New Ideas in Neutrino Detection

1. New Ideas in Neutrino Detection Mark Vagins University of California, Irvine WHEPP 9 – Bhubaneswar, India January 6, 2006

2. So, what’s new and exciting in the world of neutrino detection?

3. Solar 7Be neutrinos will be quite popular in 2006-7: SNO will drain their heavy water at the end of 2006, possibly replacing it with liquid scintillator (SNO++). KamLAND is currently purifying their scintillator in order to focus on looking for the 7Be solar neutrinos. Borexino looks like it will finally turn on this year.

4. Going even lower in energy… CLEAN - 10 tons fiducial LNe needed for pp solar neutrinos Mini-CLEAN (600 kg LNe) in SNOLAB in 2007? – R&D plus WIMP search

5. Solar neutrino Another new super low energy project: XMASS Multi purpose low-background experiment with liquid Xe • Xenon MASSive detector for solar neutrino (pp/7Be) • Xenon neutrino MASS detector (bb decay) • Xenon detector for Weakly Interacting MASSive Particles (DM search) Dark matter Double beta

6. Meanwhile, on the other end of the n energy scale:

8. ４１．４ｍ ４０ｍ My own beloved Super-Kamiokande has been taking data, with an occasional interruption, for over nine years now…

9. But what does the future hold? On July 30th, 2002, at ICHEP2002 in Amsterdam, Yoichiro Suzuki, then the newly appointed head of SK, said to me, “We must find a way to get the new physics.”

10. Taking this as our mandate, theorist John Beacom and I focused on finding some way to get new physics out of Super-Kamiokande. This partnership of theory and experiment has proven quite productive.

11. Supernova neutrinos are certainly interesting… but how could we be sure of seeing some in SK? Now, galactic supernovas may be somewhat rare on a human timescale, but supernovas are not. On average, there is one supernova explosion somewhere in our universe every second!

12. These constitute the diffuse supernova neutrino background [DSNB], also known as the “relic” supernova neutrinos [SRN]. After traveling an average distance of six billion light years, about 100,000 of these genuine supernova neutrinos pass through our bodies every second.

13. As of 2002, here’s what the various predictions of this flux looked like:

14. In 2003, Super-Kamiokande published the world’s best limits on this so-far unseen flux [M.Malek et al., Phys. Rev. Lett.90 061101 (2003)]. Unfortunately, the search was strongly limited by backgrounds, and no clear event excess was seen.

15. So, experimental DSNB limits are approaching theoretical predictions. Clearly, reducing the remaining backgrounds and going lower in energy would be extremely valuable. But how? Well, all of the events in the present SK analysis are singles in time and space. And this rate is actually very low… just three events per cubic meter per year.

16. “Wouldn’t it be great,” we thought, “if there was a way to tag every DSNB event in Super-K?” Since the reaction we are looking for is ne + p e+ + n what if we could reliably identify the neutron and look for coincident signals?

17. But we’re going to have to compete with hydrogen (p + n  d + 2.2 MeV g)in capturing the neutrons! Plus, plain old NaCl isn’t going to work… We’d need to add 3 kilotons of salt to SK just to get 50% of the neutrons to capture on the chlorine!

18. Seawater NaCl Saturation Point

19. So, we eventually turned to the best neutron capture nucleus known – gadolinium. • GdCl3 , unlike metallic Gd, is highly water soluble • Neutron capture on Gd emits a 8.0 MeV g cascade • 100 tons of GdCl3 in SK (0.2% by mass) would yield >90% neutron captures on Gd • Plus, it’s not even toxic! Man, that’s one tasty lanthanide!

21. But, um, didn’t you just say 100 tons?What’s that going to cost? In 1984: $4000/kg -> $400,000,000 In 1993: $485/kg -> $48,500,000 In 1999: $115/kg -> $11,500,000 In 2005: $4/kg -> $400,000

22. So, perhaps Super-K can be turned into a great big antineutrino detector… it would then steadily collect a handful of DSNB events every year with greatly reduced backgrounds and threshold. Also, imagine a next generation, megaton-scale water Cherenkov detector collecting 100+ per year! This is the only neutron detection technique which is extensible to such scales, and at minimal expense, too: ~1% of the detector construction costs

23. If we can do relics, we can do a great job with a galactic supernova: • The copious inverse betas get individually tagged and can be subtracted away from • the directional elastic scatter events, doubling SK’s SN pointing accuracy. • The 16O NC events no longer sit on a large background and are hence individually identifiable, as are • the backwards-peaked 16O CC events.

24. Here’s our proposed name for this water Cherenkov upgrade: GADZOOKS adolinium ntineutrino etector ealously utperforming ld amiokande, uper !

25. Oh, and as long as we’re collecting ne’s… KamLAND’s first 22 months of data GADZOOKS! GADZOOKS! will collect this much reactor neutrino data in two weeks. Hyper-K with GdCl3 will collect six KamLAND years of data in one day!

26. Here’s what the coincident signals in Super-K-III with GdCl3 will look like (energy resolution is applied): Most modern DSNB range

27. Beacom and I finally got our first GADZOOKS! paper written up as hep-ph/0309300 and sent it off to Physical Review Letters.

28. After a long wait due largely to one of the world’s slowest referees, our paper was finally published in Physical Review Letters as Phys. Rev. Lett., 93:171101, 2004 Within a year it was “topcited” on SPIRES!

