Multi quantum well structures in deep blue organic light-emitting diode
This article has been downloaded from IOPscience. Please scroll down to see the full text article.
2009 Europhys. Lett. 85 18002
(http://iopscience.iop.org/0295-5075/85/1/18002)
Download details:
IP Address: 140.116.208.56
The article was downloaded on 09/10/2009 at 07:06
Please note that terms and conditions apply.
The Table of Contents and more related content is available
HOME | SEARCH | PACS & MSC | JOURNALS | ABOUT | CONTACT US
doi:10.1209/0295-5075/85/18002
Multi quantum well structures in deep blue organic light-emitting diode
S. K. Saha1(a), Y. K. Su2(b)and W. L. Lin2
1Department of Materials Science, Indian Association for the Cultivation of Science Jadavpur, Kolkata 700032, India
2Institute of Microelectronics, Department of Electrical Engineering, National Cheng Kung University 1, University Road, Tainan 701, Taiwan, ROC
received 29 September 2008; accepted in final form 21 November 2008 published online 12 January 2009
PACS 85.60.Jb – Light-emitting devices
Abstract – Multi quantum well structures in the deep blue emitted system 2-methyl-9,10-di(1- napthyl)anthracene (α, α-MADN) are fabricated. The device structures are optimized with respect to the layer thickness and number of pairs forming the quantum wells. The Electroluminescence spectra are de-convoluted into 3 peaks with peak positions 432 nm, 458 nm and 500 nm, which are believed to arise due to the recombination of electrons and holes in the MADN, MADN/NPB interface and Alq3, respectively. It is observed that a better confinement of excitonic charges in different layers causes the efficiency to enhance by more than 25% using optimized thickness parameters and 2 pairs of quantum well structures.
Copyright c EPLA, 2009
During the last two decades organic electroluminescent devices have attracted great attention due to their wide application in full-color flat panel display technology and future solid-state lighting application [1–4]. In recent years, much attention has been paid in developing blue organic light-emitting diodes (OLED) with high efficiency, deep blue color and long operational time [5,6]. So far, a few studies have already been done using blue host materials to obtain blue emission but the efficiency is not too high [7]. Therefore, effort has been given to use a guest- host doped emitter system to enhance the efficiency of the blue emission and long operational time [8,9].
Other than host and dopant materials found to have a great impact on the recombination efficiency, the most efficient way to enhance the recombination efficiency in OLED is to balance electrons and holes in the emission layer. A couple of years back it was reported that the efficiency of green OLED could be doubled using a double-layer emitting structure [10]. Very recently, a sequential doping method has been used to enhance quantum efficiency [11]. Due to the balance of injected holes and electrons, efficiency has also been enhanced by incorporating a hole blocking layer [12].
(a)E-mail: [email protected]
(b)E-mail: [email protected]
Although many efforts have been given to enhance recombination efficiency, the field remains totally unex- plored in the area of multi quantum well (MQW) structures in which charge carriers are confined in differ- ent layers. In MQW device structures the recombination of holes and electrons occurs in different layers and efficiency could be improved due to the better balance of electrons and holes than a single-layer device. So far very few works on multi quantum well OLED structures have been reported [13,14]. All of them are Alq3-based green emission and to our knowledge there is no report on multi quantum well structures with deep blue emission.
In this letter, we report on the multi quantum well structure of the deep blue emitting material 2-methyl- 9,10-di(1-napthyl)anthracene, abbreviated asα, α-MADN to enhance the efficiency compared to the conventional MADN-based deep blue OLED.
Several device structures ITO/NPB/MADN/BCP/
Alq3/LiF/Al to optimize the thickness parameters of different layers are fabricated in both the conventional and quantum well structures. Figures 1(a), (b) and (c) give the chemical structure of α, α-MADN, conven- tional (device I) and double quantum well (device II) device structures respectively. In the conventional device structure [ITO/NPB(70 nm)/MADN(25 nm)/BCP(5 nm)/
Alq3(40 nm)/LiF/Al], NPB, MADN, BCP and Alq3 layers are used as hole transport layer, emissive layer,
S. K. Saha et al.
Fig. 1: (a) Chemical structure of (α, α-MADN). (b) Schematic of the MADN-based conventional device structure. (c) Sche- matic of the double quantum well device structure.
hole blocking layer and electron transport layer, respec- tively. In the multi quantum well structure, the first NPB layer is used as hole transport layer and MADN- NPB is used as the quantum well pair with MADN as quantum well and NPB as barrier layer for the electron.
