Most of the EDS instruments give all wt%, mole and atomic%. You probably can ask the person who took the data to send you the atomic%. In addition, I found the attached excel file which might help you on conversion of Wt% to At% and vice versa.
Mol% and wt% both are commonly used in EDS analysis. These data give the help to calculate Stoichiometric in atomic % or vice versa. It depend on observer , what you need in your results .
I do not think EDS is a good method for your purposes. Analysis of complex oxides with non-stoichiomentric formulation is a very difficult task for EDS. It requires high qualification of operator and good standards. Yes, EDS software will give a set of numbers in any case, but they can be far away from real thing.
If you have the Wt% and mole% for Ba, Sr, Co, Al, Fe, and O, you will find that these numbers are correlated through the respective atomic weights. In order not to go through the hassle of calculation, I normally take the mole% data, which directly reflect the the atomic ratios in the chemical formula. Notice that in standard BaM, for example, the theoretical mol% of Ba is (1/32)x100% = 3.1 %, and that of Fe is (12/32)x100% = 37.5%. However, I agree with Vladimir Dusevich on the issue of reliability of the technique. According to my experience, when oxygen is involved in the analysis, large deviations from theoretical values could be obtained due to the high uncertainty in the determination of the content of light oxygen element. So I hope that in your analysis oxygen is excluded, and only metal contents are analyzed. If not, in order to go around the uncertainty problems, I would
1. take only metal mole%, and normalize them to a sum of 100%. Although this seems to be a crude analysis, it is certainly much better than keeping oxygen in.
2. Compute the theoretical values and compare with those obtained in (1). Your theoretical values should be 1.56% for each of Ba and Sr, (x/32)x100% for each of Al and Co, and [(12-2x)/32]x100% for Fe.
Of course I anticipate significant deviations for Al content due to its low concentration and low Z number. The content of Co may also show significant deviations due to its low content in the sample. In addition, professional operation of the machine, and limitations of the technique as hinted by Dusevich impose significant uncertainties. Accordingly, the operator should be instructed to perform measurements at several representative spots in the sample, and measure for long enough time to improve signal to noise ratio, which is essential especially in the case of evaluating peak areas in regions of overlapping peaks, or in regions of a complex background. By the end of data analysis, divide the (sum of metal contents) by (Ba+Sr) content, and compare this ratio with the theoretical value of 12. According to my experience, if the measurements were carefully performed, the experimental ratio should fall in the range 11 - 13, with an uncertainty of less than 10% (=1/12) in this ratio.
I was so confused with three data (wt%, mole % and edx mole). I have refere your other papers to find correct analysis of EDX and ask various researchers having chemistry background but no answer satisfied my query and then thought to keep wt% only to show presence of various metal ions in prepared hexaferrites. I have refer many papers for edx analysis and found most of the authors report presence of metal ions.
OK i will do the calculation suggested by you and analyse EDX. Thanks for your detail expatiation.
If you wish please send me the data (Mole ratios) for each sample and I will provide assessment of the experimental results. (I have to know the chemical composition of the sample, of course, in order to compare)
Dear Professor, Greeting hope you are doing well. I hope these values from direct EDX spectra will not help you to know the exact chemical composition by (wt%, mole % ). The best way would be to perform an SEC analysis.