Reconstructing Nuclear Recoil Tracks in the Dark

Astrophysical evidence indicates that 23% of our universe's energy density is in the form of nonluminous, nonbaryonic matter referred to as dark matter. One theoretically appealing dark matter candidate is the Weakly Interacting Massive Particle (WIMP). Because of astrophysical dynamics, the detectable signal from the expected WIMP dark matter halo should exhibit a unique daily directional modulation for which experiments can search . The Dark Matter Time Projection Chamber (DMTPC) group aims to provide an unequivocal detection of WIMP particles through the anisotropy in the angular recoil spectrum. DMTPC uses a low-pressure time projection chamber filled with CF 4 gas to search for WIMPs via elastic collisions. Crucial to this experiment is the fidelity of nuclear recoil track reconstruction. By extracting parameters such as the angle and vector direction of nuclear recoils, DMTPC has sensitivity to the anisotropic WIMP signal. This thesis develops a new track reconstruction algorithm motivated by the physics of nuclear energy loss in a diffuse gas medium. The algorithm is applied to simulated nuclear recoils and is compared to the existing track reconstruction algorithm. The new fitting algorithm outperforms the old algorithm in determining vector direction of nuclear recoils for recoil energies between 20 and 300 keV. The algorithm shows little sensitivity to CCD read noise. The length reconstruction of the new algorithm, however, fails to outperform the old algorithm below 100 keV. Thesis Supervisor: Professor Gabriella Sciolla Title: Associate Professor of Physics Thesis Supervisor: Dr. James Battat Title: Pappalardo Fellow, Department of Physics Acknowledgments All work presented here was completed on behalf of the DMTPC group whose members include C. Deaconu, D. Dujmic, J. Battat, T. Caldwell, P. Fisher, S. Henderson, R. Lanza, A. Lee, J. Lopez, A. Kaboth, G. Kohse, J. Monroe, R. Vanderspek, T. Sahin, G. Sciolla, I. Wolf, R. Yamamoto, H. Yegorian, S. Ahlen, A. Inglis, K. Otis, H. Tomita and H. Wellenstein. I would like to thank everyone for his/her contribution to my work. A few members have had a particularly significant influence on this work and I recognize them here. First, I could not have wished for a more capable and more dedicated mentor and thesis adviser than James. Only through his countless hours of work and late nights of last-minute revisions was this thesis made possible. Furthermore, his enormous contributions and guidance to my work continued even while he welcomed his new baby girl to his family a tribute to his work ethic and commitment to his students. Of course, I need to thoroughly thank all of my labmates Shawn, Jeremy, Cosmin, and Asher. Never perturbed by my continuous stream of coding and algorithmic issues, each one offered his full attention and support. Each of them has been an incredible resource in physics and other unrelated fields and I am lucky to have been able to work with them over this last year. Furthermore, I thank Michael Betancourt, my pseudo-labmate, for his extremely insightful inputs into the algorithm development and for his constant prodding in between. And finally, I cannot fully express my gratitude towards my adviser, Gabriella, for inviting me to join DMTPC several years ago. Working with the group has been intellectually challenging and it has helped me develop a robust approach to solving complex problems. As both my thesis and academic adviser, Gabriella has been there as an expert guide for physics and general life questions and I've always admired her commitment.