Carbazole–oxadiazole containing polyurethanes as phosphorescent host for organic light emitting diodes

(1)

Carbazole–oxadiazole containing polyurethanes as phosphorescent host

for organic light emitting diodes

Cheng-Hsiu Ku

a

, Chao-Hui Kuo

a

, Man-kit Leung

a,c,*

, Kuo-Huang Hsieh

a,b,* a

Institute of Polymer Science and Engineering, National Taiwan University, Taipei 10617, Taiwan b

Department of Chemical Engineering, National Taiwan University, Taipei 10617, Taiwan c_{Department of Chemistry, National Taiwan University, Taipei 10617, Taiwan}

a r t i c l e

i n f o

Article history:

Received 26 August 2008

Received in revised form 26 December 2008 Accepted 20 January 2009

Available online 30 January 2009

Keywords:

Polymeric light emitting diodes (PLEDs) Polyurethanes (PUs)

Phosphorescent host

Phosphorescent organic light emitting diodes (PHOLED)

a b s t r a c t

New types of polyurethanes (PUs) were prepared from condensation polymerization of isophorone diisocyanate (IPDI) with various combination of 9-butyl-3,6-bis(4-hydroxy-phenyl)carbazole (Cz) and 2,5-bis(4-hydroxyphenyl)-1,3,4-oxadiazole (OXD), and end-capped with 4-tert-butyl phenol. The Cz-OXD PUs can also be used as host for phosphorescent dye. Red EL emission was obtained when Ir(btp)2(acac) or Ir(2-phq)2(acac) was used as the

phosphorescent dyes in Cz-OXD (3:1) PU. Maximum brightness of 394 cd/m2_{and EL}

efﬁ-ciency of 1 cd/A were achieved for the Ir(2-phq)2(acac) base device. In addition, white light

PLED was demonstrated when co-dopant of Ir(btp)2(acac) and Firpic were used.

1. Introduction

Organic small-molecule light emitting diodes (OLEDs) and polymeric light emitting diodes (PLEDs) have been extensively studied since the pioneering work published by Tang [1]. To obtain reasonable light emitting perfor-mance, multilayer structures were adopted in which the emissive layer was sandwiched between the hole and elec-tron transport layers[2]. Unlike the multilayer OLEDs, the prototype of PLED was established on the basis of a single-layer of PPV on ITO[3]. However, the efﬁciency and bright-ness of the output was not ideal. Many different factors might affect the PLED performance. Nevertheless, balanced carrier injection and transport properties are important and necessary conditions to achieve high performance de-vices. MEH-PPV is one of the successful examples [4,5].

Similar idea has recently been extended to OLED[6]. Poly-urethanes (PUs) have been widely used in industrial appli-cations due to their good properties in good elasticity, ﬂexibility, thermal stability, and excellent chemical resis-tance[7–12]. In addition, PU conducting polymers have also been developed [13–15]. Recently, applications of PUs on PLED have been reported. It has been demonstrated that PUs could be used as effective hole-transport matrix

[16–18] as well as applied for light emitting layers

[19,20]. Since PUs were usually prepared from condensa-tion polymerizacondensa-tion of diols and diisocyanates under me-tal-ion free conditions, [21] metal-ion contaminants could therefore be minimized. In addition, the adhesion of polar PU on ITO surface is expected to be good. This may beneﬁt for the PLED device fabrication[22,23]. How-ever, perhaps due to the ﬂuorescence quenching effect through hydrogen bonding interactions, only limited num-bers of examples about using PU as the light emitting materials were known[24,12]. Carbazole (Cz) derivatives are photoconductors with good hole-transport properties

[25–28]. Recently, carbazole derivatives have been used as efﬁcient triplet hosts in phosphorescent OLED (PHOLED)

*Corresponding authors. Present address: Institute of Polymer Science and Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei 10617, Taiwan. Fax: +886 2 3636359 (M.-k. Leung).

E-mail addresses:[email protected](M.-k. Leung),khhsieh@ntu. edu.tw(K.-H. Hsieh).

Contents lists available atScienceDirect

European Polymer Journal

(2)

[29,30]. All these intrigued us to explore the application potentials of carbazole containing PUs on PLED applica-tions. 2,5-Diphenyl-1,3,4-oxadiazole (OXD) and its deriva-tives are common electron injection and transport materials[31–34]due to their prominent electron afﬁnity, high photoluminescence, and good thermal stability

[35,36]. Since doping of the electron-deﬁcient components into the hole-transport layer sometimes may improve the performance of the OLED device,[37] we are interested in studying the performance of Cz-OXD copolymers on PLED.

