¹All non-hydrogen atoms were

reﬁned anisotropically, in 3j all hydrogen atoms were found in

diﬀerence Fourier maps, in 3h NH and OH hydrogen atoms, in 3d al

but methyl group, and these hydrogen atoms were freely, isotropically

reﬁned. All other hydrogen atoms were placed in idealized positions

and reﬁned as “riding model” with isotropic displacement parameters

set at 1.2 (1.5 for methyl groups) times U_eqof appropriate carrier

atoms. In the structure 3h the hydrogen atoms of methyl group C1

were found in two alternative positions with 50/50 site occupations.

The structures of two other compounds (5f and 5i) were also

conﬁrmed by X-ray diﬀraction, butdue to the poor quality of

crystals and severe disorderthese structures are attached as

Supporting Information. Table 1S lists the relevant information

about the crystal and reﬁnement data.

Crystallographic data for the structural analysis have been

deposited with the Cambridge Crystallographic Data Centre, Nos.

CCDC 1865881 (3d), 1892376 (3h), 1865882 (3j), 1865883 (3l),

1865883 (5f), 1865885 (5g), 1865886 (5i), 1865887 (5j) and

1865888 (5l). Copies of this information may be obtained free of

charge from The Director, CCDC, 12 Union Road, Cambridge, CB2

1EZ, UK. Fax: + 44(1223)336-033. E-mail: deposit@ccdc.cam.ac.uk.

www.ccdc.cam.ac.uk.

more promising in the application as dopants for PhOLEDs.

However, this does not include the anthracenyl substituents

since incorporation of these moieties resulted in instability of

the complex upon photo- and electro-excitation.

From the tested devices, the best performance showed the

PhOLED doped with the complex bearing 4-(1-naphthyl)-

imino-2-pentanonate as an ancillary ligand. The maximum

measured work parameters reached 12 cd/A for current

eﬃciency and 16000 cd/m²for luminance. It should be

emphasized that these values are quite high, considering that

diodes were fabricated using solution methods and taking into

account the uncomplicated device structure. For reference, a

very similar OLED based on [Ir(bzq)₂(acac)] exhibited the

maximum current eﬃciency of 4.2 cd/A. In all the examined

devices, the electroluminescence spectra demonstrated only

the dopant emission, while a signiﬁcant contribution of the

PVK/PBD matrix emission was observed in the photo-

luminescence of the same emissive layers. This indicated that

the charge trapping might be the dominant mechanism of

exciton generation in EL phenomenon. In fact, spectrally

resolved thermoluminescence studies for the most eﬃcient

host−guest system proved that the recombination centers are

created on the emitter molecules which simultaneously act as

deeper trapping states, than the pure matrix. Therefore, on the

basis of our results, it can be concluded that the introduction of

the iridium complexes into the matrix causes eﬃcient exciton

creation on the emitter molecules in EL phenomenon, proving

their applicability in PhOLED technology.

Thermogravimetric Analysis (TGA). Thermogravimetric anal-

ysis (TGA) of the prepared complexes samples was carried out using a

Q50-TGA thermobalance (TA Instruments, Inc.) under an N₂ﬂow of

60 mL min⁻¹. Samples (3−5 mg) loaded on a platinum pan were

heated from RT to 1000 °C at a rate of 10 °C min⁻¹.

Electrochemical Measurements. Electrochemical properties

were estimated from cyclic voltammetry (CV) measurement using

Electrochemical Analyzer model 620 (CH Instruments). A classic

three electrode setup was used with a platinum wire as a working

electrode, platinum spiral as a counter electrode, and silver wire as

pseudo-reference electrode. Potential was calibrated by ferrocene as

an internal standard. Measurement was carried out in THF with

Bu₄NBF₄as a supporting electrolyte. Oxidation onset potential

(E_{ox onset}) and reduction onset potential (E_{red onset}) were estimated

from the intersections of tangential lines of redox peaks and

background line. Electrochemical energy gaps (E_g) were estimated

from equation E_g= (E_{ox onset}− E_{red onset})|e⁻|. Ionization potential (IP)

and electron aﬃnity (EA) were estimated from equations IP = (5.1 +

EXPERIMENTAL SECTION

■

General Information. All syntheses and manipulations were

carried out under argon using standard Schlenk-line and vacuum

techniques. The microwave-assisted reactions were performed with

the use of a CEM Discover microwave pressure system (power max

300 W, magnetron frequency 2455 MHz, pressure max 20 bar). The

chemicals were obtained from the following sources: IrCl₃·3H₂O from

Pressure Chemicals, acetone, acetylacetone (acacH), Et₂O, MeOH,

CDCl₃, CD₂Cl₂, 1,2-dichloroethane, THF, arylamines from Aldrich,

benzo[h]quinoline (bzqH) from ABCR, THF 99.5 extra dry from

Acros Organics, Bu₄NBF₄(purity > 98) from TCI. The complex

[{Ir(μ-Cl)(bzq)₂}₂] (4)²⁸was synthesized according to the published

method. All solvents and liquid reagents were dried and distilled

under argon prior to use. The NMR spectra for 4-arylimino-2-

pentanones were recorded in CDCl₃on 300 MHz spectrometer at

298 K, using SiMe₄as internal standard for ¹H and ¹³C measurements.

For iridium complexes, the chemical shifts were referred to the

residual protonated solvent peaks (¹H δ_H= 7.26 ppm for CDCl₃and

¹H δ_H= 5.32 ppm for CD₂Cl₂). The purity of iridium materials was

determined by elemental analysis on Flash 2000 (Thermo Scientiﬁc).

HRMS data were obtained on AMD 402 two−sector mass

spectrometer of B/E geometry (EI-MS) and Waters/Micromass Q-

tof Premier mass spectrometer equipped with an electrospray ion

source (ESI-MS).

E

_{ox onset})|e⁻| and EA = (5.1 + E_{red onset})|e⁻|.³²

Computational Methods. The full optimizations of all Ir(III)

complexes in their singlet ground state were carried out. Initially, the

density functional theory (DFT)³³with B3LYPthe hybrid Becke’s

three parameter functionaland Lee−Young−Parr exchange-corre-

lation potential were used.³⁴⁻³⁶For the Ir metal center the SDD basis

set with Stuttgart−Dresden pseudopotentials³⁷were used and the 6-

31G(d) basis set for H, C, N, and O atoms.³⁸After preliminary

calculations more ﬂexible basis sets including diﬀuse and polarization

functions were used variety 6-311++G(d,p). The DFT calculations

using the B3LYP,^34,36,36M06,³⁹WB97XD⁴⁰functionals based on the

optimized S₀geometries were performed to obtain the energies of the

highest occupied molecular orbital (HOMO) and the lowest

unoccupied orbital (LUMO). To investigate solvent eﬀect, the

ground−state geometry optimization of iridium compounds was also

performed within the SCRF (self-consistent reaction ﬁeld) theory

using the polarized continuum model (PCM)⁴¹⁻⁴³in chlorobenzene

(C₆H₅Cl) solution to model the interaction with the solvent. All

calculations were performed using the Gaussian09 software package in

PL-Grid infrastructure.⁴⁴

Spectroscopic Measurements. Photophysical studies of new Ir

complexes were performed in a solution as well as for emissive layers.

All compounds were diluted (∼10⁻⁵M) in chlorobenzene and the

standard 1 cm path length quartz cuvette was used. Thin ﬁlms (60

nm) of PVK/PBD blends (7:3 weight ratio) doped with iridium

complexes were prepared on quartz substrates by means of spin−

coating. Absorption spectra were detected by a Carry 5000 (Varian)

spectrometer. The emission spectra were recorded on Edinburgh

Synthesis of Ligands and Complexes. Detailed descriptions of

the preparation of ligands and complexes, as well as spectroscopic

data, can be found in the Supporting Information.

