Chapter 1 Introduction
1.1 Background
New types of flat panel displays were developed during the past decades, such as the liquid crystal display (LCD), plasma panel display (PDP), field emission display (FED), organic electroluminescent display (organic light emitting diodes, OLEDs), etc. Among these, LCD displays has already been in mass production, and is one of the most popular display nowadays. OLEDs, which are self-luminous and do not require backlighting and module seems to be a good candidate for the next generation display.
Advantages of OLEDs over LCDs are:
1. High contrast & brightness over LCDs.
2. Fast response: The response time of OLEDs and LCD are about 10-6 and 10-3 second, respectively. This benefit might improve the blurring problem occurred in LCDs panels
3. Self emitting: No need the usage of backlight system, which significantly reducing the pigments needed to be used.
4. Thin film & robust design: OLEDs are tough enough to be used in portable devices such as cell phones and mp3 players.
5. Viewing angle: Can be viewed up to 170 degrees, OLEDs provide a clear and distinct image, even under direct sunlight.
7. High resolution: High information applications can be achieved by using an active-matrix (AM) OLEDs.
8. Flexibility: Can be produced on metal foil or plastic film.
9. Production advantages: Up to 20% to 50% cheaper than LCD processes since it doesn’t need module.
Nowadays, white organic light-emitting diodes (WOLEDs) had gained particular attention due to their promising applications in various solid-state lighting, including backlights in LCDs, automotive rear lights, airport runway lighting, and illumination light sources [1~5]. Full-color OLED displays incorporating WOLEDs with color filters are also a booming future for WOLEDs. OLEDs “installed” commercialized products are already seen in the market, and the audience all known how brilliant this technology could be. For displays, important commercialized products include:
(1) Mobile phones: Benq-Siemens S88 (2006), Samsung E428 main and sub-displays (2006), Nokia 7900 Prism (2007), Motorola U9
(3)Digital cameras: Kodak L658 (2005) and Samsung high end product NV24HD (2008).
Figure 1-1 Photographs of some OLED displays used in mobile phones (up) Nokia 7900 prism (bottom left) Nokia 8800 Sapphire
Arte (bottom right) Benq-Siemens S88
(4) TV panels: Sony XEL-1 (11-inch TV non-standard 960×540) (2007.12.1).
Figure 1-2 Sony TV XEL-1 [7]
Figure 1-3 Photographs of other OLED displays include Samsung U3, Panasonic SV-SD100, Sony EW-002 and Benq-Siemens S88
(model: Ohgo Suzuka)
For solid state lighting:
No commercialized product is announced so far. However, PictivaTM OLED Displays from OSRAM Opto Semiconductors has a team of
around 50 research engineers working on the development of OLEDs in solid-state lighting. “In the future it will be possible to use OLEDs as flexible or transparent light sources. A transparent OLED over a window in a roof would be able to allow natural light in during the day and
provide fascinating illumination for the room at night”, said Dr. Bernhard Stapp, Head of Solid State Lighting at OSRAM Opto Semiconductors. [8]
Figure 1-4 An OSRAM OLED Table lamp-- The OLED Lighting modules can rotate individually [9]
A glance in the future:
Figure 1-5 A transparent OLED over a window in a roof would be able to allow natural light in during the day and provide fascinating illumination for the room at night [10]
We believe OLED products will be continuously brought in the market and will finally be accepted by the customers.
1.2 Historical development of OLEDs
OLED was invented in 1963, Pope et al. applied bias voltage on single crystal of anthracence and they observed EL (electro-luminance) from 10-20 nm thick crystal of anthracene at voltage above 400V.
However, the operation voltage was too high for practical applications.
Progress in device performance was achieved in 1965, when Helfrich and Schneider [11] yielded electron-hole injection by using electrolytic contacts based on Ce sulphate and Na sulphite. However, to obtain significant light output, these devices had to be driven at voltage as high as 100V and the power efficiency was relatively low. A further step towards feasible organic electro-luminescent devices was made in 1970s by the usage of thin organic films prepared by vacuum vapor deposition or the Langmuir-Blodgett technique instead of single crystals [12-14].
