Bubble Chambers for Dark Matter Detection Bubble Chambers for Dark Matter Detection : COUPP (Chicagoland Observatory for COUPP (Chicagoland Observatory for Underground Particle Physics) Underground Particle Physics) Brian Odom Brian Odom
The COUPP Team The COUPP Team Principal Investigator: Principal Investigator: Wilson Fellows: Juan Collar Ilan Levine Andrew Sonnenschein (spokesperson) Undergraduates: Staff Scientists: Graduate Students: Nathan Riley Earl Neeley Peter Cooper Matthew Szydagis Tina Marie Shepherd Mike Crisler Martin Hu Undergraduates: Engineers: Erik Ramberg Luke Goetzke Hannes Ed Behnke Bob Tschirhart Schimmelpfennig Fermi National Fermi National Accelerator Accelerator Indiana University Indiana University KICP Fellows: Laboratory Laboratory South Bend South Bend Brian Odom Funding University of Chicago University of Chicago National Science Foundation Kavli Institute for Cosmological Physics Department of Energy Sept. 12, 2007 Brian Odom TAUP 2007
Wish List for a Tough Problem Wish List for a Tough Problem Dream Detector: Dream Detector: • Scalable Scalable • Cheap Cheap ~ 1 recoil / kg / day ~ 1 recoil / kg / day • Radiopure Radiopure • Good background- Good background- rejection rejection ~ 1 recoil / ton / year! ~ 1 recoil / ton / year! • Multiple active nuclei Multiple active nuclei • Easy to swap targets Easy to swap targets A A bubble chamber bubble chamber (bulk superheated liquid detector) (bulk superheated liquid detector) potentially fulfills all those wishes… potentially fulfills all those wishes…if it actually works if it actually works Sept. 12, 2007 Brian Odom TAUP 2007
Bubble Chambers for Rare Event Searches? Bubble Chambers for Rare Event Searches? • In In recent HEP, bubble chambers with recent HEP, bubble chambers with metallic parts metallic parts were stable were stable for for ~ 10 milliseconds…killer dead time ~ 10 milliseconds…killer dead time • But the many of the original chambers were But the many of the original chambers were glass glass • Condense Condensed matter research strongly suggests that both b d matter research strongly suggests that both bulk and lk and surface spontaneous nucleation surface spontaneous nucleation rates can be made low enough rates can be made low enough Roger Hildebrand at UofC holding the first Hydrogen bubble chamber Roger Hildebrand at UofC holding the first Hydrogen bubble chamber Sept. 12, 2007 Brian Odom TAUP 2007
Try This At Home? Try This At Home? A single A single microscopic microscopic interaction (nuclear recoil) gives a interaction (nuclear recoil) gives a macroscopic macroscopic signal (boiling) signal (boiling) The WIMP signal is just as large as the fork signal! The WIMP signal is just as large as the fork signal! Sept. 12, 2007 Brian Odom TAUP 2007
β and and γ Insensitivity Insensitivity • Most experiments, after heroic cleaning and shielding, still Most experiments, after heroic cleaning and shielding, still must cut many β , γ events must cut many events • Bubble chambers can be operated essentially MIP-blind Bubble chambers can be operated essentially MIP-blind Seitz (hot spike) model: Sept. 12, 2007 Brian Odom TAUP 2007
Freedom to Choose Target Freedom to Choose Target In general SUSY, channel of first In general SUSY, channel of first detection is unknown: detection is unknown: • Spin-dependent Spin-dependent: use fluor : use fluorine ne • Spin-independent Spin-independent: use : use something heavy something heavy We use CF 3 I, for excellent We use C , for excellent sensitivity in both channels. sensitivity in both channels. Other advantages of CF Other advantages of CF 3 I : I : • Inexpensiv expensive (40 USD / kg) e (40 USD / kg) • Near room temperature Near room temperature operation (~ 40 C) operation (~ 40 C) • Safe (non-toxic, fire extinguisher) Safe (non-toxic, fire extinguisher) Sept. 12, 2007 Brian Odom TAUP 2007
Freedom to Swap Target Freedom to Swap Target Dark blue: LSP (SUGRA) Light blue: LKP (Bertone, C (Bertone, Cerdeno deno, C Collar, llar, and O and Odom, om, to appear in PRL, to appear in PRL, arXiv:07 arXiv:0705 05.2 .2502) 502) Swapping Target Liquids: Swapping Target Liquids: • Upon discovery, makes ~ orthogonal cuts Upon discovery, makes ~ orthogonal cuts in in parameter parameter space pace • Alternatively, use different ta Alternatively, use different targets for neutron rejection. rgets for neutron rejection. (Response of CF3I and C3F8 to neutrons is very similar, but, for (Response of CF3I and C3F8 to ne utrons is very similar, but, for example, SI WIMP response is very different) example, SI WIMP response is very different) Sept. 12, 2007 Brian Odom TAUP 2007
2 kg Prototype Chamber 2 kg Prototype Chamber Bulk CF3I touches only the quartz Bulk CF3I touches only the quartz Movie posted at http://kicp.uchicago.edu/~odom/presentations/bubble3.avi Sept. 12, 2007 Brian Odom TAUP 2007
Leaving “Normal” Leaving “Normal” BC Mode Far Behind BC Mode Far Behind (Muon-blind) (Muon-blind) Nuclear recoil Nuclear recoil Muon tracks Muon tracks Decreasing superheat Decreasing superheat • Muon tracks Muon tracks are visible only at high superheats are visible only at high superheats • We operate at low superheat, MIP blind We operate at low superheat, MIP blind • 1 nuclear recoil = 1 bubble 1 nuclear recoil = 1 bubble Sept. 12, 2007 Brian Odom TAUP 2007
Stereo View for Bubble Position Stereo View for Bubble Position Position reconstruction is important for Position reconstruction is important for background rejection background rejection, , of both of both wall events wall events and and neutron events neutron events (multiple scattering) (multiple scattering) Sept. 12, 2007 Brian Odom TAUP 2007
Operation at 300 m.w.e., Fermilab Operation at 300 m.w.e., Fermilab Sept. 12, 2007 Brian Odom TAUP 2007
Demonstration of MIP-Blindness Demonstration of MIP-Blindness sic γ rejection • Intri Intrinsi rejection -10 at 10 better than better than 10 10 -10 at 10 keV threshold (best keV threshold (best measured MIP-rejection) measured MIP-rejection) • No cuts required! No cuts required! No need for β, γ • No need for shielding or attention to shielding or attention to β, γ radiopurity. radiopurity. E.g. E.g. 14 14 C ar C are e O O (100 / kg- (100 / kg-day) day) Sept. 12, 2007 Brian Odom TAUP 2007
Neutron calibration, Neutron calibration, in situ in situ • Switchable Am/Be Switchable Am/Be source yields source yields 5 +/- 5 +/- 0.5 n /s when 0.5 n /s when on and on and O O (0.2 n/day) (0.2 n/day) when off when off • Left: Left: blind absolute blind absolute comp compar arison on, , consistent with 100% consistent with 100% efficiency efficiency • Moderated Moderated spectrum produces spectrum produces recoils approximating rec oils approximating WIMP spectrum WIMP spectrum Sept. 12, 2007 Brian Odom TAUP 2007
Current Wall Background: α -Emitters Current Wall Background: -Emitters • Wall rate Wall rate (0.9 / (0.9 /cm cm 2 /day) OK for small vessels /day) OK for small vessels (200 /day for 2 kg chamber) (200 /day for 2 kg chamber) • γ -counting of our Ge-214 natural fused quartz -counting of our Ge-214 natural fused quartz found U (Th) at typical level of found U (Th) at typical level of 50 (30) ppb 50 (30) ppb — consistent with 100% of wall-event rate consistent with 100% of wall-event rate Commercial α -counting gives consistent rate • Commercial -counting gives consistent rate • Next-generation vessels Next-generation vessels made from synthetic made from synthetic fused silica, with expected U, Th level fused silica, with expected U, Th level < 10 ppt < 10 ppt • New vessels also undergo extra quartz etch New vessels also undergo extra quartz etch before final annealing and are then sealed, to before final annealing and are then sealed, to guard against embedded Rn progeny guard against embedded Rn progeny Sept. 12, 2007 Brian Odom TAUP 2007
Current Bulk Background: Radon & Progeny Current Bulk Background: Radon & Progeny with 100% Rn and progeny α ’s Bulk background is consistent Bulk background is consistent with 100% Rn and progeny ’s 1. 1. Flat rate with sharp onset at predicted value Flat rate with sharp onset at predicted value 2. 2. Analysis of time correlation between events supports Analysis of time correlat ion between events supports 100% Rn & progeny, 100% efficie 100% Rn & proge ny, 100% efficiency ncy Predicted α onset Predicted onset Predicted plateau Predicted plateau Next chambers: Next chambers: 1. 1. Non-thoriated welds Non-thoriated welds 2. 2. Metal seals against Rn Metal seals against Rn Sept. 12, 2007 Brian Odom TAUP 2007
Recommend
More recommend
