Fusion Neutronics Workshop

Half Day Conference Course

Neutron

Agenda

  • ⚛️ Why OpenMC
  • ⚙️ Installation options
  • 🧱 Making materials
  • 📈 Plotting cross sections
  • 🧩 Constructive Solid Geometry (CSG)
  • ☢️ Neutron source terms
  • ⚛️ Tritium Breeding Ratio simulation
  • 📊 Neutron spectra simulation
  • 🩻 Dose simulations
  • 🎯 Variance reduction
  • 🏗️ Converting CAD to DAGMC models
  • 🕸️ Unstructured mesh tally
  • 🔥 Depletion / activation / transmutation
  • ⏱️ Shutdown dose rate simulation
  • 🧑‍💻 How to get support

Simulation needed

Simulations are necessary when wanting to predict the nuclear response.

  • TBR
  • DPA
  • Gas production
  • Heating (prompt and decay)
  • Dose (prompt and decay)
  • Transmutation
  • Neutron / Photon flux spectra
  • Diagnostic responses
  • Magnet + insulator lifetimes

Why OpenMC

Code Python interface High performance / scalable Physics Permissive license CAD support Validated
MCNP ❓ ❌ ✅ ❌ ✅ ✅
Serpent ✅ ✅ ✅ ❌ ✅ ✅
FLUKA ❌ ❓ ✅ ❌ ❓ ❓
Geant4 ❓ ❓ ✅ ✅ ✅ ❓
PHITS ❓ ❓ ❌ ❓ ❓ ❓
OpenMC ✅ ✅ ✅ ✅ ✅ ✅

OpenMC community

The best way to predict the future is to create it

  • 179 contributors from across the globe have contributed to the source code.
  • Many others performing V&V, raising issues and providing feedback.

Why OpenMC

Code Python interface High performance / scalable Physics Permissive license CAD support Validated Ability to contribute
MCNP ❓ ❌ ✅ ❌ ✅ ✅ ❌
Serpent ✅ ✅ ✅ ❌ ✅ ✅ ❌
FLUKA ❌ ❓ ✅ ❌ ❓ ❓ ❌
Geant4 ❓ ❓ ✅ ✅ ✅ ❓ ❌
PHITS ❓ ❓ ❌ ❓ ❓ ❓ ❌
OpenMC ✅ ✅ ✅ ✅ ✅ ✅ ✅

Growing adoption

The relative ease of install, permissive MIT licensing and Python API have contributed to the rising popularity of OpenMC.

Installation options

Conda

  • Stable releases only
  • Older version
  • No Embree acceleration
  • Relatively easy
  • Contains MPI

Clone, compile, pip

  • Most complex, needs many commands
  • Install takes longer
  • Latest version (develop)
  • Optimised for local hardware
  • Highly customisable

Pip pre-built wheel

  • Does not contain MPI
  • Relatively easy
  • Up to date version

Install OpenMC and workshop deps

The workshop documentation is the best source of installation instructions.

https://fusion-energy.github.io/neutronics-workshop/docs/install_pip.html

Using Git

Download the workshop.

git clone https://github.com/fusion-energy/neutronics-workshop.git
cd neutronics-workshop

Download nuclear data

bash postBuild

Downloads the nuclear data cross sections and the decay data chain file to ~/nuclear_data.

Jupyter Lab / Notebook usage

  • Run a cell with shift and enter
  • Run all cells, or run in order
  • Add code cells and text cells
  • Shift and tab to get docstrings
  • Buttons for reset

Making materials

Materials can be made from:

  • Nuclides
  • Elements
  • Chemical formulas
  • Components

Making materials

Materials can be made from:

  • Nuclides
  • Elements
  • Chemical formulas
  • Components

Making materials

Materials can be made from:

  • Nuclides
  • Elements
  • Chemical formulas
  • Components

Making materials

Materials can be made from:

  • Nuclides
  • Elements
  • Chemical formulas
  • Components

Task 1 - Making materials

In Jupyter Lab / Notebook try

Plotting cross sections

  • nuclide → microscopic cross section
  • element → microscopic cross section
  • material → macroscopic cross section

Scores such as heating, heating-local (eV-barn) and damage-energy (eV-barn) are also available.

