neutrino experiments

EPJ manuscript No.(will be inserted by the editor)

Neutrino Experiments

K.T. Leskoa

Institute for Nuclear and Particle Astrophysics, Lawrence Berkeley National Laboratory, Berkeley, CA, USA

Received: date / Revised version: date

Abstract. This review examines a wide variety of experiments investigating neutrino interactions andneutrino properties from a variety of neutrino sources. We have witnessed remarkable progress in thepast two years in settling long standing problems in neutrino physics and uncovering the first evidence forphysics beyond the Standard Model in nearly 30 years. This manuscript briefly reviews this recent progressin the field of neutrino physics and highlights several significant experimental arenas and topics for thecoming decade of particular interest. These highlighted experiments include the precision determination ofoscillation parameters including θ13, θ12, ∆m2

12 and ∆m2

23 as well as a number of fundamental propertiesare likely to be probed included nature of the neutrino (Majorana versus Dirac), the number of neutrinofamilies and the neutrino’s absolute mass.

PACS. 14.60.Pq Neutrino mass and mixing – 26.65.+t Solar neutrinos – 95.85.Ry Neutrino, muon, pion,and other elementary particles; cosmic rays – 23.40.Bw Weak-interaction and lepton (including neutrino)aspects of beta decay – 13.15.+g Neutrino interactions – 14.60.St Non-standard-model neutrinos, right-handed neutrinos, etc. – 14.60.Lm Ordinary neutrinos (nue, numu, nutau) – 95.55.Vj Neutrino, muon,pion, and other elementary particle detectors; cosmic ray detectors

1 Introduction

There as been tremendous progress in the past two years inour understanding of neutrino properties and interactions.It will be challenging to adequately incorporate all of theexperimental results and interpretations into a single re-view. Fortunately many of the details describing these im-portant developments are included elsewhere within thisjournal in the contributed papers. These contain manymore of the experimental details and thorough discussionsof the analyses. This paper serves as broad-brush overviewof the recent neutrino experiments and a roadmap for themost interesting (and sometimes the most challenging) ex-periments in the coming decade.

For the past thirty years there has been intriguing hintsthat neutrinos were not correctly described in the Stan-dard Model. A variety of experiments using accelerator,solar, and atmospheric neutrinos indicated that fewer neu-trinos were detected than would have been expected fromknowledge of the source. These “disappearance” experi-ments gained in statistical and systematic significance asthe experiments matured and as there appeared multipleexperiments discovering consist neutrino suppression fromthe same neutrino sources. During the same period theneutrino source models have significantly advanced, ma-tured, and became truly robust. Increasingly, it appearedthat neutrino flavor oscillations was the most straightfor-

a This work supported by the U.S. Department of Energyunder Contract DE-AC03-76SF00098.

ward method to explain this collection of experiments. Inthis decade we have witnessed a handful of experimentsthat have effectively answered this three decade old prob-lem, convincingly demonstrated that neutrinos are mas-sive, and that neutrino undergo flavor transformations.Rapidly the field is advancing from the discovery phaseinto precision determination of neutrino oscillation param-eters. However, despite this progress there are still funda-mental neutrino properties that are only poorly known orcompletely undetermined. These little-known propertiesand parameters present the physics community with greatopportunities to continue the quest to finally understandthe neutrino.

Neutrino experiments are typically discussed and group-ed based on the source of the neutrinos. Thus we have theclassifications of experiments that primarily concentrateon neutrino oscillations from:

– solar neutrinos,– atmospheric neutrinos,– reactor neutrinos,

and– accelerator neutrinos.

Complementing these oscillation experiments there are sev-eral experiments that probe aspects of neutrinos relatingto the absolute mass of a neutrino and whether the neu-trino is a Dirac or Majorana particle. We will next examinethe progress in all these areas.

2 K.T. Lesko: Neutrino Experiments

2 Solar Neutrino and Recent Reactor

Neutrinos Experiments

The most dramatic progress in understanding neutrinosand neutrino oscillations occurred in the field of solarneutrinos. The Sun is a prodigious source of electron-neutrinos, νe, a by-product of the nuclear fusion reactionsin the core of the Sun. The description of these reactionsand experiments to detect them have developed remark-able since it was first proposed in 1963 [1] [2] that theseneutrinos could be used to probe the nuclear origins ofsolar energy into a thorough model of several complex nu-clear reaction chains. The Standard Solar Model describesthe nuclear reactions and the matter and thermal distribu-tions within the Sun. In the decade following these propos-als and as solar models were being developed, the SolarNeutrino Problem was formulated as experiments foundsignificantly lower than predicted fluxes of neutrinos.