X-ray Crystallography. Diﬀraction data were collected by the ω-

scan technique, for 5g and 5l at 100(1) K, for 3h and 3j at 130(1) K,

for 3d and 3l at room temperature on a Rigaku Xcalibur four-circle

diﬀractometer with an Eos CCD detector and graphite-monochro-

mated Mo Kα radiation (λ = 0.71069 Å), and for 5j at room

temperature on Rigaku SuperNova four-circle diﬀractometer with

Atlas CCD detector and mirror-monochromated Cu Kα radiation (λ

= 1.54178 Å). The data were corrected for Lorentz polarization as

well as for absorption eﬀects.²⁹Precise unit-cell parameters were

determined by the least-square ﬁt of reﬂections of the highest

N

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Instruments FLS980 spectroﬂuorometer equipped with an integrating

sphere used to determine the photoluminescence quantum yields.

Samples of thick layers (few μm), required for SRTL experiments,

were prepared by drop−casting from chlorobenzene solutions onto

aluminum substrates in ambient conditions. The samples for SRTL

studies were placed in the vacuum chamber on a thermostated holder

and covered by a sapphire plate. After sample photoexcitation at 15 K

for by pulsed nitrogen laser (λ = 337 nm) (PTI, model GL-3300T)

the TL measurements were carried out in the temperature range of

20−300 K with heating rate of 7 K/min. Sample thermoluminescence

was recorded by a detection system contained an optical collector, an

optical−ﬁber, a Micro HR Imaging Spectrograph and a CCD 3500

camera (Horriba Jobin Yvon).

Fabrication and Characterization of PhOLEDs. The OLEDs

were fabricated on glass substrates by means of the spin−coating

technique followed by vacuum evaporation. First, the hole injection

layer of poly(3,4−ethylenedioxythiophene) and poly-

(styrenesulfonate) mixture (PEDOT:PSS) was spin coated on an

ITO anode in ambient conditions. Second, the emissive layer of PVK/

PBD with 1 wt % Ir complex was spin-coated from a chlorobenzene

solution in a glove box. The dopant concentration inﬂuence on the

device parameters was checked only for 5h in the range of 0.5−7 wt

%. As the last step, the cathode materials were patterned through a

shadow mask by physical vacuum deposition technique. The complete

device stack can be written as ITO/PEDOT:PSS (20 nm)/PVK:PBD

+ emitter (60 nm)/Ca (20 nm)/Ag (100 nm). The device J−V−L

characteristics were determined with use of Keithley 2400 source

measurement unit connected with Minolta CS-200 camera. The EL

spectra were recorded by the CCD 3500 camera (Horiba Jobin

Yvon).

ACKNOWLEDGMENTS

■

This work was supported by the National Science Centre

(Poland) through grant no. UMO-2013/11/B/ST5/01334.

This research was supported in part by the PL-Grid

Infrastructure. E.W. acknowledges the ﬁnancial support of

the National Science Centre (Poland) through the grant no.

UMO-2017/24/T/ST5/00017.

REFERENCES

■

(1) Tang, C. W.; Vanslyke, S. A. Organic electroluminescent diodes.

Appl. Phys. Lett. 1987, 51, 913−915.

(2) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.;

Mackay, K.; Friend, R. H.; Burns, P. I.; Holmes, A. B. Light-emitting

diodes based on conjugated polymers. Nature 1990, 347, 539−541.

(3) (a) Miyata, S.; Nalwa, H. S. Organic Electroluminescent Materials

and Devices; CRC Press: London, 1997. (b) Li, Z. R. Organic Light−

Emitting Materials and Devices; CRC Press: Boca Raton, 2015.