The development of organic multilayer structures considerably improved the efficiency of light-emission by achieving a better balance of the number of charge carriers of opposite sign and further lowered the operating voltage by reducing the mismatch of energy levels between the organic materials and the electrodes. In 1987, Eastman Kodak Company announced a bi-layer organic thin film device, with electron donor and electron acceptor, via vapor deposition process. They developed low-molecular-weight organic light emitting molecular sandwiched in between two electrodes and applied current to emit light. The performance of such device was improved tremendously, and this is the
rudiment model for present OLEDs. A multilayer OLED was introduced by Tang and Van Slyke at Eastman Kodak [15], and reported highly efficient OLEDs by using Alq3. This report inspired a great interest in research and development of organic materials for optical and electronic devices. By using a hetero-layer structure of Alq3, each a few ten nanometers (10-100nm) thick, sandwiched between indium tin oxide (ITO) and Mg:Ag alloy electrodes, they could achieve high power efficiency and acceptable lifetime at low operating voltage. Since then, OLEDs have become an active field of research because of its potential application in flat panel display. In 1990, Burroughes et al. in Cambridge [16] developed other kinds of OLEDs which were based on conjugated polymers, which we call them as PLEDs afterwards while the letter “P”
stands for polymer. Spin coating is the major method for producing these kinds of devices. PLEDs overcomes OLEDs by using only one single emitting layer, which simplify the fabrication process over the later one;
but the issue to find one single material which can satisfy all requirements is difficult. Hence, PLED still doesn’t seem to make its future. Moreover, OLEDs could also be fabricated on flexible substrates such as metal foil, plastic substrate ( PET, PEN, PES….), or even paper [17]. The low cost of materials and fabrication processes are promising factors for the continuing development of OLED.
1.3 Basic Concepts of OLEDs
Positive and negative charge carriers (holes and electrons) need to transmit over organic layers and finally combine at the emission layer in order to produce the light emitting out of the OLED panel. The process
includes injection, transport, capture, and radiative recombination of the carriers. For simplicity, we will draw the spatial variation of the molecular energy levels in a band-like fashion; however, we have to bear in mind that these organic semiconductors are disordered materials without a well-defined band structure.
Figure 1-6 The mechanism of OLED panels--recombination of positive and negative charge carriers inside the organic layers must be
perfectly arranged in order to yield visible light output
The functions of each layer are explained as follows:
1. The anode:
High work function materials are preferred, for non transparent type (top emitting OLEDs), Au and Ag are the best selections, but Ag has a higher reflectivity. For bottom emission type OLEDs, the anode must be transparent for the light coupling side. ITO thus becomes the primary choice in this case. The work function of the ITO depends on its exact
stoichiometry (for ITO = 90% In2O3: 10%SnO2) and on the deposition conditions, and it’s work function is approximately 4.7 eV; further increase of the ITO work function can be achieved by treating the surface with UV-ozone, oxygen or nitrogen plasma before depositing the organic layer to levels near 5.0 eV in order to match the hole injection layer barrier. Although techniques like plasma polymerization of the ITO have been suggested [18-20], further oxidation of the top surface using O2
plasma treatment is the most widely used method to increase the work function of ITO [21, 22].
2. The hole injection (p type) side (HIL):
The hole injection efficiency can be dramatic improved with this layer.
The insertion of a thin conductive layer between the ITO and the emission layer has other advantages such as producing a smooth surface on which the ETL can be deposited (the ITO surface is too rough), and acts as a barrier for ions diffusion from the ITO into the emission layer, producing a large improvement in the device lifetime [23]. For PLEDs, PEDOT/PSS is often used as this layer.
3. The electron injection (n type) side (EIL):
For the cathode side, matching the organic lowest unoccupied molecular orbital (LUMO) level for electron injection requires the use of low work-function metals. The metals with the lowest work functions are the alkali metals [24]. However, their instability made them very difficult to evaporate in their simple metallic form. In this case, Ca and Mg are among the most widely used metals in OLEDs. A simpler and more
common way to reduce the Schottky barrier at the cathode-organic interface is to introduce a thin layer of fluoride, oxide or sulphide based on alkali or alkaline earth metals. Although LiF is the most commonly used material, other materials like CsF [25] and alkaline-earth fluorides [26] have also been successfully used. The efficiency of OLEDs can be improved by up to one order of magnitude when introducing a thin layer of fluoride between the emission layer and the cathode.
4. The emission layer (EL): emission color by using this method is also possible. For such cases, the primary conditions for such energy transfer are the amount of the overlap of the emission spectrum of the host and absorption of the guest.
Therefore, the host material always needs higher energy gap than the dopant material. In the meantime, phosphorescent materials with their intrinsic high efficiency are introduced inside the emission layer, and the EQE(%) which stands for external quantum efficiency of the device is expected to reach 20%.
5. The cathode:
Al is the most commonly used material. For top emitting OLEDs, combined materials such as Al/Ag, Ca/Ag are often selected, with a total
thickness only about 20 nm. This is because Ag is more transparent than other materials.
1.4 Motivation
WOLEDs provide various applications for displays and solid state lighting products that can compete with conventional light technologies.
My experiment focuses on producing WOLEDs devices with higher efficiencies and lighting ability OLED devices. Achieving this will require significant advances in efficiency, good CIE positions, and also acceptable CRI value.
So in order to achieve good performances, we focus on phosphorescent based device structures in order to improve the efficiencies while still remain acceptable CRI. In this thesis, several kinds of structures and devices are exanimated and the pros and cons between each kind of device structures, materials and also its effect on the CRI and %NTSC color gamut are compared.
Chapter 2 White OLEDs
2.1 The Evolution of Lighting Sources
Lighting accounts for a large portion of the energy consumed each year, therefore the new, highly efficient sources of white light sources are urgently required. Today’s incandescent and fluorescent lamps have a limited efficacy of only around 4% (about 12 lm/W) and 25% (90 lm/W) respectively. Conventional technology based on compound semiconductors (i.e. LEDs) has made great progress in solving the problem and has now demonstrated efficacies of 70–100 lm/W. Nichia cooperation already demonstrated a prototype white LED with an power efficiency about 150 lm/W [28]; But, OLEDs also offer a very promising alternative, since it might reach flat, plain and large size uniform illumination. Also, the CRI value of OLEDs is higher than that of most nowadays illumination systems due to the large FWHM1 of each emitting material. Today’s state-of-the-art phosphorescent devices (PHOLEDs) from the Universal Display Corporation (UDC) are able to offer 60 lm/W at 1,000 nits. What’s more, it is anticipated that the efficiency of PHOLED technology will be able to break 100 lm/W by 2010 [27].
1full width at half maximum
Table 2-1 Some major efficiency of light systems
Category
Power efficiency (lm/W)
Incandescent lamp 12.6-17.5 Fluorescent lamp(34W) 50-90 Osram energy saving
lamp
58-80
White LED from Nichia 70~150[28]
WOLED (Osram, UDC) 20[29]~102[30]
HID gas discharging lamps
83-203
2.2 Basic Characteristic for white lighting systems
2.2.1 The CIE diagram
Figure 2-1 The CIE 1931
The CIE 1931 diagram is often used for indexing the exact position
of the device. But this diagram has a major disadvantage, that is, the color difference is not average upon the figure. The green part of the diagram is much larger than the other half side of the picture. So we can hardly pick out the amount of color difference from this diagram.
