The Micro Features of Samsung OLED Display Pixels
University of Rochester
Department of Materials Science
The organic light-emitting diode (OLED) is a kind of light-emitting diode (LED). The OLED thin film consists of multi-layer microstructure. Typically, there are two organic-material layers between the cathode and anode layer. An OLED display uses one colored LED for each subpixel. There is no backlight like in LED, since the subpixel itself is a light source. The only thing required is power. Each LED needs to be powered separately. In an OLED display, the LEDs are powered passively using some passive CMOS resistor or something similar. This kind of OLED is also named as PMOLED which is distinguished to Active-Matrix OLED.
There are several techniques used in this project. First, the light microscope was applied to see the basic features of the pixels on the substrate panel. Then, it was necessary to do gold-coat sputtering over the whole section which would be analyed further. And then several modes are employed to see pixels from different aspects. Modes used were Secondary Electron detection, Backscattered Electron detection, Focused Ion Beam milling as well as Energy-dispersive X-ray Spectroscopy.
First, a tiny piece of the OLED part of the Samsung S6 Edge display was cutted out and placed on the glass substrate. Then, the light microscope was applied to ensure the side of pixels being obersed and view rough of the pixels arrangements and shapes. According to the organic components in OLED, gold-coat sputtering is necessary to erase the possible charging artifacts in the future imaging within SEM system. Besides, the tiny cross-section of the parts which are under the sample surface would be focued. Therefore, the conducting coat sputtered onto the sample should be very thin and sputtering time was chosen to be just 30 seconds. After initial sample preparations, the sample on the sample stab was taken into the SEM system. The first techology was utilized in SEM system is FIB milling. In order to analyze directly of the cross-section components under the pixel surface, FIB is the most direct way to provide such preconditions. With the help from Professor Brian, I complete the setting-up process for both tilting angle and writting parameters in the FIB operation interface.I have set two patterns of milling time for several pixels. One pixel has been milled for just two minutes and other pixels have been milled for ten minutes. After achieving the FIB milling to a few different pixels, SE2 detection, In-lens detection as well as BSD detection modes were applied to image the cross-section area in milled pixels. According to the various brightness comparisons of the electron detetctions, I identify a few obvious boundaries between multilayers along the cross-section area. Further, I did EDS analysis of different regions collecting the components infomation. Finally, I have done elemental mapping within EDS software and get a spatial elements distributions of pixel regions.
In Fig 1, It shous two SEM images under SE2 detection mode. They are both the images of the sample, but the left one has the high magnification as well as higher electron accelerating voltage. In both images, you can see clearly that holes milled by FIB using Ba ions. Besides, there are some contaminations in the low-mag image. It may be caused by the some remaining Ba atoms or the gas molecules. Moreover, pixels can be classified by circular shape pixels and square shape pixels. In addition, each pixel has a little tail-like line connecting to it. It might be some conducting wires or somethings to connect each pixel with other subpixel layers to function as expected.
From Fig 2, it exhibits the images of the same cross-section area taken under SE2 mode as well as BSD mode. In the SE2 image on the left, there are several layers piled together under the pixel surface. It shows the topography of the imaged area even if the secondary electron detector is not perpendicular to the sample surface. In another picture on the right, BSD electrons are emitted from the deeper layer of the interaction volume under the sample surface. Under both modes, there are cascade-like shade covering the cross-section region unde rthe pixel. They are ascribed to Ba atoms left here during the FIB milling process. The process is elastically collision and so brighter area means higher atomic-number elements in this area. In other words, we can read out the rough infomation about the elements differences within this imaged region. With regard of the BSD image, I can choose corresponding sections with different brightness to do EDS elemental analysis reasonably afterwards. In Fig 3, It also shows the same comparison between image taken under SE2 mode and under BSD mode, but the focused pixel is in a square shape. It is reasonable to do same things on a different shape pixel to investigate the structure difference and analyse the elements differences in EDS. Interesting, it shows almost the same structure as that of the circular shape pixel in Fig 2. In Fig 4, it shows another square shape pixel being milled with only 2 minutes and a shallow hole was left here. Notably, linar structure are obvios for some kind of usage.
Fig 5 Fig 6
Fig 7 Fig 8
Due to the almost similar compositional infomations of different pixels after EDS peaks comparions, the Energy-dispersive X-ray Spectra of only one circular pixel have been exhibited. According to the various X-ray emission with unique energy because of the different energy gaps between different orbitals in different elements, EDS spectra can reveal the elemental information by identifying the element peaks. In Fig 5, it mainly contains the Ag peaks, Si peak and the analyzed region is the bright part of pixel surface beside the hole. It probably tells that there is a layer of silver over silicon. In Fig 6, the box-like section was focused. The X-ray spectrum tells that it may mainly contain aluminum for connecting or elctrical conducting. In Fig 7, the imaged region is the very thin layer between Ag and Al. A sulfur peak appears into the spectrum. It may be the components of the organic materials in OLED. The very thin boundary-like layer under the aluminum box is focused and the EDS spectrum is displayed in Fig 8. Obviosuly, molybdenum peak comes to the sight. Moreover, I have checked the right-side cross-section area. The conponents are almost the same which means it is just the extension for the multilayers respectively.
In Fig 9, All these images are showing the spatial distributions of various component elements within the focused region. The dwelling time was set to 500 femtoseconds and then collecting the EDS X-ray infomation over the whole map. As infomation shown in all these images, it corresond well with the EDS peaks infomation and previous assumptions of elements distributions.