[Kaolin]

Problem :

Currently we analyse our Phosphate rock feed sample to our process using two methods  viz  UV/Vis Spectrophotometer and XRF analyser to establish the percentage of phosphate in the feed and in the report it is expressed as P2O5 in percentage. The feed is from a single source and is fed to two streams for processing. In one stream we use the UV/Vis Spectrophotometer and the other stream we XRF analyser with online auto sampler for the analysis purpose. Even though the feed is from same source we get two different results.  The UV/Vis method always reports 1% less than the XRF method. Please clarify why it happens so, as this leads to a big problem in the metal accounting.

Materials and Methods Used :

1.   Feed Material :  The phosphate in the feed sample appears in the following three forms Ca5F(PO4)3 or Ca5Cl(PO4)3 or Ca5OH(PO4)3 . The analysis is done for the phosphate or phosphorus and the result is express in the form of P2O5 in %.

2.   Chemical Standards : P2O5 Standards used in the analysis are prepared in the laboratory itself.

3.   Methods used in UV/Vis : Nitric acid dissolvement method at Wavelength  - 460 nm

4.   Method used in XRF : Pellets

[bromidewind]

My guess is that the shorter, more energetic UV rays are bumping the outermost electrons off the phosphorus resulting in the destabilization of the phosphorus. The longer, less energetic X-rays aren't as fast at achieving the ionization energy, so by the time the sample is finished running, more stable compounds exist. So basically, UV ionization occurs faster than X-ray ionization.

I'm probably completely off the wagon about this, but I thought I should at least give it a shot.

[Train]

Have you validated both of these methods?  What kind of accuracy and precision did you measure for each of them?  My guess would be the UV method has a lower recovery because of the sample prep.