Fingerprinting of mineral mixtures

This application note shows that Epsilon 4 is capable of identifying binary mineral mixtures using the powerful FingerPrinting technique.

In a FingerPrinting application, the unknown sample is measured with the same conditions as a set of raw materials which constitute an initial library of standards.

Introduction

This application note shows that Epsilon 4, a  high-performance energy dispersive X-ray fluorescence spectrometer (EDXRF), is capable of identifying binary mineral mixtures using the powerful FingerPrinting technique. The FingerPrinting technique is applicable to all materials, including liquids, solids or powders and is available as an option for the entire Epsilon 4 spectrometers.

Application background

In a FingerPrinting application, the unknown sample is measured with the same conditions as a set of raw materials which constitute an initial library of standards. Both standards and unknowns  should be presented to the spectrometer with similar sample preparation conditions (i.e. loose powders placed in liquid cups or pressed pellets, with a constant sample/binder ratio). After the measurement, the software will automatically use a specific algorithm to give the final matching with one of the standards.

Instrumentation

Measurements were performed using an Epsilon 4 EDXRF spectrometer, equipped with a 15 W, 50 kV rhodium anode X-ray tube, 6 filters, a helium purge facility, a high-resolution SDD silicon drift detector, a sample spinner and a 10-position removable sample tray. The measuring conditions are optimized per library of standards. In all measurements, the complete analysis was done in less than 7 minutes per sample.

Concentration-based FingerPrinting

The other possible option for using Fingerprinting is the concentration-based matching. Instead of making use of all information contained in the spectra, the software applies a calculation algorithm based on calibration lines. This approach works only if the library of standards has declared concentration values for at least few elements present in the library.

For this application note, a selection of binary mixtures prepared from pure limestone (CaCO3) and pure dolomite (CaMg(CO3)2) was investigated. For enhanced performance, all samples were ground using a Retsch MM 400 mixer Mill, mixed with binder (wax C, in a 10/1 = sample/binder ratio) and pressed into pellets with 32 mm diameter.

table1.PNG

Figure 1. Calibration line for Ca in binary (limestone – dolomite) mixtures 

figure1.PNG

Short-term and long-term results

The performance of the concentration-based FingerPrinting was checked on several limestone/dolomite pellets. In all trials, the first candidate in the list was the correct match to the real mixture.

Figures 2 and 3 show results obtained in one day, for single pellets from two limestone/dolomite = CaCO3/CaMg(CO3)2 mixtures. Figures 4 and 5 show the results for the long-term performance of the Epsilon 4 on multiple pellets prepared from the same binary mixtures.

figure2-5.PNG

Conclusion

The results demonstrate that Epsilon 4 can be used for rapid FingerPrint identification of minerals. It is also possible to combine both FingerPrinting and elemental quantification in one method. In this case the resolving power of the fingerprinting software can identify correctly mineral combinations with minimal differences in elemental composition.

登录

还没有注册? 创建账户