- To participate in the 911Metallurgist Forums, be sure to JOIN & LOGIN
- Use Add New Topic to ask a New Question/Discussion about Assaying, Microscopy, Mineralogy & XRF/XRD.
- OR Select a Topic that Interests you.
- Use Add Reply = to Reply/Participate in a Topic/Discussion (most frequent).
Using Add Reply allows you to Attach Images or PDF files and provide a more complete input. - Use Add Comment = to comment on someone else’s Reply in an already active Topic/Discussion.
Fertilizers Analysis by XRF (11 replies)
I haven't looked at analysing fertilisers for years now. The one thing I do remember is that the sensitivity of N then was in the range of about 3% This was caused by two main factors, namely that I was using a side windowed tube, so there was at least a 200um Be window and the flow counter window, which being mostly carbon, is still an almost ideal device to absorb NK photons.
Most important in the analysis of N in fertilizer is the sample preparation. The powder must be very fine in order to be able to have a reproducible signal. Most grinding methods will lead to a compacted sample, rather than a ground sample. Note: The analysis depth of N is extremely small so small particle size is vital. Today the spectrometer sensitivity is far better than in those times and is rather in the order of 1 kcps/% N, depending on the configuration. For N you may be obtaining <2% relative, for P and S you may be getting <1% relative, depending on the concentration levels and used equipment.
I did not get to analyze fertilizers, but I did pressed pellets and glass discs of phosphate ores for fertilizers. My precision for S and P was good enough for my needs (95%).
However, i wanted to ask you if any REE was detected in your routine and if the fertilizer P came from a sedimentary rock or igneous one.
Can you give a rough composition as there are a wide variety of fertilizers. Most are pretty easy to press without a binder but some are hydroscopic which can cause some problems.
I haven't analyzed P and S in fertilizers using XRF; however, I know looking at K is difficult. You must take care to dry the sample before you make the pellet or I have seen quite varying results if the sample preparation isn't consistent. I think the problem is not only due to the difference in x-ray scatter due to differences in the moisture content; but, for K, the valence state actually can change in the sample due to the presence of moisture causing changes in the XRF response.
To affect good grinding (without caking or balling) DuPont make a product called Vertrel XF which acts as an excellent grinding aid and can be evaporated off after grinding so leaves no residues (like methylene chloride in decaffeinate of coffee). Have you developed grinding and pressure response curves for the lightest elements of interest to make sure your preparation technique is optimal?
I am assuming we are talking about automated sample prep and an XRF in some type of automation line ? I am assuming that you are using a WDXRF instrument given the N analysis requirement. As correctly pointed out the escape depth for N is on the order of microns so minimizing particle size and surface flatness is critical. Some of your automated sample prep will be really helpful after you work out the ideal sample prep. Also be careful of some of the pellet binders that contain N. You might also want to look at the separating the precision of the instrument from that of the sample prep if you are not meeting your desired results.
Presence of moisture can indeed be a cause of poorly reproducible and also the particle size as explains so nicely. But, I don't think change of valence state has anything to do with the X-ray fluorescence response, since its an inner-shell ionization phenomenon. The change in signal intensity of low-energy characteristic X-rays of N, P or K is due to the higher mass attenuation in presence of moisture.
For the in-dept discussion. I was only talking about the K-X-ray intensity. Indeed there may be shifts with very low Z elements, which may not be discernible in EDXRF (even with ultra-thin/windowless detectors, because of the lower resolution), but in WD-XRF with certain high angular resolutions may be clearly perceptible and more so with the sophisticated multi-crystal ones (Johansson type). The L-X ray intensities (LB vs LA) may change due to Coster-Kronig transitions I guess?
If you're going to use a binder, then you'll need to be wary of carbon-based materials as they will absorb a significant amount of the NK radiation.
Are there anyone out there who analyze fertilizers by XRF on pressed pellets? If so what is the achieved precision on elements like N, P and S? and if you have any comments regarding sample prep in general. We have found that its possible to analyze pressed pellets of fertilizers by XRF, but were not completely happy about the precision we achieve. Also any knowledge on the chemical stability of the fertilizers would be great. They do decompose it seems.