2.2.2 Color temperature and correlated color temperature (CCT) 1. Definition:
The color temperature of a light source is determined by comparing its chromaticity with a theoretical, heated black-body radiator. The temperature at which the heated black-body radiator matches the color of the light source is that source's color temperature; for a black body source, it is directly related to Planck's law. However, most other light sources, such as fluorescent lamps and our OLED panels do not follow the form of a black-body spectrum (planckian locus), so in this case the correlated color temperature (CCT) is used. The CCT can be defined as the color temperature of a black body radiator which in the perception of the human eye most closely matches the light from the lamp. In Figure 2-2, we can estimate the nearly color temperature from the line projected out from the planckian locus, but for CIE positions that deviate this line too much, the CCT is meanless.
Figure 2-2 The correlated color temperature (CCT) [31]
2 . In the Space:
For the same case in the space, stars are also defined by its temperature and the color of light it emits. This is called the Stellar classification. [32, 33]
Figure 2-3 The Stellar classification
Figure 2-4 Stellar Spectral Types OBAFGKM [33]
Figure 2-5 Scorpius--Antares 3,500K [34, 36, 37]
Figure 2-6 Cygnus--Albireo (4,300K/12,100K) [34-36]
Extremely high temperatures stars will show a bluish color with temperature over 30,000 K and is classified as O, only a small amount of stars lie in this category. Most of the stars are classified as A. Our neighborhood sun which is only 1 AU (149,597,870,691 + 30 meters) away emits a yellowish-white color having surface temperatures of about 5,778 K. But while the sun crosses the sky, it may appear to be red, orange, yellow or white depending on its position. The changing color of the sun over the course of the day is mainly a result of refraction and scattering of light, which is unrelated to black body radiation. Red giants (dying stars) have relatively low surface temperatures of about 3,600 K, but they also have a high luminosity due to their large exterior surface
area.
3. Solid state lighting:
Pulling the scope back to the solid state lighting, CCT only represent the feeling of the lighting source-- blue is the "hotter" color, while red is actually the "cooler" color. The colors of 5000K and 6500K black bodies are close to the colors of the standard illuminant called D65 (Daylight, T=6500K).
2.2.3 Color Rendering Index (CRI)/ color rendition index
In the calculation of the CRI, the color appearance of 8 reflective samples is simulated when illuminated by a reference source and the test source. The reference source is a Planckian radiator (if below 5000 K) or a CIE Daylight source (if at or above 5000 K), matched to the correlated color temperature (CCT) of the test source. After accounting for chromatic adaptation with a Von Kries correction, the difference in color appearance ∆Ei for each sample between the test and reference light sources is computed in CIE 1964 W*U*V* uniform color space. The special color rendering index (Ri ) is calculated for each reflective sample by:
Ri = 100 – 4.6△Ei
The general color rendering index (Ra ) is simply the average of Ri for the first eight samples, all of which have low to moderate chromatic saturation:
A perfect score of 100 represents no color differences in any of the eight samples under the test and reference sources [31].
Figure 2-7 The measuring method for CRI
Figure 2-8 The 8 or extended CRI samples
Figure 2-9 The reflectivity’s of the 8 color samples [31]
2.2.4 Eye Sensitivities and Efficiencies
Figure 2-10 The eye sensitivities [31]
The eye shows maximum sensitivity near 555 nm (for brightness over 3 nits), which shows a greenish-yellow color, but for brightness lower than 0.003 nits, the maximum sensitivity point shifts to 505 nm. So a light system composing of a green or yellow element might raise the efficiency due to the higher of the eye sensitivity with these colors.
