VORPAL for Simulating

1
VORPAL for Simulating RF Breakdown
Kevin Paul kpaul_at_txcorp.com
VORPAL is a massively-parallel, fully
electromagnetic particle-in-cell (PIC) code,
originally developed for laser-plasma simulation.
Since it's creation in 2004, VORPAL has expanded
its capabilities to include electrostatics,
cross-section-based particle-particle
interactions, hybrid particle-fluid modeling, and
a variety of numerical models for everything from
field ionization, impact ionization, secondary
electron emission, field emission, and
particle-impact heating.
Fermilab MuCool RF Workshop III 7 July 2009
2
Tech-X Corporation Projects

• Breakdown Phase II
• Seth Veitzer
• July 2008 July 2010
• Developing VORPAL to do 3D simulations of RF
  breakdown
• Built off of a Phase I project using OOPIC
  (2D/r-z)
• eSHIELD Phase I
• Me
• July 2009 March 2010
• More VORPAL development to test magnetic
  insulation
• Will couple small-scale with large-scale
  simulations

Fermilab MuCool RF Workshop III 7 July 2009
3
VORPAL Versatile Plasma Simulation Code

• Technical Features
• Object-oriented C
• 1D/2D/3D Massively Parallel Scaling to 10,000
  Processors
• Compressed Binary Data Formatting (HDF5)
• Mac OS X / Microsoft Windows / Linux
• Multi-physics Capability
• Kinetic Plasma Model
• Field Impact Ionization
• Field Secondary Emission
• Hybrid Particle-Fluid Modeling
• Electrostatic Electromagnetic

• Uses
• Laser wake-field accelerators
• Electron cooling
• Photonic Band Gap Devices
• RF Heating in Fusion Plasmas
• Breakdown in Microwave Guides
• Simulation of Ion Sources Penning Sources
• Modeling of Plasma Thrusters
• Availability
• Consulting
• Purchase
• SBIR/STTR Collaboration
• Web interface (In development!)

Fermilab MuCool RF Workshop III 7 July 2009
4
Electrostatic Particle-in-Cell SimulationOne
Simulation Time Step
Initialization Steps...
Fields defined and initialized on a grid Ei, Bi
Particle positions velocities
initialized xa, va
Particles accelerated by the fields v'a
Particles moved based on new velocity x'a
One Time Step
New fields computed from charges Ei
Charge deposited on the grid ?i
Fermilab MuCool RF Workshop III 7 July 2009
5
Electromagnetic Particle-in-Cell SimulationOne
Simulation Time Step
Initialization Steps...
Fields defined and initialized on a grid Ei, Bi
Particle positions velocities
initialized xa, va
Particles accelerated by the fields v'a
Particles moved based on new velocity x'a
One Time Step
New fields computed from old fields E'i, B'i
Currents deposited on the grid Ji
Fermilab MuCool RF Workshop III 7 July 2009
6
Electromagnetic Particle-in-Cell SimulationOne
Simulation Time Step
New particles added (lost removed) xa, va
Particles accelerated by the fields v'a
Particles moved based on new velocity x'a
One Time Step
Collisions and interactions computed
New fields computed from old fields E'i, B'i
Currents deposited on the grid Ji
Fermilab MuCool RF Workshop III 7 July 2009
7
Electromagnetic Particle-in-Cell SimulationOne
Simulation Time Step
This is where all the interesting physics for RF
breakdown takes place!!!
New particles added (lost removed) xa, va
Particles accelerated by the fields v'a
Particles moved based on new velocity x'a
One Time Step
Collisions and interactions computed
New fields computed from old fields E'i, B'i
Currents deposited on the grid Ji
Fermilab MuCool RF Workshop III 7 July 2009
8
RF Breakdown Physics What must be modeled?

• Field emission of electrons from conductor
  surfaces
• Secondary emission of electrons from conductor
  surfaces
• Sputtering
• Neutral Desorption
• Field-induced ionization (Tunneling ionization)
• Impact ionization
• X-ray production from electron impact on
  conductor surfaces
• Surface heating due to particle impact
• Surface deformation due to melting
• Radiative cooling of ions

Fermilab MuCool RF Workshop III 7 July 2009
9
Physics Models in VORPAL/TxPhysics What can
VORPAL do now?

• Fowler-Nordheim model for field emission from
  assumed asperity
• Jensen model for field, thermal, and
  photo-induced electron emission
• Rothard model for ion-induced secondary electron
  emission (depends strongly on nuclear stopping
  power of material)
• Furman-Pivi (LBNL) model for electron-induced
  secondary electron emission
• Yamamura model for sputtering (nuclear stopping
  dependent threshold model)
• Molvik model for neutral desorption (akin to
  Rothard model)
• Tunneling ionization rates for various materials
  from Keldysh
• Parameterized impact ionization, excitation, and
  recombination cross sections for electrons and
  ions
• Diagnostics for recording energy deposited in
  absorbing boundaries
• Coronal model for computing radiated power by
  ions in a plasma (a diagnostic, no radiation
  transport)

Fermilab MuCool RF Workshop III 7 July 2009
10
VORPAL/TxPhysics Development What will VORPAL be
able to do?

• X-ray emission model for various materials due to
  electron bombardment
• Impurity radiation model for ion cooling
• Simple radiation transport
• Couple VORPAL simulations to molecular dynamics
  models for surface damage and deformation
• Temperature and emission yield diagnostic
  mapping to more easily visualize the simulations
• A web-based interface to VORPAL with the
  capability of providing computational resources
  to researchers anywhere
• Surface damage and heating model due to
  bombardment
• Multi-scale simulation capability, coupling
  fine-grain (surface asperity) simulations with
  course-grain (RF cavity) simulations

all are about 1 year away!
Fermilab MuCool RF Workshop III 7 July 2009
11
Example Impact Ionization, Elastic Scattering
Excitation

• A beam of 40 eV electrons is incident on a
  droplet of Xenon and Argon gas.
• Impact ionization, elastic scattering, and
  neutral gas excitation are all computed.

Fermilab MuCool RF Workshop III 7 July 2009
12
Example Impact Ionization, Elastic Scattering
Excitation
Fermilab MuCool RF Workshop III 7 July 2009
13
Example Impact Ionization, Elastic Scattering
Excitation
Fermilab MuCool RF Workshop III 7 July 2009