I would like to carry out a semi-quantitative or quantitative analysis of the minerals present in the laterites. I have the DRX and chemistry results (major elements).
Hello Ndzie, I try to understand better your question, did you mean that you want to determine the relationship in the stoichiometry of the minerals, or are you looking for the exact stoichiometric values and nature of the mineral? If you look for a must accurate method for a quantittive analysis you must use a DRX-Rietveld method.
Before you launch into Rietveld quantitation, as correctly stated by Jorge, it is vital that first carry out a thorough qualitative assessment. The Rietveld method will give you wt% of the mineral phases normalised to 100%, so that means that if any phases are missed out then all the values will be overestimates. This means that for the quantitation to be successful, ALL mineral phases must be identified.
Some other points to consider:
laterites are going to be high in iron, if you are using a standard lab Cu anode tube then you will have a lot of fluorescence from the sample, and your backgrounds will be unnaturally high. This will affect your ability to detect phases in low concentrations and will also affect pattern fitting. If this is the case consider using a Co tube, or using some discrimination at the detector to remove the energy from the Fe fluorescence.
Secondly: in order to do quantitative Rietveld well you need very good data. So ensure you spend a long time counting and gather the best that you can. I've seen some awful quant results based on poor data.
It is good that you have chemical data, you can use this in filtering the results from search-matching, but when filtering remember to include the elements that could not be detected e.g. H, Li, O, C, F, as these may be present in minerals. Consider also that you may have trace elements present that will not appear in your majors list, but may be substituting in the crystal structures.
If you have not done quant XRD before, there is a lot to learn...
Thank you Ian J Slipper for your answer. You understand exactly what I want to do. I will put you in an attachment an example of result that I have. One of the methods I want to use is the quantitative method. It consists in using the results of the XRD, in particular the d-spacing ([A]) to identify the minerals and the area ([cts * ° 2Th.]) To quantify them according to each peak. I sum the sum of the areas by peak identifying for each mineral that I divide by the total area of the peaks to have the percentage of the mineral in the sample. The background noise will influence the results obtained. I would like to know if this method is acceptable. The other method is a semi-quantitative analysis using the method of yvon et al (1990) used by Millogo et al. (2008). This method involves the results of the major elements. I would like to know if these methods are usable and if there are others.
As attachments I am sending you an example of XRD results, results of chemistry and Millogo thèsis (you can see the formula on page 81).
http://www.sciencedirect.com/science/article/pii/S0950061806002406
http://www.beep.ird.fr/collect/upb/index/assoc/SEA-2008-MIL-ETU/SEA-2008-MIL-ETU.pdf
Article Valorisation des argiles kaoliniques des Charentes dans le c...
- Badges
- Science method
- Similar topics
- Methodology
- Crystallography
- X-ray Diffraction