2.1 The Sudbury Neutrino Observatory

This thirty-year-old dilemma was recently and convinc-ingly solved by the Sudbury Neutrino Observatory (SNO)[3] [4]. SNO is a 1 kilotonne heavy water, D2O, Cerenkovdetector located ∼ 2-km underground in Sudbury OntarioCanada in INCO’s Creighton mine. SNO’s measurementof the total neutrino flux, independent of flavor, and ofthe electron-neutrino flux, with the use of Charged Cur-rent (CC), Neutral Current (NC), and Elastic Scattering(ES) reactions produced very strong evidence, in an ap-pearance experiment, of neutrino flavor transformation.When combined with the other experiments comprisingthe Solar Neutrino Problem – the radio-chemical exper-iments Homestake Chlorine experiment[5], SAGE[6] andGallex[7] (and GNO[8]), and the water Cerenkov exper-iments Kamiokande[9] and Super-Kamiokande[10] – theneutrino oscillation parameter space is now restricted toa single region, the Large Mixing Angle (LMA) solution.This solution invokes matter enhanced neutrino oscilla-tions (the MSW mechanism). Thus, SNO has observedflavor transformation as the electron-neutrinos created inthe solar core transform into other flavors. These resultsare presented in Figure 1.

From the global analyses of these seven experimentswe can draw even more conclusions about neutrino oscilla-tions. The LMA solution significantly favors non-maximalmixing of neutrino flavors; that the ordering of the mas-sive neutrino eigenstates is normal, mν1

< mν2, and there

is no “dark side” [12]; there there is now strong supportfor matter affects in the Sun to explain the magnitude ofthe electron-neutrino suppression. Directly comparing thetotal neutrino flux to solar model predictions improvedlimits were also obtained for sterile neutrinos and evensome proton-decay channels. A global analysis of the so-lar neutrino experiments is presented in Figure 2.

0 1 2 3 4 5 60

1

2

3

4

5

6

7

8

)-1 s-2 cm6

(10eφ

)-1

s-2

cm

6 (

10τµφ SNO

NCφ

SSMφ

SNOCCφSNO

ESφ

Fig. 1. Flux of solar νµ or ντ vs. flux of νe deduced from thethree neutrino reactions in SNO. The diagonal bands show thetotal 8B flux as predicted by the Standard Solar Model [11](dashed lines) and that measured with the NC reaction (solidband). The intercepts of these bands with the axes representthe ±1σ errors. The bands intersect at the fit values for φe andφµτ , indicating that the combined flux results are consistentwith flavor transformation.

Fig. 2. A global analysis of the solar neutrino experiments pre-sented in the text. Only the LMA solution remains as a viablesolution to this group of experiments. This figure is courtesyof H. Murayama.

K.T. Lesko: Neutrino Experiments 3

θ22sin0 0.2 0.4 0.6 0.8 1

)2 (

eV2

m∆

10-6

10-5

10-4

10-3

Rate excludedRate+Shape allowedLMAPalo Verde excludedChooz excluded

Fig. 3. Excluded regions of neutrino oscillation parametersfor the rate analysis and allowed regions for the combined rateand shape analysis from KamLAND at 95% C.L.The 95% C.L.allowed region of the Large Mixing Angle solution of solar neu-trino experiments is shown. The thick dot indicates the bestfit to the KamLAND data in the physical region: sin2 2θ = 1.0and ∆m2 = 6.9 × 10−5eV2. All regions look identical underθ ↔ (π/2 − θ) except for the solar neutrino experiment deter-mined LMA region.

2.2 KamLAND

Within a few months of SNO’s first NC publication, Kam-LAND’s first publication was released [13]. KamLAND isa ∼ 1 kilotonne liquid scintillator detector located in thecavity that formerly sited the Kamiokande experiment.Using Japan’s nuclear power reactors with a flux-weightedmean distance of ∼ 180 km, the KamLAND collaborationreported a suppression of reactor neutrinos in good agree-ment wtih and uniquely determined by the LMA solutionof the solar neutrino problem. Assuming CPT conserva-tion, this is good evidence, making use of an entirely dif-ferent source of neutrinos and with a well characterizedsource and spectrum of neutrinos, that the Solar Neu-trino Problem is answered by neutrino flavor oscillationsand with LMA parameters (∆m2 ∼ 0.7 x 10−5 , sin2(2θ) ∼0.8). An analysis of the reactor spectrum in addition to theflux rate bifurcates the LMA solution into two regions ahigher mass region, LMA2 and a lower mass region LMA1.The results of the solar experiments and KamLAND arepresented in Figure 3

2.3 Future Experiments and Improvements in Solarand Reactor Experiments

In the near term we anticipate significant improvementsin the determination of the LMA parameters and some

hope for the observation of an oscillation pattern. SNO hasnearly completed its analysis of its enhanced NC mode, us-ing 2 tonnes of NaCl dissolved in the D2O to improve itsNC sensitivity. With improved NC and CC flux measure-ments this will further refine the mixing angles. Followingthis measurement SNO plans include the deployment ofan array of 3He counters to complement the NaCl mea-surement with a method with very different systematicsand further improved sensitivity. In addition observationby SNO of spectral distortions or day/night affects (mat-ter affects due to the earth) would be important confir-mations of the MSW hypothesis. While with increasedstatistics and improved systematics KamLAND may beable to significantly better define ∆m2 if ∆m2 is below ∼1 x 10−4 eV2. So we see that solar neutrinos has rapidlyevolved from discovery mode to precision determinationof oscillation parameters.

Longer term efforts are aimed at lower energy solarneutrinos, primarily 7Be and p-p neutrinos. The 7Be flux isknown to approximately 7-10% dominated by solar modeluncertainties, but the observation of these neutrinos wouldprovide valuable confirmation of solar physics and a slightincrease in the precision of the determination of the mix-ing angle. Several liquid scintillator experiments are aimedat the 7Be flux, the conversion of KamLAND and Borex-ino being constructed in Gran Sasso. The primary flux ofsolar neutrinos, the p-p neutrinos, is known with 1 - 2%precision. This higher degree of certainty in flux and theirlower energy open a wider pallet of physics, which includes1) an improvement of θ12 by a factor of 2 to 3 and the as-sociated precision inputs to The Maki, Nakagawa, Sakata,and Pontecorvo matrix (MNSP) that describes the mixingof neutrino flavors, 2) tests of the unitary of the MNSPmatrix and in reducing covariances in future experimentssuch as long baseline measurements of θ13, 3) tests forsterile neutrinos, 4) searches for neutrino magnetic mo-ments, and 5) probes of fundamental solar physics. Thesefuture experiments are faced with even more challengingreduction of the radioactive background species including85Kr, 210Pb, 210Bi and radon.

For the longer term nuclear power reactors provide apromising source of neutrinos to probe the last unknownelement of the MNSP matrix, θ13. By probing for sub-dominate oscillation patterns superimposed on the pri-mary oscillations, it may be possible to measure θ13. Thisdetermination would provide valuable and complementaryinformation to future accelerator neutrino measurementsand important removal of parameter degeneracies faced bythe accelerator sources. If the systematics of the measure-ments can be controlled to the 0.5 - 1 % level then it maybe possible to improve our limits (or measure) θ13 by 1to 1.5 orders of magnitude. The experiment signature forsuch a long baseline, high precision reactor experiment ispresented in Figure 4.

3 Atmospheric Neutrino Experiments

High energy cosmic rays striking the earth’s atmosphereproduce pions and muons. These particles subsequent de-


Fig. 4. Subdominate oscillation patterns characteristic of θ13

terms superimposed on the dominate θ12 neutrino suppression.Figure courtesy of K. Heeger.

cay producing electron- and muon-neutrinos with energiesextending into GeV. By observing neutrinos coming fromaround the earth a variety of different path lengths canbe examined. These path lengths are directly mapped intozenith angle distributions for the neutrinos in detectors. Anumber of experiments have examined atmospheric neu-trinos over the past decade, including Kamiokande, Super-Kamiokande, MACRO, Soudan, IMB, and others.

3.1 MACRO

The MACRO detector, decommissioned several years agowas a composite detector (streamer tubes, nuclear trackdetectors, liquid scintillator and rock absorber). The de-tector was sited in Gran Sasso Underground Laboratory.The collaboration recently reanalyzed their through-goingmuon data. They improved their analysis by estimatingthe event energy using multiple Coulomb scattering inthe detector. Combining these data with their existing an-gular distribution analyses they find a suppression of at-mospheric neutrinos consistent with neutrino oscillations(νµ → ντ ) with maximal mixing and a preferred ∆m2 of0.0023 at a level of greater than 5 σ.

3.2 Super-Kamiokande

Super-Kamiokande is a massive water Cerenkov detector(23 kilotonne fiducial volume) located 2700 m.w.e. un-derground in Mt Ikenoyama Japan. They reported a newanalysis of 1489 days of data from Super-K I. This anal-ysis probes neumu to neutau oscillations and used fullycontained and partially contained events, upward goingmuons, and multi-ring events. The collaboration has up-dated their Monte Carlo simulations of their detector, im-proved their fitters and data reduction algorithms andcoupled this to their new atmospheric neutrino flux calcu-lations. The best fit to their data is for a neutrino oscil-lation hypothesis with maximal mixing and ∆m2 - 2.0 x

Prelim

inary!New analysis

Old analysis

Fig. 5. Revised analysis of Super-K’s 1489 day data set includ-ing fully contained and partially contained events, upward go-ing muons, and multi-ring events. Improving the Monte Carlosimulations, event reconstruction and using updated flux cal-culations resulted in a slight lowering of the preferred value for∆m2 that is consistent with early determinations but affects anumber of experiments probing this region of parameter space.

10−3 eV2. Their data favor νµ to ντ oscillations over νµ toνe. And there is not hint for electron neutrino appearancein their data. Importantly for many proposed or exper-iments being developed this favored solution is slightlylower in ∆m2 than previous solutions. While being sta-tistically consistent with earlier analyses, this lower valuereduces τ appear signals for several experiments. Super-K’s results presented at this conference are presented inFigure 5.

4 Accelerator and Long Baseline Neutrino

Experiments

The fourth source of neutrinos used for oscillation exper-iments are man-made neutrinos from accelerators. Thethird experiment reporting evidence for neutrino oscilla-tions was LSND a liquid scintillator detector at LAMPFat Los Alamos National Laboratory. These results are dif-ficult to reconcile with highly significant oscillations sig-natures from solar and atmospheric neutrino experimentsand the three known species of light neutrinos, which arealso indicated by LEP and SLC measurements of the de-cay of the Z-boson. The three neutrinos introduce twoseparate differences in neutrino masses. LSND apparentlyfound a third, higher ∆m2 than solar and atmosphericneutrinos. This has lead to the conjecture that the LSNDsignal involves sterile neutrinos, that is neutrinos that do


not couple to matter in the same fashion as the other threeneutrino species.

4.1 MiniBooNE

To examine this solution and produce a convincing rebut-tal or confirmation the MiniBooNE experiment at FermiLab is probing the νµ to νsterile to νe signal champi-oned by the LSND collaboration to explain their results.MiniBooNE is a oil-based dual mode (scintillating andCerenkov) liquid detector. The detector was recently com-pleted and commissioned and is collecting data. They arepursuing a blind analysis of their data and should have astatistically adequate sample to report results in 2005.

4.2 K2K

The K2K experiment combines an accelerator source ofneutrinos from KEK directed at the Super-K detector.The experiment is aimed at probing the same region of pa-rameter space indicated by the atmospheric neutrino ex-periments, including Super-K. The collaboration reportedimprovements to the near detector used to analyze the mo-mentum and angular distributions of the beam. These up-grades include the replacement of the Pb-glass calorimeterwith SciBar elements to improve tracking and energy re-sponse and to provide better particle identification. TheSciBars also have enabled studies of low energy neutrinointeractions. The K2K collaboration presented an updatedanalysis excludes the null hypothesis (no oscillations) tobe less than 1% best fit at maximal mixing 1.5 - 3.9 x10−3 eV2 consistent with atmospheric neutrino experi-ments. These results are presented in Figure 6.

4.3 ICARUS, OPERA, and CNGS

In the near term a program of beams and detectors is be-ing developed in Europe. CERN will be a source of highenergy neutrinos and several detectors will be sited un-derground at Gran Sasso (732 km baseline). These exper-iments will principally probe for tau-neutrino appearancefrom the accelerator neutrino beams. The recent best fit∆m2 solutions from Super-K present rate challenges tothese appearance experiments as the rates of τ appear-ance may be significantly reduced.

The ICARUS experiment is a liquid argon imagingTime Projection Chamber. Using a beam from CERN,the CNGS program, ICARUS will seek τ appearances inthe TPC. They have successfully operated a 300T moduleand plan on having 600 tonnes at Gran Sasso by the end of2003. By 2006 they plan on installing two additional 1200tonne modules. This mass will produce tens of events. Inaddition to neutrino oscillations, they plan simultaneousprograms in solar, atmospheric neutrino experiments andnucleon decay searches.

Fig. 6. The recent K2K analysis overlayed with Super-K at-mospheric neutrino analysis both reported at this conference.

Also planning on using the CGNS beams in 2006 isOPERA. This is a hybrid emulsion experiment, again seek-ing tau appearance in the beam. OPERA is building on ex-tensive previous experience with emulsions from the DO-NUT experiment to produce a reliable and robust detec-tor.

4.4 NuMI-MINOS

In the US the NuMI-MINOS program is being developedat FermiLAB with detectors sited at Soudan ( 750 kmbaseline) and ultimately other sites are being consideredfor additional measurements. Again this program is aimedat testing solutions of atmospheric neutrinos anomaly withaccelerator beams. If ∆m2

23 ≤ 2 x 10−3 the steel and scin-tillator detector in the Soudan Underground Laboratorywill detect an oscillation dip and can determine ∆m2

23 to ∼1 x 10−3 eV2, but will have little affect on the determina-tion of sin2(2θ23). There is the possibility of τ appearance,but again Super-K’s recent results make this experimentmore challenging. The detector has been completed andis undergoing cosmic ray calibration. the beam will becommissioned through December 2004. The near detec-tor plans are completed and they anticipate the start ofphysics in 2005 with a five year run planned.


4.5 Future Facilities

In Japan J-PARC-Kamioka program proposes to gener-ate a neutrino beam using the J-PARC facility at Jaeri.This accelerator will initially be a source of 1 GeV νµ

from 0.75 MW 50 GeV PS ( 1012 ppp, 0.275 hz,). Ulti-mately this accelerator will be upgraded to 4MW at 50GeV. Construction is scheduled for Japanese fiscal years2001 through 2006.

The first phase of the program will use Super-K as thedetector and promises roughly 100 times the statistics asthe K2K experiment. It will probe neutrino oscillationswith two channels: 1) µ to x disappearance and shouldresolve δ sin2(θ) ∼ 0.01 and ∆m2 to 1 x 10−4 in 5 yearsand 2) µ to e appearance with approximately a 20 timesimprovement over existing reactor experiments determin-ing sin2(2θ13) at 90 % C.L. to 0.006. They will include aneutral current measurement as well. In the second phasewith the increased neutrino flux and a significantly largerdetector they will pursue CP violation measurements andnucleon decay studies.

The NuMI off-axis proposal includes the developmentof a new detector to make use of existing neutrino beam.Off-axis positioning of the detector produces a kinemat-ically focused beam of neutrinos with an energy of ∼ 2GeV. This energy is below the τ production thresholdand produces relatively high rates per proton especiallyfor antineutrinos. Baselines of 700 to 1000 km are beinginvestigated. Thus, matter affects are anticipated to am-plify mass hierarchy sensitivity. Phase I, 2008-2014, willmake use of 4 x 1020 Protons On Target/year coupled toa 50 kton fiducial detector running a proposed 1.5 yearneutrinos and 5 years anti-neutrinos. Phase II, 2014-2020,would increase the POT by a factor of 25 and will run asimilar length of time in neutrino and antineutrino modesas phase I. When combined with JPARC-Super-K andNUMI on-axis data might reveal neutrino mixing param-eters, masses and hierarchy, angles, CP phase.

As mentioned above the primary European focus com-bines several detectors at Gran Sasso using neutrino beamsfrom CERN- the CNGS program. These beam are antici-pated to begin in 2006. Future improvements in the beaminclude a possible 2 GeV 4MW H− accelerator after 2010with a run plan of 2 years of neutrinos, 10 years of anti-neutrinos coupled to a long baseline detector sited in Fre-jus (130-km baseline). The detector is being discussed as400 kT water Cerenkov detector. This combination willmeasure sin2(2θ13) to 0.0025

Neutrino Factories are further in the future and are be-yond the scope of this review. This omission is not meantto reflect on or diminish the critical neutrino physics thatultimately will require such neutrino facilities to fully probeneutrino properties.

5 Neutrinoless Double Beta Decay

Double beta decay is one of the rarest decay modes knownin nature. In its usual form it is a second order weak pro-cess with the conversion of two neutrons to two protons

accompanied by two betas and two neutrinos. The finalnucleus, removed by two nuclear charge units, must bemore deeply bound than the initial nucleus and the in-termediate nucleus (separated by a single nuclear charge),must be less bound than either the initial or the final nu-cleus. If the neutrino is Majorana in nature rather thanDirac (being its own antiparticle) it is possible the twobetas to share the full decay energy with the emission ofno neutrinos. In contrast to the neutrino oscillation ex-periments presented above, which measure the differencesbetween mass eigenstates, neutrinoless double beta decaymeasures effective neutrino mass. A comprehensive reviewcan be found in [14].

