Arylvinylene phenanthroline derivatives for electron transport in blue organic light emitting diodes

Article abstract of DOI:10.1016/j.orgel.2010.12.025

A series of novel arylvinylene substituted phenanthroline derivatives has been synthesised and their application as electron transporting materials in organic light emitting diodes (OLEDs) has been investigated. One particular derivative, namely, 2,9-bis-

Full text of DOI:10.1016/j.orgel.2010.12.025

Organic Electronics 12 (2011) 666–676
Contents lists available at ScienceDirect
Organic Electronics
journal homepage: www.elsevier.com/locate/orgel
Arylvinylene phenanthroline derivatives for electron transport
in blue organic light emitting diodes
⇑
Poopathy Kathirgamanathan , Sivagnanasundram Surendrakumar, Raghava Reddy Vanga,
Seenivasagam Ravichandran, Juan Antipan-Lara, Subramaniam Ganeshamurugan,
Muttulingham Kumaraverl, Gnanamoly Paramaswara, Vincent Arkley
Organic Electronics, Wolfson Centre, Brunel University, Kingston Lane, Uxbridge UB 8 3PH, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of novel arylvinylene substituted phenanthroline derivatives has been synthesised
and their application as electron transporting materials in organic light emitting diodes
(OLEDs) has been investigated. One particular derivative, namely, 2,9-bis-(2-thiophen-2-
yl-vinyl)-[1,10]phenanthroline (C-4) shows significantly lower operating voltage (36%
reduction) and increase of power efficiency of up to 83% compared to tris (8-quinolinola-
to)aluminium (III) (Alq3). C4 also shows substantially longer (nearly 6-fold) life-time than
bathophenanthroline (BPhen) in blue devices. Doping of C4 with electron donors such as
Received 3 September 2010
Received in revised form 26 November 2010
Accepted 31 December 2010
Available online 25 January 2011
Keywords:
Phenanthrolines
Electron transporters
OLEDs
Blue materials
Device fabrication
m-MTDATA,
a-NPB and lithium quinolinolate (Liq) further reduces the turn-on voltage
and operating voltage.
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
(i) poor mobility, therefore high operating voltage and re-
duced luminous efficiency. (ii) It leaves behind consider-
able amount of residue and ash under production
condition (during vacuum thermal evaporation). (iii) Per-
ceived toxicological properties of aluminium compounds.
Considerable research effort has been expended to develop
high mobility materials which would reduce the operating
voltage without diminishing the efficiency and life-time
[8,9]. In addition to metal quinolates, oxadiazole(s), oxa-
zole(s), thiazole(s), imidazole(s), pyridyl compounds,
silole(s), perfluorinated oligophenylene(s), phenanthro-
line(s), pyrimidone(s) and triazine(s) have been reported
by several authors [3–8,10–14,19,28–31].
Organic light emitting diode (OLED) based displays
have become one of the significant flat panel display tech-
nologies as evidenced by the production of active matrix
OLEDs for mobile phones by Samsung SDI and the intro-
duction of 1100 AM-OLED TV (XEL-1) to the market by Sony.
Recent announcement by Samsung of its intention to in-
vest US Dollars 2.2 billion into active matrix OLED Televi-
sion (AM-OLED) to bring them into mainstream TV
market has given considerable impetus to the scientists
developing materials for the OLED devices [1–9].
Since Tang et al. reported an organic light emitting de-
vice (OLED) using tris (8-quinolinolato)aluminium (III)
(Alq3) as an electron transporting material (ETM) in 1987,
Alq3 has widely been employed both in commercial pro-
duction and research because of its low cost and accept-
able life-time [2]. However, Alq3 has three disadvantages:
Amongst the phenanthroline derivatives, bathophen-
anthroline (BPhen), bathocuproine (BCP) have been exten-
sively studied because of their high mobility (1 ꢀ 10ꢁ4
cm2 Vꢁ1 sꢁ1
)
compared to that of Alq3 (1 ꢀ 10ꢁ6–1 ꢀ
10ꢁ7 cm2 Vꢁ1 sꢁ1) [6–7] and deep HOMO levels (typically,
ꢁ6 eV to ꢁ6.9 eV) which make them excellent hole block-
ers. However, both virgin BCP and BPhen give significantly
poorer life time compared to other hole blockers. The for-
mer tends to crystallize (BCP films crystallize upon
⇑
Corresponding author. Tel.: +44 1895 265094; fax: +44 1895 269737.
E-mail address: p.kathir@brunel.ac.uk (P. Kathirgamanathan).
1566-1199/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.orgel.2010.12.025

P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
667
storage) and the latter has a very low Tg of 62 °C [9,15–17].
Both BCP and BPhen were found to give acceptable life-
time if they are doped or mixed with electron donors such
as low work function metals (e.g. Li and Cs) or their com-
plexes [10–13]. Kim and Im [15] reported that their devices
with pristine BCP or BPhen lasted only a few hours even at
an initial luminance of 150 cd mꢁ2. Another potential can-
didate, 1,3,5-tris(phenyl-2-benzimidazolyl)benzene (TPBI),
has similar crystallisation problem [21].
Yan-Jun Li et al. recently reported some phenanthroline
derivatives with superior performance to Alq3 [11] for their
device structure ITO/a-NPB (50 nm)/Alq3 (40 nm)/ETL
(30 nm)/LiF (1 nm)/Al where their best ETM gave a 21%
reduction in operating voltage at 1000 cd mꢁ2 and 32% in-
crease in power efficiency (lm/W) compared to Alq3 based
devices. However, no life-time information was reported.
