The current invention relates to electro-optic devices and especially, although not exclusively, to such devices suitable for use in lighting applications.
Electro-optic devices such as organic/polymeric light emitting diodes (PLEDs) have been actively investigated in recent years for display and solid state lighting due to their rapidly improving efficiency and performance1. It has been realized for these devices that the interfaces between the electrodes and emissive semiconductors play important roles in determining their operating characteristics and stability2,3. The current inventors and others have shown that ZnO can provide adequate electron injection into poly(9,9′-dioctylfluorene-co-benzothiadiazole) (F8BT)4-10. The chemical structure of which is shown in
A conventional structure for an emissive polymer LED is to have a low work function metal as the cathode, for instance a Ca/Al bi-layer, and a high work function conducing polymer as the anode, for instance and ITO/PEDTO:PSS bi-layer.2.
It is an object of the current invention to provide an electro-optic device with improved performance and/or with improved manufacturability so as to make the device easier and/or cheaper to manufacture. It is a further non-limiting object of the invention to provide devices with improved robustness over the prior art.
A first aspect of the invention provides an electro optic device comprising a first electrode and a second electrode and an emissive layer located between the first and second electrodes, the emissive layer comprising a polymeric or semiconductor material, e.g. a semiconducting and luminescent material, having a thickness of from 200 nm to 3000 nm.
The device may have a conventional or hybrid structure. Interlayers may be provided between one or other (or both) of the electrodes and the emissive layer.
The device may give luminance efficiencies of greater than 10 Cd/A for green emissive materials for a hybrid structure. The device may give luminance efficiencies of greater than 7cd/A for green emissive materials for a conventional structure.
The device may yield an EQE in excess of 4% for hybrid structures and in excess of 2% for conventional structures.
A second aspect of the invention provides a single layer polymer LED having a first electrode and a second electrode and an emissive layer of F8BT greater than 200 nm thick located between the first and second electrodes, the device comprising one or more interlayers between one or both electrodes and the emissive layer, the LED having a peak luminance efficiency of greater than 7 cd/A at a drive voltage of less than 4V and an emissive layer thickness of 200 nm and a peak luminance efficiency of greater than 15 cd/A at a drive voltage of less than 18.5V with a emissive layer thickness of about 1000 nm.
There is further provided by a third aspect of the invention, a method of forming an electro optic device, the method comprising the steps of providing a cathode on a substrate, providing an interlayer of high electron affinity on the cathode, prior to annealing the interlayer, depositing thereon an emissive layer comprising a polymeric or semiconductor material, to a thickness of from 200 nm to 3000 nm.
A further aspect of the invention provides an electro optic device comprising a first electrode and a second and an emissive layer located between the first and second electrodes the first electrode having a hole injecting material associated therewith and the emissive layer comprising a polymeric or semiconductor material having a thickness of about 1000 nm and a luminance efficiency in excess of 18 cd/A.
In order that the present invention may be more fully understood, it will now be described, by way of example only, with reference to the accompanying drawings, in which:
a is a graph showing the J-V characteristics of the first embodiment;
a is a graph showing the J-V characteristics of the second embodiment;
a is a graph showing the J-V characteristics of the third embodiment;
b is a graph showing the J-V characteristics of a device according to a fourth embodiment of the invention;
a is a graph showing the External Quantum Efficiency (EQE) of the device according to the third embodiment;
b is a graph showing the External Quantum Efficiency (EQE) of the device according to the fourth embodiment;
c is a graph showing the Power Efficiency (PE) of the device according to the third embodiment; and
d is a graph showing the Power Efficiency (PE) of the device according to the fourth embodiment.
Referring first to
In order to understand the injection and transport mechanism of electrons and holes in these LEDs, single carrier diodes were fabricated and tested according to the following protocol.
PEDOT:PSS film (thickness ˜50 nm) was spin-coated on cleaned ITO substrates and annealed at 120° C. under inert atmosphere for 30 minutes. Oxygen plasma treatment was done in order to make hole only and conventional bipolar devices. Compact ZnO layers (˜50 nm) were fabricated by employing spray pyrolysis deposition (SPD) on ITO substrates at 350 C using a Zinc acetate dihydrate (from Fluka) in methanol (80 g/l). Cs2CO3 interlayer was spin-coated on the ZnO layers using ethoxy ethanol as the solvent with a concentration of 5 mg/ml at spin speed of 6000 rpm. F8BT (Mn=97 K) was spin-coated from a p-xylene solution with the different concentration from 5 mg/ml to 60 mg/ml to achieve a wide range of polymer thickness from 31 nm to 2856 nm. We noticed polymer deposition after annealing Cs2CO3 thin film diminish device performance and polymer deposition immediately after deposition of Cs2CO3 enhances device performance. In this specification, we are showing the results for latter processed device fabrication. All samples were annealed at 155° C. under nitrogen atmosphere to improve the morphology of the F8BT. Finally, all these samples were transferred to a thermal evaporation chamber for MoO3 (10 nm) (powder, 99.999% from Testbourne) and Au (50 nm) deposition under high vacuum (1×10−6 mbar). For angular electroluminescence emission studies, devices were cut in such a way so that we could expose a small edge of the pixel (pixel area 3 mm×1.5 mm) away from top metal contact. A multimode optical fiber (core diameter=600 μm) coupled spectrograph (USB 4000: Ocean Optics, Inc) was used to collect emission at a distance of 15 mm from the LEDs at different angles, while keeping the devices stationary. Polymer films with thickness more than 1 m were found to adhere less well on the Cs2CO3 surfaces and peeled off most of the time while cutting the substrates in order to check angular emission pattern. Current density (Keithley 2400 source measurement unit) and brightness (Keithley 2000 multimeter) versus applied voltage (Keithley 2400 sourcemeter) characteristics for the LEDs were measured in air using a calibrated reference Si photodetector.
Hole-only diodes 10 were fabricated by replacing the electron-injecting layer 102, 103 of ZnO/Cs2CO3 of the conventional hybrid 100 as shown in
J-V characteristics for different thicknesses of the F8BT layer 14 show that current densities are bulk limited rather than injection limited as we model below.
In order to describe the bulk hole conduction in F8BT 14, with hole injection from the MoO3 layer 15, we use the Mott-Gurney space-charge-limited-current (SCLC) equation19 combined with field dependent mobility as described in detail in the literature20. We use a simplified form of SCLC enhanced by the Frenkel effect21 as given by the following equation:
Where ∈=∈0∈r permittivity of the polymer, μ0=the hole zero field mobility, d=thickness of polymer film, V is the applied bias (the built-in potential Vbi˜0.5 V subtracted from the applied voltage) and β is the field effect mobility coefficient, which depends upon trap depths in organic semiconductors. A value of β of 9×10−3 cm1/2 V1/2 and zero field mobilities shown in Table 1, were obtained by fitting the J-V curves of devices of different thicknesses.
a also shows fitting of the MoO3 hole injection current fitted with a field independent mobility in the SCLC model (dash line), and with the field dependent term (solid-line), which indicates that the field dependent mobility term is important. The hole mobility compares well with previously reported values for F8BT and β is found to be around one order of magnitude smaller than that reported for electron22, showing that the hole mobility is less influenced by the applied field. When the device is biased to inject the holes from PEDOT:PSS, the hole current is injection limited, this is due to high injection barrier for holes from PEDOT:PSS (work function 5.1 eV) to HOMO level (5.8 eV) of F8BT polymer.
Whilst we do not wish to be limited by any theory, we believe that there are two possible reasons for the exceptionally good hole contact from MoO3 to F8BT:
Electron-only devices 20 were fabricated by altering the prior art hybrid architecture of
We also constructed electron-only devices without the Cs2CO3 interlayer and measured the J-V characteristics, which results are also shown in
According to our electron only device data, the Cs2CO3/F8BT interface has almost identical electron injection properties as the Ca/Al electrode, which indicates favourable interfacial charge transfer process over that reported for a F8BT/CsF/AI interface, where CsF/AI electrodes are observed to have lower electron injection efficiency than Ca/Al electrodes. The electron-only J-V characteristics shown in
In this example. bipolar devices 30 were studied for LEDs characterization. The typical architecture was ITO/ZnO(70 nm)/Cs2CO3 (7 nm)/F8BT(d3 nm)/MoO3 (10 nm)/Au (50 nm). The polymer layer thickness d3 was varied from 100 nm to 3000 nm. A schematic of the device is shown in
For comparison, we also measured conventional architecture PLEDs, employing PEDOT:PSS as the hole-injecting layer and Ca/Al as the electron injecting top metallic electrode and varying the thickness d4 of the polymer layer (e.g. of the following architecture ITO/PEDOT:PSS/F8BT (d4 nm)/Ca/Al devices incorporating F8BT layers of different thickness), the results of which being shown in
In each of
We observe the operating voltages to gradually increase with increasing thickness of the polymer, e.g. F8BT, layer, but the offset between current turn-on and light turn-on is small and the luminance curves follows the current curves. This indicates injected carriers recombine efficiently and the luminous efficiency is weekly dependent on applied voltage. The luminance was collected as forward emission, as shown in
Referring now to
We observe that for conventional devices (e.g. Example 5), with increasing the thickness of the F8BT layer, the peak efficiency operating voltages are increased, which is consistent with the poorer hole-injection and lower electron mobility in the single carrier devices. However, the embodiment of Example 5 also showed improved quantum efficiencies for thicker devices. These devices also give an optimum thickness of the F8BT layer in single layer LED geometry of more than 1 μm.
We observe that on changing the thickness of the polymer (e.g. F8BT) layer, there is no significant change in the electroluminescence (EL) spectrum. This indicates that the recombination zone remains at a similar position in the device (which we consider to be close to the cathodes for the embodiment of Example 4 (and other ‘hybrids’) and close to anode for the embodiment of Example 5 (and other ‘conventional’ structures) and allows us to use the luminous efficiency presented in Table 2 to follow the effects of changing the thickness of the F8BT.
Optical losses studies with different architecture reveals the advantage of recombination zone near to lossless cathode as compare to conventional structures, as shown in Table 3.
External quantum efficiency (EQE) and power efficiency are also measured for hybrid devices, as shown in
For the hybrid bipolar devices, the bi-polar current-voltage characteristics are comparable to the hole-only diodes. This demonstrates that the current in the LED is not limited by the unipolar electron transport. If we consider ohmic injection of holes, with a high current, then these will rapidly drift across the film and build up at the electron injecting interface. This will redistribute the electric field at Cs2CO3/F8BT interface and thus enhance the injection rate for electrons. We therefore expect a large fraction of the electron-hole recombination to be near to the cathode interface. This is desirable since there is very little exciton quenching at the ZnO/Cs2CO3-polymer interface (see Table 3).
For conventional structures, improved efficiencies can be explained by adequate electron injection properties, which redistribute electrical field near to anode interface and reduces injection barrier for holes. In our studies also we found conventional structure bipolar device J-V characteristics closely follow electron only devices and for 1 μm devices we do see factor of 4 higher bipolar current for hybrid devices than conventional bipolar devices (current densities @13 V), which is in consistent with single carrier devices studies. However, the optimum LED thickness is as high as ˜1 μm, indicating that the recombination zone extends over this length scale.
We note that a balanced electron and hole mobilities under the spaced charge regime may result in devices with emissive polymeric layers which are many microns thick. We postulate that the useful range of thicknesses is likely to be in the range of from 200 nm to 4 microns, for example from 200 nm to 3000 nm, say 300 to 1500 nm, and most probably between 600 nm and 1400 nm. Whilst we have exemplified our invention using the well known polymer F8BT it is well within the ambit of the skilled person to substitute the polymer or some of it to provide a different, blended composite or laminate emissive layer. For example, poly((9,9-dioctylfluorene)-2,7-diyl-alt-[4,7-bis(3-hexylthien-5-yl)-2,1,3-benzothiadiazole]-2,2-diyl) (F8TBT) can be used as a total or partial substitute for at least some of the F8BT. Other emissive species may be used.
In summary, MoO3 provides an unprecedented ohmic contact to the high ionization potential polymer F8BT. This facilitates bright and efficient single polymer layer LEDs with peak efficiency of 18.5 Cd/A at thickness of around 1 μm. This is approximately 10 times thicker than standard LEDs. This device geometry is more advantageous in terms of fabrication cost and performance, industrially the ability to make micron thick active layers is extremely attractive since it should significantly increase reproducibility, film conformity and reduce the occurrence of shorts and black-spots. Conventional devices also found to be significantly efficient with thick layer of F8BT, due to high electron current from Ca/Al. We have also demonstrated the role of charge injection efficacy and mobility in making OLEDs with thick emissive layers, which is also advantageous for making optically pumped organic laser and future electrically injectable organic diode lasers.
Number | Date | Country | Kind |
---|---|---|---|
1000453.9 | Jan 2010 | GB | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/GB2011/050038 | 1/12/2011 | WO | 00 | 10/12/2012 |