Marco VignatiINFN Roma

NPB 2012, Shenzhen, 24 Sept. 2012

LUCIFER Low background Underground Cryogenic Installation For Elusive Rates

mmin [eV]

|m``

| [

eV]

NS

IS

Cosm

ological Limit

Current Bound

10<4 10<3 10<2 10<1 110<4

10<3

10<2

10<1

1

1m2m3m

Bolometric searches of 0νDBD

2

KK evidence -76Ge

Bilenky and Giunti, 2012CUORICINO:

11kg 130Tebkg = 0.17 c/keV/kg/y

CUORE:206kg 130Te

bkg ~ 1·10-2 c/keV/kg/y

Needs:1 ton isotope

bkg < 10-3 c/keV/kg/y

CUORICINO

• MC: most of the background in CUORICINO is due to degraded α particles which release only a part of their energy in the detector (surface contaminations, mainly in copper).

• TeO2 bolometers, per se, do not allow to discriminate β and α particles.

‣ α bkg partially reduced by cleaning the detector parts.‣ β/γ smaller in CUORE thanks also to the self-shielding geometry.

CUORE: the α nightmare

3

α’s

TeO2 TeO2

Cu

Scintillating bolometers• Scintillating crystals can be operated as bolometers. Unfortunately

TeO2 does not scintillate, other compounds must be considered.

• The simultaneous read-out of light and thermal signals allows to discriminate the α background thanks to the scintillation yield different from β particles.

4

Thermistor

BolometerEnergy Release

Light detector

• 0νDBD candidates of experimental interest:

• In general, Q > 2615 keV isotopes are preferred because they lie abovethe natural radioactivity edge.

• However the choice has been dominated so far by technology compromises.

130Te 76Ge 100Mo 116Cd

82Se

136Xe

U and Thenvironmental bkg

5

Isotope choice .(arXiv:1201.4916)

Light detectors• Germanium disks (5 cm diameter, 0.1-1 mm thick).

• Calibration with a 55Fe source: 5.9 & 6.5 keV X-rays.

‣ ΔE @55Fe: ~ 130 eV RMS‣ ΔE @baseline: 100 eV RMS

6

Candidate #1: ZnSe

7

DBD Isotope:

Q-value[keV]

isotopic abundance

Light Yield [keV/MeV]

QF

82Se

2995

9%

7-11

4

Largest crystal operated so far: 431g

Detected Light [keV]0 50 100 150 200 250 300 350 400

Ligh

t Dec

ay T

ime

[ms]

7

8

9

10

11

12

13

14 β/γ-background

α-crystal contamination

Energy [keVee]0 1000 2000 3000 4000 5000 6000 7000 8000

Ligh

t Yie

ld [k

eV/M

eV]

5

10

15

20

25

30

35

40

ZnSe• QF > 1 is odd:

‣ Observed only in this compound (CdWO4 ZnMoO4 and other crystals have QF < 1).

‣ not understood.

8

0νDBD

• Excellent separation using light energy and signal shape

Ligh

t dec

ay-ti

me

[ms]

Integral 101

/ ndf 2r 8.021 / 14

A 1.9± 13

mean 0.6± 2615

m 0.594± 5.741

Decorrelated energy [keVee]2590 2600 2610 2620 2630 2640

coun

ts /

2 ke

Vee

0

2

4

6

8

10

12

14

16Integral 101

/ ndf 2r 8.021 / 14

A 1.9± 13

mean 0.6± 2615

m 0.594± 5.741

Integral 101

/ ndf 2r 4.035 / 13

A 1.5± 10.5

mean 1.0± 2614

m 1.116± 7.938

Energy [keVee]2590 2600 2610 2620 2630 26400

2

4

6

8

10

12Integral 101

/ ndf 2r 4.035 / 13

A 1.5± 10.5

mean 1.0± 2614

m 1.116± 7.938

Decorrelated energy [keVee]2590 2600 2610 2620 2630 2640

Det

ecte

d lig

ht [k

eV]

12

13

14

15

16

17

18

19

20

21

Energy [keVee]2590 2600 2610 2620 2630 2640

Det

ecte

d lig

ht [k

eV]

12

13

14

15

16

17

18

19

20

21

ZnSe: Light-Heat correlation

9

ΔE@2615 keV:13 keV FWHM

• Operation of a tower of 32-40 Zn82Se crystals at LNGS.

‣ Option1: use the Cuoricino cryostat in hallA (presently hosting CUORE-0), if CUORE-0 stops in 2015.

‣ Option2: use the cryostat in hallC (presently running the CUORE-0 and LUCIFER R&Ds). Needs cryostat update. Cuoricino cryostat:

• Inner shield:

- 1cm Roman PbA (210Pb) < 4 mBq/Kg

• External shield:

- 20 cm Pb

- 10 cm Borated polyethylene

• Nitrogen flushing to avoid Rn contamination.

Plan for a ZnSe array in 2015

10

LUC

IFER4 crystals

per floor

ZnSe: schedule

11

2012 2013 2014 2015

Light detector R&D

Thermistors prod.

Crystal growth R&D

82Se prod. (15 kg)

Enriched crystalsproduction

Array assemblyThe most crucial part is represented by the crystal growth. The supplier (Ukraine) is presently fine-tuning the complicate procedure (that starts with metal Zn and metal Se). The request of minimizing the 82Se waste is a complicate issue. The target is to reach > 75 % efficiency.

Energy [keVee]1500 2000 2500 3000 3500 4000

Ligh

t Yie

ld [k

eV/M

eV]

9

10

11

12

13

14

15

Do we need a μ-veto?

• The event includes hits on close detectors (multi-site event)

‣ Multiple γ’s produced by μ interactions in the materials close to the detector.

• Easy to remove in this case, but what if one has one hit only?

‣ Montecarlo simulation under development.12

ZnSe data: 1 β/γ event above the 2615 keV line in 580 hours

~ 5x10-2 counts/keV/kg/y

β/γα

Candidate #2: ZnMoO4

13

DBD Isotope:

Q-value[keV]

isotopic abundance

Light Yield [keV/MeV]

QF

100Mo

3034

10%

1.5

0.2

Largest crystal operated so far: 330g

ZnMoO4 .

14 Energy [keV]0 500 1000 1500 2000 2500 3000 3500 4000

TVR

[a.u

.]

-14

-12

-10

-8

-6

-4

-2

0

2

4

Energy [keV]500 1000 1500 2000 2500 3000 3500 4000

LY [k

eV/M

eV]

0.2

0.4

0.6

0.81

1.2

1.4

1.6

1.82

2.2

2.4

Eur. Phys. J. C 72 (2012) 2142

α source

β/γ source

α source

β/γ source

0νDBD

0νDBDH

eat s

igna

l sha

pe p

aram

eter

Ligh

t Yie

ld [k

eV/M

eV]

Discrimination using the shape of the heat signal!

Excellent discrimination using the light signal

Entries 2821

Mean 2742

RMS 1449

Integral 2821

1 2 3 4 5 6 7 8 9 10310×

1

10

210

Po

Energy [keV]

Cou

nts /

9 k

eV 210

226Ra

222Rnsource_ <

Po218

Bi ï Po214 214

ZnMoO4

15

11528

1.153e+04

Energy [keV]500 1000 1500 2000 2500

coun

ts / 3

keV

0

10

20

30

40

50

60

70

80

90

Entries 11528

2.925

2590 2600 2610 2620 2630 26400

10

20

30

40

50

60

70FWHM= 6.3 keV

Energy resolution similar to CUORICINO

(2x better than ZnSe)

β/γ source

Internal contaminations:232Th < 1.4 pg/g

Simulated Čerenkov emission spectrum from

1.5 MeV γ in TeO2 at low temperatures.

Simulated emitted Čerenkov light as a

function of β/γ energy.