Task 2 - Plotting cross sections

In Jupyter Lab / Notebook try

Bonus task - nuclear data

Try plotting the neutron energy distribution of an (n,2n) reaction.

github.com/fusion-energy/neutronics-workshop/issues/228

Making CSG geometry

The simplest region is a single surface and a cell defined as below (-) that surface.

import openmc

surface_sphere = openmc.Sphere(r=10.0)
inside_sphere = -surface_sphere
cell_sphere = openmc.Cell(region=inside_sphere)
cell_sphere.fill = steel

Making CSG geometry

Cells can also be constrained by multiple surfaces. This example is above (+) one surface and (&) below (-) another.

import openmc

surf_sphere1 = openmc.Sphere(r=10.0)
surf_sphere2 = openmc.Sphere(r=20.0)
between_spheres = +surf_sphere1 & -surf_sphere2
cell_sphere = openmc.Cell(region=between_spheres)
cell_sphere.fill = steel

Making CSG geometry

The outermost surface of the model should have a boundary_type set to "vacuum" to indicate that neutrons should not be tracked beyond this surface.

import openmc

surf_sphere = openmc.Sphere(r=10.0, boundary_type="vacuum")
region_inside_sphere = -surf_sphere
cell_inside_sphere = openmc.Cell(region=region_inside_sphere)
cell_inside_sphere.fill = steel

Visualizing CSG geometry

OpenMC supports classic CSG surface families, and there are many built in ways to view the geometry.

  • Slice plots: model.plot(), geometry.plot(), cell.plot(), region.plot()
  • Voxel plots: solid and ray traced (RayTracePlot)

There are also extra packages for viewing geometry:

Task 3 - making and visualizing CSG

In Jupyter Lab / Notebook try

Neutron birth energy

  • Deuterium (D) and Tritium (T) nuclides fuse, emitting an alpha particle and a neutron.
  • The neutron energy is dominated by the binding-energy release, with the spread due to the kinetic energy of the incident D and T nuclei.
  • Our neutron travels away from the fusion event with 14.1 MeV of energy (ignoring the magnetic field).

Plot created with fusion-neutron-utils

Particle source

  • Make a simple point neutron source
  • Make a ring neutron source
  • Visualize the source
  • Visualize particle tracks
  • Make a gamma source

Tasks 4 and 5 - Define a neutron source

In Jupyter Lab / Notebook try

Simulation settings

A minimal settings object for fusion simulations.

Settings objects also accept variance reduction configurations, weight windows and volume calculation settings.

Link to OpenMC docs

settings = openmc.Settings(
    batches=10,
    particles=10000,
    photon_transport=True,
    run_mode='fixed source',
    source=my_source,
)

Tally scores

A tally can record different scores. Many scores are available:

  • H3-production
  • MT 16 (n,2n)
  • (n,total)
  • heating / heating-local
  • damage-energy
  • flux

See the documentation for more information.

Cell filters

Tally filters are used to refine how and what a tally records.

The CellFilter tells the tally to record on a specific cell.

openmc.CellFilter([cell_1])

Task 6 - TBR tally

In Jupyter Lab / Notebook try

Post processing tricks

Modify task 6 so that you can access the tally.mean using apply_tally_results.

model.run(apply_tally_results=True)

openmc.model.Model documentation

Energy and particle filters

Tally filters refine how and what a tally records.

