Lithium manganese oxide as an effective buffer layer between organic and metal layers in organic light-emitting devices

(1)

Lithium manganese oxide as an effective buffer layer between organic and metal layers

in organic light-emitting devices

Tswen-Hsin Liu

Citation: Applied Physics Letters 89, 102101 (2006); doi: 10.1063/1.2339028

View online: http://dx.doi.org/10.1063/1.2339028

View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/89/10?ver=pdfcov Published by the AIP Publishing

Articles you may be interested in

Improved performances of organic light-emitting diodes with metal oxide as anode buffer J. Appl. Phys. 101, 026105 (2007); 10.1063/1.2430511

Effect of magnesium oxide buffer layer on performance of inverted top-emitting organic light-emitting diodes J. Appl. Phys. 100, 064106 (2006); 10.1063/1.2349552

Improving the stability of organic light-emitting devices by using a thin Mg anode buffer layer Appl. Phys. Lett. 89, 103515 (2006); 10.1063/1.2345242

Effects of metal-doped indium-tin-oxide buffer layers in organic light-emitting devices J. Appl. Phys. 99, 114515 (2006); 10.1063/1.2198932

Enhancement of electron injection in inverted top-emitting organic light-emitting diodes using an insulating magnesium oxide buffer layer

Appl. Phys. Lett. 87, 082102 (2005); 10.1063/1.2033129

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 140.113.38.11 On: Thu, 01 May 2014 01:49:14

(2)

Lithium manganese oxide as an effective buffer layer between organic

and metal layers in organic light-emitting devices

Tswen-Hsin Liua兲

Department of Photonics, National Chiao Tung University, Hsinchu 300, Taiwan, Republic of China; Institute of Electro-Optical Engineering, National Chiao Tung University, Hsinchu 300, Taiwan, Republic of China; and AU Optronics Corporation, Hsinchu 300, Taiwan, Republic of China

共Received 6 March 2006; accepted 7 July 2006; published online 5 September 2006兲

Tris共8-hydroxyquinolinato兲aluminum 共Alq3兲-based organic light-emitting devices using a thermally

deposited lithium manganese oxide layer between aluminum 共Al兲 cathode and Alq3 have been

fabricated. The highest luminance efficiency obtained with a 1-nm-thick LiMnxOy layer is very

similar to that of the device with 1-nm-thick LiF. However, the device with an 18 nm LiMnxOylayer

obtained a longer operational stability although the luminance efficiency is lower. The improvements are attributed to lithium extractions of the lithium manganese oxide layer and the interfacial properties between Alq3 and Al are discussed. © 2006 American Institute of Physics.

关DOI:10.1063/1.2339028兴

Organic light-emitting devices 共OLEDs兲 are charge in-jection devices consist of organic thin layers that are essen-tially insulating materials. It is, therefore, important to re-duce barrier heights for the carrier injection at the organic/ electrode interfaces to realize low driving voltages and to balance the hole and electron injections to achieve high quantum efficiency. For improving electron injection, several cathode materials have been developed. Mg:Ag was the first alloy cathode and was introduced by Tang and Van Slyke in 1987.1Later, lithium-based cathodes such as Al:Li alloy2 and a double-layer cathode of Li/ Al 共Ref. 3兲 were developed to reduce the driving voltage. Double-layer cath-odes using a thin cathode interface layer, such as Li2O / Al

共Ref. 4兲 and LiF/Al,5

were also developed and are now widely used. Recently, Kido and Matsumoto demonstrated an effective electron-injection system by using a Li-doped organic layer with a molecular ratio of Li/ Alq3 at unity.6In

OLEDs with a configuration of indium tin oxide 共ITO兲/N,N

⬘

-bis共1-naphthyl兲-N,N

⬘

-diphenyl-1 , 1

⬘

-biphenyl-4 , -biphenyl-4

⬘

diamine 共NPB兲/tris共8-hydroxyquinolinato兲aluminum 共Alq3兲/Li-doped Alq3/ Al, the current density increases

dramatically with increasing the thickness of the doped Alq3 layer.

