Monte Carlo simulations and measurements of the radiation environment at a laser-plasma accelerator

Abstract

In current experiments at the high-power laser facility at the Lund Laser Centre, electrons accelerated to hundreds of MeV over short distances by means of laser wakefield acceleration in produced plasmas.

To determine the amount of secondary radiation generated when accelerated electrons interact with surrounding materials is of interest to ensure a safe working environment. In this thesis, the radiation levels inside the laboratory are simulated using GEANT4, a C++ class library for particle physics and particle tracking using Monte Carlo methods.

For persons directly outside the room of the vacuum chamber, conservative simulation results indicate that even for electrons accelerated to a relatively large average energy of 500 MeV, in bunches containing an average charge of 100 pC, in excess of 10 8 shots would need to be fired in a single year to reach doses of the order of the limits set by the Swedish Radiation Safety Authority. At a pulse rate of 0.1 shots per second, this corresponds to continuous operation for 27 700 hours.

As a test of the simulations, experimental dose data were collected using dosimetric instruments – measuring doses from electron, gamma and neutron radiation – showing agreement with simulated doses to within reasonable error.

Popular Abstract

At the Lund High Power Laser facility, electrons are accelerated to energies
of several hundred MeV in just a few millimeters—reaching velocities of
99.9999 % of the speed of light. When these electrons hit the surrounding
walls and equipment in the laboratory, secondary radiation is generated as
other particles are knocked out in electromagnetic and nuclear interactions.
In this thesis, this secondary radiation is simulated to make sure that persons
working in the laboratory are not subjected to harmful radiation levels.

Acceleration of electrons is achieved by sending short laser pulses of high
power onto a gaseous target. The power in a single laser pulse reaches the
order of tens of terawatt, equivalent to the power output of thousands of
nuclear power plants. As the laser pulse hits the gas, the optical fields are
strong enough to pull the electrons from the nucleus, separating negative
and positive charges from each other, thus ionizing the gas. The ionization
occurs for many atoms at the same time which means that part of the gas
is turned into a plasma. Now it no longer behaves like a gas of individual
molecules. Instead, the plasma has a more complex collective behavior
similar to that of a fluid, but strongly driven by electromagnetic interactions.
As the laser pulse travels through the plasma, electrons are subjected to a
strong force which pushes them away, both forwards and radially outwards.
This force is strong enough to create a bubble following the laser pulse which
is completely empty of electrons and therefore carries a large positive charge.
If a small fraction of electrons are placed inside this bubble, they will be
strongly attracted by the positive charge of the bubble and become trapped
inside it. Trapped electrons are now quickly accelerated to high energies
when the bubble follows the laser pulse.

Simulations of the radiation environment start with the accelerated electron beam and does not include the laser-plasma interactions themselves.
The accelerated electrons originate from the center of a vacuum chamber
made out of aluminum. When an accelerated electron enters the aluminum,
a large number of processes can occur, producing secondary particles which
in turn may undergo secondary processes, producing even more particles.
The accelerated electrons and any secondary particles that manage to pass
through the aluminum will then undergo similar processes as they hit the
concrete walls in the laboratory. All particles generated in these processes,
as well as the primary accelerated electrons, are tracked until they have lost
all their energy and stop. To simulate these particle interactions and the
production of secondary particles, so called, Monte Carlo methods are used,
where a large number of primary electrons and secondary particles (e.g. photons and neutrons) are traced using statistical methods. This means that for
each step in the simulation all the possible events are given a certain weight
according to their relative probabilities of occurring. These probabilities
are either determined from theoretical models or from experimental data.
To pick one event from the complete probability distribution of all possible
events, a random number generator is used.

Results from the simulations indicate that using the current experimental
setup, the radiation levels during one year are more than 20,000 times
below the safe limit when the correct safety precautions are taken. They also
indicate that the experiments are not without danger as a person standing directly in the electron beam would reach the yearly radiation limit within less than an hour of operation.

• author: Strand Berg, Jack
• supervisor: Martin Hansson; Olle Lundh
• organization: Atomic Physics; Department of Physics
• course: PHYM01 20161
• year: 2016
• type: H2 - Master's Degree (Two Years)
• subject: Physics and Astronomy
• report number: LRAP-529
• language: English
• id: 8893905
• date added to LUP: 2016-10-26 16:25:58
• date last changed: 2016-11-15 13:27:14