144 h if vacuum processing is employed as the manufactur-
ers would like to run their production equipment at least for
6 days in a row without interruption; (viii) good stability to
moisture and oxygen; (ix) capable of forming amorphous
film so that crystallisation induced degradation is avoided;
(x) photostability; (xi) if phosphorescent emitters are used,
the triplet energy level of the et should be higher than that of
the phosphorescent emitters; (xii) processability to produce
uniform and pinhole-free films either by evaporation (small
molecules) or by solution casting.
We have been interested in developing electron trans-
porters cum hole blockers which would give long life-
times. Here, we report the synthesis of some selected 2,9
substituted phenanthrolines, their physical properties
and their performance as ETMs in blue OLEDs [14].
A
good electron transporter (et) should have the
2. Experimental
following characteristics [7]: (i) high electron affinity to
match with the workfunction (W.F.) of the cathode, thus
reducing the energy barrier; (ii) high electron mobility
2.1. Synthesis
(l
e > 1 ꢀ 10ꢁ5 cm2 Vꢁ1 sꢁ1) to aid the transport of electrons
The compounds were synthesized according to the gen-
eral method as shown in Scheme 1. All the compounds
were purified by double sublimation.
into the emissive layer and efficiently confine the excitons in
this layer; (iii) a reversible electrochemical reduction with a
sufficiently high reduction potential;(iv) a high ionisation
potential (I.P.) to serve as a hole blocker (I.P. > 6 eV); (v) a
high glass transition temperature and melting point; (vi)
thermal stability up to at least 250 °C (high Tg/Tm/Td); (vii)
stability under evaporation conditions for a period of at least
2.1.1. 2,9-Bis(styryl)-[1,10]-phenanthroline [C-1]
A mixture of neocuproine hydrate (5.0 g; 0.024 mol)
and benzaldehyde (5.4 g; 5.2 mL; 0.050 mol) was refluxed
acetic anhydride
Ar-CHO
+
N
N
reflux, 6-18h
N
N
CH3
H3C
Ar
Ar
Scheme 1. Ar = phenyl (C-1), 4-trifluoromethylphenyl (C-2), 4-cyanophenyl (C-3) and thienyl (C-4).
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.2
1.0
0.8
0.6
0.4
0.2
0.0
C-4/ UV
C-2/ UV
C-1/ UV
C-3 /UV
BPhen/UV
C-4/FL
C-2/FL
C-1/FL
C-3/FL
BPhen/FL
300
400
500
600
700
λ/nm
Fig. 1. Thin film UV–Vis absorption and fluorescence spectra of C-1, C-2, C-3, C-4 and BPhen. Excitation wavelengths: C-1 (360 nm), C-2 (360 nm), C-3
(350 nm), C-4 (350 nm) and BPhen (350 nm).

668
P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
in acetic anhydride (20 mL) for 6 h and the cooled reac-
tion mixture was poured into ice-cold water. The product
was extracted into ethyl acetate. The organic phase was
washed with water, dried over anhydrous magnesium
sulphate and then the solvent removed. To the residue
con. HCl was added and the yellow solid that obtained
was again taken into dichloromethane and extracted with
5% NaOH solution. The organic phase was washed well
with water, dried over anhydrous magnesium sulphate
and solvent removed to give a yellow solid. The product
was purified by column chromatography on silica gel
(eluent: CH2Cl2–MeOH 49:1). The eluents containing the
product was evaporated and by addition of ether and
petroleum ether gave an analytically pure material 4.8 g
(52%). Anal. Calc. for C28H20N2: C, 87.47; H, 5.24; N,
7.28. Found: C, 87.14; H, 5.15; N, 7.20%. 1H NMR (CDCl3):
7.33(t, J = 7.4 Hz, H-1), 7.42(t, J = 7.4 Hz, H-2), 7.70(d,
J = 8.3 Hz, H-3), 7.78(d, J = 16 Hz, H-4), 7.73(d, J = 16 Hz,
H-5), 7.89(d, J = 8.4 Hz, H-6), 8.16(d, J = 8.4 Hz, H-8),
7.68 (s, H-8 and H-9).
2.1.2. 2,9-Bis(4,40-trifluoromethylphenyl)vinyl-[1,10]-
phenanthroline [C-2]
A mixture of neocuproine hydrate (5.0 g; 0.024 mol)
and 4-(trifluoromethyl)benzaldehyde (8.8 g; 0.050 mol)
was refluxed in acetic anhydride (20 mL) for 6 h and then
allowed to cool for 18 h to give light pinkish crystals.
Methanol and small amounts of water were added and
the crystals were collected by suction filtration. The prod-
uct was then washed thoroughly with methanol, water
and diethyl ether before dried under vacuum at 80 °C
for several hours to give (7.9 g; 63%) of product. Finally,
the product was purified by sublimation (270 °C at
10ꢁ6 Torr) to give an analytically pure material as a light
yellow solid, m.p. 279 °C (DSC, onset); 285 °C (DSC, peak).
Anal. Calc. for C30H18N2F6: C, 69.23, H, 3.49; N, 5.38.
Found: C, 69.54, H, 3.39, N, 5.40%; 1H NMR (d-DMF):
7.88(d, J = 8 HZ, H-3, 4H), 7.93 (d, J = 16 Hz, H-5), 8.02
(s, H-9), 8.03 (s, H-8), 8.11(d, J = 8 Hz, H-2, 4H), 8.20 (d,
J = 8.3 Hz, H-6), 8.26 (d, J = 16 Hz, H-4), 8.58 (d,
J = 8.3 Hz, H-7).
8
9
7
5
6
CHO
N
N
4
acetic anhydride
+
3
1
reflux, 6h
N
N
H
2
CH3
H3C
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.2
1.0
0.8
0.6
0.4
0.2
0.0
C-4-S/UV
C-2-S/UV
C-1-S/UV
C-3-S/UV
Batho-S/UV
C-4-S /FL
C-2-S/FL
C-1-S/FL
C-3-S/FL
Batho-S/FL
300
400
500
600
λ/nm
Fig. 2. UV–Vis absorption and fluorescence spectra of C-1 (4.16 ꢀ 10ꢁ5 M), C-2 (3.84 ꢀ 10ꢁ5 M), C-3 (5.06 ꢀ 10ꢁ5 M), C-4 (4.03 ꢀ 10ꢁ5 M) and BPhen
(6.0 ꢀ 10ꢁ5 M) in CH2Cl2. Excitation wavelengths: C-1, C-2, C-3, C-4 (all 350 nm) except BPhen (300 nm).

P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
669
Table 1
Physical properties of compounds, C-1–C-4 Compared with Alq3 and bathophen.
Compound
M.p.
(°C)
(a)
Tg
(°C)
(a)
Td
(°C)
(b)
HOMO
(eV)
(c)
LUMO
(eV)
(c)
UV (kmax/nm,
absorption)
Thin film
FL (kmax/nm,
emission)
Thin film
Comments
C-1
C-2
C-3
192
279
289
69
(d)
(d)
300
337
130
ꢁ5.8
ꢁ5.9
ꢁ5.9
ꢁ2.8
ꢁ2.8
ꢁ3.0
277 (m), 340 (sh)) 453, 493
279 (m), 345 (sh) 451, 491 (sh)
313 (m), 362 (sh) 461 (sh), 512
Decomposes on sublimation
Stable in OLED production conditions
Decomposes on sublimation. So, no device
fabricated
C-4
298
218
111
62
345
287
ꢁ6.0
ꢁ6.4
ꢁ3.0
ꢁ3.0
329(m), 385 (sh)
278 (m), 314 (sh) 385, 402 (sh)
520
Stable in OLED production conditions
Stable in OLED production conditions. Poor life-
time due to low Tg
Bathophen
(BPhen)
Alq3
415
172
352
ꢁ5.7
ꢁ2.9
265, 393
520
Stable in OLED production conditions
a: Determined by DSC; b: from TGA for 1% mass loss; c: determined from cyclic voltammetry; d: not observed; sh: shoulder.
2.0
-2.8
-2.8
-2.9
-3.0 -3.0 -3.0
3.0
4.0
5.0
E/eV
-5.7
6.0
-5.8
-5.9
-6.0
-5.9
-6.4
N
N
N
N
N
N
N
N
S
S
N
N
NC
C-3
CN
C-4
C-1
C-2
BPhen
F3C
CF3
Fig. 3. HOMO–LUMO levels of Alq3, C-1, C-2, C-3, C-4 and bathophenanthroline (BPhen).
2.1.3. 2, 9-Bis(4,40-cyanophenyl)vinyl-[1,10]-phenanthroline
[C-3]
methanol and finally with diethyl ether. The product
was dried under vacuum at 80 °C to give a light yellow
solid (7.9 g; 76%). Finally, the crude product was purified
by sublimation (at 280 °C; 10ꢁ6 Torr) to give an analyti-
cally pure material, m.p. 289 °C (DSC, onset). Anal. Calc.
for C30H18N4: C, 82.93; H, 4.18; N, 12.82. Found: C,
82.67; H, 4.07; N, 12.82%. The compound was so insoluble
in most of the organic solvents that 1H NMR could not be
carried out.
A mixture of neocuproine hydrate (5.0 g; 0.024 mol)
and 4-cyanobenzaldehyde (6.6 g; 0.050 mol) was refluxed
in acetic anhydride (30 mL) for 18 h under nitrogen
and then allowed to cool to room temperature to give a
yellow orange solid. The product was filtered off under
suction and washed with petroleum ether (40–60 °C),
9
8
7
CHO
6
N
N
4
acetic anhydride
reflux, 6h
+
5
3
N
N
2
CH3
H3C
CF3
CF3
F3C

670
P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
petroleum ether (40–60 °C), methanol followed by diethyl
2.0
3.0
4.0
-2.3
ether and dried under vacuum at 80 °C for several hours
(>10 h) to give (46 g; 48%) of product . Finally, the product
was further purified by sublimation (at 260 °C; 10ꢁ6 Torr)
to give a golden yellow solid, m.p. 298 °C (DSC, onset); Tg
111 °C. Anal. Calc. for C24H16N2S2: C, 72.70; H, 4.07; N,
7.06; S, 16.17. Found: C, 72.85; H, 4.12: N, 7.10; S, 16.17%;
1H NMR (d-DMF): 7.21(dd, H-2), 7.45 (d, J = 16 Hz, H-5),
7.52 (d, J = 3.5 Hz,H-3), 7.67 (d, J = 5 Hz, H-1), 7.95(S, H-8,
9), 8.15 (d, J = 8.3 Hz, H-6), 8.32 (d, J = 16 Hz, H-4), 8.49 (d,
J = 8.3 Hz, H-7). MS (m/z) (%): 419 (100) (M++Na), also peaks
appeared at m/z 435, 472 and 499.
-2.9
-2.9
-3.0
-2.9
-3.4 eV
LiF/Al
E/eV
ITO
5.0
CuPC
-5.3
-5.7
-5.7
6.0
-5.9
-6.0
3. Characterisation
UV–Vis absorption spectroscopy (Shimadzu, UV-1601)
and fluorescent spectroscopy (Perkin–Elmer LS55) were
performed on evaporated thin films on quartz (spectrosil)
and in dichloromethane solutions. The decomposition
characteristics were examined under nitrogen atmosphere
on a TGA equipment (Rheometric Scientific) and the Differ-
ential Scanning calorimetry was performed on Perkin
Elmer (Pyris 6 DSC). Band gap was estimated from absorp-
tion spectroscopy of thin films produced by vacuum ther-
mal evaporation.
N
O
Al
O
N
N
N
O
Perylene
α-NPB
BAlq2
Fig. 4. Energy level diagram for the device: ITO/CuPc(50 nm)/a-NPB
(75 nm)/BAlq2 + perylene(50:0.1 nm)/ETL(20 nm)/LiF(0.5 nm)/Al where
the ETL is either Alq3 or C-4.
HOMO–LUMO levels were determined by cyclic voltam-
metry (computer controlled potentiostat PAR 273 or CHI
600D, analyte = 0.5 mM, supporting electrolyte (tetrabutyl
ammonium tetrafluoroborate = 10 mM in dichloromethane
as the solvent). The working, counter and reference
electrodes were Pt foil, Pt wire and Ag/AgCl electrodes,
respectively.
2.1.4. 2,9-Bis-(2-thiophen-2-yl-vinyl)-[1,10]phenanthroline
[C-4]
A mixture of neocuproine hydrate (50 g; 0.24 mol) and
2-thiophene carboxaldehyde (46.5 mL; 0.50 mol) was re-
fluxed in acetic anhydride (90 mL) for 6 h and then allowed
to cool to room temperature over 20 h. Water (20 mL) and
methanol (130 mL) were added to the reaction mixture
and the product was collected by suction filtration to give
a light yellow brown solid. The solid was washed with
The HOMO/LUMO levels were determined from the
equation in the reference [26]:
HOMO=LUMO ¼ ꢁðE1=2vs:NHE þ 4:40Þ eV
ð1Þ
(NHE: normal hydrogen electrode)
500
105
1) C- 4
104
103
102
101
100
10-1
2) Alq
400
300
200
100
0
3
3) C-4
4) Alq
3
0
2
4
6
8
10
12
14
16
Voltage/V
Fig. 5. Current density–voltage–luminance characteristics of ITO/CuPc(50 nm)/a-NPB(75 nm)/BAlq2 + perylene(50:0.1 nm)/ETL(20 nm)/LiF(0.5 nm)/Al
where the ETL is either Alq3 or C-4.

P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
671
Table 2
Performance of blue devices (Set A).
ETM
Turn-on
voltage
(V)
Vop/V at
1000 cd mꢁ2 100 mA cmꢁ2 at
100 cd mꢁ2 100 cd mꢁ2
V at
Eff./cd Aꢁ1
Eff./lm Wꢁ1 Eff./cd Aꢁ1
at at
Eff./lm Wꢁ1
at
EQE Life-time (t1/2)/h)
(%)
initial luminance
1000 cd mꢁ2 1000 cd mꢁ2
100 cd mꢁ2
Alq3 (a)
C-4 (a)
6.0
3.5
11.0
7.0
12.5
8.0
1.2
2.3
0.5
1.2
2.2
2.5
0.6
1.1
1.54 7200
1.61 3000
(a): ITO/CuPC (50 nm)/
a-NPB (75 nm)/BAlq2 + perylene (50: 0.1 nm)/ETL (20 nm)/LiF (0.5 nm)/Al.
1.2
1.2
1.0
1.0
0.8
0.6
0.8
0.4
0.2
0.6
0.0
400
450
500
550
600
650
EL / ETL / Alq
3
0.4
0.2
0.0
EL / ETL / C-4
PL / C-4
PL /Alq
3
PL / α-NPB
PL / BAlq
2
PL/ BAlq2: Perylene [50:0.1]
400
450
500
550
600
650
λ/nm
Fig. 6. EL spectra (solid lines) of ITO/CuPc(50 nm)/a-NPB(75 nm)/BAlq2 + perylene (50:0.1 nm)/ETL(20 nm)/LiF(0.5 nm)/Al where the ETL is either Alq3 or
C-4. Thin film PL spectra of C-4, Alq3, a-NPB, BAlq2 and perylene (50:0.1).
The E1/2 employed here were determined from the
derivative of the cyclic voltammograms.
150 nm). All the devices were encapsulated with a UV cur-
able adhesive (Nagase, Japan) with glass backplates in a
glovebox under nitrogen atmosphere. The electrical and
optical measurements were carried out by computer con-
trolled Keithley 2400 Source Meter and Minolta (CS-
1000) spectrometer, respectively. Life time measurements
were carried out on encapsulated devices under constant
current.
3.1. Device fabrication
Electroluminescent devices were fabricated by thermal
evaporation method using a multi-chamber OLED pilot
plant machine (Solciet, manufactured by ULVAC, Japan).
The devices were fabricated on a patterned (pixelated
ITO substrate (40
X
per square surface resistance), Hitachi
4. Results and discussion
High Tech., Japan), cleaned with deionised water, acetone,
isopropyl alcohol and water (in that order), then dried at
150 °C for 20 min and then subjected to ozone cleaning
(UV irradiation, 185 nm) at 150 °C for 10 min. The sub-
strate was then plasma cleaned in the presence of oxygen
for 25 s and moved into the vacuum chamber from which
the hole injector (copper phthalocyanine, CuPC), hole
4.1. Physical properties
The thin film absorption and fluorescence spectra of the
compounds C-1-C-4 and bathophenanthroline (BPhen) are
compared in Fig. 1. The absorbance and fluorescence in
dichloromethane are shown in Fig. 2. All the compounds
were found to be fluorescent, light blue as thin films and
dark blue in solutions.
transporter
[N,N’-bis-(1-naphthalenyl)-N,N0-bis-phenyl-
-NPB)]], host (BAlq2 or BH-
(1,10-biphenyl)-4,40-diamine (
a
3)):dopant (perylene or BD-3), electron transporter
(20 nm) were sequentially deposited under vacuum. The
Table 1 summarises the physical properties of the com-
pounds C-1, C-2, C-3 and C-4. All the compounds were
purified by double sublimation.
evaporation rates of CuPc,
a-NPB, BAlq2, perylene, BH-3,
BD-3, Alq3 and C-4 were 0.7, 1.5, 1, 0.002, 1, 0.1, 1 and
HOMO–LUMO data are shown in Fig. 3. On substituting
the phenyl ring of C-1 with electron withdrawing groups
such as CF3 (C-2), the HOMO level is depressed by 0.1 eV
whilst LUMO level is unchanged. In contrast, both HOMO
0
1 ÅA sꢁ1, respectively.
An electron injection layer (LiF, 0.5 nm) was deposited
on top of the ETM before depositing the top contact (Al,

672
P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
and LUMO levels are lowered by 0.1 eV on substituting C-1
with the CN group (C-3). This is consistent with the postu-
lation by Bredas and Heeger [20].
evaporation. We report here the blue device performance
of only compound 4 as our preliminary results showed that
C4 was superior to C2 in terms of operating voltage, effi-
ciency and life-time [22] with green OLEDs. Thus, we fab-
ricated two sets of OLEDs. One (Set A), light blue based
on BAlq2 as host and perylene as dopant and the other
(Set B) dark blue [(CIE (x, y): (0.15, 0.14)] based on host
BH-3 (Kodak) and dopant BD-3 (Kodak).
4.2. Device performance
As C-1 and C3 decompose on sublimation, they were
not suitable for device fabrication under vacuum thermal
4.2.1. Device Set A
2.0
-2.3
-2.4
The device architecture of Set A is ITO/CuPC (50 nm)/a-
NPB (75 nm)/BAlq2 + perylene (50:0.1 nm)/ETL (20 nm)/LiF
(0.5 nm)/Al where the ETL is either Alq3 or C-4. The energy
level diagram for the device under zero bias is shown in
Fig. 4.
BH-3
-2.8
-2.9
-3.0 -3.0
3.0
4.0
-3.4 eV
LiF/Al
Fig. 5 shows the luminance vs. voltage and the current
density vs. voltage plots for the Set A. From Fig. 5, it is obvi-
ous that the current injection and transport are much more
efficient in the case of C-4 as opposed to Alq3 as evidenced
by the lower operating voltage presented by C-4 at any gi-
ven luminance or current density. For example, the voltage
required for devices with Alq3 as ETM to give a luminance
of 1000 cd mꢁ2 was 11 V as opposed to those with C-4 was
only 7 V, a reduction of nearly 36% in operating voltage.
Correspondingly, the power efficiency for devices with
Alq3 as ETL was 0.6 lm/W and those with Compound C-4
was 1.1 lm/W at 1000 cd mꢁ2, representing an increase of
83% in efficiency. The device data are also summarised in
Table 2. The colour co-ordinates obtained for devices with
Alq3 [(CIE (x, y): (0.173, 0.252)] and that for the devices
with C4 is (0.150, 0.233). This is consistent with our expec-
tation that the mobility of C-4 is much greater than that of
Alq3 and hence the electron/hole recombination would
E/eV
BD-3
ITO
5.0
CuPC
-5.3
-5.6
-5.9
-5.7
6.0
-6.0
-6.4
BD-3
BH-3
Fig. 7. Energy level diagram for ITO/CuPc (10 nm)/a-NPB(50 nm)/BH-
3 + BD-3 (30:0.5 nm)/ETL (20 nm)/LiF(0.5 nm)/Al where the ETL is either
Alq3, C-4 or BPhen.
take place closer to the
with C-4 and hence more saturated blue colour co-
ordinates. Fig. 6 compares the EL spectra of the devices
a-NPB interface for the devices
350
105
104
103
102
101
100
10-1
1) C-4
2) BPhen
300
3) Alq3
250
4) C-4
5) BPhen
200
6) Alq3
150
100
50
0
4
6
8
10
12
14
16
Voltage/V
Fig. 8. Current density–voltage–luminance characteristics of ITO/CuPc (10 nm)/a-NPB(50 nm)/BH-3 + BD-3 (30:0.5 nm)/ETL (20 nm)/LiF(0.5 nm)/Al where
the ETL is either Alq3, C-4 or BPhe.

P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
673
with C-4 and the Alq3 as ETL and the PL of
a
-NPB, BAlq2, a
relative intensity differences could be due to the slight dif-
ferences in the concentration of perylene in the host as the
control of concentration at such a low level (0.2% of pery-
lene) is not easy under the co-evaporation conditions.