Figures 2(a) and (b) show the EL spectra of multi quantum well and conventional device structures with optimized thickness parameters. Both devices show blue emission at∼ 452 nm.
To optimize thickness parameters of different layers, first of all, we have investigated the thickness variation of the NPB layer used as hole transport layer near the ITO side. Figure 3(a) shows the variation of device performance in the conventional device structure with the thickness
of NPB layers keeping other parameters fixed. It is seen that the NPB layer of thickness 70 nm gives the highest efficiency. In a similar way, fig. 3(b) shows the optimum thickness for the Alq3 layer (40 nm) corresponding to the best performance of the device. From the EL spectra, it is observed that for higher Alq3thickness the emission color shifts to a higher wavelength. Therefore, we have used 40 nm as Alq3 layer thickness, which gives on efficiency of 1.43 cd/A for the emission at 452 nm. We have also investigated the emission without the BCP layer and seen that the efficiency increases to a higher value, 1.64 cd/A, but the emission color changes towards green (500 nm) due to the injection of some holes into the Alq3layer. To limit the hole injection in to the Alq3 layer we have used the BCP layer as hole blocking for restricting the emission in the deep blue range.
Although in the present device structure we have used MADN and Alq3 with emission colors 432 nm [7] and 510 nm [15], respectively, both devices viz. conventional as well as multi quantum well structures emit color at 452 nm with optimized thickness parameters. To explain the origin of this emission color at 452 nm in our devices, we have investigated the conventional device with struc- tures [ITO/NPB(70 nm)/MADN(25 nm)/BCP(10 nm)/
Alq3(40 nm)/LiF/Al and ITO/NPB(70 nm)/MADN(25 nm)/
Alq3(40 nm)/LiF/Al. Figure 3(c) shows the EL spectra for the conventional device with BCP layers of 10 nm, 5 nm and without the BCP layer. From the figure, it is seen that emission colors for 3 devices are at 436 nm, 456 nm and 500 nm, respectively. This is due to the fact that without the BCP layer holes cannot be blocked in the MADN layer and holes are injected into the Alq3 layer; as a result, emission takes place mostly in the Alq3 layer. Using the BCP layer of thickness 10 nm, the emission occurs at 436 nm as it is confined within the MADN layer due to hole blocking at the BCP layer.
Therefore, in the present device, instead of emitting a single color from the MADN layer, multi colors are produced in other layers like Alq3and also at the interface (MADN/NPB) of two layers. To investigate the emission from the MADN/NPB interface, we have de-convoluted the EL spectra by two peaks as shown in fig. 3(d). In this de-convolution process peak positions are obtained at 439 nm and 477 nm, which are unphysical as these are neither from MADN nor from Alq3 layers. The degree of fitting by two peaks is also not good enough.
Therefore, we have de-convoluted the EL spectra with 3 peaks, and the following peak positions at 432 nm, 458 nm and 500 nm are obtained. The first peak (432 nm) arises due to the recombination of electrons and holes in the MADN layer, the second one (458 nm) arises due to exciplex emission [16] with MADN as electron acceptor and NPB as donor at the MADN/NPB interface and the third (500 nm) due to recombination in the Alq3 layer.
For clarity, it is mentioned that in this de-convolution process we have chosen only the number of peaks but not peak positions, which are obtained by the de-convolution
Fig. 2: (a) EL spectra of the two-pairs multi quantum well device. The EL spectra (with peak value at 452 nm) are de-convoluted into 3 peaks with peak positions at 434 nm, 458 nm and 500 nm. (b) EL spectra of the conventional device with peak positions at 432 nm, 457 nm and 498 nm after de-convolution.
process. The remarkable agreement of the estimated peak positions (432 nm, 458 nm and 500 nm) confirms that in the present devices emissions occur from MADN, Alq3and the interface of MADN/NPB.