Herein we report the preparation and characterization of novel PUs, named as CX10, CX31, CX11, CX13 and CX01 (Scheme 1). The abbreviation of the polymers indeed reﬂects the composition of the polymers, in which C and X represent carbazole and oxadiazole units, respectively; and

the following number represents the feeding ratio of these two components in the PUs. Among the PUs, carbazole based CX10 and the oxadiazole based CX01 were homo-polymers and CX31–CX13 were carbazole–oxadiazole based co-polymers. Although PU polymers usually dis-solved in polar aprotic solvent such as DMF, this solvent is not completely compatible with PEDOT-PSS, a common water-soluble hole-injection polymer, because the PED-OT-PSS layer would be slightly etched by DMF during mul-ti-step coating process[16]. To make use of the PU solution compatibly with the PEDOT-PSS layer, we have adjusted the alkyl substituent on the PUs, so that the ﬁnal PUs would have good solubility in a relatively non-polar sol-vent such as THF.

In this study, we discovered that the PU could be used as host material for red phosphorescent emitter. White

N Bu O O HN O NH O O N N O O CX10 ( x : y =100 :0 ) (50%) CX31 ( x : y =75 :25 ) (55%) CX11 ( x : y =50 :50 ) (40%) CX13 ( x : y =25 :75 ) (60%) CX01 ( x : y =0 :100 ) (45%) HN O NH O O x y H N H N Br Br Pd(PPh3)4 B(OH)2 MeO N Bu MeO OMe CH2Cl2 BBr3 N Bu HO _OH HO C O NHNH2 C O OPh + N N O HO OH HO 270oC -PhOH and H2O NBS 1 (92 %) 5 (45 %) 4 (70 %) 3 (70 %) N Br Br Bu [(C4H9)4N]HSO4 KOH, reflex 2 hrs C4H9Br DMF, 0oC 2 (65 %) NCO OCN 4-Me3CC6H4OH (5%) DMF, 50oC + N N O HO OH X N Y Bu HO OH 4 5 IPDI a b c d e f g

(3)

light emitting device has also been fabricated to study. To our knowledge, successful examples of using PU as host for the light emitting layer are rare.

2. Experimental section

2.1. Fabrication of phosphorescent PLED device

The devices with a standard conﬁguration of ITO/PED-OT-PSS (30 nm)/PU (30 nm)-Ir complexes/BCP (10 nm)/ Alq3(40 nm)/LiF (1 nm)/Al (150 nm) were fabricated. ITO

plates were first cut into appropriate sizes and the ITO sur-face was cleaned by sonication, rinsing in deionized water, Triton-100 water solution, deionized water, acetone and ethanol treatment then dried by a dry nitrogen flow and finally cleaned by oxygen plasma. The PEDOT-PSS solution were prepared by mixing the commercially available PED-OT-PSS solution with diluted Triton-100 solution (5% in H2O) in a 10: 1 (V/V), spin-coated at a rate of 4000 rpm,

and dried at 90 °C for 30 min under vacuum. After then, the emitting polymer layers (2000 rpm, 30 nm) were spin-coated from a solution of Ir complex (4 mg) and CX31 (46 mg) in THF (2 mL) on top of the PEDOT-PSS layer at a spinning rate of 4500 rpm. The layer was baked (50 °C for 10 min), followed by thermal vacuum evaporation of a layer of 2,9-dimethyl-4,7-diphenyl-1,10-phenanthro-line (BCP) (10 nm), and Alq3 (40 nm) at pressure below

5 106_{torr with deposition rate of 3 Å/s. The LiF and Al}

contacts were ﬁnally deposited under the same vacuum conditions. The deposition rates for LiF and Al cathodes were 1 and 4 Å/s, respectively. The thickness of each layer was calibrated, respectively, by an Alpha step instrument.

2.2. Synthesis of PU’s CX10–CX01

The preparation of the homopolymers CX10 and CX01, and copolymers CX31, CX11, and CX13 were summarized inScheme 1. The actual ratio of the carbazole/oxadiazole units in the copolymers were determined by 1_{H NMR}

(Fig. 1). The data were summarized in Table 1and de-scribed as follows.

2.2.1. CX10

A typical procedure: To a two-necked round bottom ﬂask (50 mL), equipped with a magnetic stir under nitro-gen, was loaded a solution of 9-butyl-3,6-bis(4-hydroxy-phenyl)carbazole (4) (2.04 g, 5.0 mmol), isophorone diisocyanate (1.16 g, 5.25 mmol), and dried DMF (25 mL). After reaction for 24 h, 4-tert-butylphenol (0.075 g, 0.5 mmol) was injected into the reaction mixture as the terminating agent. The reaction was further reacted for an-other 24 h. When reaction was completed, the polymer was precipitated by drop-wise addition of the reaction mixture into methanol. The polymer was redissolved in DMF and re-precipitated in MeOH for at least 5–10 times in order to obtain an essentially pure sample. The PU was collected and washed with toluene and chloroform to re-move any DMF. The ﬁnal PU was collected and dried under vacuum conditions for about 12 h at 100–105 °C (2.85 g, 50%): Anal. Calcd based on the calibrated composition listed inTable 1: CX10: C, 74.4; H, 6.7; N, 6.5%. Found: C, 72.9; H, 7.4; N: 7.4%; 1H NMR (400 MHz, DMSO-d6) d

