Development of tandem p-i-n di-chromatic WOLED device

For applications of WOLEDs, it is important that WOLEDs possess high brightness and EL efficiency at a lower current density and stable spectral characteristics in a wide range of injection current. In the past years, many monochrome tandem OLEDs have been reported to be useful for providing high luminance [26,27], the luminance at a fixed current density increases linearly with the number of stacked and independent OLED elements. This can lead to a

Device Voltage (V)

Current Eff.

(cd/A)

Power Eff.

(lm/W)

E. Q. E.

(%) CIE_x,y

Conventional 6.4 9.0 4.4 3.5 (0.32, 0.41)

I 4.4 9.1 7.1 3.0 (0.37, 0.47)

II 3.6 7.4 6.4 2.7 (0.33, 0.43)

III 3.8 9.9 8.2 3.6 (0.32, 0.43)

58

significant improvement in lifetime by reducing the degradation that accompanies the high drive currents required to achieve similarly high brightness in a single-unit device.

In this section, we consolidate both the structural features of p-i-n technique and that of the tandem device concept into one WOLED device to modify the high drive voltage issue of reported tandem WOLEDs [28,29]. The major challenge in tandem OLEDs in general is to prepare the effective connecting layer between emitting units so that the current can smoothly flow through without facing substantial barriers. The organic doped bilayer in this study encompasses an n-doped organic layer and a p-doped organic layer to form a doped organic p-n junction at their contact interface, and offers several advantages, including excellent optical and electrical properties, as well as the ease of fabrication by thermal evaporation.

The architectures of the WOLED devices are shown in Figure 3-11. Device III is the standard p-i-n WOLED unit with a DEML system as we discussed in last section, which gives rise to a balance white emission in a thin thickness of 15 nm. The DEML system comprises one co-dopant emitting layer with MADN:

5% NPB: 3% DSA-Ph: 0.2% Rb and one blue emitting layer of MADN: 5%

NPB: 3% DSA-Ph. Device IV is the tandem p-i-n WOLED with stacking two WOLED units by a bilayer of organic doped thin film with optimized thickness consist of BPhen: 2% Cs₂CO₃/NPB (20 nm): 50 % (v/v%) WO₃ (70 nm). The

n-type doped layer and the p-type doped layer are in contact with each other to

59

Figure 3-11 Device architecture of devices III and IV.

Figure 3-12 shows the transmittance of BPhen: 2% Cs2CO3 (20 nm)/NPB:

50 % (v/v%) WO3 (70 nm) bilayer thin film and the connecting layer is transparent in the visible region from 420 nm to 750 nm, which is essential to achieve an efficient tandem WOLED. As shown in Figure 3-13(a), devices III and IV achieved 2105 cd/m² and 4.5 V, 4780 cd/m² and 9.6 V at 20 mA/cm², respectively. As expected, the luminance and drive voltage increases with the increasing number of active units. Both devices III and IV show near flat current efficiency versus current density response as shown in Figure 3-13(b). Device IV with tandem structure achieved a current efficiency of 23.9 cd/A and an E. Q.

E. of 8.5% at 20 mA/cm², which is about 2.3 times greater than those of device III (10.5 cd/A and 3.9%). The enhanced EL efficiency is attributed to the effectiveness of the conductive p-n junction in electrically connecting two emitting units. Both light emissive units can efficiently produce light under the same current driving. It can also be observed that the power efficiency of device

60

(a) (b)

IV (7.8 lm/W at 20 mA/cm²) is much higher than those of the reported fluorescent tandem WOLEDs [28,29] due to the p-i-n structure of device IV, which can effectively reduce the high drive voltage issue of tandem OLED device.

Figure 3-12 Transmittance spectrum of BPhen: Cs2CO3 (20 nm)/NPB: WO3 (70 nm) thin film.

Figure 3-13 (a) L-J-V and (b) current efficiency and external quantum efficiency

vs current density characteristics of devices III and IV.

Figure 3-14(a) shows the EL spectra of devices III and IV at 20 mA/cm² in the forward direction. With respect to device III, device IV also simultaneously emits a balanced white color and an essentially identical EL spectrum with

61

CIEx,y coordinates of (0.30, 0.43) and no unexpected peak-shift was observed, except the relative intensity of blue emission is slightly reduced and the full width of half maximum of device IV (136 nm) is smaller than that of device III (144 nm). This phenomenon is not expected to result from shifting of the RZ because these devices were all driven at a fixed current. We attribute this slight spectrum change to the optical interference and minor microcavity effect in multilayer devices, which has been reported in tandem OLEDs [30]. As shown in Figure 3-14(b), under different levels of luminance, device IV also reveals a stable EL color with ΔCIEx,y of ±(0.024, 0.030) from (0.303, 0.430) to (0.297, 0.400) at a broad range from 4700 cd/m² to 67800 cd/m² and nearly no current-induced quenching was observed either as depicted in Figure 3-13(b).

Figure 3-14 (a) EL spectra at 20 mA/cm² and (b) CIEx,y coordinates vs luminance characteristics of devices III and IV.

Another important factor of tandem WOLED devices is to obtain high EL efficiency and stable EL color with acceptable angular dependency characteristics. Figure 3-15(a) shows the normalized EL efficiency versus the viewing angle characteristics of tandem device IV. It is generally assumed that

(a) (b)

62

the EL emission pattern from OLEDs is approximately Lambertian [31,32] and it is clear that the angular dependence of device IV is well fitted by a Lambertian distribution. The EL spectra of device IV under different viewing angles of 0°, 30°, and 60° were also shown in Figure 3-15(b). It is noteworthy that the emission shows less angular dependence in device IV as two main peaks of white emission at different viewing angles remain the same. The shifts in CIE x and y coordinates of device IV from the viewing angle of 0° to 60° are only 0.024 and 0.01, respectively. In a strong microcavity effect, OLED devices potentially yield a non-Lambertian emission profile and cause a large angular-dependent color shift. According to our observation of the angular dependence of intensity and color, the microcavity effect is minor in this tandem device which is in agreement with previous report [30].

Figure 3-15 (a) Normalized EL intensity vs viewing angle characteristics of devices III and IV. (b) EL spectra of device IV under viewing angles of 0°, 30°, and 60° off the surface normal.