Author Topic: Quant results as tables for large batches of samples  (Read 904 times)

orlandin

  • Professor
  • ****
  • Posts: 39
Quant results as tables for large batches of samples
« on: March 02, 2021, 06:50:46 AM »
Hello, all!

First, thank you so much for producing and maintaining this wonderful program Dr. Ritchie! Our lab has a lot of probe work in the <1,000 ppm range, and as our 8200 ages I have big plans for turning our humble little EDS-SEM into a highly productive and user-friendly microprobe-lite using DTSA-II. I've worked through a lot of the learning curve and am starting to get results that are really exciting, but am running into a small quality of life issue when processing many spectra at once: the format of the results is beyond my skills to easily export and display, especially when processing many elements for many spectra at once (which DTSAII is doing beautifully!). One example is that the +/- in the results are awesome to have, but the presence of that character in the cells significantly complicates statistical analysis of the results for my weak and unimaginative Excel skills.

I know I am spoiled by PFE, but I wanted to ask how does everyone here get the results from many spectra into their preferred format? And if possible, how to get DTSAII results into a PFE-esque delimited format?

Lastly, and unrelated, I want to comment here that learning to use DTSAII has really been a fun and informative process coming from a mostly WDS background. Lots of familiar concepts that need to be explored from a slightly different angle, not to mention learning a few things about my standard collection that I might have missed more easily on the probe. I think this will be a fantastic teaching and analytical tool.

Best,

Phil

Nicholas Ritchie

  • Moderator
  • Professor
  • *****
  • Posts: 155
    • NIST DTSA-II
Re: Quant results as tables for large batches of samples
« Reply #1 on: March 03, 2021, 04:40:37 AM »
Phil,
    I'm glad to hear that DTSA-II is working for you.
    There is function on the scripting side that might do what you want - tabulate(...)

     First, you need to know that the spectra currently being displayed can be accessed using the `selected()` function.  Then
> tabulate(selected())
will tabulate the *measured composition* as mass fractions for the selected spectra.  To display, the normalized mass, fraction
> tabulate(selected(), normalize=True)
To display, the atom fractions
> tabulate(selected(), massFracs=False)

If you need more sophisticated output or a different format, you can use the the Python code for tabulate as a template to develop your own output function.

"Do what you can, with what you have, where you are"
  - Teddy Roosevelt

orlandin

  • Professor
  • ****
  • Posts: 39
Re: Quant results as tables for large batches of samples
« Reply #2 on: March 03, 2021, 10:56:26 AM »
Hello, Dr. Ritchie!

Thank you so much for this! Yes, this is all of my spreadsheet dreams come true. I am anticipating crushing ideological opposition to quantified EDS with an overwhelming volume of results.

Best,

Phil

orlandin

  • Professor
  • ****
  • Posts: 39
Re: Quant results as tables for large batches of samples
« Reply #3 on: March 24, 2021, 08:57:46 AM »
Hello, all!

As my standards-based DTSA-II EDS adventure continues, I wanted to follow up here with a note: Nicholas' script is the best solution, but you can also copy the results from the last screen (where there is a 'Finish' button) straight into Excel, and then with the application of the functions LEFT and RIGHT very quickly extract results and one-sigma uncertainty as you please. Either way, it is very easy to get a large batch of results into whatever format a user prefers.

Best,

Phil