Re: [Ifeffit] Calculation of metal concentration (and hence density) from Edge Jump

Dear Alexey, Yes it certainly helped. I was almost on the correct lines but was not completely aware of the ion chamber consideration. I will check the values for amplification and will take the proper values into the consideration. Sincerely yours Pushkar On Mon, Feb 27, 2017 at 3:53 PM, Alexey Boubnov wrote:

Dear Pushkar,

1) µ is the total absorption at a given energy. If you are above the edge, it will be higher than if you are below. Test it by changing the energy. NOTE that the absorption value you have in Athena is calculated form the ion chamber values, which are amplified. If I2 is amplified 10x higher than I1, -log(I2/I1) will be -log(10) = -1, i.e. lower by 1 relative to the real value. Knowing the amplification factors, you can take this into account. 2) yes. Athena determines it for you during normalisation. It is the difference between the pre-edge and post-edge fit curves at the position of E0. This is the proper way of determining the edge step. 3) yes. This is a statistical thing. Try to normalise using different ranges and you will get slightly different values of Delta µ. The standard deviation you can give as delta µ. If the uncertainty does not interest you, you can probably set this value to 0.

Hope this helps. Alexey

2017-02-27 5:56 GMT+01:00 pushkar shejwalkar :

...

Dear Alexey, Thank you for your insightful suggestions and very useful information. Now that I have got the XAFSmass program running (?) it is giving me some values and no more error. So I am assuming that I am doing everything right. However, I still have some questions regarding the READING and UNDERSTANDING of the data input and output from the program. From the manual, it is not very clear and difficult to understand some of the input values. In the unknown concentration tab 1) u = high u?? meaning highest absorption? 2) DELTA ud is = to deltaU i.e. typically what we get as edge step in Athena?? 3) deltaut = uncertainty? For the novice learner like me could you please tell me how exactly can we calculate these values from the single spectra (preferably). I am aware of the crude ways to find it but would like to more proper and scientific way. By the Way, in order to establish these values, should we use the crude data file or properly processed data file e.g. K-weight corrected, background adjusted, aligned and calibrated data? Sorry for the inconvenience and thank you very much in advance for the details. Sincerely yours Pushkar

On Thu, Feb 23, 2017 at 6:18 PM, Alexey Boubnov wrote:

...

Dear Pushkar,

1) M%x is the mass percentage. 2) This function is for calculating the element concentration from the absorption spectrum. XAFSmass does not have problems with long formulas. You wrote the formula correctly, but maybe you did not indicate also the total absorption, edge jump with uncertainty and energy?

All the best. Alexey

2017-02-23 2:54 GMT+01:00 pushkar shejwalkar :

...

Thank you, Alexey, for very useful information. Currently, I am working on XAFSmass. Technically it looks much more flexible than the software provided by SPring-8 since almost all the variables can be changed and adjusted as per the requirements. 1) However, sample information is very tedious to put in and confusing as well. Do you happen to know what exactly does M%x (is this mass % or mole %). 2) I am working on the tab with the determination of unknown concentration Label and I am interested in finding out the Fe concentration only. So I should write Fe%x (I assume). However, since I use organosilica's the formula is not so straightforward. My empirical formula for the catalyst is FeC580H382O152Si98N10F6S2. Do you think the program can handle such information? I wrote Fe%xC_580H_382O_152Si_98N_10F_6S_2. But it is giving an error that 'can not calculate concentration'. Am I doing something wrong? I have already read the pdf in the software and as such it looks not so complicated but just the syntax of writing chemical formula is different and con-conventional.

Sincerely yours Pushkar

On Wed, Feb 22, 2017 at 9:52 PM, Alexey Boubnov < alexeyboubnov@gmail.com> wrote:

...

Dear Pushkar,

regarding your last question: the absorption edge jump (Delta µd) is the difference in absorption before the edge and after the edge, e.g. Pt L3-edge is at 11564 eV, so Delta(µd) = µd at 11580 eV - µd at 11550 eV. The total absorption (µd) takes all atoms into account. You only need the difference. I and I0 do not contain information on the sample, so they are useless for this purpose.

Regarding your original question (calculation of the edge jump knowing the composition of your sample): there are programs available to calculate the edge jump and amount of sample from the composition and total absorption, saving you the effort with formulas and letting you focus on your sample. Try XAFSmass (https://intranet.cells.es/Bea mlines/CLAESS/software/xafsmass.html).