Lithium manganese oxide, LiMn2O4, is one of the most

prominent materials used as cathode for rechargeable lithium batteries.7 The lattice electronic conductivity of amorphous LiMn2O4 can reach values of 2⫻10−5– 5⫻10−5S cm−1,7,8

which is well described by a hopping conduction model with an activation energy of about 0.16 eV, declaring that the electron exchange between Mn3+ and Mn4+ ions should be responsible for the conductive property.7,9The melting point of commercial LiMn₂O₄powder is around 400 ° C; hence, it is appropriate to form film by thermal evaporation, which is commonly used in OLEDs. In this letter, we report the pos-sibility of employing thermally deposited LiMnxOy as an

electron-injection layer for efficient and stable OLEDs. The devices used in this study have a multilayer struc-ture of ITO/copper phthalocyanine 共CuPc兲/NPB/Alq3,

where CuPc is the injection layer, NPB is the hole-transport layer, and Alq3 is the emission as well as

electron-transport layer. All organic layers were prepared by conven-tional vapor deposition at ambient temperature.10 The thicknesses were 15, 60, and 75 nm, for CuPc, NPB, and Alq3, respectively. A layer of LiMnxOy or LiF was

subse-quently deposited from a tungsten boat. Without breaking vacuum, the top electrode was prepared by sequential depo-sition of a thick Al overlayer using resistive heating. The electrochemical grade 共99.99%兲 LiMn₂O₄ powder in this study was purchased from Aldrich and used without further purification. The thickness of each layer was monitored by a calibrated quartz thickness monitor. In particular, the tooling factor of LiMnxOywas calculated according to the thickness

measured by scanning electron microscope.

The current density and luminance as a function of op-erating voltage for OLEDs that contain the LiMnxOy

thick-nesses共0, 1, 6, 18, 30, and 42 nm兲 or 1 nm LiF are shown in Figs. 1共a兲 and 1共b兲, respectively. It can be seen that by in-serting the LiMnxOy layers between Alq3 and Al both the

current-voltage 共I-V兲 and luminance-voltage 共L-V兲 curves are shifted towards a lower voltage. However, there are no significant differences in I-V and L-V characteristics in de-vices with different LiMnxOylayer thicknesses. As the anode

contact for hole injection in all these devices was the same, this also indicates that the presence of a LiMnxOy layer

be-a兲_{Author to whom correspondence should be addressed; electronic mail:}

[email protected]

FIG. 1. I-V and L-V characteristics of a series of device with different LiMnxOythicknesses and 1 nm LiF.

APPLIED PHYSICS LETTERS 89, 102101共2006兲

(3)

tween Alq3 and Al in Alq3-based OLEDs greatly improves

the electron injection over a wide range of thicknesses. Figure 2 exhibits the plot of luminance efficiency along with the applied current density. It is clear that the device with 1-nm-thick LiMnxOy displayed the highest luminance

efficiency共⬃5 cd/A兲 among the devices, which is approxi-mately five times higher than the LiMnxOy-free device. Gap

states, which act as quenching centers, resulting from chemi-cal bonding between Al and Alq3 at the Al/ Alq3 interface,

have been suspected to be one of the causes behind the poor performance seen in OLEDs based on Al cathode.11–13The results have indicated that a LiMnxOythickness of only 1 nm

is sufficient to remove these states completely by avoiding a direct contact between Al and Alq3.14 However, a further

increase in the LiMnxOy layer thickness may result in a

gradual decrease in efficiency. It is understood that the addi-tion of a LiMnxOy layer between Alq3and Al will alter the

internal electric field distribution leading to a change in both the hole and electron injections. The attenuation in lumi-nance efficiency of the devices due to the presence of LiMnxOy layers between Alq3 and Al might results from a

less balanced charge injection. On the other hand, Fig. 3 shows the optical absorption spectrum of a deposited LiMnxOy film in the wavelength range of 250– 800 nm. A

sizable absorption was observed at visible light regime. The inset of Fig. 3 also shows the electroluminescent spectra of the devices driven at 20 mA/ cm2_{. Obviously, the optical}

length between the reflective cathode and the half mirror ITO for these devices has been changed by inserting different thicknesses of LiMnxOy films. The spectrum shifts to the

longer-wavelength regime indicate that electroluminescence

was enhanced at red wavelengths and was suppressed at green wavelengths. Therefore, the deterioration of luminance in the thicker LiMnxOy domain might also result from the

destructive interference of green light in the half-cavity OLEDs incorporated with the light absorption property of LiMnxOy.

At present, the physical origin of the improved electron-injection characteristics is not certain. Previous studies on the LiF / Al double-layer cathode have revealed that the ther-modynamically allowed reaction, expressed as 3LiF + Al + 3Alq3→AlF3+ 3Li+Alq3−,

15–17

can induce electron injection due to the formation of a thin n-doped Alq₃ layer as in the cases of using Li-doped Alq3 as the electron-injection layer

or using Li as the low-work-function cathode.6,18Similarly, the chemical reactions 关e.g., 3LiMnxOy+ Al+ 3Alq3

→Al共MnxOy兲3+ 3Li+Alq3

−_{兴 at the interface of Al, LiMn}

xOy,

and Alq3 to generate 3Li+Alq3 −

are also possible. However, the results suggest that the high injected current was also achieved at larger LiMnxOythicknesses. It is unlikely that Al,

LiMnxOy, and Alq3will contact each other to undergo

reac-tions considering the coverage of LiMnxOyat the thicknesses

of 18, 30, and 42 nm. Since electrochemical extraction of Li ions in LiMn2O4 cathode occurs at 4 V in rechargeable

lithium batteries,7 therefore, it is more likely due to the lithium extraction of LiMnxOyat LiMnxOy/ Al interface, and

forms a Li–Al alloy at the interface. Nevertheless, when a bilayer cathode of LiMnxOy共18 nm兲/Al 共200 nm兲 is acting

as an electron-injection contact on an Alq3layer, the Li ions

extracted at Alq3/ LiMnxOyinterface can form a thin n-doped

Alq₃ layer and is advantageous for reducing the electron-injection barrier to Alq3 layer.

Figure 4 shows the operational stability driven at 40 mA/ cm2 _{for devices that contain typical LiMn}

xOy

thick-nesses 共0, 1, and 18 nm兲 as well as device with 1 nm LiF. One noteworthy feature in the inset of Fig. 4 is the occur-rence of two distinctly separate time scales in the luminance decay when there is no buffer layer between Al and Alq3,

with an early rapid decay to 50% of the initial luminance at the initial 5 h. This instability may be attributed to a number of factors, including the formation of deep carrier traps in the bulk, interface degradation, and mismatch between the Fermi level of the Al layer and the lowest unoccupied molecular orbital edge of the Alq3 layer.19 However, by adding a

LiMnxOyor LiF buffer layer of a suitable thickness between

Alq₃ layer and Al cathode, reliability can be significantly improved. As mentioned in Fig. 2, the highest luminance FIG. 2. Absorption spectrum of a deposited LiMnxOyfilm taken at room

temperature in the wavelength range of 250– 850 nm. Inset: Normalized electroluminescent spectra of a series of device with different thicknesses of LiMnxOyat 20 mA/ cm2.

FIG. 3. Luminance efficiency vs current density curves of devices made with different LiMnxOythicknesses and 1 nm LiF.

FIG. 4. Normalized luminance as a function of operational time for the devices with 0, 1, and 18 nm LiMnxOyand 1 nm LiF. Inset: The device

without the lithium manganese oxide or LiF layer shows distinctly separate time scales in the luminance decay.

102101-2 Tswen-Hsin Liu Appl. Phys. Lett. 89, 102101共2006兲

(4)

efficiency is obtained with a LiMnxOy thickness of 1 nm,

which is very similar to that of the device with 1-nm-thick LiF. But the device incorporating a 1-nm-thick LiMnxOy

shows a lower operational stability than the conventionally used LiF device. The longer operational stability can be ob-tained by the device with 18-nm-thick LiMnxOy, which

shows a somewhat lower efficiency than the one using 1 -nm-thick LiF. The work of Zhan et al. shows that a sodium stearate buffer layer can improve current injection and device thermal stability of the OLEDs, as a result of the decrease of the interfacial roughness at Alq3/ Al.20Since the electron

in-jection is nearly independent of the LiMnxOythickness over

the range of 1 – 42 nm in Fig. 1, we suspect that the differ-ence in the interfacial stability between Alq₃, LiMnxOy, and

Al owing to morphology change of the thicker LiMnxOymay

very well play a role to enhance the durability. In our atomic force microscopy共AFM兲 studies, an Alq3film with a

thick-ness of 75 nm was evaporated on an ITO-coated glass sub-strate, and then 1-nm-thick or 18-mm-thick LiMnxOy was

deposited on the top. AFM images of the 1-mm-thick LiMnxOy/ Alq3film showed flat surfaces with a mean

rough-ness of approximately 0.85 nm. However, the AFM image of the 18-nm-thick LiMnxOy/ Alq3 film shows a relatively

rough and uneven surface with a mean roughness of approxi-mately 3.02 nm, which cannot add support to the observation of Zhan et al. The improvement indicates that some other mechanisms are operating.

Since the electrochemical properties of LiMn₂O₄ might be affected by surface morphologies which are dependent on preparation methods, to shed more light on understanding the mechanism 共or mechanisms兲 of thermally deposited LiMnxOyfilms resulting in improved performance and

life-time, it is important to know the nature of the interfaces, as well as the deposited LiMnxOy chemical content. In this

re-gard, studies using time-of-flight secondary ion mass spec-troscopy, and angle-resolved high-resolution x-ray photo-electron spectroscopy, of the deposited LiMnxOy film, as

well as Al/insulator contact, are currently in progress. The author would like to thank Shih-Feng Hsu, Chi-Hung Liao, and Chieh-Wei Chen for discussion and their valuable help in this work.

1_{C. W. Tang and S. A. Van Slyke, Appl. Phys. Lett. 51, 913}_共1987兲. 2_{L. S. Hung, Thin Solid Films 47, 363}_共2000兲.

3_{J. Kido, K. Nagai, and Y. Okamoto, IEEE Trans. Electron Devices 40,}

1342共1993兲.

4_{T. Wakimoto, Y. Fukuda, K. Nagayama, A. Yokoi, H. Nakada, and M.}

Tsuchida, IEEE Trans. Electron Devices 44, 1245共1997兲.

5_{L. S. Hung, C. W. Tang, and M. G. Mason, Appl. Phys. Lett. 70, 152}

共1997兲.

6_{J. Kido and T. Matsumoto, Appl. Phys. Lett. 73, 2866}_共1998兲.

7_{Y. Shimakawa, T. Numata, and J. Tabuchi, J. Solid State Chem. 131, 138}

共1997兲.

8_{H. Kawata, M. Nagatab, H. Kageyamac, H. Tukamoto, and A. R. West,}

Electrochim. Acta 45, 315共1999兲.

9_{C. V. Ramana, M. Massot, and C. M. Julien, Surf. Interface Anal. 37, 412}

共2005兲.

10_{S. A. Van Slyke, C. H. Chen, and C. W. Tang, Appl. Phys. Lett. 69, 2160}

共1996兲.

11_{Y. Gao, K. T. Park, and B. R. Hsieh, J. Appl. Phys. 73, 7894}_共1993兲. 12_{Y. Gao, K. T. Park, and B. R. Hsieh, J. Chem. Phys. 97, 6991}_共1992兲. 13_{S. A. Jeglinski, O. Amir, X. Wei, Z. V. Vardeny, J. Shinar, T. Cerkvenik,}

W. Chen, and T. J. Barton, Appl. Phys. Lett. 67, 3960共1995兲.

14_{F. Li, H. Tang, J. Anderegg, and J. Shinar, Appl. Phys. Lett. 70, 1233}

共1997兲.

15_{L. S. Hung, R. Q. Zhang, P. He, and G. Mason, J. Phys. D 35, 103}_共2002兲. 16_{M. G. Mason, C. W. Tang, L. S. Hung, P. Raychaudhuri, J. Madathil, L.}

Yan, Q. T. Le, Y. Gao, S.-T. Lee, L. S. Liao, L. F. Cheng, W. R. Salaneck, D. A. dos Santos, and J. L. Brédas, J. Appl. Phys. 89, 2756共2001兲.

17_{Q. T. Le, L. Yan, Y. Gao, M. G. Mason, D. J. Giesen, and C. W. Tang,}

Appl. Phys. Lett. 87, 375共2000兲.

18_{E. I. Haskal, A. Curioni, P. F. Seidler, and W. Andreoni, Appl. Phys. Lett.} 71, 1151共1997兲.

19_{K. L. Wang, B. Lai, M. Lu, X. Zhou, L. S. Liao, X. M. Ding, X. Y. Hou,}

and S. T. Lee, Thin Solid Films 363, 178共2000兲.

20_{Y. Q. Zhan, Z. H. Xiong, H. Z. Shi, S. T. Zhang, Z. Xu, G. Y. Zhong, J.}

He, J. M. Zhao, Z. J. Wang, E. Obbard, H. J. Ding, X. J. Wang, X. M. Ding, W. Huang, and X. Y. Hou, Appl. Phys. Lett. 83, 1656共2003兲.

102101-3 Tswen-Hsin Liu Appl. Phys. Lett. 89, 102101共2006兲