Win X-ray

[Ben Buse]

Hi

Does any one know if you can do multiple layers in win x-ray - I persumed you could but when I loaded it number of regions was greyed out - was it a feature never added.

I realise there are many programs I could do it in - Casino, MC-Xray, dtsa2, penepma etc.

MAny thanks

Ben

[Probeman]

Hi

Does any one know if you can do multiple layers in win x-ray - I persumed you could but when I loaded it number of regions was greyed out - was it a feature never added.

I realise there are many programs I could do it in - Casino, MC-Xray, dtsa2, penepma etc.

MAny thanks

Ben

Hi Ben,
What are you trying to do?

Calculate concentrations from intensities or intensities from concentrations or are you trying to generate a full spectrum from a given composition?
john

[hdemers]

Hi Ben,

winxray is only for bulk sample.

Like John mention, it depend what you try to do. Only characteristic x-ray, CASINO v2 work well. Full x-ray spectrum with complex geometry mcxray can do it. In the last version available for download I have added examples of geometry to help the user build their one sample.
http://montecarlomodeling.mcgill.ca/download/download.html

Contact me if you need help.

Regards,
Hendrix

[Probeman]

Hi Ben,

winxray is only for bulk sample.

Like John mention, it depend what you try to do. Only characteristic x-ray, CASINO v2 work well. Full x-ray spectrum with complex geometry mcxray can do it. In the last version available for download I have added examples of geometry to help the user build their one sample.
http://montecarlomodeling.mcgill.ca/download/download.html

Contact me if you need help.

Regards,
Hendrix

Hi Hendrix,
Are MCXray and Casino 3 the same application?  Or does MCXray replace Casino 3?

Also, this is the link I have on my Technical web page (http://probesoftware.com/Technical.html)

http://www.memrg.com/#!programs-download/c143t

Should I replace this link with the one you have above?
john

[Mike Matthews]

winxray is only for bulk sample.

Hi Hendrix,

Can you only have a single element bulk sample? I can't seem to get it to accept a second element - I can increment the number of elements and enter the data but clicking 'OK' doesn't accept the updated configuration.

[Ben Buse]

Hi Mike,

The trick is when you change the number of elements to > 1. The default weight fraction and atomic fraction is 1 for each new element which is clearly wrong - when the total must equal 1.

eg. Al 1 Oxygen 1 atomic fraction - so need to change it to Al0.5 Oxygen 0.5 or what ever hitting enter - then it will accept it - again I think the total must equal 1 not 0.97 or whatever, although I might be wrong.

Ben

[Ben Buse]

Hendrix, John

Thanks for clarifying that. Reasons I was interested in win x-ray - bse appeared better - but as Hendrix points out on seperate post can use casino v3. Also win x-ray simulate charge in insulating samples - but this assumes a grounded coat i.e. charge at surface = 0: so don't need to add it as geometry. - although I surpose you can't model x-rays passing through gold coat vs carbon coat under electric field.#

Ben

[Probeman]

Hendrix, John

Thanks for clarifying that. Reasons I was interested in win x-ray - bse appeared better - but as Hendrix points out on seperate post can use casino v3. Also win x-ray simulate charge in insulating samples - but this assumes a grounded coat i.e. charge at surface = 0: so don't need to add it as geometry. - although I surpose you can't model x-rays passing through gold coat vs carbon coat under electric field.#

Ben

Hi Ben,
The Penepma GUI in Standard prints out the BSE coefficient here (see the BSE Frac. field in the lower right of the GUI):

http://probesoftware.com/smf/index.php?topic=202.msg1519#msg1519

Note also that you can optimize for BSE production using the 2nd "Optimize Production for..." option near the top.

The Penepma GUI also supports a bi-layer model (film and substrate).  And you can even do two layers on a substrate using the trilayer model, but you'll need to tweak the Penepma input as described here:

http://probesoftware.com/smf/index.php?topic=57.msg478#msg478

[Ben Buse]

Hi Ben,
The Penepma GUI in Standard prints out the BSE coefficient here (see the BSE Frac. field in the lower right of the GUI):

http://probesoftware.com/smf/index.php?topic=202.msg1519#msg1519

Note also that you can optimize for BSE production using the 2nd "Optimize Production for..." option near the top.

The Penepma GUI also supports a bi-layer model (film and substrate).  And you can even do two layers on a substrate using the trilayer model, but you'll need to tweak the Penepma input as described here:

http://probesoftware.com/smf/index.php?topic=57.msg478#msg478

John

That's good to know and bse optimize sounds good. Plus its easy to do films in Penepma with the pre-made files (bilayer) shipped with the standard version as you point out. So all I now need to do is run a test and see how many showers are needed to get the right answer.

John, - I'm intrigued how do you optimize it for bse - what do you modify and how does thsi make it better for bse.

Also regarding the bilayer files, just a word of caution I discovered when I last used them - I don't know much about penepma but I understand from the penepma.pdf that the DSMAX (maximum step length) should be a 1/10th or less of a thin body. The default (in bilayer.in) is 1nm so for bilayer files of < 10nm would need editing I guess - but this is easy in the penepma gui by clicking on "edit input file" button. If this is important to make it obvious would it be worth shipping it with different bilayer.in for 1,2 and 5nm or maybe commenting the bilayer.in file or bilayer_readme.txt file. Is there anything else that's important.



