Hi Everybody!
This is Ely Bordt and Josh Biales! This semester our objective was to determine mineral compositions of Sichuan Basin clay samples collected in 2015. We are hoping to see if these minerals are consistent with those found in prehistoric lake beds. In order to determine these compositions, we performed X-Ray Diffraction Analysis on 10 samples to obtain diffractograms of various peaks that may provide insight as to what the mineral makeup of these clays are.

Having ran our samples in the XRD which produces these diffractograms, we then had to somehow decipher what compositions the peaks indicate. To do this, we analysed each diffractogram using PDXL, a software that produces a list of prospective minerals that our clays could contain, ranking them from best match to worst. An example of a diffractogram that was run through PDXL can be seen in figure 1.

Figure 1: A diffractogram for sample YJ-011 that was analysed using PDXL.
The top 3 best-fit minerals are listed in the right-middle box.

PDXL produces a list of  the best fit minerals that is likely to make up our sample: SiO2 (quartz low), LiD (Lithium Deuteride), and Fe6Si4O10(OH)8 (Greenalite). Because we cannot say that PDXL is 100% accurate, we run the diffractogram again using another program called Match!* 3. By running the plot through this second program, we can double check if PDXL is producing accurate results. The output from our Match!* 3 analysis can be seen in figure 2. Notice that the top best fit mineral is once again SiO2 (quartz low).

Figure 2 shows the best-fit minerals for YJ-011.
The percent match is displayed on the right of the bottom panel in green. 

Having run our sample through both PDXL and Match!* 3, we can use the Open Crystallography Database to obtain .CIF files for our best fit mineral. CIF files are ideal diffractograms for pure minerals. We can search for our minerals (for example, quartz low) in the Open Crystallography Database, obtain its CIF, and overlay it on top of our sample's diffractogram. If the minerals produced by PDXL and Match!* 3 are in fact in our clay, then their CIF should lie neatly on top of our diffractogram.

The program used to overlay our minerals' CIFs is MAUD. MAUD is able to take our minerals' diffractograms, adjust them on top of our own sample, and depending on how much adjustment was needed, produce percent compositions for our sample. After importing the CIFs for Quartz Low, Lithium Deuteride, and Greenerite, the final diffractogram can be seen in figure 3.

Figure 3 shows our 3 best fit minerals (red peaks) overlaid on top
of our sample (black peaks).  

The CIFs matched up amazingly well for YJ-011, and for each of our other 9 samples. For YJ-011 in particular, the percent compositions as indicated by MAUD are 16.7% Quartz Low, 82% Lithium Deuteride, and 0.55% Greenerite. 

Of course, these values are not completely accurate. Since we are dealing with clays, there are likely  immeasurable amounts of trace minerals among the material. However, based on the consistency between our samples, getting some amount of Quartz low and Lithium Deuteride for most if not all of them, we are confident that we are headed in the right direction in terms of determining the compositions of our clays.

At this point, we still need to take our findings and cross reference them with minerals that are commonly found in prehistoric lakes.

Over the course of the semester, we learned a great deal about how to obtain data from the XRD and analyze it using various software. Perhaps we could take what we learned about the analyze portion, and apply this to the Dominica samples that we tried XRDing in the fall.

We can proudly say that we are leaving this semester with new-found skills in compositional analysis.

Thanks and happy geomorphologing!
Ely and Josh