A Synthesis and Luminescence Study of Ir(ppz)3 for Organic Light-Emitting Devices

Article abstract of DOI:10.1246/bcsj.77.751

Tris(1-phenyl-κC¹-pyrazolato-κN²)iridium (Ir(ppz)₃) was prepared and its luminescence properties were investigated for the application to organic light-emitting devices (OLEDs). The photoluminescence (PL) spectra of Ir(ppz)₃ in dichloromethane showed a peak at 437 nm at room temperature. The luminescent lifetime of an Ir(ppz)₃ film doped in CBP was found to be 218 ns, which indicated that its emission is phosphorescent. OLEDs were fabricated with doped films of Ir(ppz)₃ in several hosts, and the electroluminescence (EL) peak was observed at 450 nm. The luminance of OLEDs was pure blue, with the CIE coordinates of x = 0.158, y = 0.139 at 100 cd/m², but luminous efficiencies were low since the LUMO of Ir(ppz)₃ is higher than those of the hosts used.

Full text of DOI:10.1246/bcsj.77.751

Ó 2004 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 77, 751–755 (2004)
751
A Synthesis and Luminescence Study of Ir(ppz)3 for Organic
Light-Emitting Devices
Eun Jeong Nam,1;3 Jun Ho Kim,2 Bong-Ok Kim,3 Sung Min Kim,3 No Gill Park,3
ꢀ;1;3
Young Sik Kim,1;3 Young Kwan Kim,1;3 and Yunkyoung Ha
1Department of Science, College of Engineering, Hongik University, Seoul, Korea
2Department of Electrical Engineering, Hongik University, Seoul, Korea
3Research Institute for Science and Technology, Hongik University, Seoul, Korea
Received September 8, 2003; E-mail: ykha@hongik.ac.kr
Tris(1-phenyl-ꢀC1-pyrazolato-ꢀN2)iridium (Ir(ppz)3) was prepared and its luminescence properties were investigat-
ed for the application to organic light-emitting devices (OLEDs). The photoluminescence (PL) spectra of Ir(ppz)3 in di-
chloromethane showed a peak at 437 nm at room temperature. The luminescent lifetime of an Ir(ppz)3 film doped in CBP
was found to be 218 ns, which indicated that its emission is phosphorescent. OLEDs were fabricated with doped films of
Ir(ppz)3 in several hosts, and the electroluminescence (EL) peak was observed at 450 nm. The luminance of OLEDs was
pure blue, with the CIE coordinates of x ¼ 0:158, y ¼ 0:139 at 100 cd/m2, but luminous efficiencies were low since the
LUMO of Ir(ppz)3 is higher than those of the hosts used.
Since Tang and co-workers reported organic light-emitting
devices (OLEDs) with a multi-layer structure, materials for
the devices have been extensively studied over the past dec-
ade.1,2 Luminescent materials for these devices are generally
classified into two types, fluorescent ones and phosphorescent
ones. Fluorescent materials internally experience a quantum ef-
ficiency loss, however, due to non-emissive triplet exciton cor-
responding to 75% of the excited states. Recently, to achieve a
high efficiency in OLED, electrophosphorescent devices were
developed by Forrest and Thompson.3 An internal quantum ef-
ficiency approaching 100% was obtained by utilizing both sin-
glet and triplet excitons occurring in the emitting layer doped
with a phosphorescent dopant.
was reported that Ir(ppz)3 was employed as an exciton blocking
layer or as a dopant in the polymer host in other devices, but to
best of our knowledge, this is the first time a device containing
the emitting layer of Ir(ppz)3 in monomer hosts was fabricated
using a vacuum deposition method.7 A theoretical calculation
indicated that the ligand, ppz, would lead to blue emission upon
coordination with the iridium center. The excited state lifetime
of Ir(ppz)3 was found to be more than 200 ns, indicating that the
emission was phosphorescence. Electroluminescent (EL) de-
vices were fabricated with a doped film of Ir(ppz)3 in several
hosts, and their EL characteristics were discussed.
