Hi all,
I have a work related problem, which is calculating the dose rate from a certain waste barrel with cemented nuclear waste at a certain distance in order to estimate the dose a worker would be exposed to if he/she has to handle the barrel. I already succeeded in reproducing the gamma detector from the sample files.
As I understand, you need conversion factors for each energy group to convert from tallies into, say muSv/h. There is a table in ICRP74, for example, which has some numbers for H*(10)/flux in [Sv*cm^2] (page 179).
I know there was a similar topic on here, but the linked sample files are 404 now.
Can someone point me in the right direction how to tackle this problem with OpenMC?
I want to preface this by saying that I have a minor in Health Physics, and I have taken a class in my graduate studies covering external dosimetry in which I used MCNP to calculate doses, but that I am not yet an engineer and that nothing said here should be used to directly to inform policy related to dose calculations to individuals and that I take no responsibility for actions being completed using this guidance (I don’t want liability). This is just for your better understanding, and you should research the topic further.
I would recommend looking into specific codes for this application such as Visual Monte Carlo (Although i don’t know if this is validated you might want to look into it) as they include a lot of the framework you would desire such as reference phantoms and whatnot.
First, I would always complete a simplified model to evaluate dose as you might find the values are so low that they are insignificant even if you considered a more in depth model. Be conservative with all assumptions here. The important tally for dose is the absorbed energy as you want to find the units Gray, which is J/kg, which you can then calculate the equivalent dose rate and effective dose rate. I used to use MCNP, and the tally we would use was the F6 tally which gave units Mev/g, I think the equivalent tally in openmc is the heating tally, although I would verify this. Depending on geometry of your problem you might be able to consider it as a point source, but since you said they are handling it I would assume you couldn’t do this, and I would model the source using the geometry of the waste fuel barrel. The first sim I would probably do is look at the contact dose at the surface of the barrel as this would be the highest possible dose, and if its within your regulatory and operational bounds you won’t need to calculate the rest. I used to do this through the software Microshield, although once again I don’t know if this is validated.
If you determine it to be significant and you want to continue using OpenMC, I would then model a human phantom using the ICRP reference phantoms (female and male), at different expected positions that a worker would be at during work/transport. Try to make your source as accurate as possible, especially with any shielding (make sure to consider self-shielding by making the source in the actual material of the nuclear waste, if the waste has multiple materials and cannot be easily determined the exact composition make sure to use the least attenuation material for a conservative estimate). Once you have your source defined, and your phantoms completed, I would begin the simulation measuring the heating within the relevant organs, as well as, the overall total heating to the entire person. You will need to make sure you are in neutron-photon transport and not just neutron (if you only define a neutron source you won’t track photons, I would just explicitly state photon transport always to prevent wrong assumptions), and filter the tallies specific for each type of radiation present (photons, beta, neutrons). Depending on your amount of shielding around the waste, betas may be ignored. For neutrons you will need to filter by energy of the source particle as they have varying radiation weighting factors for different neutron energies, the rest of the particles do not require this. After you get the heating within the total body, you convert to units of gray (j/kg), and then convert to Sv using the radiation weighting factor. These will all be relative to a unit of time, and therefore you will need to do a best estimate of the expected exposure time to multiply by to get the overall equivalent and effective dose to the person/organs. The equivalent does can be compared to regulatory/occupational limits for a year, and the effective dose compared to regulatory occupational limits to specific organs over a year.
This was all assuming that the radiation being emitted would not have acute effects (deterministic) as it was not significantly high enough for these effects, and instead lies within the stochastic range. If this is not the case the regulatory limits differ, and you will need to evaluate against these. As a heads up, I am a little rusty with the process so I would verify the specifics of everything, but i hope this can be used for guidance. Personally I would look into software specific to dose calculations as there are a lot out there with phantoms and whatnot already modeled that use either/both deterministic simulation and stochastic simulations to determine dose rates instead of doing all the work in OpenMC, but if you must use OpenMC this might help a little with understanding the problem. If you want to go over things in more depth, you can message me directly and we could set up a meeting at some point to discuss this stuff, I might also still have course materials related to the subject I could share.
ICRP74 is good to get familiar with for this topic, although I believe ICRP 116 is the newer version. I would also look at ICRP 92 for dose conversion coefficients, but there is also a command in OpenMC for these coefficients,
openmc.data.dose_coefficients¶
and there was a support topic related to this problem but instead converting from flux instead of deposited heat here,
1 Like
Hi, thank you so much for taking your time replying to my topic. We are also using MicroShield here, but since I am still contract staff, I have not the necessary rights to use it. I will look further into it, currently reading ICRP, and will soon give feedback how it goes.