In the quantum well device structure, to enhance the efficiency by electron confinement in different MADN layers we have investigated the thickness variation of each layer of quantum well pairs (MADN-NPB). Figures 4(a) and (b) show the EL spectra of the quantum well devices with NPB layer thicknesses of 15 nm and 5 nm. Figure 4(c) shows the energy band diagram of the device showing the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) levels for all organic layers. From the band diagram it is seen that the LUMO level of MADN is lower than the NPB LUMO level. Therefore, the MADN layer is used as quantum well where the electrons are confined and the NPB layer is used as barrier. From Fig. 4(a) it is seen that there is a red-shift with increasing voltage for the 15 nm NPB layer thickness at all operating voltages applied in the experiment, whereas no such shift in EL spectra is observed in the case of the 5 nm NPB layer.
As mentioned above, in the present multi quantum well (MADN-NPB) device there are three types of emissions:
one due to the recombination of electrons and holes in the MADN layer, another due to exciplex emission at the interface (MADN/NPB) and the third in the Alq3 layer. Three devices are fabricated with a variable thickness of NPB layers as 15 nm, 10 nm and 5 nm, respectively. The EL spectra for the 15 nm NPB layer
is de-convoluted into three peaks: one corresponding to MADN emission (432 nm), another corresponding to exciplex emission (457 nm) and the third due to Alq3 emission (near 500 nm). It is seen that at lower voltage (7 V) the emission color is at 436 nm and at higher voltages there is a red-shift of 20 nm up to voltage 18 V.
This is due to the fact that at lower voltage, most of the holes are confined within the NPB and the MADN layer to give exciplex and MADN emission and very few holes can cross the BCP layer for Alq3emission. But as voltage increases, more and more holes can cross the BCP layer and give rise to Alq3 emission. Figure 4(d) gives the relative variation of the intensity ratio of two emissions, at 457 nm and 493 nm, with respect to 430 nm. From the figure it is seen that the slope of the Alq3 emission with respect to the MADN emission (432 nm) is larger for the 15 nm NPB layer than for the 5 nm NPB layer. This means that at higher voltages the Alq3 emission increases with respect to the MADN (432 nm) emission; as a result, a red-shift in EL spectra is observed at higher voltages.
For the 5 nm NPB layer thickness, no such red-shift with voltage is observed and the emission color is at 436 nm. In this device also, we have de-convoluted the EL spectra into three peaks and seen that the Alq3 emission is less compared to the other two. This is due to the fact that for a lower thickness of the NPB layer, holes are accumulated (space charge effect) at the NPB layer and forms a barrier to limit the hole injection from the ITO side to the MADN layer. Therefore, the number of holes that reach the Alq3 layer does not depend on
S. K. Saha et al.
(c) (d)
Fig. 3: (a) Relative variation of efficiency with thickness of different NPB layers (used as hole transport layer near the ITO side). (b) Relative variation of efficiency with thickness of different Alq3 layers. (c) EL spectra for 3 devices: device A without the BCP layer, device B with the 5 nm BCP layer and device C with the 10 nm BCP layer. (d) EL spectra of devices shown in fig. 1(a) de-convoluted into two peaks. De-convoluted peak positions are at 439 nm and 477 nm.
the applied voltage, the emission in Alq3 layer remains constant with voltage and no such red-shift with voltage is observed for the 5 nm NPB device.
We have also studied the variation of the MADN layer thickness and seen that 10 nm layer thickness is the optimum for better efficiency. For a MADN layer thickness higher than 10 nm, efficiency increases but the operating voltage also increases to a higher value and, therefore, we limit the MADN layer thickness to 10 nm.
