Author Topic: MAC calculated for Pu Mb  (Read 2578 times)

remidelville

  • Student
  • *
  • Posts: 4
MAC calculated for Pu Mb
« on: August 18, 2014, 05:48:17 am »
Hello,
We are dealing with quantification of MOX (PuUO2) samples. For Pu we are using its Mb line.
When calculating k-ratio and ZAF correction factors for Pu we have noticed a very low absorption correction for Pu. Further analysis showed that the MAC value of Pu Mb (3.53 keV) in Pu, not tabulated in CalcZaf but calculated using McMaster expression, was a factor 10 too large (5627) to what is expected. It is also inconsistent with the MAC value for Pu Ma which is tabulated at 650.8(LINEMU, Henke) and which has an energy (3.35 keV) only slightly smaller. There should not be any influence of an absorption edge. In fact, we made our own calculation of MAC value of Pu Mb in Pu and we found 571 which is consistent with Ma line.

Interestingly the McMaster expression does give consistent value for the MAC of PuMb in U (528). The problem seems specific to PuMb in Pu. Any ideas why?

It would be good to add tabulated value for Mb in CalcZaf (none available at the moment).

Thanks for your feedback,

Rémi Delville
SCK-CEN (Belgian Nuclear Research Center)
« Last Edit: January 23, 2018, 08:29:03 am by John Donovan »

Probeman

  • Emeritus
  • *****
  • Posts: 1672
  • Never sleeps...
    • John Donovan
Re: MAC calculated for Pu Mb
« Reply #1 on: August 18, 2014, 04:04:53 pm »
Hi Rémi,
Yes, you are exactly correct.  When I put these tables together I found data for up to uranium easily, the higher actinides are more sparse. Philipp Pomel has been working on putting together a more complete data set for these elements.

By the way, to facilitate this there is a user defined USERMAC.MDB file you can create (based on any of the existing MAC databases) and then edit it to your liking- it will not be overwritten by subsequent updates. Most labs doing nuclear work have done this.

Also, please note that one can easily compare the MACs from all 6 MAC tables in CalcZAF by using this menu:



Then enter you the emitter/absorber pair like this:



and you'll get what is available in all the MAC tables like this:



As you said already, the MACs for the Z > 92 are sparse in the default tables and McMaster is from 1969 so that was never intended for these elements. In the meantime I believe Philipp Poeml will have something for you nuclear people soon.
« Last Edit: January 23, 2018, 08:29:14 am by John Donovan »
The only stupid question is the one not asked!

wrigke

  • Professor
  • ****
  • Posts: 30
Re: MAC calculated for Pu Mb
« Reply #2 on: August 19, 2014, 12:10:05 pm »
While I am not certain why what you observe happens with Pu, I can say that the Mb MACs compiled by Farthing and Walker note that energies for that emitter-absorber pair are "below edge M5 of absorber" and are not considered safe estimates.  In their table, Pu mb in Pu has a MAC of 565, similar to what you computed and similar to the other actinides around it except for Pa and Th, which are substantially greater.

From what I understand, ITU uses the MACs compiled by Farthing and Walker for their work.  And in fact if you use Cameca Peak Sight with an SX100R, you have those MACs too as that is the database it uses.

Karen Wright
« Last Edit: January 23, 2018, 08:29:27 am by John Donovan »

Philipp Poeml

  • Professor
  • ****
  • Posts: 186
Re: MAC calculated for Pu Mb
« Reply #3 on: September 01, 2014, 09:42:57 am »
Yes, in fact we are using the MACs from Farthing and Walker. At the time, we supplied Cameca with these MACs so that is what people find in the PeakSight software now.
« Last Edit: January 23, 2018, 08:29:37 am by John Donovan »

Probeman

  • Emeritus
  • *****
  • Posts: 1672
  • Never sleeps...
    • John Donovan
Re: MAC calculated for Pu Mb
« Reply #4 on: September 03, 2014, 06:09:22 pm »
We are dealing with quantification of MOX (PuUO2) samples. For Pu we are using its Mb line.
When calculating k-ratio and ZAF correction factors for Pu we have noticed a very low absorption correction for Pu. Further analysis showed that the MAC value of Pu Mb (3.53 keV) in Pu, not tabulated in CalcZaf but calculated using McMaster expression, was a factor 10 too large (5627) to what is expected. It is also inconsistent with the MAC value for Pu Ma which is tabulated at 650.8(LINEMU, Henke) and which has an energy (3.35 keV) only slightly smaller. There should not be any influence of an absorption edge. In fact, we made our own calculation of MAC value of Pu Mb in Pu and we found 571 which is consistent with Ma line.

Interestingly the McMaster expression does give consistent value for the MAC of PuMb in U (528). The problem seems specific to PuMb in Pu. Any ideas why?

It would be good to add tabulated value for Mb in CalcZaf (none available at the moment).