8.4–8.6 (bs, 2H, Ha), 8.0–7.5 (m, 8H, Hb, Hc, and Hdon

car-bazole), 7.3–7.1 (bs, 4H, He), 4.4–4.3 (bs, 2H, N–CH2–),1.8–

0.7 (m, 25H, alkyl protons). 2.2.2. CX31 (55%)

When reaction was completed, the polymer was precip-itated by drop-wise addition of the reaction mixture into methanol. The polymer was dissolved in DMF and re-precipitated again in MeOH for several times, followed by precipitation in MeOH/H2O = 4:1. The collected PU was

ﬁ-nally washed with toluene and chloroform to remove residual DMF. Anal. Calcd based on the calibrated composi-tion listed inTable 1. (Cz)74.7–(OXD)25.3: C, 72.0; H, 6.4; N,

7.7% Found: C, 71.5.; H, 7.4; N, 7.5%.1_{H NMR (400 MHz,}

DMSO-d6) d 8.7–8.5 (bs, Ha), 8.11–8.13 (bs, Hg), 8.05–7.60

(m, Hb, Hc, and Hd), 7.4–7.3 (bs, Hf), 7.2–7.1 (bs, He), 4.5–

4.4 (bs, N–CH2–), 1.7–0.8 (m, alkyl-protons). The ratio of

2.96/1.00 for the carbazole/oxadiazole units was deter-mined by their integrations at d 7.2–7.1 and 7.4–7.3. 2.2.3. CX11 (40%)

ﬁ-nally washed with toluene and chloroform to remove residual DMF. Anal. Calcd based on the calibrated composi-tion listed inTable 1. (Cz)52.3–(OXD)47.7: C, 69.1; H, 6.3; N,

8.8%. Found: C, 69.2.; H, 6.9; N: 9.7%.1_{H NMR (400 MHz,} Fig. 1. The1 H NMR spectra of CX10–CX01. Table 1 Compositions of CX10–CX01 by1 H NMR. PU Integration Ratio Carbazole at d 7.1–7.2 OXD at d 7.3–7.4 CX10 1 0 CX31 2.009 0.679 2.96/1.00 CX11 1.452 1.305 1.11/1.00 CX13 0.355 0.995 1.00/2.80 CX01 0 1

(4)

DMSO-d6) d 8.6–8.5 (bs, Ha), 8.4–8.2 (bs, Hg), 8.05–7.60 (m,

Hb, Hc, and Hd), 7.4–7.3 (bs, Hf), 7.2–7.1 (bs, He), 4.5–4.4

(bs, N–CH2–), 1.7–0.5 (m, alkyl-protons). The ratio of

1.11/1.00 for the carbazole/oxadiazole units was deter-mined by their integrations at d 7.2–7.1 and 7.4–7.3.

2.2.4. CX13 (60%)

When reaction was completed, the polymer was pre-cipitated by drop-wise addition of the reaction mixture into methanol. The polymer was re-dissolved in DMF and re-precipitated again in MeOH for several times, fol-lowed by precipitation in MeOH/H2O = 1:1. The collected

PU was ﬁnally washed with toluene and chloroform to re-move residual DMF. Anal. Calcd based on the calibrated composition listed inTable 1. (Cz)26.3–(OXD)73.7: C, 66.3;

H, 5.9; N, 10.1% Found: C, 67.8.; H, 6.6; N, 9.9%.1_{H NMR}

(400 MHz, DMSO-d6) 8.7–8.6 (bs, Ha), 8.3–8.2 (bs, Hg),

7.8–7.5 (m, Hb, Hc, and Hd), 7.4–7.3 (bs, Hf), 7.2–7.1 (bs,

He), 4.5–4.4 (bs, N–CH2–), 1.7–0.5 (m, alkyl-protons).

The ratio of 1.00/2.80 for the carbazole/oxadiazole units was determined by their integrations at d 7.2–7.1 and 7.4–7.3.

2.2.5. CX01 (45%)

ﬁ-nally washed with toluene and chloroform to remove residual DMF. Anal. Calcd based on the calibrated composi-tion listed inTable 1. CX01: C, 63.40; H, 5.73; N, 11.38% Found: C, 63.3; H, 6.4; N, 11.3%. 1H NMR (400 MHz, DMSO-d6) d 8.3–8.1 (bs, 4H, Hg), 7.5–7.3 (bs, 4H, Hf), 1.5–

0.7 (m, 18H).

3. Results and discussion

3.1. Synthesis of the monomers and the corresponding PUs

Scheme 1shows the synthetic pathways for 9-butyl-3,6-bis(4-hydroxyphenyl)carbazole (4) and 2,5-bis(4-hydroxy-phenyl)-1,3,4-oxadiazole (5). Carbazole 4 was synthesized through a four-step reaction sequence. Bromination of 9H-carbazole with NBS afforded 3,6-dibromo-9H-9H-carbazole (1) in 92% yields[38]. Butylation was performed by reacting 1-bromobutane with 1 in the presence of phase transfer agent under basic conditions to give 2 in 65% yield[39]. Su-zuki coupling[40]of 2 and 4-MeOC6H4B(OH)2led to 3. The

protective methyl groups were ﬁnally removed by treat-ment with BBr3to give bisphenol 4[41].

Monomer 5 was prepared from condensation of 4-hydroxybenzoic acid hydrazide with 4-4-hydroxybenzoic acid phenyl ester [42]. Phenol, the by-product, was re-moved through distillation. Recrystallization of the crude product from a mixed solution of N,N-dimethylacetamide and deionized water provided 5 in 45% yield.

Condensation polymerization of the isocyanate mono-mer IPDI with 4 and 5 was performed in dried DMF[43– 46] and end-capped with 4-tert-butylphenol to afford

CX10–CX01. The polymers were puriﬁed by precipitation in MeOH for several times before use. Their1_{H NMR}

spec-tra of the aromatic regions were enlarged and shown in

Fig. 1.

CX10 showed 1_{H NMR at d 8.4–8.6, a broad singlet}

which was assigned to Ha. On the other hand, resonance

signals showed up at d 8.0–7.5 were assigned to Hb, Hc,

and Hd. These signals merged to show multiplets with

the integration equal to 8 protons. Protons He, which are

ortho to the electron-donating oxygen atoms, showed up at the relatively up-ﬁeld region of d 7.3–7.1 with the inte-gration equal to 4 protons. The N–CH2– protons showed

up at d 4.4–4.3 while other alkyl protons all appeared at d 1.8–0.7.

On the other hand, CX01 clearly showed two sets of broad1_{H NMR signals. One set appeared at d 8.3–8.1 were}

assigned to Hg, which were relatively down-ﬁeld shifted

due to the the presence of the adjacent electron-withdraw-ing oxadiazole groups. On the other hand, the other set of signals at 7.5–7.3 was assigned to Hfthat were ortho to the

electron-donating oxygen atoms.

Copolymers CX31, CX11, and CX13 showed all proton signals arising from the carbazole and oxadiazole units. Their ratios could be simply determined by using the signal integrations at d 7.3–7.1 and 7.5–7.3 and were summarized inTable 1. The deviation of the NMR ratios from their feed-ing ratios were small. These results were also consistent with the data of elemental analyses.

3.2. Physical properties of CX10–CX01

The Mn, Mw, and PDI values, were determined by

gel-permeation chromatography (GPC) in THF against polysty-rene standard (Table 2). The GPC results were in good agreement with the NMR integrations. Their thermal char-acteristics, including Td and Tg, were collected with the

scanning rate of 20 °C/min.

The Mwof the PUs were all ranged between 9800 and

6400, as for the number average molecular weight (Mn)

values of the polymers were in the range of 6600–4600. Although Mwand Mn were relatively low in comparison

to other PLED polymers, the presence of inter-chain hydro-gen bonding interactions between the urethane groups al-lowed high degree of inter-chain entanglement, leading to good ﬁlm quality from the spin-coating process. This point of view was further proved in the AFM study discussed in latter sections.

The high Tgand Tdof CX10–CX01 reﬂected their high

glassy-state durability and thermal stability. These results

Table 2

Yields, molecular weights, and thermal properties of CX10–CX01. PU Yield (%) Mw Mn PDI Td(°C)a _Tg_(°C) CX10 50 9800 6600 1.48 270 178 CX31 55 6400 4600 1.41 264 167 CX11 40 8400 5800 1.45 257 153 CX13 60 6600 4700 1.42 272 146 CX01 55 7600 5600 1.36 301 135

(5)

were attributed to the inter-chain hydrogen bond interac-tions as well as the high rigidity of the carbazole and oxa-diazole moieties. The Tgof the PUs fell into the region of

135–178 °C and the 5% weight loss of the polymer (Td)

were all above 257 °C. In particular, CX01 showed a notice-ably high Tdof 301 °C.

3.3. Photophysical properties of the PUs

Normalized UV-Vis absorption and photo-luminescent spectra of the PUs in solution and solid thin-film were shown inFigs. 2 and 3. The photophysical data were sum-marized in Table 2. The UV-Vis absorption spectrum of CX10 showed two major absorption bands peaking at 260 and 300 nm, and one weak absorption band at 340 nm. When the weight of the OXD portion in the PU in-creased, the relative absorption intensities at 260 and 340 nm decreased and completely disappeared in the spec-trum of CX01. On the other hand, CX01 showed a major absorption band at 300 nm and two other absorption bands at 240 and 270 nm. No charge-transfer absorption band was observed in the spectra of CX31–CX13, indicat-ing that charge transfer interactions between the carbazole and oxadiazole groups were insignificant. Similar observa-tion was also obtained in the solid-film sample.