Here you can enter the bulk composition (e.g. Fe%10SiO_2 for a silicate with 10% Fe-content), energy (e.g. slightly above Fe K-edge: 7150 eV), pellet area (e.g. 1.3 cm²), total absorption (e.g. µd = 1) and press "calculate". This gives you the mass of your sample (e.g. 12 mg [hypothetical value]) and the edge jump due to the Fe K-edge (e.g. Delta(µd) = 0.2 [hypothetical value]). The mass and edge jump are approximately proportional. Consequently if your pellet in reality contains 60 mg iron silicate (5x the calculated number), absorption values will also scale: Delta(µd) = 1.0 and µd = 5, which is a little too high. Play around with the value of µd in the program to see the effect on mass and edge jump.

I wish you successful calculations and experiments! Alexey

2017-02-22 5:54 GMT+01:00 pushkar shejwalkar < pshejwalkar2004@gmail.com>:

...

Thank you, Bruce and Alexey for your kind information. Bruce the communication, unfortunately, is not available on our Internet connection from university to download for free. Do you think it would be possible (under the copyrights act off course) for you send a PDF copy? The communication seems really interesting and I will keep the procedure in mind (for March Dates when we have BL for us again) to make sure that we have a calibration curve. (off course XRF is more reliable and easy). Alexey, I read the section (also some other information in that book). The book is really great. Thanks for sharing. I have already sat down with the pen and paper. but just want to know one thing. Is Edge Jump = Absorbance?? or we need to calculate the absorbance by actually practically finding out the I and Io? e.g. if edge jump = absorbance I can modify the equation I = Io^e-upd into its A form. If not I will use ion chamber gas ratio to calculate the I and Io respectively. Which way is more correct? Best Regards Pushkar

On Tue, Feb 21, 2017 at 10:15 PM, Bruce Ravel wrote:

> > > This might help: https://doi.org/10.1107/S1600577514001283 > > Note that we measured a sequence of standards of known thickness and > concentration /prior/ to the rest of the experiment, so the methodology in > that paper might not be useful after the fact in the absence of that prep > work. > > B > > > On 02/20/2017 10:19 PM, pushkar shejwalkar wrote: > >> Dear All, >> First of all, is it possible to calculate the concentration of >> metal >> (loading, therefore) from just the edge jump? Is there a simpler >> equation like knowing the Flux values or some basic instrumental >> information we can calculate the concentration term using Io and It >> values of beamline (the simplest form of Beer-lambert law)? if so >> what >> formula should we use. >> I searched in the mailing list archive and have found one such >> archive >> but it discusses how to use calculation to find out how much sample >> would be needed to get the edge jump. I am rather interested in the >> reverse way. I have an experimental edge jump and want to calculate >> (if >> possible with as much accuracy as possible) the concentration of >> metal. >> I know with only edge jump it will be difficult and XRF will be a >> better >> and more reliable way. But for better understanding and cross >> checking >> between different samples, I wish to do such calculations. >> Can anyone guide me to the right direction and equations to do such >> calculations? >> Thank you all in advance >> Warmest regards >> Pushkar >> >> -- >> Best Regards, >> Pushkar Shejwalkar. >> Post-doctoral -Researcher, >> Tokyo Engineering University, >> Tokyo-to >> Japan >> >> >> >> _______________________________________________ >> Ifeffit mailing list >> Ifeffit@millenia.cars.aps.anl.gov >> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit >> Unsubscribe: http://millenia.cars.aps.anl.g >> ov/mailman/options/ifeffit >> >> > > -- > Bruce Ravel ------------------------------------ bravel@bnl.gov > > National Institute of Standards and Technology > Synchrotron Science Group at NSLS-II > Building 743, Room 114 > Upton NY, 11973 > > Homepage: http://bruceravel.github.io/home/ > Software: https://github.com/bruceravel > Demeter: http://bruceravel.github.io/demeter/ > > _______________________________________________ > Ifeffit mailing list > Ifeffit@millenia.cars.aps.anl.gov > http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit > Unsubscribe: http://millenia.cars.aps.anl.g > ov/mailman/options/ifeffit >

-- Best Regards, Pushkar Shejwalkar. Post-doctoral -Researcher, Tokyo Engineering University, Tokyo-to Japan

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

-- Best Regards, Pushkar Shejwalkar. Post-doctoral -Researcher, Tokyo Engineering University, Tokyo-to Japan

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

-- Best Regards, Pushkar Shejwalkar. Post-doctoral -Researcher, Tokyo Engineering University, Tokyo-to Japan

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

-- Best Regards, Pushkar Shejwalkar. Post-doctoral -Researcher, Tokyo Engineering University, Tokyo-to Japan