(4) (a) Yersin, H. Highly Eﬃcient OLEDs with Phosphorescent

Materials; Wiley−VCH Verlag GmbH and KgaA: Weinheim, 2008.

(b) Zysman-Colman, E. Iridium(III) in optoelectronic and photonics

applications; John Wiley & Sons: Newark, 2017; pp 753.

(5) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.;

Kwong, R.; Tsyba, I.; Bortz, M.; Mui, B.; Bau, R.; Thompson, M. E.

Synthesis and characterization of phosphorescent cyclometalated

iridium complexes. Inorg. Chem. 2001, 40, 1704−1711.

(6) (a) Ko, S. H. Organic light emitting diode−material, process and

devices; InTech: London, 2011. (b) Pereira, L. F. R. Organic light

emitting diodes: the use of rare earth and transition metals; CRC Press,

Hoboken, 2012. (c) Chi, Y.; Chang, T.-K.; Ganesan, P.; Rajakannu, P.

Emissive bis-tridentate Ir(III) metal complexes: Tactics, photophysics

and applications. Coord. Chem. Rev. 2017, 346, 91−100.

(7) Perepichka, D. F.; Meng, H.; Ling, M.-M. Phosphorescent

Polymer Light-Emitting Diodes. In Organic Light-Emitting Materials

and Devices; Li, Z., Meng, H., Eds.; Taylor and Francis: Boca Raton,

2007; pp 413−449.

(8) Xu, H.; Chen, R.; Sun, Q.; Lai, W.; Su, Q.; Huang, W.; Liu, X.

Recent progress in metal−organic complexes for optoelectronic

applications. Chem. Soc. Rev. 2014, 43, 3259−3302.

(9) (a) Chen, S.; Wu, Q.; Kong, M.; Zhao, X.; Yu, Z.; Jia, P.; Huang,

W. On the origin of the shift in color in white organic light-emitting

diodes. J. Mater. Chem. C 2013, 1, 3508−3524. (b) Glowacki, I.;

Szamel, Z.; Wiosna-Salyga, G. Spectrally resolved thermolumines-

cence versus electroluminescence spectra of PVK doped with 1 wt %

of Ir(btp)₂(acac). Org. Electron. 2016, 31, 127−135.

(10) Witkowska, E.; Wiosna-Salyga, G.; Glowacki, I.; Orwat, B.; Oh,

M−j.; Kownacki, I.; Kubicki, M.; Gierczyk, B.; Dutkiewicz, M.;

Cieszko, P.; Luszczynska, B.; Ulanski, J.; Grzelak, I.; Hoffmann, M.;

Ledwon, P.; Lapkowski, M. Effect of fluorine substitution of the β-

ketoiminate ancillary ligand on photophysical properties and electro-

luminescence ability of new iridium(III) complexes. J. Mater. Chem. C

2018, 6, 8688−8708.

(11) (a) Radwan, Y. K.; Maity, A.; Teets, T. S. Manipulating the

excited states of cyclometalated iridium complexes with β-ketoiminate

and β-diketiminate ligands. Inorg. Chem. 2015, 54, 7122−7131.

(b) Maya, R. A.; Maity, A.; Teets, T. S. Fluorination of cyclometalated

iridium β-ketoiminate and β-diketiminate complexes: extreme redox

tuning and ligand-centered excited states. Organometallics 2016, 35,

2890−2899. (c) Lai, P.-N.; Brysacz, C. H.; Alam, M. K.; Ayoub, N. A.;

Gray, T. G.; Bao, J.; Teets, T. S. Highly efficient red-emitting bis-

cyclometalated iridium complexes. J. Am. Chem. Soc. 2018, 140,

10198−10207.

ASSOCIATED CONTENT

■

S

* Supporting Information

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.inorg-

chem.9b02785.