The photoluminescence (PL) of the thin ﬁlm of CX10 showed the kmax(PL)at around 400 nm. On the other hand,

the PL of CX01 showed emission peaking at 422 nm. When comparing the spectra of CX10 and CX01, ones would per-ceive that both Cz and OXD ﬂuorophores emitted in a sim-ilar region. In this situation, the Förster type energy transfer between two ﬂuorophores would be poor due to mismatching of their absorption–emission spectra. There-fore, the emission spectra of CX31–CX13 were composed

225 250 275 300 325 350 375 400 0.0 0.2 0.4 0.6 0.8 1.0

In

te

n

s

ity

(a

.u

.)

wavelength (nm)

CX10 CX31 CX11 CX13 CX01

Fig. 2. UV spectra of the PU-polymers in THF: CX10 (j); CX31 (h); CX11 (N); CX13 (4); CX01 (d). 350 400 450 500 550 0.0 0.2 0.4 0.6 0.8 1.0 Normaliz ed intensity (a.u.) CX10 CX31 CX11 CX13 CX01 Wavelength (nm)

Fig. 3. PL spectra of PU co-polymers in solid ﬁlm of CX10 (j); CX31 (h); CX11 (N); CX13 (4); CX01 (d).

Table 3

UV–visible absorption, photoluminescence properties, and the estimated HOMO–LUMO levels of CX10–CX01 in solid ﬁlm.

PU kabsmax(nm) kPLmax(nm) Egap(eV) HOMO (eV) LUMO (eV)

CX10 296 (303)a _{403 (396/0.21)}b _3.45 5.28 1.83 CX31 295 (302) 404 (396/0.18) 3.44 5.28 1.84 CX11 295 (300) 406 (396/0.51) 3.52 5.31 1.79 CX13 295 (306) 406 (397/0.48) 3.53 5.31 1.78 CX01 296 (305) 420 (399/0.40) 3.71 5.54 1.83 a

In the parentheses is noted the solution kabsmax(nm). b

In the parentheses is noted the solution kPLmax(nm) and the relative quantum yield by comparing against coumarin 1 in THF[47]. The concentration of the PU solutions was 105_{g/mL in THF. For the ﬁlm preparation, PU solution (10}2_{g/mL) was spin-coated onto a quartz plate with spin-rate of 2000 rpm} and was further dried with spurt nitrogen.

0 2 4 6 8 10 12 14 16 18 20 22 24 1 10 100 1000 Lu m inance [ c d/ m 2 ] Voltage [V] Ir(2-phq)2acac Ir(btp)2acac

Fig. 4. The BV plots of the red color PLED device when CX31 as phosphorescent host: Ir(btp)2(acac) (d); Ir(2-phq)2acac (j).

(6)

of the emissions from the Cz and OXD units even in solid ﬁlm. It is noteworthy to mention that many red-light emitting phosphorescent dyes would have weak bands at 400–500 nm that were assigned to spin-allowed 1MLCT transitions and weaker absorption tails above 500 nm, which are attributed to spin-forbidden 3_{MLCT and} 3_LC

transitions[48,49]. In the present cases, the Cz–OXD PUs have ﬂuorescent emission peaking at around 400 nm and tailing to 500 nm. The good spectral overlap properties would be beneﬁcial to the energy transfer process from the polymer matrix to the red-light emitters.

3.4. Electrochemical behavior of the PUs

The oxidation potential of CX10–CX01 was determined by cyclic voltammetry (CV) against ferrocene as the inter-nal standard (Table 3). Since the PUs did not show revers-ible waves in their CV and E1/2 could not be accurately

determined, we adopted their onset potentials (Eonsetox ) for

estimation of the HOMO energy level. The HOMO and LUMO levels were calculated according to an empirical formula of EHOMO= e(Eonsetox + 4.8) (eV),[50]and ELUMO=

EHOMO Egin which Egis the optical band gap. The value

400 450 500 550 600 650 700 750

Wavelength [nm]

Intensity [normalized]

10V 12V 14V 15V 20V 10 11 12 13 14 15 16 17 18 19 20 0.30 0.31 0.32 0.33 0.34 0.35 0.36 0.37

Voltage[V]

Color Coordinate(X)

0.30 0.31 0.32 0.33 0.34 0.35 0.36 0.37

Color Coordinate(Y)

a

b

c

Fig. 5. The EL spectrum (top) of the white light emitting PLED device in different voltages, 10 V (j); 12 V (h); 14 V (d); 15 V (s); 20 V(N). The plot shows the shift of the CIE coordinates along with increasing applied electrical voltage (middle). The PLED device shows write emission (bottom).

(7)

of Eg was estimated by using the equation of Eg= 1240/

konset of UV(eV). According to our estimation, their HOMO

energy levels lies between 5.28 and 5.54 eV, which have good matching with that of ITO. The optical gaps of 3.44–3.71 eV for CX10–CX13 were calculated according to their spectral data. The LUMO levels of 1.78–1.84 eV were accordingly derived.