After discussion with Philipp Poeml I realized that you already have the ITU "blue book" values for the transuranium elements in the file USERMAC.TXT which is included in the CalcZAF.msi distribution. 

See the CalcZAF Xray menu and simply create a new USERMAC using the MAC30.DAT file as its “basis” and then update it (again from the Xray menu) using the USERMAC.TXT file from CalcZAF. The USERMAC.TXT file contains the database values from the Institute of Transuranium Elements. The reference is:

I. R. Farthing and C. T. Walker, Heinrich's Mass Absorption Coefficients for the K, L and M Lines, The European Commission, Institute for Transuranium Elements, Karlsruhe, Report K0290140,
1990
« Last Edit: January 23, 2018, 08:30:15 am by John Donovan »
The only stupid question is the one not asked!

John Donovan

  • Administrator
  • Emeritus
  • *****
  • Posts: 2259
  • Other duties as assigned...
    • Probe Software
Re: MAC calculated for Pu Mb
« Reply #5 on: January 23, 2018, 08:28:35 am »
This is an old thread but I should bring it up to date a bit.  First I wanted to note that our latest Probe for EPMA/Probe Image customer is the Belgian Nuclear Research Center (SCK-CEN) in Belgium on their SX100R:

http://probesoftware.com/Customer.html

This is a shielded instrument designed by Cameca specifically for analysis of nuclear materials. Philippe Poeml was instrumental in helping to set up this installation up in their controlled area laboratory, since access is difficult for nuclear sites such as this.  In addition Philippe provided for them his carefully prepared x-ray and MAC x-ray tables which have been optimized for actinide characterization. I have attached them as a ZIP file below (remember to login to see attachments).

To take advantage of Philipp's actinide optimized files (which will work for CalcZAF as well), simply unzip the files and copy them to the Probe for EPMA ProgramData folder (usually C:\ProgramData\Probe Software\Probe for EPMA) and overwrite the existing files. Note that normally Microsoft makes the ProgramData folder a hidden folder, so one might need to make this hidden folder visible using the View | Folder Options | View tab in Explorer.

Also, because the actinide MAC values are tabulated in the USERMAC.DAT file, one needs to select this as the default MAC file for calculations of nuclear materials.  This is easily done from the Analytical menu in PFE (or CalcZAF), but if one works with nuclear materials often, one can set the default in the Probewin.ini file (also in the ProgramData folder), by editing this line as seen here:

MACTypeFlag=7         ; default MAC file (1 = LINEMU, 2 = CITZMU, 3 = McMaster, 4 = MAC30, 5 = MACJTA, 6 = FFAST, 7 = USERMAC.DAT

Thank-you to Philipp Poeml for his expertise and hard work in developing these nuclear optimized x-ray and MAC files.
« Last Edit: January 23, 2018, 01:13:25 pm by John Donovan »
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

Ben Buse

  • Professor
  • ****
  • Posts: 363
Re: MAC calculated for Pu Mb
« Reply #6 on: May 02, 2018, 07:12:55 am »
Hi John,

Within empmac.dat, what are the empirical MACs "Xavier PENELOPE (2017)", is there a paper reference, do you know how they were calculated

Thanks

Ben

Probeman

  • Emeritus
  • *****
  • Posts: 1672
  • Never sleeps...
    • John Donovan
Re: MAC calculated for Pu Mb
« Reply #7 on: May 02, 2018, 08:10:40 am »
Hi John,

Within empmac.dat, what are the empirical MACs "Xavier PENELOPE (2017)", is there a paper reference, do you know how they were calculated

Thanks

Ben

Hi Ben,
Sorry, I do not have any reference to Xavier PENELOPE (2017) in my default empmac.dat file. Someone else must have edited that entry.
john
The only stupid question is the one not asked!

Ben Buse

  • Professor
  • ****
  • Posts: 363
Re: MAC calculated for Pu Mb
« Reply #8 on: May 02, 2018, 08:21:40 am »
Hi John,

This was in the nuclear.zip materials, maybe Philipp knows?

Ben

Probeman

  • Emeritus
  • *****
  • Posts: 1672
  • Never sleeps...
    • John Donovan
Re: MAC calculated for Pu Mb
« Reply #9 on: May 02, 2018, 09:03:48 am »
Hi John,

This was in the nuclear.zip materials, maybe Philipp knows?

Ben

Ah, yes.  Philipp probably knows since he put this zip together.
The only stupid question is the one not asked!

Philipp Poeml

  • Professor
  • ****
  • Posts: 186
Re: MAC calculated for Pu Mb
« Reply #10 on: May 27, 2018, 11:32:58 pm »
Yes, Philipp knows, but was not aware somebody wanted to know.
It is something I am working on with Xavier Llovet. It is data we are planning to publish this year.
So there will be a paper reference soon.