Thanks

Ben

[Probeman]

Hi Ben,
The Penepma GUI in Standard prints out the BSE coefficient here (see the BSE Frac. field in the lower right of the GUI):

http://probesoftware.com/smf/index.php?topic=202.msg1519#msg1519

Note also that you can optimize for BSE production using the 2nd "Optimize Production for..." option near the top.

The Penepma GUI also supports a bi-layer model (film and substrate).  And you can even do two layers on a substrate using the trilayer model, but you'll need to tweak the Penepma input as described here:

http://probesoftware.com/smf/index.php?topic=57.msg478#msg478

John, - I'm intrigued how do you optimize it for bse - what do you modify and how does thsi make it better for bse.
 DSMAX
Also regarding the bilayer files, just a word of caution I discovered when I last used them - I don't know much about penepma but I understand from the penepma.pdf that the DSMAX (maximum step length) should be a 1/10th or less of a thin body. The default (in bilayer.in) is 1nm so for bilayer files of < 10nm would need editing I guess - but this is easy in the penepma gui by clicking on "edit input file" button. If this is important to make it obvious would it be worth shipping it with different bilayer.in for 1,2 and 5nm or maybe commenting the bilayer.in file or bilayer_readme.txt file. Is there anything else that's important.

Hi Ben,
Good point on the bi-layer DSMAX parameter.  I've only modeled thicker films so this has not been an issue for me, but I should probably add a check for that in the GUI code.

As to BSE optimization just check the "template" files. The Cu_Cha.in file is the default for characteristic x-ray optimization. The BSE optimization template is Cu_back.in. You can compare these two files to see the differences.  You might want to refer to the Penepma documentation.
john

Edit: I just uploaded v. 11.4.8 which warns the user about this issue.

[hdemers]

Hi Hendrix,
Are MCXray and Casino 3 the same application?  Or does MCXray replace Casino 3?

Also, this is the link I have on my Technical web page (http://probesoftware.com/Technical.html)

http://www.memrg.com/#!programs-download/c143t

Should I replace this link with the one you have above?
john

Hi John,
Your link are OK, I just link directly to the download page for the programs. http://www.memrg.com is the new website for Gauvin's group.

Now a brief history of electron trajectory simulation Monte Carlo software from Quebec.

• CASINO 1 was a dos and unix version available for free. It was created during Pierre Hovington and Dominique Drouin PhD at Sherbrooke University. The code was base on program from Prof Gauvin.
  
• CASINO 2 was the same code, but with a windows GUI and created  by a student (Alexandre Real Couture) under Dominique Drouin as a new Professor at Sherbrooke University. Main features: simple geometry, characteristic x-ray, nice interface.
  
• 1 or 2 years later (I am not sure of the release date), I (Hendrix Demers) create winxray program  for my master thesis and make it available. Main features: x-ray spectrum, only bulk sample, simple charging model with coating at the surface.
  
• Another student Pierre Michaud at McGill University created a 3D sample geometry program (mcxray) with full x-ray spectrum. It is more or less replace winxray.
  
• In the meantime, CASINO 3D (v3) was developed at Sherbrooke University by a long list of students. It was a complete rewrite if the code and, sadly, the x-ray simulation was dropped.
  

So mcxray and casino v3 are not the same software. If you want to do x-ray simulation, use mcxray; if you want to simulate electron trajectories (BSE, SE, STEM) use CASINO v3.

Regards,
Hendrix
PS: Keep in mind, I have only mention the program from Quebec, i.e., Gauvin and Drouin groups. During the same period, a lot of development was done by other groups: like PENELOPE or DTSA and other.

[hdemers]

Hendrix, John

Thanks for clarifying that. Reasons I was interested in win x-ray - bse appeared better - but as Hendrix points out on seperate post can use casino v3. Also win x-ray simulate charge in insulating samples - but this assumes a grounded coat i.e. charge at surface = 0: so don't need to add it as geometry. - although I surpose you can't model x-rays passing through gold coat vs carbon coat under electric field.#

Ben

Hi Ben,
For winxray, I use a simple model of charging with a thin coating at the surface. The main results were published in this paper:
Demers, H. & Gauvin, R. X-ray Microanalysis of a Coated Nonconductive Specimen: Monte Carlo Simulation Microscopy and Microanalysis, 2004, 10, 776-782
But if you can read french I can send you my master thesis with more detail.

If you want to model insulator with the effect of Au versus C coating. You can just add the x-ray absorption from the layer to the results given by winxray (if you are at high accelerating voltage, this approximation should be good). Also PENELOPE can simulate magnetic and electric field, but I never try it. If you are interested look at Sorbier PhD thesis.

Regards,
Hendrix

[Ben Buse]

Hi Hendrix,

Thanks. Do you have a copy of Sorbier PhD thesis - is it in english (my French is terrible)?

Ben