2 of 4

KEY CONCEPTS

PHYSICS OF THE SUPERHEATED STATE

BUBBLE CHAMBER

Bubble nucleation efficiency modelled

through simulations

Superheated state of liquid argon in pressure vessel.
Incoming particle deposits its energy through elastic scattering.
A bubble nucleation occurs when the energy deposited is above energy threshold.
Thus bubble chambers are threshold detectors.

The Model

The Data

UV scintillation

Bubble

Intro:

Hi everyone! Welcome to my poster presentation. I study the bubble nucleation efficiency in bubble chambers, which is a dark matter detector. Astronomical and cosmological observations strongly suggest the existence of a “new” type of particle that does not interact electromagnetically, so we can’t see it, therefore it’s named dark matter. But it exert gravitational influence on the known matter. So when it comes into a bubble chamber, it would collide with the liquid molecules and deposit energy, causing bubbles to form: this is called nucleation.

My research is on finding the threshold energy to create a bubble in Superheated Liquid Argon. SLA is used in bubble chambers, because it's at a metastable state. That means the argon is supposed to be a gas at this temperature and pressure, but it's still a liquid because it needs to cross an energy barrier to vaporize, as shown here. We trap it in this local minimum by steadily lowering the pressure from the liquid state. This is great because if we see a bubble in the chamber, that tells us a particle deposited at least this amount of energy over a certain length. In theory, when a particle deposit less than the threshold energy, a bubble wouldn't form, and a bubble would always form when energy deposited is greater than threshold energy. So we expect the nucleation efficiency curve to be a step function, which goes from 0 to 100 at the threshold energy as you can see here in this green line. But experimental data shows that bubbles don't nucleate 100% of the time when energy deposited is greater than the threshold energy! To explain this, a previous study proposed that some energy is lost to the rotational and vibrational excitation of the molecule, since the data is taken from c3f8, a complex molecule with many such degrees of freedom. Based on this model, an efficiency curve here is constructed through simulations. It also shows this similar gap. My project is to model the efficiency curve for argon, for two reasons. 1. Argon doesn't have the rotational and vibrational degrees of freedom, so if the curve is closer to a step function, then we have further evidence to support this model. Two, we'd be able to compare the simulation results with experimental data from argon, since we are building a bubble chamber with superheated liquid argon. It’ll be great if the experimental data looks closer to a step function, but if it doesn’t, then we will look for another reason that explains the energy loss. �

3 of 4

SIMULATION PARAMETERS

ARGON BUBBLE FORMATION SIMULATION RESULTS

Snapshots of the growing bubble in our simulation

Argon superheated at 130 K and 20 PSIA
LJ parameters determined and calculate

via Molecular Dynamics (MD) simulations.

Bubble growth evolution obtained over time and from different directions.
Similar results obtained for the simulation and the Seitz Model (theory).

Evolution of bubble radius over time seen from different directions

Lennard-Jones (LJ) Potential

Theoretical bubble evolution

Simulation: Conditions and Results

Rmax

Effective radius over time

To model the bubble nucleation efficiency in argon, we need to determine the threshold energy, which depends on the temperature and pressure. We use LAMMPS for our simulations, and we use the Lennard Jones potential to model argon’s intermolecular interactions. We first find the LJ parameters for argon through matching certain parameters in molecular dynamics simulations. After superheating the liquid to 130 Kelvins and 20 PSIA, we deposit energy within a cylinder in our liquid, then leave the system to evolve on its own. These graphs here shows the initial stages of bubble formation, and as you can see it starts from a cylinder and grows into a spherical bubble. This graph shows how the bubble radius grow over time seen from different directions, and then finally this graph is the result of overall bubble radius over time. It fits quite well with the theory, the first stage being linear, and then slows down after the bubble reaches a local maximum radius. The red line here is the theoretical prediction for the second stage bubble growth, and it fits quite well with our simulation data. This bubble here is an above threshold bubble, so since the Rmax is greater than the critical radius of argon, it expands beyond that point. If we have an below threshold bubble, then it would continue to collapse. We then repeat the simulation with different values of energy deposited, over different track length, to find the Rmax for each dE/dx.

4 of 4

SIMULATION RESULTS AND ANALYSIS

BUBBLE EFFICIENCY NEXT STEPS

Next Steps with SRIM + PHITS:

Obtain the Nucleation efficiency for Argon.

Results obtained with superheated Argon

Rotational excitation spectrum PHITS

ESeitz = 40 eV

Results obtained with superheated C₃F₈

Results with C₃F₈

Which is plotted here. Once we fit a function through the points, we would take the energy corresponding to the critical radius to be our threshold energy. The graph below is the result for c3f8. And the threshold energy is found to be 2.14keV. Once we have that, we will use SRIM and PHITS in the next steps to obtain the bubble nucleation efficiency curve. With SRIM, we can give a particle a certain amount of energy, send it into a slab of molecules, and see how far it goes before it loses all its energy. Since the process is randomized, each particle would reach different depths depending on how many collision it had etc. Then for each energy level, we can graph the distribution of depth of penetration. We compare it to the critical length for each energy level to determine whether bubbles form or not. For 6keV here, all particle had a track of less than the critical length, therefore dE/dx is greater than needed and the bubble nucleation efficiency would be 100%. We also use PHITS to find the amount of energy that went into rotational excitation of the molecule. So for argon, we expect this to be zero. And then we repeat the SRIM simulation for many different energy levels to obtain this graph here. And that’s my final goal. Any questions?

1 of 4

2 of 4

3 of 4

4 of 4