3.5. Surface analysis of the PU ﬁlm on ITO surface

High performance OLED devices usually require having the organic layers deposited onto a smooth ITO surface. Therefore, the roughness control of the ITO surface is essential for improving the OLED performance. Literature reported that spin-coating of a layer of conducting polymer such as PEDOT-PSS onto the ITO surface would signiﬁ-cantly reduce this problem. In the present section, the sur-face conditions of the ITO substrate before and after PU modiﬁcation were compared. The roughness of the surface was monitored by atomic force microscopy (AFM) using a non-contact Mac mode.

Before surface modification, there were many irregular prominences on the ITO surface. The original roughness r.m.s. (Root-mean-square) of ITO glass surface was 0.429 nm. After spin-coating a layer of CX10, with the thickness of around 30 nm on top, the roughness dropped to 0.217 nm. Contrary to the bare surface, the surface be-came more flat after CX10 modification. The height profile with good flatness has been observed.

3.6. The use of CX31 as phosphorescent host in the light emitting layer

Since the PUs have good charge-transport characters, we were interested in applying them as bipolar hosts for PLED. The Cz–OXD PU copolymer could be used in princi-ple as host matrix for phosphorescent light emitting purpose, with balanced hole and electron transport proper-ties [51–53]. To ﬁnd the appropriate component ratio, PHOLED from CX10, CX31, and CX13 have been fabricated and compared. The device structure of ITO/PEDOT-PSS(30 nm)/Ir complex (8%) in PU (30 nm) /BCP(10 nm)/ Alq3(40 nm)/ LiF(1 nm)/ Al(150 nm) was adopted in these

study. However, only CX31 was proved an effective host. When a red phosphorescent dye-dopant of bis(2-ben-zo[b]thiophen-2-yl-pyridine) (acetylacetonate) iridium, denoted as Ir(btp)2(acac), was used, the PHOLED device

turned on at 5.7 V (1 cd/m2_{) with the highest brightness}

of 169 cd/m2 (Fig. 4). The EL spectrum is identical with the PL of Ir(btp)2(acac), with the peak emission at

615 nm. When bis(2-phenylquinoline)(acetylacetonate) iridium(III), denoted as Ir(2-phq)2acac, was used, the

de-vice turned on at 9 V (1.0 cd/m2_{) and reached the}

maxi-mum brightness of 394 cd/m2_{at 21 V with the highest EL}

efﬁciency of 1.0 cd/A.

White phosphorescent light emitting device was com-posed by co-doping bis(3,5-diﬂuoro-2-(2-pyridyl)phenyl-(2-carboxypyridyl)iridium (FIrpic) and Ir(btp)2(acac) in

CX31 with the weight ratio of 2:5:80. The CIE shown was (0.34, 0.34) at 14 V, which is very close to the pure white light CIE of (0.33, 0.33) (Fig. 5). The brightness performance

with the device was 45 cd/m2and the turn on voltage was below 10 V with the maximum efﬁciency of 0.21 cd/A. However, the relative intensities of the blue and red chan-ged slightly when the applied voltage increased.

To further understand whether there were any unusual polymer-Ir complexes interactions occurring in the light emitting matrix, the polymer-Ir complexes blend was subjected to IR analysis.Fig. 6showed the IR spectra of CX31, Ir(bpt)2acac, and their blending. The IR of CX31

showed strong absorptions at 3250–3500 cm-1_{and 1650–}

1700 cm-1_{that were assigned to absorption of the –NH–}

and the C@O stretching modes. No obvious new features were observed after Ir(bpt)2acac (8 wt%) was mixed into

the polymer matrix, indicating that the light emitting layer was composed of a physical blend of the PU and the Ir complexes.

4. Discussion and conclusion

The present work successfully demonstrated the use of PU based carbazole–oxadiazole co-polymers as host for PHOLED applications. To our knowledge, no example has been reported in this research direction. Although PUs are well known engineering polymeric materials with good mechanical properties, applications of PU on PLED are uncommon. Particularly, the applications of PU as the light emitting layer are very rare. This is probably due to the hydrogen bond quenching effects[54]. It has been reported that vibrational deactivation through hydrogen bond inter-actions between the protic environment and the fluoro-phore would facilitate the fluorescence quenching. For example, hydrogen bonding of Nile-Red, an important red-emission fluorescence dyes for laser and OLED applica-tions, with alcohols occurred both in the ground state and the excited states [55]. The fluorescence quantum yield was significantly diminished with increasing hydrogen bond donating power of the medium.

Similar phenomena were reported on the arylnitrile derivatives [56]. The exceptionally efﬁcient ﬂuorescence quenching was observed for 4,-N,N-dimethylaminobenzo-nitrile (DMABN) in polar protic solvents. A mechanistic

4000 3500 3000 2500 2000 1500 1000 0 20 40 60 80 100 120

cm

-1 Ir(bpt)₂acac in CX31 Ir(bpt)₂acac CX31

(8)

explanation based on the CN HO conﬁguration has re-cently been evidenced.

