Monte Carlo Simulation of Nuclear Physics

PRA, BeqMoni, Theremino, Fitzpeaks and beyond!
Post Reply
Rob Tayloe
Posts: 110
Joined: 10 Nov 2020, 12:00
Contact:

Monte Carlo Simulation of Nuclear Physics

Post by Rob Tayloe » 12 Aug 2023, 05:37

At some time in the past there was an interest in performing monte carlo simulations on nuclear physics. Such simulations can be useful in the design of radiation shielding, examining the performance of radiation detectors, and understanding how fissile and other nuclear systems might operate. These programs simulate particle behavior through the use of a large number of random numbers whose distribution functions are controlled by physics such as interaction probabilities (referred to as cross-sections) and system geometry.

In the US, a program from the Los Alamos National Laboratory (LANL) called MCNP (for monte carlo neutral particle) is quite widely used. The distribution of MCNP (outside of LANL) is through the Radiation Safety Information Computational Center (RSICC) at the Oak Ridge National Laboratory (ORNL). If one is associated with a US university nuclear program or certain nuclear programs outside of universities the software is free. For others there is a fee and for everyone there is a licensing process.
https://mcnp.lanl.gov/
https://rsicc.ornl.gov/default.aspx

Folks at the European Nuclear Research Center (CERN) have developed a monte carlo nuclear simulation program called GEANT4. I believe that GEANT4 can be freely downloaded and used.
https://geant4.web.cern.ch/

I have not used GEANT4. I did cover use of MCNP for radiation shielding when I taught graduate nuclear engineering courses at the Ohio State University. We spent a number of weeks covering the theory and basic operation of the MCNP code and made some calculations of theoretical and real (meaning experimental) conditions. Following is a link to an unclassified, open-source report discussing a past activity in which I was involved where MCNP was used to determine the response function for a neutron detector.
https://www.osti.gov/biblio/471366/

There is a fairly large amount of material that one must know in order to effectively use these programs. There are YouTube videos that purport to cover monte carlo nuclear simulations, MCNP, and GEANT4. I have not watched all of these videos. I have begun watching some of them. Links are provided below.

Series on monte carlo nuclear simulations -
https://www.youtube.com/watch?v=lO6vUfg ... DkpQLpOtlb

Series on MCNP -
https://www.youtube.com/watch?v=qyFUH8Y ... Up&index=1

Series on GEANT4
https://www.youtube.com/watch?v=Lxb4WZy ... 9vqg4KXeVL


Rob Tayloe
Posts: 110
Joined: 10 Nov 2020, 12:00
Contact:

Re: Monte Carlo Simulation of Nuclear Physics

Post by Rob Tayloe » 26 Aug 2023, 01:21

If one wishes to perform monte-carlo or other radiation transport calculations it is very useful to be able to access appropriate data on compositions and mixtures for materials (including radiation detectors). Links are provided for compendium documents and calculators -

https://compendium.cwmd.pnnl.gov/

https://www.pnnl.gov/publications/compe ... modeling-1

May 14, 2021
Report
Compendium of Material Composition Data for Radiation Transport Modeling

View/Download report
Abstract
In 2011, Pacific Northwest National Laboratory (PNNL) produced a document known as the Materials Compendium, or Compendium of Material Composition Data for Radiation Transport Modeling, PNNL 15870, Rev. 1, that contains material information useable for modeling purposes for properties of 372 materials. This information is used in several modeling programs used by the radiological/nuclear community, though it is primarily tailored for the Monte-Carlo-N-Particle code produced by Los Alamos National Laboratory. This new document Revision 2 includes a complete review and update of all materials data and references, addressing discrepancies and changes in materials data or references that have occurred since the first revision, an additional 40 materials have been added, primarily newer detector materials developed since the last revision, and isotopic specificity.
Citation
Detwiler R.S., R.J. McConn, T.F. Grimes, S.A. Upton, and E.J. Engel. 2021.
Compendium of Material Composition Data for Radiation Transport Modeling
Richland, WA: Pacific Northwest National Laboratory.

User avatar
ColoRad-o
Posts: 33
Joined: 18 Oct 2019, 08:27
Location: Colorado, USA
Contact:

Monte Carlo for calculating gamma spectrometer internal efficiency, and data for a 2"x2" NaI:Tl detector

Post by ColoRad-o » 31 Oct 2023, 10:43

Hi Rob! I'm sorry to have overlooked your postings on Monte Carlo for so many months.

I had thought about using MCNP, or even Geant4 in order to predict, from the detailed geometry and properties of a simple NaI:Tl scintillator/PMT detector (like the one Steven sells here), the actual gamma ray detection efficiency. This is at least partly because I don't want to pay for well-calibrated gamma sources. I found the hoops one needs to jump through--even for a retired academic--to get MCNP access to be daunting and annoying. FLUKA/FLAIR looked promising as well two years ago when I last checked.. InterSpec provides *support* for several ways to parametrize internal efficiency, FYI.

