With the rising popularity of OLED displays, some specific challenges emerge in mobile and micro-display domains. Existing manufacturing options, such as fine metal mask (FMM) and inkjet printing (IJP), are not compatible with the requirements for upcoming generations of applications, for example, in reaching higher resolutions and pixel densities or in allowing for transparent displays with sufficient image quality. Now, imec demonstrates the viability of photolithography as a prime candidate to overcome the shortcomings of FMM and IJP. As such, it introduces a new process into a domain where hardly anyone expected it would ever have been possible.
Trends in mobile OLED displays
OLED displays are becoming the dominant technology for mobile phones and watches. There is barely a need to further increase the resolution (~500ppi in the state of the art), but the trend is to increase the screen-to-body ratio and remove all visible bezels. Consequently, the sensors that are integrated with the front screen (fingerprint sensor, selfie camera…) need to disappear in the display. For this, technologies are required that allow for display transparency without image degradation.
Pixel size and density become relevant in other domains such as micro-OLED displays that can be used in augmented reality devices, for example. Several types of ‘augmented reality’ glasses are being researched. Real 3D projection, however, remains elusive. A super-high-resolution 2D display equipped with a lenslet array could be a device that produces satisfactory imaging cubes. To achieve this, pixel size needs to be scaled down by an order of magnitude compared to the state of the art— aiming for 3000 to even 6000ppi. Specifically, it would mean bringing the typical ~70µm smartphone pixel size down to ~7µm or even down to the one-micron scale.
Where existing solutions fail and lithography was thought to be impossible
In the current organic and hybrid electronic fabrication, fine metal masks (FMM) are widely used to create red, green, and blue emission-layer (EML) patterns in the OLED stacks. As such, displays with more than 500ppi pixel definition are already being commercialized. Also, +2000ppi directly patterned OLED pixels that avoid the use of fine metal masks are under development. Nonetheless, some drawbacks prevent the technology from moving towards large substrate sizes and higher aperture ratios at high resolution. Take, for example, mask sagging in the case of FMM and shadowing effects in the context of direct printing[ii]. Also, it is not possible to create a hole structure via FMM methods. And the cleaning steps and mask replacements make it hard to reduce this manufacturing route's running cost.
The industry also investigates manufacturing routes that use solution printing processes such as inkjet printing (IJP). AMOLED demonstrators with 150ppi and above are shown by multiple companies[iii]. These approaches are favorable because they are not limited by the substrate size. And there are several research directions in improving the printing resolution, such as aerosol jet or electrostatic jet printing. However, there are still challenges such as layer-thickness uniformity in large-area deposition and the lower reliability than evaporated OLEDs[iv].
For a long time, lithography was not considered a viable alternative. Specifically, because of the complex and harsh processing conditions that were assumed not to be compatible with OLED stacks. Take, for example, the sensitivity of OLED materials to humidity, oxygen, UV radiation, organic solvents, and plasma treatments. These are all elements that are part of a standard lithography flow, and that could cause unwanted OLED degradation.
The imec solution and results
Because of imec’s persistence, partnerships, and in-house expertise, a continued effort of several years resulted in a 365nm i-line lithography process that allows the creation of industry-compatible OLED-pixel stacks. No fundamental limitations are expected regarding substrate size for mobile displays or towards 200mm and 300mm wafer sizes for microdisplays. These achievements result from combined optimizations in the device structure, photoresist, and process environment.
Specifically, imec demonstrated two critical results at the International Displays Workshops (IDW) conference in December 2020[v]. Firstly, the creation of patterned devices with 10µm size and 20µm pitch, showing no degradation in their electroluminescence (EL) spectra and lifetime compared to non-patterned devices. The only degradation measured that occurred after consecutive lithography steps was an increase in drive voltage. Being in the order of 6.6V, this should be lowered to a more acceptable range of 3.8V at 1000nit, which is the FMM benchmark. However, the fact that this drive voltage increases after the first lithography step and remains stable after the second may indicate that degradation is due to the direct exposure of photoresist products on top of the organic semiconductor during the litho process, which gives the researchers valuable insights and a feasible route for further optimization.
A second result was the demonstration of high-resolution hole arrays through functional OLED stacks, an essential enabler for in-display sensing. In a first step, a test structure with a full array of holes was created, resulting in a surface with an aperture ratio of 81% that showed a transparency increase from 20-70% depending on the recipes that were used. In a second stage, holes were created within functional OLED devices themselves. The device characteristics before and after hole opening showed no considerable degradation, except for again a slight increase in drive voltage. However, with only 0.6V increase, this was much lower than in the context of the OLED pixel creation itself.
In terms of industrial relevance, the above results cannot be underestimated. For relaxed scenarios in which only one color is needed and drive voltage is not an issue, the solution can already be considered for industrial implementation. This also applies to applications where you only need single patterning to create a hole array. For full-color mobile applications where drive voltage should be low or for scenarios in which more complex hole structures are needed, further development is required. Yet, considering the current results are the effort of a relatively small group of people and organizations, these developments can move at an accelerated pace in case more industry players tune in. Imec therefore warmly invites actors from the entire value chain (materials, equipment, processes, devices) to actively participate in the further development towards industrial maturity. It took the industry fifty years from the first 5-micron CPU in 1970 to arrive at the 5nm chips in the latest iPhones. Imec, with its partners, reduced the OLED pixel size from 20 to 10 micron in less than five years. 3,000ppi displays are already possible with color-by-white. To enable the 6,000ppi displays and beyond, which the industry is looking for, pixel sizes of around one micron need to be achieved. The further and joint development of photolithography for the domain of OLED displays could make this goal closer to reality than ever expected.
Written by Tung-Huei Ke, R&D project lead at imec.
[ii] C. Hwang et al., “Novel Plane Source FMM Evaporation Techniques for Manufacturing of 2250ppi flexible AMOLEDs,” SID Symp. Dig. Tech. Pap., vol. 49, no. 1, pp. 1003–1006, 2018.
[iii] Z. Wu et al., “Development of 55-in. 8K AMOLED TV based on coplanar oxide thin-film transistors and inkjet printing process,” J. Soc. Inf. Disp., vol. 28, no. 5, pp. 418–427, 2020.
[iv] T. W. Lee et al., “Characteristics of solution-processed small-molecule organic films and light-emitting diodes compared with their vacuum-deposited counterparts,” Adv. Funct. Mater., vol. 19, no. 10, pp. 1625–1630, 2009.
[v] Tung-Huei Ke et al., “Island and Hole Fabrication on OLED Stack for High-Resolution Sensor in Display Application”, IDW2020.