Finally, in the optimized condition, we have investigated the multi quantum well structures with 2 and 4 (MADN- NPB) pairs and compared them with the conventional
device structures. It is seen that the two-pairs device gives the best performance with the efficiency of 1.4 cd/A, an enhancement of more than 25% over the conventional device. The relative enhancement of the efficiency of both the conventional and multi quantum well device is shown in fig. 5. This enhancement of the efficiency in the MQW device is attributed to the excitonic charge confinement in different MADN layers and to the better balance of electrons and holes.
In summary, multi quantum well device structures with the deep blue emitted system 2-methyl-9,10-di(1- napthyl)anthracene (α, α-MADN) have been fabricated.
Fig. 4: (a) Electroluminescence spectra of the device with the 15 nm NPB layer (used as quantum well pair). The spectra show a red-shift of 20 nm with increasing voltage from 7 V to 18 V. (b) Electroluminescence spectra of the device with the 5 nm NPB layer (used as quantum well pair). The spectra show no red-shift with increasing voltage. (c) Energy band diagram of the device showing the HOMO (lower numbers) and LUMO (upper numbers) levels of different organic layers. All values are in eV.
(d) Relative variation of EL intensity of two emissions at 493 nm and 456 nm with respect to 430 nm for devices with 15 nm and 5 nm NPB layers.
Fig. 5: Variation of efficiency with different device structures viz. conventional, two pairs and four pairs.
The emission spectra consist of three colors, at 432 nm, 458 nm and 500 nm, respectively. The device parameters like the thickness of each layer and the number of quantum well pairs are optimized to tune the emission color and to achieve a better performance. It is seen that using two pairs and optimized thickness parameters the multi quantum well device gives an enhancement of efficiency of more than 25% over the conventional device.
REFERENCES
[1] Tang C. W. and Vanslyke S. A., Appl. Phys. Lett.,51 (1987) 913.
[2] Burroughes J. H., Bradley D. D. C., Broun A. R., Marks R. N., Mackay K., Friend R. H., Burn P. L.
and Holmes A. B., Nature (London),347 (1990) 539.
[3] Baldo M. A., Thomsom M. E. and Forrest S. R., Nature (London),403 (2000) 750.
S. K. Saha et al.
[4] Vanslyke S. A., Chen C. H. and Tang C. W., Appl.
Phys. Lett.,69 (1996) 2160.
[5] Arakane T., Funahash M. I., Kuma H., Fukuoka K., Ikada K., Yamamoto H., Moriwaki F. and Hosokawa C., SID Int. Symp. Dig. Tech. Pap.,37 (2006) 37.
[6] Lee M. T., Liao C. H., Tsai C. H. and Chen C. H., Adv. Mater.,17 (2005) 2493.
[7] Ho M. H., Wu Y. S., Wen S. W., Lee M. T., Chen C. H., Kwork K. C., So S. K., Yeung K. T., Cheng Y. K. and Gao Z. Q., Appl. Phys. Lett., 89 (2006) 252903.
[8] Lee M. T., Chen C. H., Liao C. H., Tsai C. H. and Chen C. H., Appl. Phys. Lett.,85 (2004) 3301.
[9] Tao S., Lee C. S., Lee S. T. and Zhang X., Appl. Phys.
Lett.,91 (2007) 013507.
[10] He G., Pfeiffer M., Leo K., Hofmann M., Brinstock J., Pudzich R.and Salbeck J., Appl. Phys.
Lett.,85 (2004) 3911.
[11] Divayana Y. and Sun X. W., Appl. Phys. Lett,90 (2007) 203509.
[12] Okumoto K. and Shirota Y., Chem. Mater.,15 (2003) 699.
[13] Qiu Y., Gao Y., Wei P. and Wang L., Appl. Phys.
Lett.,80 (2002) 2628.
[14] Kim S. H., Jang J., Hong J. M. and Lee J. Y., Appl.
Phys. Lett.,90 (2007) 173501.
[15] Tang C. W., Van Slyke S. A. and Chen C. H., J. Appl.
Phys.,65 (1989) 3610.
[16] Cocchi M., Virgili D., Giro G., Fattori V., Di Marco P., Kalinowski J.and Shirota Y., Appl. Phys.
Lett.,80 (2002) 2401.
