DOI: 10.1002/adma.200702595 On the Limited Operational Lifetime of Light-Emitting Electrochemical Cells** By Thomas Wa˚gberg, P. Ralph Hania, Nathaniel D. Robinson, Joon-Ho Shin, Piotr Matyba, and Ludvig Edman* Emissive devices based on conjugated polymers (CPs) as the active material are of interest for a range of display and solid-state lighting applications, because they are predicted to offer an attractive combination of good device performance and low-cost fabrication on large and flexible surfaces. [1] A major portion of the research efforts in this field has been directed towards the development of polymer light-emitting diodes (pLEDs), which today, via clever device engineering and improved material properties, are able to deliver a relatively impressive performance. [2] The drawbacks of pLEDs, however, are that they are critically dependent on the use of a low-work-function and reactive cathode material (typically Ca) and a very thin CP film (ca. 100 nm thickness) to function properly. An alternative CP-based emissive device is the light- emitting electrochemical cell (LEC), [3] which addresses the disadvantages of pLEDs by allowing for efficient light emission at low applied voltages from devices comprising air-stable electrodes (e.g., Au) and very thick CP-based active layers (up to 1 mm in thickness). [4] Despite these important advantages, LECs have so far only attracted very limited attention, most probably owing to an operational lifetime that currently is far from adequate for most commercial applications. In this context, it is quite surprising that only a small number of publications have directly addressed the issue of the mechan- ism behind the limited operational lifetime of LECs. [5] In this Communication, we pinpoint the spatial position at which device degradation occurs and identify the mechanism behind the chemical degradation reaction. We have employed a planar surface cell configuration with a millimeter-sized interelectrode gap and with the active material – a mixture of poly(2-methoxy,5-(2 0 -ethylhexyloxy)- p-phenylene vinylene) (MEH-PPV), poly(ethylene oxide) (PEO), and LiCF 3 SO 3 – directly exposed to the environment. Figure 1a and b shows chemical structures of the active material components and a schematic of the device config- uration, respectively. The direct access to the active material, together with the fact that it is possible to operate such wide-gap LECs with significant light emission at low voltage, [4] makes this configuration well-suited for studies of the irreversible chemical and/or electrochemical side reactions that take place during long-term operation of LECs. By employing a combination of direct optical probing of device operation, photoluminescence (PL) measurements, and Raman spectroscopy we are able to demonstrate that the main lifetime-limiting process in LECs, somewhat unexpect- edly, occurs in the thin emission zone (the light-emitting p–n junction positioned between the doped regions) and not at the electrode interfaces. In post-mortem devices allowed to relax by short-circuiting the electrodes, we find that a clearly observable ‘‘degradation line’’ has formed at the previous location of the light-emitting p–n junction, and that the PL intensity and the Raman signatures of the MEH-PPV polymer in this region are strongly and irreversibly affected. These results, in combination with a follow-up study on spectroscopic changes in active-material films as a function of photo- excitation intensity and temperature, suggest that the observed irreversible chemical changes in the p–n junction region are due to photoinduced degradation of the vinyl group of the MEH-PPV polymer, and that this photodegradation process is much more severe at elevated temperatures. Figure 2 shows the current (open squares) and the brightness (solid circles) as a function of time for a representative planar Au/MEH-PPVþPEOþLiCF 3 SO 3 /Au surface cell with a 1 mm interelectrode gap during operation at an applied voltage of V ¼ 5V and a temperature of T ¼ 360 K. The high temperature is necessary to allow for significant ionic conductivity of the active material and a related low-voltage operation of the wide-gap device. [4a,6] Note also that all device operation was performed under high-vacuum conditions on extensively dried samples, as specified in the Experimental section. While the current was measured, the wide-gap devices were monitored through the optical window of the vacuum cryostat in order to probe the detailed position and intensity of the light emission. The inset of Figure 2 presents such a photograph. During the initial operation of an LEC, the active material is electrochemically doped p-type next to the anode and n-type next to the cathode. With time, these doped regions grow in COMMUNICATION [*] Prof. L. Edman, Dr. T. Wa ˚gberg, Dr. J.-H. Shin, P. Matyba Department of Physics Umea ˚ University 90187 Umea ˚ (Sweden) E-mail: ludvig.edman@physics.umu.se Dr. P. R. Hania Department of Physical Chemistry Lund University 22100 Lund (Sweden) Dr. N. D. Robinson Linko ¨ping University 60174 Norrko ¨ping (Sweden) [**] This research at Umea ˚ University is supported by Vetenskapsra ˚det, Magn. Bergvalls stiftelse and stiftelsen Lars Hiertas Minne. T.W. thanks the Wenner Gren foundation and Carl Tryggers Foundation for support. 1744 ß 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2008, 20, 1744–1749