Well, 6 elements with atomic numbers lower than 10 are probably H, Li, Be or B, C, N, O or F. H content can be measured accurately with elemental analyser, provided that your sample can be burnt in oxygen. The same applies to C, N and O. Li, Be and B can be quantified by atomic absorption spectroscopy. Only F requires other techniques. XRD can help only if you are sure that the phases you are looking for are crystallised, which is not obvious in a solid waste. But even in this case, I am not sure that you can easily quantifiy the related elements. Better using an electronic (Castaing) microprobe.

Alain

How you can be sure what do you see, hear, taste etc.? We are continuously learning about the physical background of all these techniques and the technology used. I must admit that it is some kind of believe, mixed with a critical view on the results. Even if the technique used is described by a certain accuracy (in spec sheets or publications), it is only valid for very specific conditions. Often they are listed together with the specifications. If this isn't you should be sceptic!

The accuracy of EDX can be quite high for certain element combinations (fraction of a percent), but e.g. for carbon it is not better than several percents, i.e. one or two orders of magnitude difference. A similar situation exists for XRD. You can never be perfectly sure...but you can do your best evaluating the result critically...even if it obviously fits to your expectations. Maybe even then it is very important to doubt...and crosscheck it by another technique.

The accuracy of XRD results vary with the type of the radiation employed. Normally EDAX, PIXE and XPS are around 5% accuracy,

There are many parameters that can be obtained from XRD measurements; the accuracy will be determined by what measurement you're performing, your sample (geometry, uniformity, composition, strain, microstructure, defects... etc), your measurement (optics alignment, sample alignment, optics used, measurement parameters...), your instrument (monochromator, detector resolution...) your data analysis (assumptions, background, fitting...) and more. When the right measurement is done correctly on a proper sample you can obtain quantitative results with very high accuracy, well below 1% for lattice parameters, for example.

EDX's accuracy is different for different elements and spectral lines. Elemental standards can be used to improve the accuracy. Some lines overlap, increasing the errors and can sometimes prevent quantification. Uniform samples allow better accuracy than non-uniform samples, thin films etc. As mentioned above, light elements are harder to quantify (say, lighter than ~oxygen). It is crucial to examine the spectra, verify a proper separation between the lines of interest, verify that the signal-to-noise ratio is adequate and that the fitting of the peaks and background was done correctly. When all the above is done correctly on a proper sample you can have errors as small as 1~2% percent in the total composition. When done incorrectly (e.g. by assuming that the automatic fitting/quantification is golden without looking at the spectrum) your error can be infinite.

I agree with the previous comments, there is no absolute nor single answer to your question, as it depends on what you are measuring. The best way of being sure of your results is referring to purified standards, having elemental composition close to what you expect in your own sample. For example, in elemental analysis, if you expect a nitrogen content close to 5%, then you should use at least 2 pure standards having N contents of 1 and 10%. And this should be done frequently.

Alain

Thank you all for your valuable suggestions.. as the real problem I am facing now is that quantitative analysis of the elements as well as phases present in my sample. As I am working on a solid waste it contains around 12 elements (detected till now). I want to know the quantity of each element as 6 elements are having atomic no. less than 10. whether quantitative analysis by XRD will be handful or not?

Well, 6 elements with atomic numbers lower than 10 are probably H, Li, Be or B, C, N, O or F. H content can be measured accurately with elemental analyser, provided that your sample can be burnt in oxygen. The same applies to C, N and O. Li, Be and B can be quantified by atomic absorption spectroscopy. Only F requires other techniques. XRD can help only if you are sure that the phases you are looking for are crystallised, which is not obvious in a solid waste. But even in this case, I am not sure that you can easily quantifiy the related elements. Better using an electronic (Castaing) microprobe.

Alain

Hello there,

Insert replicates in all your analyses and calculate the average values. 5% variation is the acepted norm for any analyte.

Hello there,

Add relicates in all your analyses and calculate the average value for an analyte. 5% variation is the accepted norm

Dear Sidharth,

If you are analyzing solid wastes with organics and interested in non-intrusive bulk chemical speciation, try FTIR/Raman. You will not get elemental quantification but at least the functional groups/species present may help you in quantification.

In SEM-EDX with ultrathin window detectors you will get only semi-quantitative or relative quantification down to Carbon...and if your sample is inhomogeneous (which it seems) you will not get representative results.

For CHNS Alain already advised you above.

Respected Dhruba Sir,

I have analyzed my sample with FTIR also in search of finding any C-N bond streaching but results were negative. The solid waste termed as "Spent Pot Lining " is mostly comprises of inorganic fractions like NaF,Na3AlF6,CaF2,NaCN,Al2O3,SiO2 etc. To get the exact composition (Quantitative ) details I have done all the above mentioned studies.

So you must have already done XRF too.?...It seems ideal for the major part of your waste quantification although you will not get elements below Na.

Hello Sidharth,

Frst try to quantify the elements with XRF analysis using powder sample as well as fusion as a glass disc. Then digest a known weight of sample in perchloric/HNO3/HF medium @ 150oC (use teflon ware, done in triplicate, first fume and then dissolve the resulting gel in nitric acid) and make it up to a known volume in deionised water, so that the final acid concentration is 3%. Then analyse the solution using ICP-OES or ICP-MS. For concentration > 500 ppb use ICP-OES, and for concentrations < 500ppb use ICP-MS. Perchloric and HF are lethat so safety gear such as face shield / glasses, safety shoes, double gloves and laboratory coat are mandatory. Then compare results.

Respected Dhruba Sir,

I have not performed XRF studies till now. I have conducted SEM-EDX AND CHNS analysis 3 times and I have found at most 1.2% in deviation of results in chemical compostion of elements. From the literature i have found the details of elements and with those selected elemets i have done the EDX analysis as well as the CHNS analysis.

Respected Aly Sir,

Thank you very much for your suggestion i will definitely perform the above mentioned technique for the characterization.

When you perform ICP-MS or OES take care about the possible halide interference since your sample has halide and of course safety aspects as Aly advised.

Dear colleage Sidharth,

Energy Dispersive X-ray Analysis (EDX) can be used to investigate the properties and compositions of a broad range of sample types. EDX is used for chemical identification of elements and their concentration.

X-Ray Diffraction (XRD) is used for Phase identification. It is used to identify spatial arrangements of atoms in crystalline phases.

In EDX, electrons knock out electrons from atoms, producing X-rays of characteristic wavelength. These X-Rays are then detected to identify the element from which they were produced. In XRD, X-rays of known wavelength are used to probe the structure of the material, using the lattice as a diffraction grating.

To perform an EDX you need an electron source. For XRD you need an X-ray source.

If you are looking for chemical composition, concentration gradients, or solute segregation, use EDX.

If you are looking for phase distribution, lattice strain fields or stored defect content, use XRD.

hope this helping 

Have a very good day