Experimental
The heavy metal, such as Ir or Pt, in the complexes was
known to induce the intersystem crossing by strong spin–orbit
coupling, leading to mixing of the singlet and triplet excited
states.4,5 The spin-forbidden nature of radiative relaxation from
the triplet excited state was then possible, resulting in high
phosphorescence efficiencies. Thus, iridium complexes and
platinum complexes have extensively been introduced to the
emitting layer as dopants. In addition, iridium complexes were
known to have high photoluminescence efficiency and relative-
ly short excited state lifetime, which minimizes quenching of
triplet emissive states.6
Color tuning of the emission of the iridium complexes can be
achieved by a suitable choice of the ligand, and thus iridium
complexes chelated with various kinds of ligands have been re-
ported, demonstrating red, green, and blue phosphorescence.
Suitable ligands can be designed from theoretical calculations
and synthesized to be complexed with the Ir center, yielding
Ir complexes displaying the desirable color emission.
General Methods. All the procedures for the synthesis of
Ir(ppz)3 were carried out in an N2 or Ar atmosphere. Though the
resulting complex was found to be air stable, the stability of the in-
termediate species was a concern, and thus precautions against air
and water were taken during the synthesis. Solvents for the synthe-
sis were pre-dried according to purification methods, or anhydrous
solvents packaged under nitrogen were purchased and used with
standard Schlenk techniques. The ligand, 1-phenylpyrazole
(ppz), was purchased from Aldrich. UV–vis spectra were obtained
on a 8452A Hewlett Packard spectrophotometer with 1 ꢁ 10ꢂ4
M
(1 M = 1 mol dmꢂ3) of Ir(ppz)3 in CH2Cl2. PL spectra were ob-
tained on a LS50B BSM spectrophotometer with an excitation
wavelength of 350 nm. The concentration of Ir(ppz)3 in CH2Cl2
for the PL solution spectra was 1 ꢁ 10ꢂ3 M. 1H NMR spectra were
obtained on Bruker 200 spectrometers. FAB/MS spectra were
measured on a Jeol JMS-AX505WA at Seoul National University
in Korea. Elemental analyses were also performed by an EA1110
of CE instruments at Seoul National University in Korea.
Herein, we report the synthesis of an iridium complex,
Ir(ppz)3, and investigation of its luminescent properties. It
Synthesis of the Ir(ppz)3. Ir(ppz)3 was synthesized from reac-
tion of ppz with Ir(acac)3, according to the procedure previously
Published on the web April 2, 2004; DOI 10.1246/bcsj.77.751

752 Bull. Chem. Soc. Jpn., 77, No. 4 (2004)
Synthesis and Luminescence Study of Ir(ppz)3
reported for Ir(ppy)3 and its derivatives.8 Ir(acac)3 (100 mg) and
more than three molar amounts of ppz (0.95 mL) were dissolved
in degassed glycerol (10 mL), and the mixture was heated at reflux
under nitrogen for 10 h. After cooling, 1 M HCl (50 mL) was added
to the reaction mixture resulting in the precipitation of the crude
product. The precipitate was collected by filtration and dissolved
in hot dichloromethane (30 mL). The volume of the dichloro-
methane solution was reduced, and the concentrated solution was
flash-chromatographed on silica and eluted with CH2Cl2 to remove
the dark impurities. The yellow solution was concentrated and re-
crystallized from a CH2Cl2/MeOH solution to give a yellow pow-
der of Ir(ppz)3 with a yield of 25%.
Fig. 1. A synthetic scheme of Ir(ppz)3.