The life-time of the devices with Alq3 as ETM has signif-
icantly longer life-time than C-4. We attribute this to the
high Tg (172 °C of Alq3) as compared to C-4 (111 °C) and
the fact that Alq3 is a mixture of at least two isomers thus
retaining a high degree of amorphicity [27].
mixture of BAlq2 with perylene (50:0.1) and Alq3 and C-4.
It is clear from Fig. 6 that the peak positions of the EL spec-
tra and the PL of mixture of BAlq2 and perylene are identi-
cal, but the relative ratios of the intensity of emission are
slightly different. The second peak at 483 nm from the EL
of the devices with C-4 as the ETL is slightly more intense
than the first peak at 455 nm when compared with the EL
from devices where Alq3 is ETL. The EL spectra of devices
with Alq3 has a longer emission tail than the EL spectra
of devices with C-4 as the ETL. This explains the differences
in colour co-ordinates obtained here. Furthermore, the
It is well known that the nature of current injection of
an electron transporter is dependent on its interfacial char-
acteristics with the adjacent layer, namely, host + dopant.
6
5
4
3
2
1
0
1) BPhen
2) Alq3
3) C-4
1
10
100
Luminance / cdm
1000
10000
-2
Fig. 9. Current efficiency vs. luminance characteristics of ITO/CuPc (10 nm)/a-NPB (50 nm)/BH-3 + BD-3 (30 + 0.5 nm)/ETL (20 nm)/LiF (0.5 nm)/Al, where
the ETL is either Alq3, C-4 or BPhen.
1) BPhen
2) Alq3
3) C-4
3
2
1
0
0
1
2
3
4
-2
Luminance / cd m
Fig. 10. Power efficiency vs. luminance characteristics of ITO/CuPc (10 nm)/a-NPB (50 nm)/BH-3 + BD-3 (30 + 0.5 nm)/ETL (20 nm)/LiF (0.5 nm)/Al, where
the ETL is either Alq3, C-4 or BPhen.

674
P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
Table 3
Performance for Blue Devices from Set B.
ETM
Turn-on
voltage
(V)
Vop/V at
V at
Eff./cd Aꢁ1
Eff./
Eff./cd Aꢁ1
at
Eff./lm Wꢁ1
at
CIE (x, y) at
EQE(%) Life-time
1000 cd mꢁ2 100 mA cmꢁ2 at
lm Wꢁ1 at
1000 cd mꢁ2
(t1/2) at
100 cd mꢁ2 100 cdmꢁ2
1000 cd mꢁ2 1000 cd mꢁ2
1000 cd mꢁ2
Alq3
6.0 0.2 11.5 0.2
14.0
12.0
12.0
3.5
4.5
4.6
1.3
2.0
2.0
4.0
4.4
4.4
1.2
1.4
1.6
(0.152,
0.145)
(0.147,
0.157)
(0.151,
0.144)
3.3
3.4
3.6
1500
BPhen 5.0 0.2
C-4 4.3 0.2
9.5 0.2
9.5 0.2
40
250
1.2
1.0
0.8
0.6
0.4
0.2
0.0
BH3 / FL
BD3 / FL
BH3+BD3=30:0.5 / FL
EL Spectra
400
450
500
550
600
λ/nm
Fig. 11. Fluorescence and EL spectra of ITO/CuPc (10 nm)/ a-NPB (50 nm)/BH-3 + BD-3 (30 + 0.5 nm)/C-4 (20 nm)/LiF (0.5 nm)/Al.
400
300
200
100
0
105
104
103
102
101
100
1) C-4+m-MTDATA(20+0.1)
2) C-4+a-NPB(20+0.1)
3) C-4+EI-101(18+2)
4) C-4
5) C-4+m-MTDATA(20+0.1)
6) C-4+a-NPB(20+0.1)
7) C-4+EI-101(18+2)
8) C-4
0
2
4
6
8
10
12
14
16
Voltage/V
Fig. 12. Current density vs. voltage vs. luminance for ITO/CuPc (10 nm)/a-NPB (50 nm)/BH-3 + BD-3 (30 + 0.5 nm)/C-4 (20 nm)/LiF (0.5 nm)/Al where C-4
was doped with m-MTDATA (0.5%), a-NPB (0.5 %) and Liq (10%).

P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
675
The question is whether C-4 would perform well in an-
other system. Further, the comparison of the electron
transporter, C-4 with Alq3 alone is not sufficient as it is
known that bathophenanthroline (BPhen) has much higher
electron mobility and deeper HOMO level than Alq3, any
new ETM has to be compared with BPhen as well. Thus,
we designed the device structure Set B to compare C-4
with BPhen and Alq3.
[23]. Okumoto and Kanno reported that doping of the elec-
tron transporters with hole transporters such as m-MTDA-
TA,
increased life-time [18]. We fabricated devices of the Type
(ITO/CuPc (10 nm)/ -NPB (50 nm)/BH-3 + BD-3
(30:0.5 nm)/C-4 (20 nm)/LiF (0.5 nm)/Al) where the com-
pound 4 is doped with m-MTDATA (0.5%), -NPB (0.5%)
a-NPB resulted in reduction in operating voltage and
C
a
a
and Liq (10%) and found that the turn-on voltage and the
operating voltages are indeed reduced (Fig. 12). The perfor-
mances of the devices are summarised in Table 4.
