When a negative isn't a positive

So I think I have posted before about how I have been working on generating X-ray diffraction data and that it was giving me some trouble. I have been more than a little frustrated. For those that do not know already (and who doesn't?), here is a basic XRD lesson:

1. polish your sample
2. insert it into the machine
3. tell the machine what to do
4. cheer victoriously when your data comes pouring off

I've been having a little trouble with that fourth step (and it has nothing to do with my current cold and loss of powerful victorious voice). The XRD works by shooting a fine beam of X-rays at your sample. If your sample is crystalline, like mine, X-rays will be refracted back out. Think of a prizm that takes white light and breaks it out into many colors. Same idea. We then move a camera around an arc to find where those "colors" are and record what we see. Most of the arc is dark because the X-rays are not being refracted evenly. In certain spots, it is very intense. You end up generating peaks like the ones below. It's really very simple. It's one of the most basic tools in materials science and yet I've had trouble getting "clean" data from my material. Look at the difference between the green and purple curve and the orange curve. The orange one is not very useful for accurate analysis. I've recently discovered that part of my problem is due to some misdirection by some people that were helping me here at UF. I've had my fill with the XRD guys up here. They are very smart, but don't think outside of the box at all. My work is very straightforward, but they just don't get it and would not tell me the whole story. They only told me how to prepare samples for our machines, not what the basic requirements for XRD are. In the end, I made my samples to thin and now we can not get a strong enough refracted peak to see anything. So it's back to Orlando next week to, hopefully, finish my data set by using thick specimens (I had been told I had to get them as thin as possible because that's what our machines require, however, if the samples are too thin X-rays will "leak out the back and be useless, lowering your intensity).

If you look at the purple curve up above, I've hand-computer-drawn some smaller peaks underneath it. The smooth line that covers the whole range is the raw data. I then have to process it to produce 4 peaks. Basically there are two spots where X-rays refract out and they are overlapping. Each of these spots has to be further broken down into 2 peaks (for a total of 4) because the radiation that we use is not clean and is a mix of 2 types (very difficult to separate without traveling to a special beam line [particle accelerator] in the Midwest). So, mathematically I have to break down 1 peak into 4 peaks (groups 1a, 1b and 2a, 2b). The center of 1a and the center of 2a are all I care about. Once I have those two numbers I'm done with the graph. I then take those two numbers, work my mathematical magic on it, and I can calculate what is called misfit.

By now you may be wondering about the name of the post (When a negative isn't a positive). It is based on a misconception that I had convinced myself of for months. I was stupid. I don't know how else to say it. Because the two peaks overlap in my material, the misfit is very small (less than 1%). And, strangely enough, I had convinced myself it was NEGATIVE. Which is why I was severely frustrated when my experimental data came back and it was positive. I thought I was back to square one. Thankfully, yesterday I found the error of my ways. I was working with a thermodynamic prediction software package (it can simulate things like misfit). I became frustrated again because the software was predicting positive misfit. Since the software is not the end all-be all of materials I wasn't too worried. Just in case, Dr. Fuchs and I started digging into publications from other groups, including those that I based my erroneous idea on. Turns out the misfit is actually POSITIVE. So everything that I've been doing for the last month or two IS valid. I am on the right track and my data is falling in line. I have been so frustrated lately that it wasn't working that I started to question whether I could graduate or not (my XRD data is that important). I now think I can be done on Tuesday with one more trip to Orlando. That will only leave TEM and I think I have found a solution to my problem there in the name of Ben, a fellow student. Ben is very, very good at making TEM samples and looking at them on a microscope and he is willing to help me. The next few weeks are going to be crazy, but hopefully I am finally approaching the end.

Sorry for the long, technical post. I tried to write it in a way that would make sense. I'm still sick and have been a little slow as a result. Hopefully I can recover before Monday, because it's going to be a busy week.