Ir(ppz)3: Tris(1-phenyl-ꢀC1-pyrazolato-ꢀN2)iridium, mp > 250
ꢃC. UV–vis (ꢁ/nm, ("/Mꢂ1 cmꢂ1)): 242 (4:4 ꢁ 104), 292 (1:7 ꢁ
104), 325 (1:4 ꢁ 104). 1H NMR (200 MHz, CDCl3) ꢂ 6.76–7.96
(m, 18H), 6.37 (d, 3H). FAB/MS calcd for C27H21N6Ir 621.7,
found 622. Anal. Calcd for C27H21N6Ir: C, 52.16; H, 3.40; N,
13.52%. Found: C, 52.05; H, 3.33; N, 12.82%.
Theoretical Calculation.
Computationally, the electronic
ground states of Ir(ppz)3 and its host materials (CBP, MCP,
BCP) were calculated using the B3LYP density functional meth-
od.9 LANL2DZ and 6-31G(d) basis sets were employed for Ir
and other atoms, respectively. In the case of Ir(ppz)3, time-depend-
ent DFT (TD-DFT) calculations using the B3LYP functional meth-
od were performed at the respective ground-state geometry, while
the basis set of ppz ligands were changed to 6–31+G(d) for the first
excited-state calculation of the complex.10–12 Typically, the lowest
10 triplet and 10 singlet roots of the nonhermitian eigenvalue equa-
tions were obtained to get the vertical excitation energies and com-
pared with the absorption spectrum to examine each peak.
Fabrication of OLED. OLEDs were fabricated by high vacu-
um (ꢄ1:3 ꢁ 10ꢂ4 Pa) thermal deposition of organic materials onto
the surface of an indium tin oxide (ITO, 30 ꢀ/Ã, 80 nm) coated
glass substrate, chemically cleaned using acetone, methanol, distil-
lated water, and isopropyl alcohol. The organic materials were de-
posited in the following sequence: 40 nm of N,N0-bis(1-naphthyl)-
N,N0-diphenyl-1,10-biphenyl-4,40-diamine (NPB) was used as a
hole transporting layer, followed by a 20 nm thick emissive layer
consisting of various hosts and Ir(ppz)3 as a dopant. The hosts used
were 4,40-N,N0-dicarbazole-1,10-biphenyl (CBP), 2,9-dimethyl-
4,7-diphenyl-1,10-phenanthroline (BCP), 1,3-bis(N-carbazolyl)-
benzene (MCP), and 1,3,5-tris(N-carbazolyl)benzene (TCP), and
the doping concentration was 8 wt %. Ten nm thick BCP and 30
nm thick tris(8-quibolinato)aluminum (Alq3) layers were deposit-
ed as an exciton blocking layer and as an electron transporting lay-
er, respectively. The typical organic deposition rate was 0.2 nm/s.
Finally, 150 nm of Al:Li was deposited as a cathode. All organic
materials used were supplied by Gracel Display Incorporation in
Korea. The active area of the OLED was 0.09 cm2. Current–volt-
age–light intensity of OLEDs was measured with a source measure
unit (Keithley 236), an electrometer (Keithley 617), and a photo-
diode. The transient PL was measured with a photo-excitation at
337 nm from a pulsed nitrogen laser. The pulse width was 4 ns.
The response was detected by a fast PMT and digitized by a 500
MHz digital storage oscilloscope (Tektronix TDS 654C).
Fig. 2. UV–vis and PL spectra of Ir(ppz)3.
um complexes. We think that it is due to extra nitrogen capable
of a different coordination in ppz, and due to a lack of ligand
stability under the reaction conditions.
The melting point of Ir(ppz)3 was above 250 ꢃC, while free
1
ppz was liquid at room temperature. The H NMR spectra of
these complexes showed proton peaks of the coordinated ligand
at 7.96–6.76, and 6.37 ppm, shifted upfield compared to those
of the free ligand at 7.96–7.29, and 6.47 ppm, respectively.
FAB/MS showed the molecular ion peak at m=e 622 for the
complex. Elemental analysis results of Ir(ppz)3 were also found
to be satisfactory.
The photophysical properties of Ir(ppz)3 were investigated
using UV–vis absorption spectra and photoluminescence (PL)
spectra. The absorption spectra for Ir(ppz)3 are shown in
Fig. 2. The absorption patterns of the spectra indicated that
ꢀ
ꢃ–ꢃ ligand-centered (LC) transitions occurred below 250
nm, and metal-to-ligand charge transfer (MLCT) transitions oc-
curred above 270 nm, similar to known cyclometallated iridium
complexes.13
The results of the DFT calculation on the ground state of
Ir(ppz)3 indicated that 5d orbitals of Ir and orbitals of ppz li-
gands were strongly mixed in the three highest occupied molec-
ular orbitals (HOMOs), which have been labeled as d2, d1a and
d1b following Hay’s notation.14 The 5d character in those orbi-
tals was 50% for d2 and 48% for d1a and d1b from the popula-
tion analysis. On the other hand, the three lowest unoccupied
ꢀ
ꢀ
ꢀ
molecular orbitals (LUMOs), labeled as ꢃ1 , ꢃ2a , and ꢃ2b
did not show any 5d–ꢃ mixing, indicating the 5d character
,
ꢀ
ꢀ
ꢀ
was 0% for ꢃ1 and 3% for ꢃ2a and ꢃ2b .
As shown in Table 1, the calculated excitation energy for the
lowest excited triplet state (T1) in Ir(ppz)3 was 3.14 eV, with
two higher triplet states extremely closed in energy. It was
0.55 eV higher than that of Ir(ppy)3.14 All corresponded to ex-
citations from an electron in an occupied orbital of d2, d1a, and
Results and Discussion
Synthesis and Characterization of Ir(ppz)3. Ir(ppz)3 was
prepared from the one-step reaction of Ir(acac)3 with the ppz
ligand as shown in Fig. 1. The yield of Ir(ppz)3 was 25%, which
was relatively low compared to those of Ir(ppy)3 or other iridi-
ꢀ
d1b containing significant Ir 5d character in the lowest ꢃ orbi-
tals of ppz ligands. Because the ground states of d2, d1a, and d1b

E. J. Nam et al.
Bull. Chem. Soc. Jpn., 77, No. 4 (2004)
753
Table 1. Calculated Excitation Energies (E), Dominant Orbital Excitation, and Oscillator Strengths (f)
from TD-DFT Calculations for Ir(ppz)3
Triplet states
T1(3A)
T2(3E)
Excitation
E/eV
Wavelength/nm
ꢀ
d2 ! ꢃ1
3.14
3.15
3.15
3.43
3.45
3.45
3.58
3.59
3.59
3.64
394.5
394.3
394.3
361.4
359.2
359.2
345.7
345.3
345.3
340.2
ꢀ
ꢀ
d1a ! ꢃ1
d1b ! ꢃ1
T3(3A)
T4(3E)
d2 ! ꢃ1
d2 ! ꢃ2a
d2 ! ꢃ2b
ꢀ
ꢀ
ꢀ
T5(3A)
T6(3E)
d2 ! ꢃ1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
d1a ! ꢃ1 , d2 ! ꢃ2a
ꢀ
d1b ! ꢃ1 , d2 ! ꢃ2b
T7(3A)
d2 ! ꢃ2a , d1b ! ꢃ1
ꢀ
Excitation
Singlet states
S1(1A)
S2(1E)
E/eV
Wavelength/nm
f
ꢀ
d2 ! ꢃ1
3.56
3.62
3.62
3.66
3.66
3.70
3.76
3.76
3.85
4.56
348.1
342.1
342.1
338.8
338.8
335.5
329.9
329.9
322.2
271.8
0.0141
0.0275
0.0277
0.0053
0.0052
0.0039
0.0556
0.0556
0.0622
0.0090
ꢀ
ꢀ
ꢀ
ꢀ
d1b ! ꢃ1
d1a ! ꢃ1
d2 ! ꢃ2a
d2 ! ꢃ2b
S3(1E)
S4(1A)
S5(1A)
d1a,1b ! ꢃ2a,2b
d1a,1b ! ꢃ2a,2b
d1a,1b ! ꢃ2a,2b
d1a,1b ! ꢃ2a,2b
ꢀ
ꢀ
ꢀ
ꢀ
S6(1A)
S7(1A)
ꢀ
d2 ! ꢃ4a
orbitals have very similar energy, it is thought that any excita-
tion of an electron starts from one of them. Since the occupied
orbital has a strong 5d component and the ligand ꢃ virtual or-
bital predominates in the unoccupied orbital, we assigned the
triplet (T1–T7) and the singlet (S1–S7) states in Table 1 to
MLCT (metal-to-ligand charge transfer) states.