Procedures for the synthesis of ligands and complexes,

NMR spectra of the new compounds, X-ray analysis

detailed data, thermal analysis curves, cyclic voltammo-

grams, theoretical and experimental data correlation,

frontier orbitals’ energy levels and distributions,

optimized geometry coordinates, and additional photo-

physical data (PDF)

Accession Codes

CCDC 1865881−1865888 and 1892376 contain the supple-

mentary crystallographic data for this paper. These data can be

obtained free of charge via www.ccdc.cam.ac.uk/data_request/

cif, or by emailing data_request@ccdc.cam.ac.uk, or by

contacting The Cambridge Crystallographic Data Centre, 12

Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

AUTHOR INFORMATION

■

Corresponding Authors

*E-mail: irekk@o365.amu.edu.pl.

*E-mail: gabriela.wiosna-salyga@p.lodz.pl.

ORCID

(12) Citti, C.; Battisti, U. M.; Ciccarella, G.; Maiorano, V.; Gigli, G.;

Abbate, S.; Mazzeo, G.; Castiglioni, E.; Longhi, G.; Cannazza, G.

Analytical and preparative enantioseparation and main chiroptical

properties of Iridium(III) bis(4,6-difluorophenylpyridinato)-

picolinato. J. Chromatogr. A 2016, 1467, 335−346.

Ireneusz Kownacki: 0000-0002-5445-760X

Maciej Kubicki: 0000-0001-7202-9169

Przemyslaw Ledwon: 0000-0001-6769-8739

Notes

(13) Da Silva, M. A. V. R.; Da Silva, M. D. M. C. R.; Paiva, J. P. A.;

Nogueira, I. M. C. S.; Damas, A. M.; Berkley, J. V.; Harding, M. M.;

The authors declare no competing ﬁnancial interest.

O

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Akello, M. J.; Pilcher, G. Thermochemical and Crystallographic

Studies of some p-Ketoimine Derivatives. J. Chem. Soc., Perkin Trans. 2

1993, 1765−1769.

(14) (a) Gavezzotti, A. Are Crystal Structures Predictable? Acc.

Chem. Res. 1994, 27, 309−314. (b) Gavezzotti, A.; Filippini, G. J.

Geometry of the intermolecular X-H···Y (X, Y = N, O) hydrogen

bond and the calibration of empirical hydrogen-bond potentials. J.

Phys. Chem. 1994, 98, 4831−4837.

(15) Chepelin, O.; Ujma, J.; Wu, X.; Slawin, A. M. Z.; Pitak, M. B.;

Coles, S. J.; Michel, J.; Jones, A. C.; Barran, P. E.; Lusby, P. J.

Luminescent, Enantiopure, Phenylatopyridine Iridium-Based Coordi-

nation Capsules. J. Am. Chem. Soc. 2012, 134, 19334−19337.

106, 6647−6653. (b) Nonoyama, M. Benzo[h]quinolin-10-yl-N

Iridium(III) Complexes. Bull. Chem. Soc. Jpn. 1974, 47, 767−768.

(29) CrysAlisPro (Version 1.171.38.41), Rigaku OD, 2015.

(30) Sheldrick, G. M. SHELXT − Integrated space-group and

crystal-structure determination. Acta Crystallogr., Sect. A: Found. Adv.

2015, A71, 3−8.

(31) Sheldrick, G. M. Crystal structure refinement with SHELXL.

Acta Crystallogr., Sect. C: Struct. Chem. 2015, C71, 3−8.

(32) Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G.

C. Electrochemical Considerations for Determining Absolute Frontier

Orbital Energy Levels of Conjugated Polymers for Solar Cell

Applications. Adv. Mater. 2011, 23, 2367−2371.

(33) Becke, A. D. Density-functional thermochemistry. III. The role

of exact exchange. J. Chem. Phys. 1993, 98, 5648−5652.

(34) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J.

Ab Initio Calculation of Vibrational Absorption and Circular

Dichroism Spectra Using Density Functional Force Fields. J. Phys.