4.2.2. Device Set B
We found that m-MTDATA as the dopant was most
ITO/CuPc
(10 nm)/a-NPB
(50 nm)/BH-3 + BD-3
effective in reducing the operating voltage, then
a-NPB
(30:0.5 nm)/ETL (20 nm)/LiF (0.5 nm)/Al where the ETL is
either Alq3, C-4 or BPhen. The energy level diagram for
the devices in Set B is shown in Fig. 7.
and Liq. However, there was no significant gain in either
the current efficiency or power efficiency.
Fig. 13 shows the energy level diagram for device Type
C. The turn-on voltage of the doped systems simply follow
the energy difference between LUMO level of the dopants
and LUMO level of the host ETM (C-4), The difference of en-
ergy between the LUMO levels of the dopants, m-MTDATA,
Figs. 8–10 show the luminance vs. voltage, current den-
sity vs. voltage, current efficiency (cd/A) vs. luminance and
power efficiency (lm/W) vs. luminance for Set B. Reduction
in operating voltage with C-4 is again substantial. The
turn-on voltage for the devices with C-4, BPhen and Alq3
are 4.3, 5.0 and 6.0 V, respectively (Table 3). A 17% reduc-
tion in operating voltage and 33% increase in power effi-
ciency has been achieved by replacing Alq3 with C-4 at
1000 cd mꢁ2. The current injection from C-4 is similar to
that of BPhen.
We attribute the low turn-on voltage and operating
voltage of C-4 to its high mobility. The EL and PL spectra
of BH-3, BD-3 and a mixture of BH-3 and BD-3 (30.0:0.5)
are shown in Fig. 11. The PL spectra of the mixture is blue
shifted by 4 nm compared to virgin BD-3 as expected. The
EL spectra of the devices in Set B are identical to the PL of
the mixture. Thus, the same excited state is involved in
both the cases.
a-NPB and Liq and C-4 are 1.1, 0.7 and 0.5 eV which corre-
late with the turn on-voltage of 3.8, 4.1 and 4.2 V, respec-
tively. The diminution of the turn on voltage and operating
voltage could be attributed to charge transfer from the do-
nor molecules to the acceptor (C-4), thereby increasing the
conductivity [24,25]. Life-time measurements of the doped
systems are on going and will be reported elsewhere.
-1.9
2.0
-2.3
-2.3
-2.4
-2.5
BH-3
-3.0
3.0
4.0
-2.8
BD-3
-3.4 eV
LiF/Al
It has to be stressed that interfacial effects dominate
OLED performance and the electron transporter/host +
dopant interface is no exception. Thus, a good electron
transporter for one particular system may not be as good
in another system even if the mobilities of the ETM’s are
high. What is required is a balanced injection of electrons
and holes.
E/eV
ITO
5.0
CuPC
-5.1
-5.3
C 4
-5.3
The life time at 1000 cd mꢁ2 for C-4 is 6.25 times longer
than that for BPhen as ETL. We attribute the enhanced life-
time from devices with C-4 to its higher Tg of 111 °C as op-
posed to the lower Tg of 62 °C for BPhen.
-5.4
-5.6
6.0
-5.9
-6.0
Though, the life-time of devices with C-4 is longer than
that with BPhen, they are not as long as Alq3. It is general
practice to dope the electron transporters with lithium
quinolinolate (Liq) to improve the life time of devices
Fig. 13. Energy level diagram for ITO/CuPc (10 nm)/a-NPB (50 nm)/BH-
3 + BD-3 (30 + 0.5 nm)/C-4 (20 nm)/LiF (0.5 nm)/Al where C-4 was doped
with m-MTDATA (0.5%), a-NPB (0.5%) and Liq (10%).
Table 4
Device Performance of C-4 with m-MTDATA,
a
-NPB and Liq as dopants.
ETM
Turn-on
Vop/V at
V at
Eff./cd Aꢁ1 at Eff./lm Wꢁ1 at Eff./cd Aꢁ1 at
Eff./lm Wꢁ1 at EQE CIE (x, y) at
voltage (V) 1000 cd mꢁ2 100 mA cmꢁ2 100 cd mꢁ2
100 cd mꢁ2
1000 cd mꢁ2
1000 cd mꢁ2
(%) 1000 cd mꢁ2
Compound 4
(C-4)
C-4 + m-MTDATA 3.8
(0.5%)
4.3
9.5
8.7
8.7
9.0
12.0
10.5
10.5
11.5
4.6
4.8
4.0
4.3
2.0
4.4
1.6
3.6 (0.151, 0.144)
2.4
2.2
1.8
4.6
4.2
4.4
1.7
1.6
1.6
3.6 (0.152, 0.155)
3.5 (0.146, 0.139)
3.6 (0.148, 0.143)
C-4 +
a-NPB
4.1
(0.5%)
C-4 + Liq
(10%)
4.2

676
P. Kathirgamanathan et al. / Organic Electronics 12 (2011) 666–676
[8] D. Tanaka, T. Takeda, T. Chiba, S. Watanabe, J. Kido, Chem. Lett. 36
5. Conclusion
(2007) 262.
[9] P. Kathirgamanathan, V. Arkley, S. Surendrakumar, Y.-F. Chan, S.
Ravichandran, S. Ganeshamurugan, M. Kumaraverl, J. Antipan-Lara,
G. Paramaswara, V.R. Reddy, SID Digest 32.2 (2010) 465.
[10] S. Reinnke, F. Lindner, G. Schwartz, N. Seidler, K. Walzer, B. Lussem,
K. Leo, Nature 459 (2009) 234.
[11] Y.-J. Li, H. Sasabe, S.-J. Su, D. Tanaka, T. Takeda, Y.-J. Pu, J. Kido, Chem.
Lett. 38 (2009) 712.