Similar effects were observed for arylamine derivatives. Signiﬁcant quenching of the intramolecular charge sepa-rated state of N,N-dimethylaminophenylacetylene was ob-served in protic solvents, and both the triplet and ﬂuorescence yields were much lower than those in aprotic polar solvents[57]. Again, radiationless relaxation through the vibrational modes of hydrogen bond complexation of ArNMe2 HO has major contributions to the quenching

process.

Since the PU systems formed from diisocyanate and diol contains polar protic –NH(CO)– components and many fluorescent dyes usually contain the hydrogen bond accepting groups such as nitrile, pyridyl, or amino groups, these observations lead to a suspicion on whether PU is an appropriate matrix for the light emitting layer. Ha reported the first attempt on using PU-DCM as the matrix for the red-emission PLED[19,20]. In these cases, the maximum brightness of the devices was 50–60 cd/m2(Red emission). We attributed the low brightness of the PLED devices to the radiationless deactivation of the exciton through of the hydrogen bond interactions. Chien has attempted using PU as the scaffold to tether oligo-PPV units and used them as fluorescent dopant in PVK-PBD for electroluminescent devices. In this situation, when the hydrogen bond array was highly diluted, 500 cd/m2_{of brightness (blue-green)}

could be reached[58]. In our cases, we adopted the carba-zole–oxadiazole PUs as the bipolar host for phosphores-cent OLED (PHOLED) and successfully demonstrated that it could be a good host for red PHOLED (400 cd/m2_{, and}

1 cd/A). The performance is one order of magnitude higher than the ﬂuorescent system. Although radiationless relax-ation would still be one of the major pathways for conver-sion of the triplet exciton back to the ground state, in particularly, for the red light emitters because of the lower energy gap that close to the IR region, our observations suggested that PU would still be one of the candidates for PLED applications.

The present research pioneered the work of using car-bazole–oxadiazole based PUs as the phosphorescent host for PLED applications. The successful use of CX31 as the phosphorescent host for PLED, indicating that PUs can also be used as appropriate materials for the light emitting layer.

Acknowledgment

This work was supported by Ministry of Economic Af-fairs (grant no. 93-EC-17-A-08-S1-0015) and National Sci-ence Council of Taiwan (NSC 95-2113-M002-021-MY3), Academia Sinica (Thematic Project) and Advanced Polymer Nano-technology Research Center.

References

[1] Tang CW, Vanslyke SA. Appl Phys Lett 1987;51:913. [2] Shirota Y, Kageyama H. Chem Rev 2007;107:953.

[3] Burroughes JH, Bradley DDC, Brown AR, Marks RN, Mackay K, Friend RH, et al. Nature 1990;347:539.

[4] Braun D, Heeger AJ. Appl Phys Lett 1991;58:1982.

[5] Lee YZ, Chen X, Chen SA, Wei PK, Fann WS. J Am Chem Soc 2001;123:2296.

[6] Huang TH, Jiann T, Lin JT, Chen LY, Lin YT, Wu CC. Adv Mater 2006;18:602.

[7] Randall D, Lee S, editors. The polyurethanes book. New York: Wiley; 2002.

[8] Li Y, Kang W, Stoffer JO, Chu B. Macromolecules 1994;27:612. [9] Suresh KI, Kishanprasad VS. Ind Eng Chem Res 2005;44:4504. [10] Green RJ, Corneillie S, Davies J, Davies MC, Roberts CJ, Schacht E,

et al. Langmuir 2000;16:2744.

[11] Hasegawa M, Ikawa T, Tsuchimori M, Watanabe O, Kawata Y. Macromolecules 2001;34:7471.

[12] Crenshaw BR, Weder C. Macromolecules 2006;39:9581. [13] Njuguna J, Pielichowski K. J Mater Sci 2006;39:4081.

[14] Wang YZ, Hsu YC, Chou LC, Hsieh KH. J Polym Res 2004;11:127.

[15] Sanjai B, Raghunathan A, Natarajan TS, Rangarajan G. Phys Rev B 1997;55:10734.

[16] Kuo CH, Peng KC, Kuo LC, Yang KH, Lee JH, Leung Mk, et al. Chem Mater 2006;18:4121.

[17] Ku CH, Kuo CH, Chen CY, Leung Mk, Hsieh KH. J Mater Chem 2008;18:1296.

[18] Lin KR, Kuo CH, Kuo LC, Yang KH, Leung Mk, Hsieh KH. Eur Poly J 2007;43:4279.

[19] Lim H, Noh JY, Lee GH, Lee SE, Jeong H, Lee K, et al. Thin Solid Films 2000;363:152.

[20] Jeong H, Zou D, Tsutsui T, Ha CS. Thin Solid Films 2000;363:279. [21] Li M, Daniels ES, Dimonie V, Sudol ED, El-Aasser MS.

