Characterizing Metal Oxide Coated Carbonyl Iron (CI) Particles
using Microscopic Techniques

Sivan Salzman
Scanning Electron Microscopy Practicum MSC 507
University of Rochester, Spring 2009

In this project we describe the process of characterizing coated and uncoated CI particles using the scanning electron microscopy (SEM) and other microscopic techniques acquired in the MSC 507 course. The CI particles are an integral part of the Magnetorheological finishing (MRF) process carried at the Laboratory for Laser Energetics (LLE) at the University of Rochester. The results shown here help us to get better understanding of the modification process of the CI particles and their behavior.

Magnetorheological finishing (MRF) is a deterministic polishing process developed at the University of Rochester for polishing optical materials [1]. The magnetorheological (MR) fluid consists of ~80 wt% carbonyl iron (CI) particles. New MR fluid based on modified CI is studied these days at the LLE. Here we characterize modified CI particles coated with a thin layer of metal oxide (XyO2) using microscopic techniques (The specific description of the metal oxide used is not revealed here due to sensitivity of patent manners). Fresh CI and coated CI particles were viewed in the SEM and were characterized using X-ray technique. The results show the change in the CI particles shape and texture after the modification process. As part of this course requirement, other techniques were used, such as stereo pairs (3D) image and colorization of an image.

Materials and methods
Two types of CI particles were tested in this project: 1) Fresh CI, and 2) Metal oxide coated CI. All the samples have a powder form and they were milled with mortar and pastel before attached to the SEM stub. In addition, two other samples were prepared in a form of puck by mixing CI particles powder with epoxy powder under controlled pressure and temperature. The pucks were polished using the MRF machine in order to cut the particles and obtain their cross section. The samples were prepared and examined using six techniques described in the results and discussion paragraph.


Results and discussion
This paragraph describes the techniques used to characterize the coated and uncoated particles. The differences between the two types of CI particles are shown in the attached figures and disused as well.

1. Sample coating

All the samples, except of the fresh CI powder sample, were sputtered with a thin layer of gold/palladium alloy on the surface to improve their conductivity. This process allows us to reduce / prevent charging on the sample surface and improve the signal and the quality of the images taken on the SEM. The thickness of the gold/palladium conductive layer is about 20 (the samples were sputtered for 20 seconds using a current of 20mA).


2. Secondary electrons

The following images were taken using the secondary electron detector (SED). There are two kinds of SE detectors: SE2 and InLens. The images shown in Figures 1-3 were taken using the InLens detector (will be explained in the next paragraph). The accelerating voltage and the working distance are given at the bottom of the each image. Figure 1 shows two SEM images of fresh CI particles. Figure 2 shows two SEM images of Metal oxide coated CI particles. Figure 3 shows SEM images of the cross section of the particles that was obtained from the epoxy pucks after they were polished with the MRF machine. From the first two figures (1 and 2) we can see the change in the CI particles texture and shap. The surface of the particles in Figure 2 is rougher due to the metal oxide coating layer. Figure 3 presents two images of particles cross section. From this figure we see that the coating layer appears to be uniform, and that the particles diameter increased due to the modification process. This suggests that the coating process is accurate. in addition, the scratching and grooving marks on the cross section area of the particles are due to the polishing process of the pucks. These marks are the MR fluid signature on the surface.

Figure 1: SEM image of fresh CI particles using the InLens detector. (a) Magnification: X 29.17K; Accelerating voltage: 5KV; Working distance: 7.9mm. (b) Magnification: X 183K; Accelerating voltage: 15KV; Working distance: 4mm.


Figure 2: SEM image of Metal oxide coated CI particles using the InLens detector. (a) Magnification: X 17.40K; Accelerating voltage: 10KV; Working distance: 5.5mm. (b) Magnification: X 20.58K; Accelerating voltage: 10KV; Working distance: 4.4mm.

Figure 3: SEM image of the cross section obtain from the polished pucks. (a) Cross section of fresh CI particle. Magnification: X 53.51K; Accelerating voltage: 10KV; Working distance: 3.9mm. (b) Cross section of Metal oxide coated CI particles. The metal oxide layer is obtained here as amorphous dark halo around the original CI particles. Magnification: X 31.16K; Accelerating voltage: 10KV; Working distance: 3.9mm.


3. Back scattered electrons

One image of the Metal oxide coated particles was taken with the back scattered electron detector (BSD). Because the BS electrons are emitted from the inner part of the sample and not from its surface it is difficult to obtain the surface texture of the CI particles using these electrons signal. We present in Figure 4 a set of three images taken with the BS, SE2 and InLens detectors. This presentation allows us to judge the three detectors in our settings. By using the SE2 detector we see details that we barely see in the image taken with the BS detector (for example: the pattern of the metal oxide layer). By using the InLens detector we see more details more clearly than how we see them in the image taken with the SE2 detector. It seems that the most informative image for our purposes is the image taken with the InLens detector. Therefore, in this project most of the images were taken with the InLens detector in a short working distance.




Figure 4: SEM images of the metal oxide coated CI particles taken with different detectors: (a) BSD, (b) SED-SE2, and (c) SED-InLens. All images were taken under the same conditions: Magnification: X 6.62K; Accelerating voltage: 20KV; Working distance: 10.8mm.



4. X ray diffraction pattern

The X-ray diffraction for both fresh and coated (w/metal oxide) particles is shown in Figure 5 (a) and (b), respectively. The first image shows that the fresh CI sample consists of: iron (Fe), carbon (C) and aluminum (Al). The Fe element is the sample composition, while the C and the Al elements are from the carbon tape and the SEM aluminum stub. The second image (Fig. 5(b)) shows the coated CI sample that consists of: iron (Fe), carbon (C), aluminum (Al), oxygen (O) and a metal component (Xy). The additional elements in this case are the oxygen and the metal composites that appear due to the coating layer: XyO2.



Figure 5: X-ray diffraction pattern of: (a) Fresh CI sample, and (b) Metal oxide coated CI sample.



5. Stereo pairs - 3D image

One of the techniques studied in the class was the stereo pairs. This technique integrates two images that were taken in different tilting angles of the stage: image 1 with tilting angle of 0 , and image 2 with tilting angle of 4 . Then, by using Photoshop, one of the images was colored in red and the other one was colored in cyan. The combination of these two images with the different angle and color causes the 3D effect. Two examples are shown in Figure 6 (a) and (b).

Figure 6: Stereo pairs image of the fresh CI sample.




6. Colorized images

Another technique studied in the class was image colorization. Figure 7 and 8 show images that were colorized using Photoshop.


Figure 7 : Colorized image of the fresh CI sample.

Figure 8: (a) Colorized image of the metal oxide coated CI sample, (b) the original image.



The techniques used in this project provided us information about the CI particles shape and texture before and after the metal oxide coating process. Better understanding of the modification process was achieved thanks to the SEM images and the X-ray analysis. In addition, working on this project improved my microscopic skills and experience.

[1] S. D. Jacobs et al., Magnetorheological finishing: A deterministic process for optics manufacturing, in Optical Fabrication and Testing (SPIE, Tokyo, Japan, 1995), Vol. 2576, pp. 372382.

<#form method=POST action="">

Please enter any comments, criticisms, questions, etc. below.

Your name:

Email address: