1. COUPP-60 NuMI Commissioning Andrew Sonnenschein All Experimenters Meeting, Oct 18, 2010

2. Searching for Dark Matter with Bubble Chambers (COUPP) Exploit physics of bubble nucleation to discriminate against background gammas and alphas: successful 2009 run of small chamber demonstrated potential to reach zero background. Scalable: 60-kg being commissioned now; 500-kg and 16-ton (DUSEL) versions under discussion. Neutron scattering background event in 4-kg chamber Construction of 60-kg chamber Physics reach for spin- dependent Weakly Interacting Massive Particle (WIMP) interactions on protons– COUPP is now sensitivity leader in this channel, will soon probe SUSY models MSSM

3. COUPP-60 Commissioning Ran from July 28 to August 30 Goal: Test fully operating detector before moving to Snolab – Stability of mechanical systems, DAQ, photography – Backgrounds due to internal radioactivity Analysis in progress, rates appear to be low. – Acoustic alpha/ nuclear recoil discrimination Earlier talk by Eric Dahl (Aug. 16) is on All Experimenter’s Meeting web site. Run ended Aug. 30 due to a combination of problems that will be discussed in following slides. Bottom line: some of the problems are quite serious– significant work needed before move to Snolab. This talk

4. The Easier Stuff Rust on outside of stainless pressure vessel. – Damage is superficial so far, due to limited time under water. Causes mostly evident and easy to correct (a few non-stainless items in a system which should be 100% stainless steel, small stainless parts not properly cleaned and passivated). Leaks in pressure vessel and inner vessel fill plumbing. – Repairs in progress. Tightening bolts has reduced or eliminated pressure vessel leak. Control system– minor software issues. DAQ- need to change operating system on digitizer crate to improve driver support. Vibrations. Unexpected vibrations of outer and inner vessels. May cause excess noise in acoustic sensors, possible surface bubble nucleation. – May need to add isolation materials.

5. Uneven lighting, hard shadows 5 bubbles (neutron multiple scatter) 6th bubble- harder to see due to uneven lighting Dark region at bottom, poor photography, poor triggering Shadows- a challenge for image analysis software

6. Photography and Illumination Scheme bubble Propylene Glycol Inner Vessel Superheated CF 3 I Scotchlite Reflector CAMERA OPTICAL FIBER BUNDLE diffuser LED BOX GLASS PRESSURE WINDOW shadow Illumination is provided by 2 optical fibers (only 1 shown below for illustration). Small number of point-like light sources cast hard shadows and give poor uniformity. We need a larger number of light sources, better pointing and diffusion.

7. Bright Spots Bright Spot Appears in Camera 0 Image After Filling. Shape changes as chamber pressure is cycled.

8. Cause of Bright Spots: Broken Glue Joint Between Lenses and Pressure Window CAMERA PLANOCONCAVE LENS BONDED TO PRESSURE VESSEL WINDOW WITH OPTICAL EPOXY DEPRESSURIZED PRESSURIZED 0.015 inches LENS PULLS AWAY FROM WINDOW WHEN PRESSURIZED

9. Excess Bubbling at Water/ CF 3 I Interface Problem has existed at some level in all COUPP chambers, but COUPP-60 is the worst. Cause is unknown. Ideas: Dissolved gas in water or particulate floating on interface. Consequence is reduction in live time fraction due to 30-60 second compression cycle after each detected bubble. Live fraction reduced to 25% at 40 degrees C. CF3I/ water interface

10. Darkening of Video Images Progressive from beginning of run. Eventually made data taking impossible Cause was a mystery until we viewed the chamber with white light… Average Pixel Intensity

11. CF 3 I Is Decomposing and Changing Color Photo by Raidar Hahn Iodine dissolved in water (orange) Iodine dissolved in CF3I (purple) “ring around the collar” near water/ CF3I interface. Solid material? Bubbles? Could be cause of surface boiling? Color change indicates the presence of free iodine (I 2 ) dissolved in chamber liquids.

12. Photodissociation of CF 3 I Destruction of CF3I molecule by light is well measured at UV wavelengths. Produces free iodine by E.g. 2  + CF 3 I -> C 2 F 6 + I 2 We have seen this before in samples exposed to ambient light, but not with red light illumination. Light Absorption by CF 3 I Solomon et al., 1994

13. Possible Causes of CF 3 I Decomposition A light leak? So far we know that – Light tightness of the camera package : <10 -4 of ambient light levels. – Some light from two green (530 nm) “power on” LEDs on the back of the video cameras. – We are exposing samples to a green LED now– should soon know if this is the problem. Incompatibility with chamber materials? Gold O-ring? – Gold wire used for stainless to quartz seal for the first time in this chamber. Tests in 2006 at U. Chicago did not see any problems, but may not have been sensitive enough. Chemical impurities in CF3I? – This is a new batch and analysis indicates purity level is less than what we had in the past. – If this is the problem, it should also show up in recently started Snolab run which was filled from same batch of CF3I.

14. New Safety Issues Full face respirator with organic vapor/ acid filter Chemical hazard suit to protect from acidified water leaks Forced air ventilation Confined space entry procedures for work on bubble chamber inside water shield tank In principle, a number of hazardous substances can be produced by reactions between CF 3 I and H 2 O (HF, HI, I 2, COF 2 ). Likely not present in significant quantity and should not be released from system.

15. Next Steps PPD has formed a COUPP-60 Task Force to help us call on appropriate resources. Working groups: Optics/Illumination, Chemistry, Mechanical/Operations. Core group from collaboration will be supplemented as necessary by outside experts. Key issue is access to chemistry expertise, which is in short supply at Fermilab. Collaborators at Indiana University are analyzing samples by GC/MS. Argonne has expertise in Chemical Sciences and Engineering Division which we have started to tap. We are recovering samples of gases and fluids from bubble chamber. Analysis of decomposition products may suggest mechanism. Will attempt to reproduce the problem in a test tube. If this is a light exposure problem, we should know soon. In parallel, we can make progress on better understood optics and mechanical issues.