If the ββ(0ν) decay is mediated by a light massiveMajorana neutrino, the half-life for this decay is:

[T 0ν1/2(0

+ → 0+)]−1 = G0ν(E0, Z)

∣

∣

∣

∣

M0νGT −

g2V

g2A

M0νF

∣

∣

∣

∣

2

〈mν〉2 ,

(1)where G0ν is the exactly calculable phase space integral,〈mν〉 is the effective neutrino mass and M0ν

GT , M0νF are the

nuclear matrix elements, defined in

|M0ν | ≡M0νGT −

g2V

g2A

M0νF = (2)

〈f |∑

lk

H(rlk, A)τ+l τ+

k

(

σl · σk −g2

V

g2A

)

|i〉 . (3)

In the usual sense any neutrino species (or flavor) canbe expressed as a superposition of mass eigenstates as de-scribed by the MNSP matrix. For example, the electronneutrinos are superpositions,

νe =

N∑

i

Ueiνi , (4)

and the rate of the ββ(0ν) decay is proportional to theeffective neutrino mass

〈mν〉2 =

∣

∣

∣

∣

∣

N∑

i

U2eimi

∣

∣

∣

∣

∣

2

=

∣

∣

∣

∣

∣

N∑

i

|Uei|2eαimi

∣

∣

∣

∣

∣

2

, (all mi ≥ 0) .

(5)This quantity depends, as indicated, on the N −1 Ma-

jorana phases αi/2 of the matrix U which are irrelevantin neutrino oscillation experiments that do not change thetotal lepton number.

There are several indicators and experiments that sug-gest that the next generation of neutrinoless double betadecay experiments are well positioned to access the neu-trino mass range of interest. The neutrino oscillation ex-periments presented above indicate the difference in neu-trino masses, however if we assume the lightest mass issmall we can estimate the mass range of interest fromthe atmospheric neutrino experiments, M2 ∼ ∆m2 ∼ 2 x10−3 eV2 . While the solar neutrino experiments suggestM2 ∼ ∆m2 ∼ 7 x 10−5 eV2. The WMAP observationsfavor massive neutrinos of M ∼ 0.2 eV. Importantly, the


solar neutrino experiments measure the mixing angle, θ12

to be less than maximal so there can not be accidentalcancellation of solar neutrino phases.

To distinguish normal two neutrino from neutrino-lessdouble beta decay in practice, energy resolution is criticalexperimental consideration in addition to the ubiquitousconcerns over backgrounds and contamination. Some ex-periments are exploring additional experimental handlesincluding imaging the charge and paths of the betas. So farthere exist limits on this process (and one much discussedobservation). In addition to the semi-conductor diode ap-proach used for many years, new technologies are beingapproached including cryogenic calorimeters. Several newexperiments are in final commissioning phases and thereare a variety of proposals and collaborations forming topursue this next frontier in neutrinos.

5.1 NEMO

The NEMO experiment is currently undergoing commis-sioning in the Frejus Underground Laboratory with 4800m.w.e. over burden. Their target consists of 6.9 kg of en-riched 100Mo and they employ both event tracking andcalorimetry. The tracking is provided by 6180 drift cellsoperating in Geiger mode with an external magnetic field.They obtain calorimetry information from 1940 plasticscintillators coupled to low activity PMTs. In additionthere are extensive shields for magnetic fields, radon, neu-trons and cosmic rays. After five years of operation theyhope to reach t1/2 < 8 x 1024 years or 〈mν〉 < 0.1 to

0.4 eV for 100Mo and 〈mν〉 < 0.6 to 1.2 eV for 182Se.The collaboration has plans for introducing a variety ofdouble beta decay targets into NEMO to obtain multiplesimultaneous measurements of nuclei such as 82Se, 116Cd,130,natTe, 150Nd, 96Zr, and 48Ca. They anticipate ββ(0ν)results in 2004.

5.2 Cuoricino

The Cuoricino experiment is an innovative cryogenic ex-periment pursuing ββ(0ν) physics. The initial detector,Cuoricino, consists of 42 kg of TeO detectors. The detec-tors are divided into 14 modules in each tower with four760 g detectors in each module operating at a temperatureof 50 mK. The decay energy contained in each module isdetected by NTD Ge thermistors. Several modules withnine detectors are included to investigate surface contam-ination issues. After successful operation of the Cuoricinodetector the collaboration plans upgrades and increasingdetector mass by adding additional detector towers untila final mass of 760 kg is reached for the CUORE experi-ment. With the Cuoricino detector, assuming contamina-tion can be controlled sensitivity of 〈mν〉 < 0.24 to 0.4eV may be obtained. This range of sensitivity would sig-nificantly overlap the claim by [15]. Assuming bulk andsurface contamination are adequately controlled the finalCUORE sensitivity for ββ(0ν) would be 〈mν〉 < 37 to140 meV x (T[y])1/4, where T is the running time of thefull experiment.