blue phosphorescence.
Electroluminescent properties of Ir(ppz)3 in OLEDs were
studied in detail by using Ir(ppz)3 as an emitter doped in vari-
ous hosts. The device structure was ITO/NPB/host:Ir(ppz)3/
BCP/Alq/Al:Li. Hosts used were 4,40-N,N0-dicarbazole-1,10-
biphenyl (CBP), 2,9-dimethyl-4,7-diphenyl-1,10–phenanthro-
line (BCP), 1,3-bis(N-carbazolyl)benzene (MCP), and 1,3,5-
tris(N-carbazolyl)benzene (TCP). The EL spectra of the
OLEDs with various hosts are shown in Fig. 3. All spectra
showed very similar features regardless of the host, which indi-
cates that the emission spectra are from Ir(ppz)3. Furthermore,
the emission must originate from Ir(ppz)3, not from the host,
since the emission was found to be phosphorescence. The EL
emission peak was shown at 450 nm, slightly shifted from
the PL solution emission at 437 nm. The CIE coordinates for
the device were measured to be x ¼ 0:158, y ¼ 0:139 at 100
cd/m2, which corresponds to pure blue.
ꢀ
It is known that an evidence for significant mixing of singlet
and triplet excited states is seen in both absorption and emission
spectra of phosphorescent materials which show strong spin–
orbit coupling. The strong spin–orbit coupling on the Ir center
3
thus allows for the formally spin-forbidden MLCT transition
with an intensity comparable to the allowed 1MLCT transi-
tion.15
From the comparison with the absorption spectrum shown in
Fig. 2, peaks distributed from 280 to 400 nm were assigned to
MLCT absorptions. Singlet excited states (S1–S7) and triplet
excited states (T1–T7) are separately located in the regions
270–350 nm and 340–400 nm, respectively. The peaks located
at ca. 325 nm are believed to originate from S5 and S6, whose
oscillator strengths are dominant.
The powder of Ir(ppz)3 exhibited green luminescence under
365 nm light. On the other hand, the solid and solution PL spec-
tra showed somewhat different results. The PL solution spectra
of Ir(ppz)3 in dichloromethane were obtained, as shown in
Fig. 2. Ir(ppz)3 showed an emission at 437 nm. The spin-coated
neat solid film of Ir(ppz)3 on a quartz plate also exhibited PL
peaks at 410 nm with broad and weak intensity, and at 485
nm with relatively strong intensity.
The solid film of the complex doped in CBP (4,40-N,N0-di-
carbazole-1,10-biphenyl) was prepared by spin coating on a
quartz plate, and its emission lifetime at 437 nm was obtained.
The excited state lifetime was measured to be 218 ns, indicating
that the emission from the film of Ir(ppz)3 doped in CBP was
Fig. 3. EL spectra of Ir(ppz)3 with various hosts.

754 Bull. Chem. Soc. Jpn., 77, No. 4 (2004)
Synthesis and Luminescence Study of Ir(ppz)3
Fig. 4. J-V-L characteristics of OLEDs using 8% Ir(ppz)3:various hosts. The device structure is ITO/NPB/8% Ir(ppz)3:various
hosts/BCP/Alq/Al:Li.
one with MCP could be related to the better PL efficiency of
Table 2. HOMO, LUMO, and Energy Gap of Ir(ppz)3 and of
CBP than that of MCP.