Chem. 1994, 98, 11623−11627.

(35) Kim, K.; Jordan, K. D. Comparison of Density Functional and

MP2 Calculations on the Water Monomer and Dimer. J. Phys. Chem.

1994, 98, 10089−10094.

(36) Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P. V.

R. Efficient diffuse function-augmented basis sets for anion

calculations. III.* The 3-21+G basis set for first-row elements, Li−

F. J. Comput. Chem. 1983, 4, 294−301.

(37) Dunning, T. H.; Hay, P. J. Gaussian Basis Sets for Molecular

Calculations. In Applications of Electronic Structure Theory; Schaefer

III, H. F., Eds.; Plenum: New York, 1976; pp 1−28.

(38) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.;

Gordon, M. S.; DeFrees, D. J.; Pople, J. A. Self-consistent molecular

orbital methods. XXIII. A polarization-type basis set for second-row

elements. J. Chem. Phys. 1982, 77, 3654−3665.

(39) Zhao, Y.; Truhlar, D. G. The M06 suite of density functionals

for main group thermochemistry, thermochemical kinetics, non-

covalent interactions, excited states, and transition elements: two new

functionals and systematic testing of four M06-class functionals and

12 other functionals. Theor. Chem. Acc. 2008, 120, 215−241.

(40) Chai, J.−D.; Head-Gordon, M. Long-range corrected hybrid

density functionals with damped atom−atom dispersion corrections. J.

Chem. Phys. 2008, 128, 084106.

̀

́

̈

(16) (a) Rafols-Ribe, J.; Will, P.-A.; Hanisch, C.; Gonzalez-Silveira,

M.; Lenk, S.; Rodríguez-Viejo, J.; Reineke, S. High-performance

organic light-emitting diodes comprising ultrastable glass layers. Sci.

Adv. 2018, 4 (5), 1−9. (b) Kwak, K.; Cho, K.; Kim, S. Analysis of

thermal degradation of organic light-emitting diodes with infrared

imaging and impedance spectroscopy. Opt. Express 2013, 21, 29558−

29559.

(17) Chen, Z.-Q.; Shen, X.; Xu, J.-X.; Zou, H.; Wang, X.; Xu, Y.;

Zhu, D.-R. Iridium(III) complexes based on 5-nitro-2-(2′,4′-

difluorophenyl)pyridyl: Syntheses, structures and photoluminescence

properties. Inorg. Chem. Commun. 2015, 61, 152−156.

(18) Hirva, P.; Haukka, M.; Jakonen, M.; Moreno, M. A. DFT tests

for group 8 transition metal carbonyl complexes. J. Mol. Model. 2008,

14, 171−181.

(19) Skorka, L.; Filapek, M.; Zur, L.; Malecki, J. G.; Pisarski, W.;

Olejnik, M.; Danikiewicz, W.; Krompiec, S. Highly Phosphorescent

Cyclometalated Iridium(III) Complexes for Optoelectronic Applica-

tions: Fine Tuning of the Emission Wavelength through Ancillary

Ligands. J. Phys. Chem. C 2016, 120, 7284−7294.

(20) Kodate, S.; Suzuka, I. Assignments of Lowest Triplet State in Ir

Complexes by Observation of Phosphorescence Excitation Spectra at

6 K. Jpn. J. Appl. Phys. 2006, 45, 574−578.

(21) (a) Li, J.; Djurovich, P. I.; Alleyne, B. D.; Yousufuddin, M.; Ho,

N. N.; Thomas, J. C.; Peters, J. C.; Bau, R.; Thompson, M. E.

Synthetic Control of Excited-State Properties in Cyclometalated

Ir(III) Complexes Using Ancillary Ligands. Inorg. Chem. 2005, 44,

1713−1727. (b) Yersin, H.; Rausch, A. F.; Czerwieniec, R.; Hofbeck,

T.; Fischer, T. The triplet state of organo-transition metal

compounds. Triplet harvesting and singlet harvesting for efficient

OLEDs. Coord. Chem. Rev. 2011, 255, 2622−2652.