5.3 C0BRA

C0BRA is another innovative approach to double betadecay. Concentrating on double electron-capture nucleirather than double beta decay the collaboration hopes tocombine several interesting isotopes in a single detector.Pursuing room temperature semi-conductors, they are as-sembling CdTe (CdZeTe) detectors. Thus a single detec-tor would permit multiple measurements of (106, 108)Cd,64Zn, and 120Te EC modes and 70Zn, 116Cd, and 130Tedouble beta decay studies. The collaboration has installedprototype detectors in Gran Sasso (having established sev-eral world records for limits on EC decay modes), and aredeveloping a proposal for a more ambitious detector. Theproposal is planned for 2004.

5.4 Majorana and Genius

Pursuing well developed semi-conductor technology andinnovative shielding concepts there are two proposals tobuild large arrays of Ge detectors. The Majorana collabo-ration is developing a proposal to build an array of ∼ 500kg of isotopically enriched 76Ge detectors. These detec-tors would be highly segmented and mounted in ultra-lowbackground cryostats in a deep underground laboratory.Sites at SNOLAB in Canada and possible National Under-ground Scientific and Engineering Laboratory in the U.S.are being investigated. The GENIUS proposal is also cen-tered on ∼ 500 kg of enriched Ge, but would submerge thediodes in a large cryogenic bath to reduce external back-grounds. Both collaborations are forming and developingcomprehensive proposals.

6 Tritium Endpoint Measurements

Precise measurements of beta-spectra permit, to date, themost accurate direct limits to be set on neutrino masses.In particular, tritium beta decay has provided the most ac-curate limits on νe. Experiments spanning nearly twentyyears and using both magnetic and electrostatic spectrom-eters have reduced the limit on the mass of νe from ∼ 40eV down to the current limit of ∼ 2 eV. Heroic efforts toovercome experimental difficulties, the development of avariety targets (frozen, gaseous atomic and molecular tar-gets) and elegant spectrometer designs have made theseadvances possible. The Mainz experiment has collected tri-tium data from 1994 through 2001. Since 1998 they signifi-cantly improved their signal to background and eliminatedor reduced many of their systematics. These improvementshave eliminated the negative m2 problems that plaguedearlier experiments. A new analysis of the 1998/99 dataplus the 2001 data looking specifically in the last 70 eV in-terval yield the following limits: m2

ν = -0.7 ± 2.2 ±2.1 eV2

and mν ≤ 2.3 eV at 95% C.L. With this limit the intrinsicsensitivity limit of the Mainz spectrometer is reached.

A new collaboration based at Karlsruhe is developinga new experiment, KATRIN. The proposal includes a pre-spectrometer with a fixed retarding potential followed by


transport solenoids and ending in a main spectrometerwith variable retarding potentials. The pre-spectrometerwould be 1.7 m in diameter and 3.5 m long with an energyacceptance of ∼ 70 eV. The main spectrometer would be10 m in diameter and 22 m long with an energy accep-tance of 1 eV. By using a stronger tritium source, longerexperimental runs, improved energy resolution, and bet-ter experimental calibration the KATRIN collaborationhopes to improve the limit on mν from 2 eV to 0.2 eVwith 5 σ discovery potential to mν ≤ 0.35 eV2. Fundingproposals are current being advanced. Commissioning ofthe main spectrometer is anticipated in 2006 while thefirst measurements would begin in 2007.

7 Other Neutrino Properties

There are additional neutrino properties that are currentlybe addressed including fundamental quantities such as themagnetic moment of the neutrino. The Texono collabora-tion using the Kuo-Sheng reactor have recently deriveddirect experimental limits of µν < 1.3 x 10−10µB at 90 %C.L.

8 Conclusions

As this review briefly presents, the progress in under-standing neutrino properties has made tremendous ad-vances in the past two years. The solar neutrino problemhas been solved. We have very convincing evidence fromSNO’s appearance experiment that neutrinos undergo fla-vor transformations and are massive. While their mass isnot enough to comprise the entire Dark Matter, neutrinosare the first confirmed source of this intriguing materialin the universe. The solar neutrino experiments have con-verged on the Large Mixing Angle solution to matter en-hanced oscillations – in contrast to the quark sector lep-tons appear to mix with at least two large angles. Thissolution was confirmed by a reactor experiment, Kam-LAND, assuming CPT conservation. The evidence fromatmospheric neutrinos and long base line experiments pro-vide a consistent picture of oscillations between νµ and ντ

with maximal mixing between these flavors. In the briefspan of two years these experiments have advanced fromthe discovery phase in convincing demonstrating that neu-trinos are massive and transform between species to thephase of precision determination of neutrino oscillationand mixing parameters. A global compilation of results ofthese oscillation experiments are illustrated in Figure 7.This figure highlights the positive signatures for oscilla-tions seen by the solar neutrino experiments and recentlyby a long baseline reactor experiment, the atmosphericneutrino anomaly recently confirmed by K2K, the LSNDexperiment, and a number of searches with different modesof oscillations that did not detected flavor transforma-tions.

