Taking OIM Analysis To Heart

Matt Nowell
EBSD Product Manager, EDAX

One of the fun perks of working at EDAX is the easy availability of scientific characterization equipment. I’ve had Scout groups in to look at bugs and Velcro with SEM imaging and have helped science projects along by looking at the chemistry of the lettering on M&M candy and the change in composition of coins as a function of time. A few years ago though I had the opportunity to analyze something very near and dear to my heart, and to my son’s.

My youngest son Finn was a micro-premie, born at only 23 weeks gestation. He was just over 12” long and weighted 770 grams. Obviously his care at this stage required a great deal of medical care and intervention. One of the complications he experienced is termed Patent Ductus Arteriosus (PDA), where a portion of the heart fails to close after birth. For Finn, the treatment was to close the dutus arteriosus surgically using a small clip, and this is where the microstructural characterization story starts to get interesting.

After the surgery, the doctor came in to explain the procedure. He positions a small piece of titanium wire around the DA, and clips it shut to close it. He also had a few of the clips attached to a little gauze pad to show us. They looked pretty much like half of a staple, and in essence that is exactly what they are. I asked him if I could keep the samples, and he agreed. I did not look at the Explanation of Benefits though to know if I was charged for them.

I took a clip, polished it up for OIM work, and put it in the SEM. The first dataset I collected was using ComboScan to image the entire clip. Here is the resulting combined image quality and orientation map (IQ + IPF), covering roughly a 5 mm by 1.5 mm area, with a 2.5 µm step size:

Matt_Fig 1

Satisfied I was getting data, I focused my attention on the bend region, where I expected microstructural changes to occur due to the bending. This dataset covered a 390 µm x 403µm area with a 300 nm step size to better resolve the microstructural detail. The resulting IQ + IPF map looks like this:

This data provides a lot of information as to what is happening in the material during the bending process. When the wire is bent, the inner diameter (right hand side of the image) is put into compression while the outer diameter (left hand side) is put into tension. These localized stress states cause permanent or plastic deformation, which results in a shape change of the material. This plastic deformation can be measured using a Local Orientation Spread (LOS) metric within our OIM Analysis software. The LOS image is shown here:

In the LOS image, the plastic deformation levels are plotted on a thermal scale, where the regions of lower LOS are colored blue, and as LOS values increase the coloring becomes warmer into the yellows and reds. The inner diameter stressed in compression has the greatest LOS field, while the outer diameter in tension also shows some higher LOS values. These higher LOS values would suggest that at least some of the deformation induced during bending is accommodated by the formation of dislocations, which manifest themselves as the small orientation rotations measured by the LOS metric. Click here for more information about LOS and measuring plastic strain with EBSD.

EBSD and OIM is especially useful for characterizing grain boundary structure, and this example is no exception. Here is the misorientation angle distribution for the bend area map:

The red line shows the measured misorientation distribution while the blue line shows the distribution expected for a random set of titanium measurements. Clearly there is some preference for specific grain boundary misorientations. OIM Analysis provides some nice tools to investigate this. First I will plot only the misorientations between 50° and 90°.

The highest peaks can then be selected and colored.

Finally, this color coding can be applied to any of the maps in OIM Analysis. Here I have selected an image quality map for this interactive highlighting:

Here we can see the spatial distribution of these specific grain boundaries, with some correlation with the regions of higher stress and strain. Typically these peaks of misorientations are either evidence of phase transformations or twinning. In this case, it is expected that these are twin boundaries that are introduced into the material during bending.

In the end, I did not learn too much about the material that I did not expect. When you apply enough force to something, my expectation is that it will either bend or break. As a microstructural characterization guy, I am happy to see measurable data showing regions that bent. As a father, I am happy to see a material that works as designed. Once in a while when we are at the hospital for some follow up visit, I keep my eye out for the surgeon, ready to tell him his excellent choice of material and strain rate, feeling confident I have the data to back that up.