(22) Hsu, C.-C; Lin, C.-C.; Chou, P.-T.; Lai, C.-H.; Hsu, C.-W.; Lin,

C.-H.; Chi, Y. Harvesting Highly Electronically Excited Energy to

Triplet Manifolds: State-Dependent Intersystem Crossing Rate in

Os(II) and Ag(I) Complexes. J. Am. Chem. Soc. 2012, 134, 7715−

7724.

̀

(41) Cances, E.; Mennucci, B.; Tomasi, J. A new integral equation

formalism for the polarizable continuum model: Theoretical back-

ground and applications to isotropic and anisotropic dielectrics. J.

Chem. Phys. 1997, 107, 3032−3041.

(42) Mennucci, B.; Tomasi, J. Continuum solvation models: A new

approach to the problem of solute’s charge distribution and cavity

boundaries. J. Chem. Phys. 1997, 106, 5151−5158.

(43) Cossi, M.; Barone, V.; Mennucci, B.; Tomasi, J. Ab initio study

of ionic solutions by a polarizable continuum dielectric model. Chem.

Phys. Lett. 1998, 286, 253−260.

(23) You, Y.; Nam, W. Photofunctional triplet excited states of

cyclometalated Ir(III) complexes: beyond electroluminescence. Chem.

Soc. Rev. 2012, 41, 7061−7084.

(44) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson,

G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.;

Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J.

V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams-Young, D.; Ding, F.;

Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson,

T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.;

Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;

Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;

Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.;

Bearpark, M. J.; Heyd, J. J.; Brothers, E. N.; Kudin, K. N.; Staroverov,

V. N.; Keith, T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.;

Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;

Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.;

Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J.

Gaussian 16, Revision B.01; Gaussian, Inc.: Wallingford CT, 2016.

(24) Kawamura, Y.; Goushi, K.; Brooks, J.; Brown, J. J.; Sasabe, H.;

Adachi, C. 100% phosphorescence quantum efficiency of Ir(III)

complexes in organic semiconductor films. Appl. Phys. Lett. 2005, 86,

071104−3.

(25) Wu, L−L.; Tsai, S−H.; Guo, T−F.; Yang, Ch−H.; Sun, I−W.

Synthesis and green electrophosphorescence of a novel cyclo-

metalated iridium complex in polymer light-emitting diodes. J.

Lumin. 2007, 126, 687−694.

(26) Tang, H.; Li, Y.; Wei, C.; Chen, B.; Yang, W.; Wu, H.; Cao, Y.

Novel yellow phosphorescent iridium complexes containing a

carbazole−oxadiazole unit used in polymeric light-emitting diodes.

Dyes Pigm. 2011, 91, 413−421.

(27) Glowacki, I.; Szamel, Z. The impact of 1 wt% of Ir(ppy)₃on

trapping sites and radiative recombination centres in PVK and PVK/

PBD blend seen by thermoluminescence. Org. Electron. 2015, 24,

288−296.

(28) (a) Sprouse, S.; King, K. A.; Spellane, P. J.; Watts, R. J.

Photophysical effects of metal-carbon σ bonds in ortho-metalated

complexes of iridium(III) and rhodium(III). J. Am. Chem. Soc. 1984,

P

DOI: 10.1021/acs.inorgchem.9b02785

Inorg. Chem. XXXX, XXX, XXX−XXX

Article Doi

Effect of β-ketoiminato ancillary ligand modification on emissive properties of new iridium complexes

DOI: 10.1021/acs.inorgchem.9b02785

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.inorgchem.9b02785

Full text of DOI:10.1021/acs.inorgchem.9b02785

Products guided by the article

R&D Labs maybe for 20010-38-2

Relevant to this article

Hot Product