[12] S.R. Forrest, V. Bulovic, P. Peumans, US Patent 6451,415 B1 (17-09-
2002).
[13] H.W. Lee, J.-G. An, H.-K. Yoon, H. Jang, N.G. Kim, Y. Do, Bull. Korean
Chem. Soc. 26 (2005) 1569.
[14] P. Kathirgamanathan, S. Surendrakumar, PCT Patent WO 2008/
078115 A1 (03-07-2008).
[15] V. Kim, W.B. Im, Phys. Status Solidi A 201 (2004) 2148.
[16] C.H. Hsiao, S.W. Liu, C.T. Chen, J.H. Lee, Org. Electron. 11 (2010) 1500.
[17] S. Naka, H. Okada, H. Onnegawa, T. Tsutsui, Appl. Phys. Lett. 76
(2000) 197.
[18] K. Okumoto, H. Kanno, US Patent 2006/0051563A1 (09-03-2006).
[19] J.P. Chen, X.C. Li, US Patent 6713,781 B1 (30-03-2009).
[20] J.L. Bredas, A.J. Heeger, Chem. Phys. Lett. 217 (1994) 507.
[21] N. Li, P. Wang, S.L. Lai, W. Liu, C.S. Lee, S.T. Lee, Z. Liu, Adv. Mater. 22
(2010) 527.
[22] P. Kathirgamanathan, S. Surendrakumar, S. Ravichandran, V.R.
Reddy, J. Antipan-Lara, S. Ganeshamurugan, M. Kumaravel, G.
Paramaswara, V. Arkley, Chem. Lett. 39 (11) (2010) 1222.
[23] E. Bohm, C. Pflumm, F. Voges, M. Flammich, H. Heil, A. Busing, A.
Parham, R. Fortte, T. Mijica, in: IDW’ 09 Proceedings, Miyazaki,
Japan, 2009, p. 431.
[24] P. Kathirgamanathan, D.R. Rosseinsky, J. Chem. Soc., Chem. Commun.
(1980) 306.
[25] D. Fox, M.M. Labes, A. Weissberger (Eds.), Physics and Chemistry of
the Organic Solid State, Interscience Publishers, New York, 1967.
[26] B.W. D’Andrade, S. Datta, S.R. Forrest, P. Djurovich, E. Polikarpov, M.
Thompson, Org. Electron. 6 (2005) 11.
We have demonstrated that it is possible to produce
phenanthroline derivatives for example, 2,9-bis-(2-thio-
phen-2-yl-vinyl)-[1,10]phenanthroline (C-4) which can
act as good electron transporters and hole blockers with
reasonably long life-times even in their virgin form. Fur-
ther synthetic work is in progress to identify phenanthro-
line derivatives which would give much longer life-time
than Alq3 or doped BPhen.
Acknowledgements
We thank Drs. T. Yuki and Mr. S. Nakajima (Pioneer,
Japan), Dr. C. Domoto (Kyocera), Dr. H. Ogawa (Kyocera)
and Dr. C.H. Lee , Dr. J.H. Lee and J.H. Shin (Samsung
Displays, Korea) for stimulating discussions. We also thank
Dr. H.H.-Y. Oh (LG Displays, Korea) for fruitful discussions.
We also thank Professor C.H. Chen (Shanghai Jiao Tong
University) for thought provoking questions and Dr.
David R. Rosseinsky of Exeter University for constant
encouragement.
We are grateful for the support from Brunel University
and encouragement from Professors C. Jenks, M. Sarhadi, G.
Rodgers and J. Silver.
References
[27] M. Cole, W. Brutting, Physics of Organic Semiconductors, Wiley-VCH
Verlag GmBH, ISBN-13:978-3-527-40550-3, 2005, pp. 95–128.
[28] H.Y. Chen, W.Y. Lam, J.D. Luo, Y.L. Ho, B.Z. Tang, D.B. Zhu, M. Wong,
H.W. Kwok, Appl. Phys. Lett. 81 (4) (2002) 574.
[29] H.F. Chen, S.J. Yang, Z.H. Tsai, W.Y. Hung, T.C. Wang, K.T. Wong, J.
Mater. Chem. 19 (2009) 8112.
[30] Z. Ge, T. Hayakawa, S. Ando, M. Ueda, T. Akiike, H. Miyamoto, T.
Kajita, M. Kakimoto, Org. Lett. 10 (3) (2008) 421.
[31] S.J. Su, H. Sasabe, Y.J. Pu, K. Nakayama, J. Kido, Adv. Mater. 22 (2010)
3311.
[1] T. Matsushima, C. Adachi, Appl. Phys. Lett. 89 (2006) 253506.
[2] C.W. Tang, S.A. Van Slyke, Appl. Phys. Lett. 51 (1987) 913.
[3] H. Sasabe, T. Chiba, S.J. Su, Y.-J. Pu, K. Naayama, J. Kido, Chem.
Commun. (2008) 5821.
[4] M. Ichikawa, N. Hiramatsu, N. Yokoyama, T. Miki, S. Narita, T.
Koyama, Y. Taniguchi, Phys. Status Solidi RRL 1 (1) (2006) R37.
[5] P. Kathirgamanathan, WO 00/32717 (08-06-2000).
[6] T. Sano, M. Fujita, T. Fujii, Y. Nishio, Y. Hamada, US Patent 5456988
(10-10-1995).
[7] A.P. Kulkarni, C.J. Tonzole, A. Babel, S.A. Jeneke, Chem. Mater. 16
(2004) 4556.