To develop an efficient blue electrophosphorescent device, it
Its Host Materials Calculated with B3LYP Density Func-
tional
is really necessary to have a host material with proper energy
levels and a wide band gap.
Ir(ppz)3
CBP
MCP
BCP
LUMO
HOMO
Gap
ꢂ0:5537
ꢂ5:0251
4:4714
ꢂ1:2289
ꢂ5:3167
4:0878
ꢂ0:7498
ꢂ5:4555
4:7057
ꢂ1:2828
ꢂ5:7863
4:5035
Conclusion
Tris(1-phenyl-ꢀC1-pyrazolato-ꢀN2)iridium (Ir(ppz)3) was
prepared and its luminescent characteristics were investigated.
UV–vis spectra exhibited absorption below 400 nm. Theoreti-
cal calculations revealed that the absorption peak distributed
from 270 to 400 nm originated from an MLCT transition of sin-
glet and triplet states. PL solution spectra showed an emission
peak at 437 nm and the excited state radiative lifetime of
Ir(ppz)3 was 218 ns, which indicates the emission was blue
phosphorescence. The luminance of OLED using Ir(ppz)3
was pure blue, but luminous efficiencies were low, considering
the fact that the LUMO of Ir(ppz)3 is much higher than those of
the hosts used. To develop an efficient blue electro-phosphores-
cent device, it is necessary to have a host material with proper
energy levels and a wide band gap.
There is a report that a device containing Ir(ppz)3 doped in a
polymer film showed an emission at 547 nm.7c This EL wave-
length is significantly shifted from the PL wavelength of
Ir(ppz)3. It is believed that the device structure influences the
EL wavelength in that patent since the device structure is quite
different from ours.
J-V-L characteristics of OLEDs of Ir(ppz)3 with various
hosts are shown in Fig. 4. The device with CBP showed a max-
imum luminance of 365 cd/m2 at 216 mA/cm2 and a maximum
luminous efficiency of 0.23 cd/A at 10 V.
It was considered that the low luminous efficiencies of
OLEDs with Ir(ppz)3 were caused by the energy level mis-
match between the hosts and Ir(ppz)3. The calculated LUMO
of the hosts are shown in Table 2. There are two energy transfer
mechanisms—Dexter transfer and Forster transfer. For the
Forster transfer, the relevant energy gap is the excitation ener-
gy, while for the Dexter transfer the relative position of the ex-
cited state energy level of the host to that of the dopant is im-
portant. Since the LUMO of Ir(ppz)3 is higher than that of hosts
as shown in Table 2, the Dexter energy transfer from the host to
Ir(ppz)3 does not seem to be effective.
The authors thank the Korea Science and Engineering Foun-
dation for financial support (KOSEF R04-2001-000-00026-0)
of this work.
References
1
(1987).
2
C. W. Tang and S. A. Van Slyke, Appl. Phys. Lett., 51, 913
a) C. W. Tang, Information Display, 10, 16 (1996). b) Y.
Since all of the hosts used were not proper for the Dexter en-
ergy transfer from the host to Ir(ppz)3, the main energy transfer
mechanism would be a Forster transfer. In fact, the calculated
triplet excitation energy of Ir(ppz)3 is 3.14 eV, while the singlet
excitation energy of CBP is 3.50 eV, which means CBP could
be used as a host for Ir(ppz)3. Forster transfer is known to be
governed by the spectral overlapping between the absorption
of the dopant and the emission of the host. The emission peaks
of MCP, CBP, and BCP were 360 nm, 380 nm, and 450 nm, re-
spectively. CBP and MCP show good energy overlapping,
while BCP is not good enough to be used as a host. The reason
why the device with CBP showed higher EL efficiency than the
Hamada, IEEE Trans. Electron Devices, 44, 1208 (1997). c) C.
H. Chen and J. Shi, Coord. Chem. Rev., 171, 161 (1998).
3 M. A. Baldo, M. E. Thompson, and S. R. Forrest, Nature,
403, 750 (2000).
4
a) S. Lamansky, P. Djurovich, D, Murphy, F. Abdel-
Razzaq, H.-E. Lee, C. Adachi, P. E. Burrows, S. R. Forrest, and
M. E. Thompson, J. Am. Chem. Soc., 123, 4304 (2001). b) S.
Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq, R. Kwong,
I. Tsyba, M. Bortz, B. Mui, R. Bau, and M. E. Thompson, Inorg.
Chem., 40, 1704 (2001). c) T. Tsutsui, M.-J. Yang, M. Yahiro,
K. Nakamura, T. Watanabe, T. Tsuji, Y. Fukuda, T. Wakimoto,
and S. Miyaguchi, Jpn. J. Appl. Phys., 38, L1502 (1999).

E. J. Nam et al.
Bull. Chem. Soc. Jpn., 77, No. 4 (2004)
755
5
a) M. A. Baldo, M. E. Thompson, and S. R. Forrst, Pure
8
K. Deadbeat, P. I. Djurovich, F. O. Garces, G. Carlson, and
R. J. Watts, Inorg. Chem., 30, 1685 (1991).
A. D. Becke, J. Chem. Phys., 98, 5648 (1993).
Appl. Chem., 71, 2095 (1999). b) S.-C. Chan, M. C. W. Chan, Y.
Wang, C.-M. Che, K.-K. Cheung, and N. Zhu, Chem.—Eur. J.,
19, 4180 (2001). c) J. Brooks, Y. Babaya, S. Lamansky, P. I.
Djurovich, I. Tsyba, R. Bau, and M. E. Thompson, Inorg. Chem.,
41, 3055 (2002).
9
10 a) P. J. Hay and W. R. Wadt, J. Chem. Phys., 82, 270
(1985). b) J. Hay and W. R. Wadt, J. Chem. Phys., 82, 284
(1985). c) J. Hay and W. R. Wadt, J. Chem. Phys., 82, 299 (1985).
11 W. J. Hehre, R. Ditchfield, and J. A. Pople, J. Chem. Phys.,
56, 2257 (1972).
12 a) R. Bauernschmitt and R. Ahlrichs, Chem. Phys. Let., 256,
454 (1996). b) M. E. Casida, C. Jamorski, K. C. Casida, and D. R.
Salahub, J. Chem. Phys., 108, 4439 (1998).
13 a) A. P. Wilde, K. A. King, and R. J. Watts, J. Phys. Chem.,
95, 629 (1991). b) M. C. Colombo, A. Hauser, and H. U. Gudel,
Top. Curr. Chem., 171, 143 (1994).
14 P. J. Hay, J. Phys. Chem., A106, 1634 (2002).
15 S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq,
H.-E. Lee, C. Adachi, P. E. Burrows, S. R. Forrest, and M. E.
Thompson, J. Am. Chem. Soc., 123, 4304 (2001).
6
K. Dedeian, P. I. Djurovich, F. O. Carlson, and R. J. Watts,
Inorg. Chem., 30, 1685 (1991).
a) P. I. Djurovich, A. Tamayo, J. Li, B. D. Alleyne, I. Tsyba,
7
R. Bau, and M. E. Thompson, 2002 ACS National Meeting at
Boston, Division of Inorganic Chemistry, Technical Program
#93. b) V. Adamovich, J. Brooks, A. Tamayo, A. M. Alexander,
P. I. Djurovich, B. W. D’Andrade, C. Adachi, S. R. Forrest, and
M. E. Thompson, New J. Chem., 26, 1171 (2002). c) The referee
gratefully mentioned the US patent application 20010019782
which describes the EL devices containing the iridium complexes
including Ir(ppz)3. However, those devices were fabricated with
the polymer blend method and thus different from the devices
fabricated in this study.