In spite of these rapid advances there remain funda-mental questions about neutrinos, some of which nearterm experiments are well positioned to address.

100

10–3

∆m

2 [

eV

2]

10–12

10–9

10–6

10210010–210–4

tan2θ

LMA

LOW

SMA

VAC

SuperKCHOOZ

Bugey

LSNDCHORUS

NOMAD

CHORUS

KA

RM

EN

2

PaloVerde

νµ↔ν

τ

νe↔ν

X

νe↔ν

τ

NOMAD

νe↔ν

µ

CDHSW

KamLAND

BNL E776

LMA

Fig. 7. This figure, courtesy of H. Murayama, indicates thedifferent experimental signatures for neutrino oscillations: thesolar and reactor neutrino region, atmospheric neutrino exper-iments and recent long base line accelerator experiments, andfinally the LSND results. Limits for other oscillation modesare also presented. The region labeled LMA is the preferredregion obtained by a global analysis of all solar radiochemicalexperiments, Super-K, and SNO.

– Is θ13 finite? If so, what is its magnitude?A number of proposals using long base line reactor ex-periments are being developed to address this in thenear in parallel with longer term accelerator proposals.The reactor and accelerator experiments provide com-plementary information necessary to begin analyzingpotential CP violation effects with leptons. Ultimatelyneutrino factories and intensive neutrino sources maybe required to fully address the issue of CP violationwith neutrinos.

– Is the neutrino Dirac or Majorana in nature?

Again a number of near term experiments are being de-veloped and proposed to address this issue by pushingthe limits of ββ(0ν) down by several orders of magni-tude.


– What is the absolute mass of a neutrino?

Both direct endpoint measurements and ββ(0ν) exper-iments are being developed to address this question.

– Are there sterile neutrinos?Solar neutrino experiments are beginning to place in-teresting bounds on possible sterile neutrinos. In ad-dition the miniBooNE experiment will specifically ad-dress the LSND results that appear to require the pres-ence of sterile neutrinos.

– Is the MNSP matrix unitary? What are theprecise determination of its elements? Are theremore neutrinos than the three known species?

In the very near term SNO and KamLAND should sig-nificantly refine sin2(2θ12) and ∆m2

12. The atmosphericneutrino oscillation parameters, sin2(2θ23) and ∆m2

23are being addressed both by more refined analyses ofexisting data and by a new generation of long base lineexperiments in the U.S., Japan, and Europe.

A thorough understanding of neutrino mixing, neu-trino oscillations and possible CP violation may well re-quire the development of new accelerator facilities andexperimental techniques. However, we are currently pre-sented with an exciting frontier to explore with even moredramatic discoveries and new physics potentially with ourreach in the coming decade.

References

1. J.N. Bahcall Phys. Rev. Lett. 12, 300 (1964);2. R. Davis Phys. Rev. Lett. 12, 303 (1964)3. Q.R. Ahmad et al. (the SNO Collaboration), Phys. Rev.

Lett. 89, (2002) 011301 .4. Q.R. Ahmad et al. (the SNO Collaboration), Phys. Rev.

Lett. 89, (2002) 011302.5. B. T. Cleveland et al., Astrophys. J. 496, (1998) 505.6. J.N. Abdurashitov et al., J.Exp.Theor.Phys. 95, (2002) 181.7. W. Hampel et al., Phys. Lett. B447, 127 (1999).8. M. Altmann et al., Phys. Lett. B490, (2000) 16.9. Y. Fukuda et al., Phys. Rev. Lett. 77, (1996) 1683.10. S. Fukuda et al., Phys. Lett. B539, (2002) 179.11. J.N. Bahcall, M.H. Pinsonneault, and S. Basu, Astrophys.

J. 555 (2001) 990.12. A. de Gouvea, A. Friedland, and H. Murayama, Phys. Lett.

B 490 (2000) 125.13. Eguchi em et al. Phys. Rev. Lett. 90 (2003) 021802-1.14. S.Elliott and P. Vogel Annu. Rev. Nucl. Part. Sci. 52

(2002) .15. H.V. Klapdor, et al. Mod. Phys. Lett. A 16 (2001) 2409.

neutrino experiments

Documents