Meanwhile, I stumbled on what appeared to be more-or-less what I needed (to the extent that any 2"x2" NaI:Tl detector with PMT is similar). Is published in the article
Development and calibration of a real-time airborne radioactivity monitor using gamma-ray spectrometry on a particulate filter, R. Casanovas, J. J. Morant, M. Salvadó, IEEE Transactions on Nuclear Science, January 2014. Let me know if you want a CSV file of their data points, digitized from their figure.

Best wishes--DMW
Attachments
2x2NaIScintillEffic.pdf
(38.38 KiB) Downloaded 92 times
D. M. Wood, retired physics professor
Arvada, Colorado (USA)
SAFECAST member (bGeigie Nano)

jneilson
Posts: 32
Joined: 21 May 2024, 05:08
Location: Oxfordshire, UK
Contact:

Re: Monte Carlo Simulation of Nuclear Physics

Post by jneilson » 25 May 2024, 05:09

Rob Tayloe wrote:
12 Aug 2023, 05:37
At some time in the past there was an interest in performing monte carlo simulations on nuclear physics.
Indeed, but not just some time ago - Monte Carlo radiation transport modelling has been and still is very much a great tool for this kind of work, and is regularly used in industry, academia and research.

I'd add another couple to your list of MC codes; Serpent (https://serpent.vtt.fi/serpent/) and OpenMC (https://docs.openmc.org/en/stable/). I'd note OpenMC, being open source, might be a good route for the amateur crowd as there's no access and licensing issues to work through - I've had students download and run it pretty much straight away; although we did find the documentation is very much centred around k-eigenvalue neutron criticality modelling rather than fixed-source photon modelling.

Otherwise, your comprehensive list of links is great - especially the PNNL Materials Compendium which is one of my go-to resources as well. It is focused on materials you're likely to find around a reactor though, so for some of the more random materials you might want to look at it can be a bit limited. When something isn't included there, I often find myself searching for Materials Safety Data Sheets to find out the chemical constituents of certain materials so I can model them.
ColoRad-o wrote:
31 Oct 2023, 10:43
I had thought about using MCNP, or even Geant4 in order to predict, from the detailed geometry and properties of a simple NaI:Tl scintillator/PMT detector (like the one Steven sells here), the actual gamma ray detection efficiency. This is at least partly because I don't want to pay for well-calibrated gamma sources.

Meanwhile, I stumbled on what appeared to be more-or-less what I needed (to the extent that any 2"x2" NaI:Tl detector with PMT is similar)...
Of course, it always depends what's "good enough" for your use case. My issue with Monte-Carlo calibrations is that they generally don't have the information to take into account the individual uniquenesses of each exact detector - little variations of densities, degradation over time, chamfering of the corners of the crystal making a smaller detection volume, varying thicknesses of things like casings, endcaps, dead-layers, light-proofing etc. These all can have an effect on efficiency, so when we do Monte-Carlo efficiency modelling in industry and need high accuracy and precision, we typically still need to use calibrated sources at some point to benchmark the model and fine-tune the model parameters until simulation consistently matches the experimental measurements.

Rob Tayloe
Posts: 110
Joined: 10 Nov 2020, 12:00
Contact:

Re: Monte Carlo Simulation of Nuclear Physics

Post by Rob Tayloe » 18 Jun 2024, 07:36

I have spent a little time playing with a demo version of the MCC-MT monte-carlo program. The demo version (for a 30 day trial) can be requested from -
https://www.tals.eu/mcc-mt

The program developers are in Finland. In the short while that I've been playing with the program (using supplied examples) I have changed the size of the NaI detector (to be consistent with something that I have). And I have changed the source to Cs-137. The demo version limits the number of "histories" that one can track. I thought that my brief fooling around illustrated the ease with which problems can be modified. I like the graphics showing the particle tracks. The figure below shows a some of the these tracks. The source is a Cs-137 point above the 25 x 25 mm detector. Alas the quoted price is very high for my purposes (4200 euros) - I don't know if this is a one-time charge or an annual subscription.
det-src-cs137-01.jpg

jneilson
Posts: 32
Joined: 21 May 2024, 05:08
Location: Oxfordshire, UK
Contact:

Re: Monte Carlo Simulation of Nuclear Physics

Post by jneilson » 22 Jun 2024, 06:44

Thanks for adding that one, Rob, that's a package I haven't encountered before. The interface looks stylistically quite close to Canberra/Mirion's ISOCS/LabSOCS software though, which does similar things for modelling detectors and calculating detector-item geometry efficiencies, and is also somewhat prohibitively expensive for hobby use. It's quite a bit more "black box" code that hides what is actually going on; I think a lot of the calculations it does are deterministic rather than Monte-Carlo, but I think there is some Monte Carlo involved, especially in the initial detector characterisation they do - I should probably read the manuals again.
Image
Image

AlexF
Posts: 1
Joined: 26 Jun 2024, 19:38
Contact:

Re: Monte Carlo Simulation of Nuclear Physics

Post by AlexF » 26 Jun 2024, 19:52

MCC-MT software uses the Monte Carlo method to play out particle interaction events. To speed up the simulation, the calculation is parallelized into threads processed by PC processor cores. When using a multi-core processor, the calculation speed can be increased many times. I think ISOCS/LabSOCS takes a different approach.

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests