Monte Carlo Testing of a Gamma Veto System

Monte Carlo Testing of a Gamma Veto System for LUX Ryan Sacks Mentors: Daniel Akerib and Michael Dragowsky Department of Physics, Case Western Reserve University Abstract Dark matter detection requires effective reduction and rejection of background events to accurately detect dark matter interactions. The LUX detector consists of a volume of liquid xenon with photo multiplier tubes inside of a sealed container shielded by a medium such as water or liquid scintillater. Upgrading the current LUX water shield would allow for effective detection of high energy gammas in the Me. V range that have gone through small angle Compton scattering to produce an event in the main detector. Detection of the high energy gammas in the shield region will allow vetoing of the 10 ke. V signals produced by the Compton scattering. Modeling of the detector will be carried out to ascertain the location of and energy deposition of the gamma events. Determining the ratio of events that are not detectable in the detector to those that are is of critical component to the feasibility of the upgrade. The final product will be to determine what energy level outside of the active Xenon volume will be required to effectively veto an event. Additional PMTs will be required to detect these events as the current set up does not have the required sensitivity to detect gammas in the shielding region. The purpose of the study is to simulate the efficiency of detecting WIMP-like events caused by gamma from contaminants in the PMTs. The study will consider variations in the design to improve veto efficiency. Introduction Results In the 1930’s Frank Zwicky [1] made observations that galaxies in the Coma cluster had a theoretical mass that is not consistent with the observable mass. The basis of this is derived from the brightness of the clusters which has a direct correlation to the cluster mass. A number of other experiments [2, 3] have also confirmed through the measurement of radial velocity that the luminous mass does not yield enough mass to force the rotational velocity of stars on the outer arms of galaxies to be at their observed values. These results were corroborated by a second type of observation, gravitational lensing. Einstein’s theory of General Relativity [4] states that light coming from a distant object will be deflected by the gravitational well of a massive object. Measurements of Galaxy Cluster Abell 2218 show that a disparity between the mass of the lensing cluster that is observed and the amount required to create the affect seen. Measurements have shown that there is a missing part of the mass of the universe and that observations account for ~4% of the mass-energy of the universe [5]. The hypothesis of dark matter is to account for the “missing mass” that is required for the described observations. Figure 3: A graph of energy deposited by each scattering event g Figure 5: A graph of the location within the detector where energy was deposited, top PMT bank. 187 cm Conclusions and Continuing Work • The copper housing around the PMTs suppresses the distance the gammas travel after they have scattered in the detector medium. e- Figure 1: A WIMP (purple) interacts with Xenon atom (light blue) • The copper shields above the upper PMTs (Figure 5) and below the lower PMTs (Figure 4) act as gamma absorbers. Figure 2: The LUX Detector as modeled in MCNP • For the top PMTs there a large number of events that occur at the gas/liquid interface. This could be used as a possible veto area. The LUX detector uses liquid Xenon to experimentally verify the existence of Weakly Interacting Massive Particles (WIMPS). The detection method depends on a collision between a WIMP and the Xenon (see Figure 1). The interaction produces two distinct signals, primary scintillation signal from gamma rays and a secondary electron signal [5]. The electrons are drifted through the liquid medium via electrically charged grids and are detected by arrays of photomultiplier tubes (PMTs). The use of both signals gives LUX the ability to pinpoint in 3 -D space where the collision took place. The focus of this project is to develop a system through which signals from gamma rays released from radioactive decay within the PMTs can be vetoed. By undergoing small angle Compton Scattering, 1 Me. V gamma rays can mimic a dark matter signal. Method The modeling for this project was done using a Monte Carlo code called MCNPPolimi. The code is based on MCNP 5 but instead of dealing with masses of particles tracks the interactions of single particles. The geometry was modeled using MCNP (see Figure 2) and included all areas of liquid Xenon, PMTs, copper shields and all other relevant geometry. After modeling of the detector was completed, runs using a 1 Me. V gamma source were preformed. This data was complied and then fed into a specialty written MATLAB code. The code preformed cuts based on energy and geometry in order to single out those interactions that could mimic dark matter signals. After this the code will then determine which of these signals scatters only once in the liquid Xenon. Current measures in place already veto any multiple scatters in the xenon itself, the current focus is to determine a way to veto gamma rays that scatter in the detector volume only once. • Determine exactly how many of the scatters that deposit energy are single scatters instead of multiple scatters. • Investigate the difference between solid copper housings around the PMTs and copper sheets around the PMTs. • Investigate the affect of a 1 mm gap between the Teflon/Polyethylene wrapper on the inside of the detector and the Titanium wall. Figure 4: A graph of the location within the detector where energy was deposited, bottom PMT bank. The test results for the detector in Figure 2 are given above. Of the 4000 test particles used in the simulation, 651 deposited energy consistent with a dark matter signal. The energy range of interest is between 0 and 5 ke. Vee. This represents the same energy scale that a dark matter event will give off. Figure 3 shows the distribution of energy depositions within the detector in the target range. This is a fairly even distribution which is to be expected as any large spike or drop off will occur at the Compton Peak which is not in this energy range for Xenon. Figure 4 shows the location of the energy depositions within the detector when the gammas are released at the bottom PMT bank. The blue dots are those that take place in the liquid Xenon and could cause a false-positive. Note that very few of the gammas make it far from the point of origin (yellow dot). This is due to the copper housing around the PMTs. Figure 5 shows the distribution when the gammas are released from the top. Note that there is a gap from the source to a large concentration of blue dots. This is due to the fact there is a 10 cm gap of gaseous Xenon between the source and the liquid Xenon detector. • Investigate the affect of different materials for the shields above and below the PMT banks. Acknowledgements I would like to thank the following people for their help on the project: Daniel Akerib, Mike Dragowsky, Tom Shutt and Alex Bolozdynya. References [1] F. Zwicky, Helv. Phys. Acta, 6, 110 (1933). [2] V. Rubin, W. K. Ford, Astrophysical Journal, 159, 379 (1970). [3] J. F. Navarro, C. S. Frenk and S. D. M. White, Astrophysical Journal, 490, 493 (1997). [4] R. Hennings-Yeomans, First 5 Tower WIMP-Search Results from the Cryogenic Dark Matter Search with Improved Understanding of Neutron Backgrounds and Benchmarking. Thesis, Case Western Reserve University (2009) [5] LUX Collaboration, Construction of the LUX Dark Matter Experiment at the Sanford Underground Science and Engineering Laboratory. NSF and DOE proposal. (2007)