Polishing marks on ITO glass from MRF process
Laser laboratory of Energetic, University of Rochester, Rochester, NY 14623
ITO (Indium-Tin-Oxide)-coated glass is manufactured by sputtering a thin layer of ITO coating on a glass substrate. Due to its special properties such as high electrical conductivity and optical transparency, ITO glass has been used in display applications, microstructure application, flat antennas for mobile communication, etc. For most of these applications, however, the dimensional structure and microstructure, especially the thickness and surface roughness of the ITO thin layer, play a key role in the properties of the resulting device. The manufacture of an ITO coating with a smooth surface is important for producing high quality ITO glass.
Magnetorheological finishing (MRF) technique is a precision finishing technique which can remove surface damage off the optical surface with magnetorheological (MR) fluid and result in a very smooth surface with surface roughness down to only a few nanometers. The MRF technique is a potential technique that can be applied to polish the ITO surface.
The purpose of this project is to use SEM instrument to investigate the surface properties of ITO coating after polished with MRF technique. With the aid of SEM, we expect to observe the microstructure evolution, trail mark and debris resulting from the interaction between the MR fluid and the ITO coating during the polishing process.
We will take MRF spot on the ITO layer of the glass with a spot taking machine (STM) located at LLE/UR. The STM is a prototype machine for MRF with only Z-axis of motion. The polishing process with STM will remove ITO material off the surface and result in a D-shaped spot on the surface, as schematically shown in Fig.1. The ITO coating is very thin, 100nm thick, which will make the spot taking process very challenging. We examined two spots characteristic of the interaction during materials removal. The spotting time of the first spot (Spot #1) is 45s resulting in the depth of deepest penetration (ddp) of 0.066mm in the spot area. The second one (Spot #2) is 150s having the ddp of 0.060mm.
Fig.1 Spot image (spot #1) from Zygo Mark IV: (a) 2-D; (b) 3-D;
ITO glass is supposed to be conductive and doesn¡¯t need a conductive coating. In our experiment, we tried to take image of ITO spot without sputtering a conductive coating but failed to produce any image. Then we coated a thin layer of gold onto the ITO surface using a sputter coater. The sample was sputtered for 30seconds and the thickness of the resulting gold coating is about 30nm. We examined the sample with the gold coating using SEM in the following experiments.
3.3 Scanning electron microscopy and other characterization techniques
We used LEO 982 field emission (FE) scanning electron microscopy (SEM) to obtain secondary electron (SE) image and backscattered electron (BSE) image and X-ray diffraction pattern of the spotted area on ITO coating. Electron interaction modeling was performed for elements of interest. Light microscopy (LM) was performed with Leica DMR light microscope (Nomarski objectives) with a built-in camera model 300 from Leica.
4. Results and discussion
4.1 Light microscopy image
Fig.2 is the LM image of the spot area on the ITO coating. The polishing marks produced during the MRF process due to the interaction between polishing abrasives and ITO coating are clearly displayed on the polished surface.
Fig.2 Light microscope image of the spot area on the ITO coating (spot #1)
4.2 Secondary electron image
Fig.3 shows the topography of the polished ITO coating. All the images were taken with short working distance and under low accelerating voltages. As shown in Fig.3(a), the surface has obvious polishing marks generated during the polishing process due to the interaction between the abrasive particles and ITO layer which corresponds to the results observed from LM image. Additionally, the surface exhibits some particles which might be the hard polishing abrasive particles embedded in the ITO surface. Fig.3(b) magnifies the embedded particles which display an onion-ring like structure possibly caused by the MRF polishing process. Unfortunately, we were not able to further confirm the composition of this embedded particle. Fig.3(c) shows the surface topography of the polished area of spot #2 where the surface exhibits groove-like structure with one component embedded in between the other one.
Fig.3 SEM image of polishing marks within the spot area on ITO surface: (a) polishing marks (spot #1); (b) enlarged view of abrasive particles embedded in ITO surface (spot #1); (c) enlarged view of polishing marks (spot #2)
4.3 Backscattered electron image in a mixed mode
Fig.4 shows the mix mode image of spot area of ITO surface (spot #2). The mix mode, by name, is a method containing both SE and BSE information. From Fig.4, we can see clearly two different phases exist on the ITO surface. One is supposed to be the ITO coating (grey area in Fig.4). The other phase, existing between the grooves left on the ITO surface during polishing (white area in Fig.4), might be the MR fluid leftover and dried on the surface.
Fig.4 Mix mode image of the spot area on ITO surface (spot #2)
4.4 X-ray diffraction pattern
Fig.5 shows the X-ray diffraction pattern of the spot area on the ITO surface (spot #2). Elements such as In and Sn are detected which are the composition of ITO coating on the glass. As we know, the ITO is about only 60nm thick. The accelerated X-ray electron will easily penetrate through the ITO coating and interact with the glass. So, metal elements like Na, Mg and Ca are also observed. Pd is from the conductive coating we generated using the sputter coater.
Fig.5 X-ray diffraction pattern on the ITO surface (spot #2)
4.5 Energy distribution mapping
Fig.6 is the element mapping result of the spot area of the ITO surface (spot #2). The energy dispersive X-ray characterization process not only detected elements of glass including Ca, Na, Si, O, but also iron element (Fe) which is the main element of carbonyl iron particles we used in the MR fluid. The observation of iron element confirms the results in Fig.3(b) that some carbonyl iron particles were left and embedded in the surface of polished ITO coating. We didn¡¯t detect elements of ITO coating such as Sn and In because the incident electrons are of high energy and easily penetrate through the thin ITO coating into the glass layer and not able to produce any information of ITO layer.
Fig.6 Element mapping of the spot area on ITO surface (spot #2): (a) Ca; (b) Fe; (c) Na; (d) O; (e) Si
4.6 Electron interaction modeling
The following is an electron interaction model of a 10kV accelerating voltage electron beam incident on to ITO glass. This model is based on the sample of thin film structure with a thin ITO coating on a glass substrate. The thickness of the ITO coating is 60nm for the modeling. Fig.7(a) is the modeling result for In which is the main element in ITO coating. It is reasonable that the electron interaction modeling of In element mainly stay in the ITO layer. Fig.7(b) is the modeling result for O which exist both in the ITO coating and the glass substrate. As shown in Fig.7(b), the interaction from oxygen electron disperses from the surface of ITO coating down to the glass layer.
Fig. 7 Electron interaction model of element: (a) In; (b) O
With a combination of several characterization techniques associated with SEM, we examined the spot area of the ITO coating using the LEO 982 FE-SEM working with a short distance and under low accelerating voltage. The SE images clearly show the polished surface is characteristic of polishing marks resulting from the interaction between abrasives and ITO coating. Also, the presence of abrasives embedded in the polished spot area is observed from SE images. The mix mode of SE and BSE indicates the existence of two phases in the polished ITO coating. One is supposed to be ITO coating itself; the other phase might be the MR fluid leftover in between the polishing marks. X-ray diffraction detects both the elements of ITO coating and of glass. The element mapping demonstrates the iron element on the ITO surface which further confirms the carbonyl iron particles embedded in the ITO coating during polishing.
I would like to thank Brian McIntyre for his assistance in operating SEM. I am very grateful to have the advice from my advisor Stephen Jacobs. I really appreciate the help from Anne Marino, Henry Romanofsky who are always ready to offer me a hand. My appreciation also goes to the LLE for providing me with the availability of a couple of instruments and facilities.