2.3 Different methods for achieving white OLED
2.3.1 Multi Emissive layer structure (MEML)
White light emission can be obtained from multiple-emissive-layer (MEML) structures, in which different layers emit in different regions of the visible spectrum. Compared with other white OLED structures, such as the multi-doped emissive layer, the MEML is less complicated in fabrication process and more flexible in color tuning because not only doping concentration but also layer thickness can be used to modulate the device chromaticity. As a result, a merit of white OLEDs with a MEML is that its chromaticity can be tuned just by varying the thicknesses of the EMLs without changing the concentrations of phosphorescent dopants from their optimal values. Therefore, considering that the efficiency of doped OLEDs is sensitive to its concentration, white OLEDs with a MEML should have a higher efficiency than those with a multi-doped EML, in which at least one dopant is far away from its optimal concentration to achieve white emission. This is also the most common way to produce white light in the meantime—most of the reports use a bi- or tri- chromatic structures in order to produce white light. For a
bi-chromatic fluorescent WOLED, a yellow plus blue or magenta plus sky blue emitting layer is often seen. These structures demonstrated high efficiency (10-17 cd/A) and excellent stability at a reasonable voltage [38]. For the case of using phosphorescent materials, due to the lack of an efficient yellow emitting material, the multi emissive layer structure is often made into three layers composing of red, green and skyblue materials, but these kind of structure produces white color near the illumination point A [39]. The color is different from the white point, but might seem warmer for most illumination places. Some recent research is listed in the following table.
Table 2-2 Multi-emission layers publications
Device Structure EL performances CIE 1931 ref.
2002
CBP:FIrpic(6%)(B)/CBP:Btp2Ir(acac)(8%)(R) 11 cd/A(Max) 6.4 lm/W(Max) (0.37,0.40) @10 mA/cm2 40
2006
Ir(ppy)3(G)/Ir(ppq)2(acac)(R)/Firpic(B) 9.9 lm/W (0.44, 0.43)@1000nits 39
2007
mCP:Firpic(B)/CBP:Ir(piq)2(acac)R/mCP:Firpic(B) 10.8 cd/A EQE=7.6% (5000 nits) (0.33, 0.35) 41
[Firpic(B) /TL/(btp)2Ir(acac)(R)/ 17.2 cd/A@10834 nits (0.38, 0.42) 43
Ir(flpy)3(Y)/Ir(ppy)3(G)]
TCTA:PQIr(R)/mCP:Ir(ppy)3(G)/UGH2:FIr6(B) 32 lm/W EQE=16.6% (0.38, 0.39) 42
NPB(B)/Ir(Y):Ir(R):CBP/CBP/Ir(G):CBP 21.6cd/A 10.3 lm/W (0.39, 0.41) 44
CBP:Ir(ppy)3:Rb/CBP:Firpic/CBP:Ir(piq)2(acac) 27 cd/A EQE=12.4% (100 nits) (0.40, 0.42) 47
2008
4p-NPD:Ir(MDQ)2(Acac)/TPBI:Ir(ppy)3 31.6 lm/W EQE=15.2% (1000 nits) (0.49, 0.41) 45 CBP:Btp2Ir(acac)/CBP:Ir(ppy)3/CBP:DPAVBi/ 10.0 lm/W EQE=8.5% (100 nits) (0.32, 0.46) 46
So, in short we conclude some pros and cons of this structure:
The structure is more flexible in color tuning because not only the structure can be controlled by inserting blocking layers, but also each layers’ doping concentration and layer thickness. Some disadvantages include it needs relatively high operating voltage due to combined thicknesses of the many layers. Although BPhen, F4-TCNQ which are electron & hole injection layer materials (p-i-n device), may assist in reducing the voltage, but leads to deteriorate of the lifetime. In the meantime, some improvement trying to combine fluorescent and phosphorescent materials in order to produce white color is also introduced [46], because that the host material CBP or blocking layer BCP used in phosphorescent structures both have high HOMO values over 6.0 eV, leading to higher voltage issue. So, by using a blue fluorescent material, the voltage will drop and also reach a more saturated blue emission.
Di-chromatic Tri-Chromatic Figure 2-11 Multi-emission layer structure
2.3.2 Single Emission Layer Structure (Multiply doped emission layer) Single emission layer structure means that it’s emission layer is only composed of one layer. To ensure that all emission originates from a single thin layer, several dopants required to produce white light emission are mixed into a single host layer [48].
Table 2-3 Single emission layer publications
Device Structure EL performances CIE 1931 ref.
Device Structure EL performances CIE 1931 ref.