Dear Nur Farahiyah,

A model calculation is done in the attached Excel Sheet. Hope this is self explanatory for you.

Good Luck!

LRK

Dear Nur Farahiyah,

Get the molecular weight of each oxide. Then for each molecular weight of oxide, find the weight of each element as a percentage for that molecule (eg. amount of Ca and O in a CaO molecule, amount of P and O in a P2O5 molecule etc.). Then divide the percentage of each oxide in your XRF analysis by its molecular weight (eg. CaO). Then you will get the molecular proportions of each oxide in your analysis. Since you know how much Ca (or P) is present as a percentage value in one molecule, you can now calculate the total amount Ca (or P) contributed by the number of CaO (or P2O5) molecules in your XRF analysis and also the amount of O contributed by various oxide molecules. From the proportions of theses weights of elements and oxygen you can calculate elemental percentage for the whole rock or mineral.

Hope this will help

Dear Mrs. Perera,

I have attached the parameters to this anwer for your case.

Best regards

H.G.Dill

Dear Dr.L. R. K. Perera , I did not really understand the explanation.

Here is my calculation:

The molecular weight  (MW) of CaO is 56.077, MW of P205 is 141.943. Then for CaO,  I calculated the weight percent of Ca= (40.078/56.077) x 100 = 71.47 %. for O = (15.99 /56.077) = 28.53%.

For P205, weight percent of P = ((2x30974)/141.943 ) x100 = 43.64%, for O = (5x15.999 /141.943) = 56.357%. 

Then i divide all the value  percentage of each oxide in my XRF analysis by its molecular weight  . For example CaO = 2%/ 56.077 = 0.0357.

So from these values what i need to do next to calculate the elemental percentage of Ca?

Dear Nur Farahiyah,

I have sent a message to your mail box and will get in touch very soon.

LRK

From the % of Cao you know the wt of it in the ore

Then as we know by a simple way

CaO - - - - - - - - - - > ca

Mwt                           At. Wt

Wt (gm)                      x???

 So,  you can calc. The wt of ca in CaO

Then; Ca % in sample =(wt of Ca / wt of total sample) x100

Dear Nur Farahiyah,

A model calculation is done in the attached Excel Sheet. Hope this is self explanatory for you.

Good Luck!

LRK

The above answers are perfectly correct, when you know the coordination number of the cation.

Pay attention that there is uncertainty in the case of elements which coordination number is variable:i.e. Iron, that can be either Fe +3 or Fe+2.

Dear Nur Farahiyah,

First of all did you fuse your sample? If you didn't,  you will get a relative weight percentage from the conversion and not the absolute mass fractions. The XRF calculates considering the sum of oxides of only the detected elements as 100%. Keeping that in mind if you prepare sample by fusion (without loss of Ca and P),,,you can hope to get a more accurate estimate

Thank you all for the sugesstions.

If your EDX machine is working properly, and the operator knows what he is doing, then the elements are listed as % oxides, and can be shown to add up to 100% or not, depending on the particular analysis. However, some elements could be present for example as sulfates of sulfides, or simply not oxides, so a clear image can be obtained if all elements are analyzed and unravelled to give the total elements, and even then, porous products will not add up to 100%, so it can be said that EDX is only a qualitative analysis, chemical analysis is better for getting chemical results that are reliable, and real (EDX could give the result for surface hydrates or hydroxides, that do not reflect the inside composition of any sample). the instruments are also programmed to give you the elements as percentages  directly, so you do not have to worry about converting oxides to elements if that is a problem for you......

Hi there,

why do we convert elements to oxides e.g. Ca to Cao.

Please reply

Not sure how you ran you samples, but typically 10000ppm is equal to 1% weight. This is a general assumption on my part and not sure about separating values of chemicals into individual elemental ppm or weight %. Good luck! 

elements like Ca2+ live with their counterparts like négative anions, the oxides being the simplest and most usual, specially after calcinations. So XRF gieves analytical results as oxides and this also helps to see that no majors elements are omitted for examples holes in the sampls. Thus if the toalt gives 100% it is OK. Think for example about NaS or Na SO4, the amount of Na is different!

You can apply law of definite proportion for the conversion.

Dear Nur,

Generally, XRF cannot detect elements below Na. So if these elements are present in your sample it cannot do the absolute quantification. These low-Z elements are called the dark matrix.The fundamental parameters usually quantifies in relative, assuming the total concentration of detected elements as 100 % if you run in elemental mode. On the other hand, if you run the quantification in oxide mode (usually used for fused sample pellets and may be default in your instrument) corrects automatically assuming the sum of oxides = 100 % for the elements detected.

You can check whether the elemental mode is available and whether there is an option to provide composition of the dark or unknown matrix. As you already know the stoichiometry (of low-Z elements) of the apatite, you may feed the atomic % composition. It may give you the corrected elemental concentrations.

If the empirical mode of quantification or alpha coefficients are used you have to use similar matrix certified reference materials with known elemental composition (usually several CRMs with range of concentrations for each element) to calibrate and you can get more accurate quantification of elements for your unknown sample.

All the best.

We need this kind calculations quite often as some customers like to have the results in element-% and some in oxide-%. There are some programs  which will do the calculations for you. For example ElemToOx from www.open.ac.uk.

All you need to do, is apply the conversion factor. The conversion factor for CaO and P2O5 are 1.399 and 2.2914 respectively.

 element percent = oxide percent /conversion factor

or

Assuming you know the element percent and you need to convert it the oxide percent,

then, oxide percent = element percent  x conversion factor

what will be the Fe wt% if both FeO and Fe2O3 is found from xrf analysis.Can I simply add up the Fe content found from both Oxides.

You will have to divide their oxide percent by their conversion factions, 0.777311

0.699433 for FeO and Fe2O3 respectively

In one of the research paper, the Fe2O3 (%) in a sample was 5.03. According to the conversion factor , the Fe2O3 % should be divided to 0.699433 to get the Fe %. It comes out to be 7.191 But in the paper it is given 6.97. Why the difference is??

This is very simple, go to the software and change the units instead of oxides you can get ppm for low concentration of minerals and % ofrthe higher ones.

Each oxide has its own conversion factor to be used. 

Generally: 

Oxide Percent *  Conversion factor = Element Percent

Element Percent ÷ Conversion factor = Oxide Percent

You can use the attached excel sheet to convert the oxides to its element and vice versa

Examples of conversion between wt and mole % can be found in lots of books. This is basic chemistry. I give examples in my book, Introduction to Glass Science and Technology, including those conversion factors mentioned above.

Is there any simple test to show the Fe (Iron scrap) content of EAF slag? (We are not looking for all Fe elements of all combinations such as various FE oxides and so on, we only need to know how many percent of the slag (EAF steel slag) is Iron scrap?

Usually in the test results of XRF the Fe(total) and Fe2O3 or FeO are reported, but how can we find the Fe metal (as Iron scrap) content of a sample of slag.

For example the test results of a portable XRF from about 30 points of a EAF steel slag pile, has attached, how can I know how much is the weight % of the Fe metal (as iron scrap) in this EAF steel slag pile?

Thank you all in advance.

you can examine it by XRD BURKER 2D phase , it's qualitative and quantitative but also it's very complex due to overlapping of peaks , actually i am searching for the same .. but in other words,I have three samples and the presence of FeO,Fe2O3,Fe3O4 is different and i want to know the relation or why it appeared in one sample while the other one is disappeared

i need reference for method describe in Excel sheet.

Dear Fahad,

The calculation in the Excel sheet is based on the understanding of basic chemistry of oxides and what chemical analysis of a mineral means. It is not coming from a text book. However, any basic mineralogy book dealing with calculation of mineral formula from a chemical analysis is useful to understand the calculation. Ex. Manual of Mineralogy by Cornelius Klein and Cornelius S. Hurlbut (21st or 22nd Edition) published by John Wiley would be a useful reference.

Thank you Perera for your useful reply

I recommend Dhrubajyoti's answer

Best Regards

http://www.marscigrp.org/oxtoel.html

How about using the molar weight of the oxide and the atomic weight of the elemnt and then estimate the atomic weight percent? Because you already have oxide weight % from XRF...

Please, send your email and I will send to you a file contains all the convert factors

XRF occurs following a Loss on Ignition or calcination at 1,000 C - and all elements are measured as atomic % but as the partial analysis cannot total 100%, conversion factors are applied to allow for oxygen on the basis of the most common oxide in a calcined sample. Hence regardless of the actual mineralogical starting composition, after calcining all Fe will be converted to Fe2O3 - the most logical ferrous oxide and Na to Na2O, K to K20, Al to Al2O3 etc - which should now all total to 100%. If it adds to more than 100%, a normalisation factor (designated N in the method) is used to reduce all components to a result which totals to 100.00%. This is designed to overcome common matrix interferences. Note that XRF is measuring the components mixed into a fused lithium borate glass after calcination, not the actual mineral sample - a trap for the unwary.

This attachment (oxides to element) could be useful

PLS I WILL APPRECIATE IF SOMEONE CAN EXPLAIN IN DETAIL THE PROCESSES INVOLVES TO INTERPRET RAW RESULT FROM LASER DIFFRACTION EQUIPMENT (MALVEREN 2000). MY Email address is [email protected] in case you want to contact me

THNK YOU

you can use it

I am also interested to understand this fully. The thoughts and suggestions presented here are great, but I am still not looking into a thorough explanation. e.g., how much would be wt% of Fe if were to determine from 10% Fe2O3?

Thanks

You should not rely on one instrumental method to always provide the ‘correct‘ answer. XRF is effective but has limitations, just as XRD and ICP-ES/MS do. The best approach is to understand the matrix and make allowances for the errors inherent in each of the methods, then you can move closer to the truth. I’m still waiting for the ‘holy grail’ of instrumental methods that always ‘pops out’ the answers I need! 😂

Go to the pages 32, 33 and 34 of the book "Basic Analytical Petrology" of Paul C. Ragland.

There it is explained what you are looking for. Calculate elemental weight percent from oxide weight percent given by XRF, and vice versa. It is well explained with examples and very easy to understand

Thank you.

You guys are truly amazing!

Alternatively, you can convert your data online. Use the following link.

http://www.marscigrp.org/oxtoel.html

Regards

Dear Carlos...waouuuuuh that is amazing in converting oxides...thank you!

Hello, the calculation are easy to be made : you calculate the molecular weight of CaO (40.08 + 16= 56.08). To calculate Ca and O individually, you cross multiply the weight % of CaO, the molecular weight of CaO and the one for Ca then for O using their molecular weight ...

Regards,

you use this convert table