DTSA script to quantify k-ratios

[Ben Buse]

Hi,

Just found my script to quantify k-ratios, so I thought I'd post it before I lose it again

Code: [Select]

import dtsa2 as dtsa2
import dtsa2.mcSimulate3 as mc3


f = open('C:/Users/Ben/Documents/FeKR.csv', 'r')
FeLines = f.readlines()
print(FeLines)
f.close()
FeLines4 = [float(FeLines[0].strip()),]
for x in range (1,9):
FeLines4.append(float(FeLines[x].strip()));
print(FeLines4)
f = open('C:/Users/Ben/Documents/NiKR.csv', 'r')
NiLines = f.readlines()
f.close()
NiLines4 = [float(NiLines[0].strip()),]
for x in range (1,9):
NiLines4.append(float(NiLines[x].strip()));
print(NiLines4)
for x in range (0,9):
print quant(kratios = {transition('Fe K-L3'):FeLines4[x],transition('Ni K-L3'):NiLines4[x]})


The files are as below: txt files - list of k-ratios separated by return.

0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1

[Philippe Pinard]

Here is my modified version of DTSA2 with a special class (MassAbsorptionCoefficientCalculator) to calculate MACs from X-ray intensities measured at different kVs:
https://www.dropbox.com/s/bf3wx3n5f7rg2r4/epq.jar?dl=0

and in attachment, an example of a script, which we used for this paper Llovet, X.; Pinard, P. T.; Heikinheimo, E.; Louhenkilpi, S. & Richter, S. Electron Probe Microanalysis of Ni Silicides Using Ni-L X-Ray Lines Microscopy and Microanalysis, 2016, 1-11 and I used in my PhD thesis.

Note that you need to fit intensities, not k-ratios.

I don't know about DTSA-II, but what you describe is exactly what Pouchou's XMAC software does.

One difference with DTSA-II is that you can use different PRZ models and fit "unconventional lines" like Ll, etc. but I agree that XMAC also works well. The main reason I wrote a special class was to be able to fit the MAC of more than one X-ray line, e.g. Ni La and Fe La from a binary Fe-Ni sample.

[Ben Buse]

Hi Philippe,

Thanks that's great -yes i've read the paper and that's the kind of thing I want to do. Do you know which version of DTSA its compatible with - I replaced Iona epq but it no longer loads

Thanks

Ben

[Probeman]

Hi Ben,
I found that proceedings volume and scanned it, so the Pouchou paper you wanted is attached below (please login to see attachments).
john

[Ben Buse]

Hi John,

Thanks for finding the right paper! and scanning it

Method is easy - take generated phi-rho-z curve and modify mac so absorbed phi-rho-z curve matches data. I've just done this in excel using solver and casino for the generated phi-rho-z curves. In sheet 1 - can calculated mac used in casino 2159 for Fe La in Fe - i.e. similar to MAC30 not FFAST.

In sheet 2 - took Pouchou 1985 data for Ni at 10,15,20,25,30kV. Solve it and get mac of 3368 similar to Pouchou value of 3300 for Ni La in Ni and a long way from book value due to white lines - (peak in absorption spectra).

Ben

[Ben Buse]

Hi,

Does anyone know Pouchou & Pichoir expression for ionisation cross section, - I found a reference to J. L. Pouchou, F. Pichoir, in: 11th ICXOM, London Ontario, 1986 (J. D. Brown, R. H. Packwood, eds.), London, Ont, 1987, p. 249. Does anyone have a copy

Thanks

Ben

[Brian Joy]

Hi Ben,

It's also contained in their 1991 paper in Electron Probe Quantitation.  I've attached a copy.

Brian

[Ben Buse]

Hi Brian,

Thanks, Oh good I've good the green book, I'd missed the section - found it now

Ben