Macromolecules 2005;38:4183.

[22] Ishii H, Hayashi N, Ito E, Washizu Y, Sugi K, Kimura Y, et al. Phys Stat Sol (A) 2004;201:1075.

[23] Ishii H, Sugiyama K, Ito E, Seki K. Adv Mater 1999;11:605. [24] Deepa P, Jayakannan M. J Polym Sci Polym Chem 2008;46:5897. [25] Aoki A, Tamagawa Y, Miyashita T. Macromolecules 2002;35:3686. [26] Kimoto A, Cho JS, Higuchi M, Yamamoto K. Macromolecules

2004;37:5531.

[27] Lu J, Tao Y, D’iorio M, Li Y, Ding J, Day M. Macromolecules 2004;37:2442.

[28] Hwang SW, Chen Y. Macromolecules 2002;35:5438.

[29] You Y, Kim SH, Jung HK, Park SY. Macromolecules 2006;39:349. [30] Jiang J, Jiang C, Yang W, Zhen H, Huang F, Cao Y. Macromolecules

2005;38:4072.

[31] Xin Y, Wen GA, Zeng WJ, Zhao L, Zhu XR, Fan QL, et al. Macromolecules 2005;38:6755.

[32] Wu CW, Tsai CM, Lin HC. Macromolecules 2006;39:4298. [33] Zhan X, Liu Y, Wu X, Wang S, Zhu D. Macromolecules 2002;35:2529. [34] Peng Z, Zhang J. Chem Mater 1999;11:1138.

[35] Kim JH, Park JH, Lee H. Chem Mater 2003;15:3414.

[36] Lee DW, Kwon KY, Jin JI, Park Y, Kim YR, Hwang IW. Chem Mater 2001;13:565.

[37] Blochwitz J, Hofmann PM, Leo MK. Synth Met 2002;127:169. [38] Zhu Y, Rabindranath AR, Beyerlein T, Tieke B. Macromolecules

2007;40:6981.

[39] Paliulis O, Ostrauskaite J, Gaidelis V, Jankauskas V, Strohriegl P. Macromol Chem Phys 2003;204:1706.

[40] Pan X, Liu S, Chan HSO, Ng SC. Macromolecules 2005;38:7629. [41] Lau CK, Belanger PC, Dufresne C, Scheigetz J, Therien M, Fitzsimmons

B, et al. J Med Chem 1992;35:1299.

[42] Connell JW, Hergenrother PM, Wolf P. Polymer 1992;33:3507. [43] Mehta R, Dadmun MD. Macromolecules 2006;39:8799. [44] Viswanathan S, Dadmun MD. Macromolecules 2002;35:5049. [45] Liu Z, Wu X, Yang X, Liu D, Jun C, Sun R, et al. Biomacromolecules

2005;6:1713.

[46] Chiang LY, Wang LY, Kuo CS. Macromolecules 1995;28:7574. [47] Rusalov MV, Druzhinin SI, Uzhinov BM. J Fluorescence 2004;14:

193. The quantum yield of coumarin 1 in THF is 0.85. For reference, see.

[48] Xu Y, Guan R, Jiang J, Yang W, Zhen H, Peng J, et al. J Polym Sci Part A: Polym Chem 2007;46(2):453.

[49] Zhao Q, Jiang C, Shi M, Li FY, Yi T, Cao Y, et al. Organometallics 2006;25(15):3631.

[50] Liu MS, Jiang X, Liu S, Herguth P, Jen AKY. Macromolecules 2002;35:3532.

[51] Xia H, Zhu Y, Lu D, Li M, Zhang C, Yang B, et al. J Phys Chem B 2006;110:18718.

[52] Carlson B, Phelan GD, Kaminsky W, Dalton L, Jiang X, Liu S, et al. J Am Chem Soc 2002;124:14162.

[53] Hwang FM, Chen HY, Chen PS, Liu CS, Chi Y, Shu CF, et al. Inorg Chem 2005;44:1344.

(9)

[54] (a) Prasanna de Silva A, Nimal Gunaratne HQ, Gunnlaugsson T, Huxley AJM, McCoy CP, Rademacher JT, et al. Chem Rev 1997;97:1515. For other examples, see;

(b) Martin MM, Bréhéret E. J Phys Chem 1982;86:107; (c) Zhao GJ, Han KL. J Phys Chem A 2007;111:9218.

[55] Nagy K, Biczik L. Magyar Kemiai Folyoirat, Kemiai Kozlemenyek 2004;109–110(1):29–33.

[56] (a) Köhler G, Grabner G, Rotkiewicz K. Chem Phys 1993;173:275; (b) Pál K, Kállay M, Köhler G, Zhang H, Bitter T, Kubinyi M, et al. Chem Phys Chem 2007;8:2627.

[57] Hirata Y, Okada T, Nomoto T. J Phys Chem A 1998;102:6585. [58] Blumstengel S, Sokolik I, Dorsinville R, Voloschenko D, He M,