29. Choubey and Petcov consider the reactor signal of GADZOOKS! From Phys. Lett. B594: 333, 2004: • “The upper limit on Dm221 is determined solely by the SK-Gd data.” • “The lower limit on sin2q12 is determined by the SK-Gd data.” • “The results of our analysis show that the SK-Gd experiment has a remarkable potential in reducing the uncertainties in the values of Dm221and sin2q12.” • “We find that the SK-Gd experiment could provide one of the most precise (if not the most precise) determinations of the solar neutrino oscillation parameters Dm221and sin2q12.” Here are their main results…

30. Choubey and Petcov consider the reactor signal of GADZOOKS!

31. Therefore, we could very well be just a few years away from the world’s first true precision measurement (<1 % uncertainty) of a fundamental neutrino parameter!

32. So, adding 100 tons of GdCl3 to Super-K would provide us with at least two brand-new, guaranteed signals: • Precision measurements of the • neutrinos from all of • Japan’s power reactors • (~5,000 events per year) 2) Discovery of the diffuse supernova neutrino background [DSNB], also known as the “relic” supernova neutrinos (~5 events per year)

33. In addition to our two guaranteed new signals, it is likely that adding GdCl3 to SK-III will provide a variety of other interesting (and not yet fully explored) possibilities: • Solar antineutrino flux limit improvements (X100) • Full de-convolution of a galactic supernova’s n signals • Early warning of an approaching SN n burst • (Free) proton decay background reduction • New long-baseline flux normalization for T2K • Matter- vs. antimatter-enhanced atmospheric n samples(?)

34. For example, Odrzywodek et al. note that late-stage Si burning in very large, very close stars could provide a two day early warning of a core collapse supernova if neutron detection is possible. In SK with GdCl3 this would mean an increase in the low energy singles rate… a factor of 10 increase in the case of Betelgeuse. [Odrzywodek, Misiaszek, and Kutschera, Astropart.Phys. 21:303-313, 2004] [SK with Gd is the only detector which can do this]

35. One other thing you can do with this technology… Two months ago I co-authored a modest (~$2,000,000) proposal to use GdCl3-based neutron detection as part of the Office of Nonproliferation Research’s Active Interrogation Program.

36. At NNN05, before I had even given my talk, John Ellis suddenly stood up and demanded of the SK people in attendance: Why haven’t you guys put gadolinium in Super-K yet? Our GADZOOKS! proposal has definitely been getting a lot of attention recently: As I told him, studies are under way…

37. You see, Beacom and I never wanted to merely propose a new technique – we wanted to make it work! Now, suggesting a major modification of one of the world’s leading neutrino detectors may not be the easiest route…

38. …and so to avoid wiping out, some careful hardware studies would be needed. • What does GdCl3 do the Super-K tank materials? • Will the resulting water transparency be acceptable? • Any strange Gd chemistry we need to know about? • How will we filter the SK water but retain GdCl3?

39. As a matter of fact, early on I made two discoveries regarding GdCl3 while carrying a sample from Los Angeles to Tokyo: • GdCl3 is quite opaque to X-rays • Airport personnel get very upset when they find a kilogram of white powder in your luggage

40. Two years ago I was awarded a U.S. DoE Advanced Detector Research Program grant to support the study of key gadolinium R&D issues. Work immediately got under way at UC Irvine [UCI] building a small version of the SK water filtration system, while at Louisiana State University [LSU] a materials aging study was put together under the supervision of Bob Svoboda.

41. Example of Soak Sample Tank Weld Joint: Room temperature soak in 2% GdCl3 We will inspect surface every three months via SEM, optical, and XRD Now at 30 years of equivalent exposure!

42. These bench tests have been very encouraging. In a single pass through our UCI water filtering system we can remove ~99.99% of the Gd and return it to our 500 liter holding tank. But can this be scaled up to kilotons of Gd-enriched water? With the help of this big detector at KEK (and a new 2005-6 DoE ADR grant) we’re in the process of finding out!

43. After two years of small-scale testing at UCI and LSU, in order to study the GdCl3 concept in a “real world” setting it was suggested that we use a beautiful piece of equipment – the old one kiloton [1KT] detector from the K2K experiment. This 2% model of Super-K and Super-K itself are quite similar, but they are not completely identical… • Three key differences: • the SK tank is high grade stainless steel • while the 1KT tank is iron • the SK water is degasified but the 1KT’s is not • surface/volume ratio 6X higher for 1KT

44. So, starting two months ago, I have begun using K2K’s old one kiloton water Cherenkov detector to study the GdCl3 technique: K2K’s 1 kiloton tank is being used for “real world” studies of • Gd Water Filtering – UCI built and maintains this water system • Gd Light Attenuation – using real 20” PMTs • Gd Materials Effects – many similar detector elements as in Super-K

45. This September, 4,000 kg of GdCl3 (enough for two fills) was shipped from Shanghai to Yokohama. Bringing thousands of kilos of mysterious, high-purity white powder into the country from China was not entirely trivial… …but at least this time I managed to avoid being arrested!

46. 4,000 kg of GdCl3 at KEK...

47. …contained in 80 barrels.

48. On November 10th, 2005, after 3.5 years of work on the idea, the first scoop of GdCl3 goes into our 500 liter pre-treatment tank. After a few hours of filtering the solution, 200 kg (0.02% by mass) of dissolved GdCl3 was injected into the main 1KT tank.

49. Proposed SK Level Initial 1KT level

50. So, what did we see? The Gd filtering system worked perfectly, producing four tons/hour of Gd-free water while at the same time returning the gadolinium to the main tank. Two days after being injected from the bottom, the GdCl3 had reached the top of the tank. Collected light from downward muons had dropped by less than 1%. So now we know - the filtering works, and the GdCl3 will not significantly degrade SK’s water transparency!