  • The EnergyFilter tells the tally to record the score on provided energy bins.
  • The ParticleFilter can be used to record photons or neutrons.
openmc.EnergyFilter(np.linspace(0, 15e6, 719))
openmc.ParticleFilter(['neutron'])

Task 7 - neutron spectra tally

In Jupyter Lab / Notebook try

Energy function filters

The EnergyFunctionFilter tells the tally to multiply by a factor when scoring the particle, for example to convert flux to dose.

energy_function_filter = openmc.EnergyFunctionFilter(
    energy=energy_bins,
    y=dose_coeffs,
    interpolation="cubic",
)

Task 8 - cell dose tally

In Jupyter Lab / Notebook try

Meshes and mesh filters

Two options for making the mesh, then use it in the tally as another filter type.

mesh = openmc.RegularMesh().from_domain(
    domain=geometry,
    dimension=[100, 100, 100],
)
mesh_filter = openmc.MeshFilter(mesh)
flux_tally.filters = [mesh_filter]

Task 9 - mesh dose tally

In Jupyter Lab / Notebook try

Variance reduction

OpenMC supports:

  • Implicit capture
  • Weight windows
  • Importance values

OpenMC can generate weight windows using:

  • the MAGIC method
  • Random Ray + FW-CADIS

Task 10 - weight windows

In Jupyter Lab / Notebook try

DAGMC

DAGMC lets OpenMC transport particles through CAD geometry converted to a surface mesh.

CAD overview

Today simulation on CAD in OpenMC uses:

  • DAGMC for surface mesh geometry
  • DAGMC for volume mesh (unstructured mesh)
  • LibMesh for surface mesh
  • LibMesh for volume mesh (unstructured mesh)

In the future, XDG will offer another option.

Task 11 - convert CAD to DAGMC

In Jupyter Lab / Notebook try

Visualizing CAD to DAGMC

In addition to slice plots, the OpenMC plotter and the OpenMC geometry plot, there are two options specifically for DAGMC geometry viewing:

Task 12 - Simulating with DAGMC

In Jupyter Lab / Notebook try

Unstructured meshes

Unstructured meshes are currently available via two libraries.

  • MOAB: tet mesh only, file is h5 or vtk format
  • LibMesh: tet, hex and mixed meshes, file is exodus format

In the future XDG will offer another option.

Task 13 - Simulating with unstructured mesh

In Jupyter Lab / Notebook try

Activation pathways

  • Activation reactions can result in stable and unstable isotopes.
  • Neutron induced reactions tend to not increase the proton number.
  • Irradiation of 5 stable Zr isotopes produces 20 unstable isotopes via direct reactions.

Decay pathways

  • Unstable isotopes decay.
  • Beta- decay transmutates the nuclide +1 proton, -1 neutron, and gammas are also emitted.
  • Unstable isotopes can have multiple decay routes, each with a different probability (branching ratio).
  • Each has a characteristic half life.
  • Stable and natural abundances are not the same thing.

Task 14 - transport free transmutation

In Jupyter Lab / Notebook try

Task 15 - transmutation with transport

In Jupyter Lab / Notebook try

Radioactive decay

  • Decay can produce different particles: Beta+, Beta-, alpha, neutron, gamma.
  • We are unlikely to see neutron decay products in fusion (with the exception of Be).
  • The range of alpha and beta is typically short and often blocked within the material itself.
  • The primary interest is the gamma.

Task 16 - transport with decay gammas

In Jupyter Lab / Notebook try

Getting support with OpenMC

Perhaps the best part of OpenMC is the community.

Other features we have not covered

  • R2S shutdown / residual dose simulations
  • Tally triggers
  • OpenMC lib C++ interface
  • Volume calculations
  • Track simulations
  • Coupled depletion

Many more topics are covered in the neutronics workshop.

Other OpenMC courses

  • NEA databank
  • Conferences
  • OSSFE

We are also considering making a 2 day fusion neutronics analysis workshop depending on demand and availability.

How did we do

  • ⚛️ Why OpenMC
  • ⚙️ Installation options
  • 🧱 Making materials
  • 📈 Plotting cross sections
  • 🧩 Constructive Solid Geometry (CSG)
  • ☢️ Neutron source terms
  • ⚛️ Tritium Breeding Ratio simulation
  • 📊 Neutron spectra simulation
  • 🩻 Dose simulations
  • 🎯 Variance reduction
  • 🏗️ Converting CAD to DAGMC models
  • 🕸️ Unstructured mesh tally
  • 🔥 Depletion / activation / transmutation
  • ⏱️ Shutdown dose rate simulation
  • 🧑‍💻 How to get support

![bottom-right](logo)