Ion motion in conjugated polyelectrolyte electron transporting layers.

In organic light-emitting diodes (OLEDs), electrons are injected from the cathode into the π*-band (lowest unoccupied molecular orbital) of an organic semiconductor and holes are injected from the anode into the π-band (highest occupied molecular orbital). If the work function of the cathode is greater than the π*-band or the work function of the anode is less than the π-band, these differences lead to charge injection barriers. Because of these barriers the current is limited to a first approximation by a combination of Fowler-Nordheim tunneling and thermionic emission mechanisms.1 These barriers reduce device power efficiencies by increasing the turn on voltage and creating unbalanced charge injection. This problem provides a substantial obstacle to using OLEDs in solidstate lighting applications. A challenge thus lies in reducing the electron injection barrier. Using low work function metals, such as Ca or Ba, can effectively reduce the barrier, but these metals tend to be environmentally unstable, which lowers device lifetime, and require strict encapsulation processes. OLEDs with multiple layer architectures can display improved efficiencies.2 Electron transport/injection layers (ETLs) reduce the electron injection barrier by a variety of mechanisms, including placing a dipole adjacent to the cathode,3 band bending using doped ETLs to create ohmic contacts,4 or through a series of energetically cascading layers.5 ETLs with high electron mobilities can move the charge recombination region away from the cathode, where excitons can be quenched.6 ETLs can also be used as blocking layers that prevent holes from migrating across the device without recombining, forces the recombination profile away from the cathode, and alters the internal field distribution.7 Polymer LEDs (PLEDs) offer the opportunity of device fabrication using solution methods. However, multilayer fabrication is challenging if all the components display similar solubility characteristics. Depositing a new polymer layer can lead to removal of the underlying layer and/or mixing of the components.8 Conjugated polyelectrolytes are helpful in this context since their charged groups increase their solubility in polar solvents, such as water or methanol. The differences in solubility, compared to neutral conjugated polymers, are advantageous for fabricating the desired multilayer architectures. Additionally, by using conjugated polyelectrolytes as the ETL one can circumvent the restrictions of having to match cathode work function to the semiconductor energy level.9,10 Despite the successful application of conjugated polyelectrolytes as ETLs, the exact mechanism by which electron injection barriers are reduced remains poorly understood. It has been proposed that a permanent dipole between the cathode and the polymer lowers the injection barrier,11 possibly due to self-assembly and alignment of the polymer at the interface.12 However, it has also been shown that the nature of the charge compensating ion also influences the turn-on voltage of the device and the maximum efficiency.13,14 In this contribution we reveal that the reduction in the injection barrier involves a rearrangement of the counterions within the ETL under the applied bias. The device test structures used for our measurements (Scheme 1) were fabricated by spin casting a chlorobenzene solution of poly(9,9-dioctylfluorenyl-2,7-diyl) (PFO) on top of an ITO electrode that was cleaned, UV/ozone treated, and then passivated with 80 nm of poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) (PEDOT). The resulting PFO thickness was approximately 100 nm, as determined by atomic force microscopy (AFM) measurements. After the structure was annealed at 120 °C for 30 min, a layer of poly[9,9′-bis[6′′-(N,N,N-trimethylammonium)hexyl] fluorene-altco-phenylene] (PFN, see Scheme 1 for molecular structure) with tetrakis(imidazolyl)borate (BIm4) counterions was deposited from methanol to serve as the ETL. This counterion was previously shown to give excellent performance in related PLEDs.14 The PFNBIm4 thickness was controlled by varying the spin rate and was measured using profilometry by comparison with a control PEDOT/ PFO layer. Before applying the Al cathode by evaporation at 10-6 Torr, the devices were kept under vacuum at 10-4 Torr for 12 h to remove solvents. Fabrication and testing were carried out inside a N2 atmosphere drybox. Figure 1a shows how the current density (J) develops over time at different applied voltages for a device containing a 10 nm thick PFN-BIm4 layer. A low current is initially observed, which increases up to a limiting value. Faster response times and higher steady state J values are attained when the applied voltage is increased. Defining the response time as the time when J is 50% of its maximum value, the transients in Figure 1a have response times of 0.5 s for 6 V, 2.5 s for 5 V, 6.1 s for 4.5 V, and 20 s for 4 V. The luminance (L) characteristics for these devices are shown in Figure 1b. Similar behavior to the J vs time response is observed: higher brightness and lower response times occur as one increases the applied bias. Lowering electron injection barriers leads to higher current densities and more excitons via charge recombination events. Also shown in Figure 1 are the characteristics of a device without PFN-BIm4 at an applied bias of 5 V, that is, the Al cathode was deposited on the PFO layer. Under these conditions one observes no significant emission and a current density that is over 2 orders of magnitude smaller than that obtained with the ETL layer and invariant after the first measurement is taken (approximately 80 ms). The primary action of PFN-BIm4 is thus to improve electron injection. Scheme 1. Device Structure and Molecular Structure of PFN-BIm4 Published on Web 08/17/2007