Candidate #3: TeO2

16Photon Energy (keV)

0 500 1000 1500 2000 2500

Che

renk

ov li

ght (

eV)

0

100

200

300

400

500

600

700

800

900

1000

/ ndf 2r 3.723 / 5

Prob 0.5899

p0 1.143e-09± -3.111e-08 p1 4.448e-06± 0.0001678

p2 0.00428± 0.1222

p3 0.8393± -10.43

/ ndf 2r 3.723 / 5

Prob 0.5899

p0 1.143e-09± -3.111e-08 p1 4.448e-06± 0.0001678

p2 0.00428± 0.1222

p3 0.8393± -10.43

Prod

uced

Che

renk

ov li

ght[e

V] 867 eV

0νD

BD

β/γ energy [keV]

m)µ (h0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

hdN

/d

0

100

200

300

400

500

600

700

800

dN/dλ

[μm

-1]

λ [μm]

TeO2 does not scintillate, however MeV β’s emit Čerenkov light, unlike α’s [ T. Tabarelli de Fatis, Eur. Phys. J. C 65 (2010) 359].

Heat energy [keV]0 1000 2000 3000 4000 5000

Ligh

t ene

rgy

[keV

]

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4 / ndf 2r 11.9 / 6 [keV] thE 64.47± 360.3

Yield [eV/MeV] 2.846± 48.33

/ ndf 2r 11.9 / 6 [keV] thE 64.47± 360.3

Yield [eV/MeV] 2.846± 48.33

210Po-α contamination

β/γ source

Detected β/γ light: 48 eV/MeV 105 eV @2.527 MeV

0νDBD

Čerenkov from a CUORE crystal

17

• Detected light not sufficient to discriminate α’s event by event.

‣ Needs light detector development: light collection, energy resolution.

Signal/Noise of light detector0 1 2 3 4 5 6 7 8

y]

26D

BD se

nsiti

vity

[10

i68

% 0

0

2

4

6

8

10

12

14

16

18

bkg. [counts/keV/kg/y]_

0.01 with 90% enr.0.04 with 90% enr.0.010.04

CUORE + Čerenkov + enrichment

18

γ bkg. (MC)= 0.001

Signal/Noise of light detector0 1 2 3 4 5 6 7 8

y]

26D

BD se

nsiti

vity

[10

i68

% 0

0

2

4

6

8

10

12

14

16

18

bkg. [counts/keV/kg/y]_

0.01 with 90% enr.0.04 with 90% enr.0.010.04

CUORE + Čerenkov + enrichment

18

Pres

ent:

S =

105

, N =

75

eV γ bkg. (MC)= 0.001

Signal/Noise of light detector0 1 2 3 4 5 6 7 8

y]

26D

BD se

nsiti

vity

[10

i68

% 0

0

2

4

6

8

10

12

14

16

18

bkg. [counts/keV/kg/y]_

0.01 with 90% enr.0.04 with 90% enr.0.010.04

CUORE + Čerenkov + enrichment

18

Pres

ent:

S =

105

, N =

75

eV

Targ

et

γ bkg. (MC)= 0.001

• ZnSe

‣ Target: build a bolometric experiment of ~ 10 kg of 82Se‣ Status: Isotope in production, crystal growth under optimization.

• ZnMoO4

‣ Target: build a bolometric experiment of ~ 10 kg of 100Mo ‣ Status: MoU with IN2P3 and ITEP in consideration.

• TeO2

‣ Target: develop light detectors with S/N improved by a factor 4.‣ Status: KIDs and Neganov-Luke detectors under development.

Overview until 2015

19

• ZnSe

‣ Target: build a bolometric experiment of ~ 10 kg of 82Se‣ Status: Isotope in production, crystal growth under optimization.

• ZnMoO4

‣ Target: build a bolometric experiment of ~ 10 kg of 100Mo ‣ Status: MoU with IN2P3 and ITEP in consideration.

• TeO2

‣ Target: develop light detectors with S/N improved by a factor 4.‣ Status: KIDs and Neganov-Luke detectors under development.

Overview until 2015

19

baseline

CUORICINO data

α’s (+ β/γ’s?)

130Te 82Se 100Mo

α’s + β/γ’sso far:82Se and 100Mo arrays will answerCUORE (130Te) will measure

the reduction

Ultimate background: β/γ• So far, we do not have a measure of the β/γ background above

2615 keV with bolometric arrays à la CUORICINO, and of the CUORE β/γ background at the 130Te Q-value.

20

• To cover the entire inverted hierarchy of neutrino masses a bolometric experiment operated in the CUORE cryostat will require:

Crystal R&D Status ΔE(keV)

Bkg. reach

Enrichment cost (€/g)

Zn82Se

Zn100MoO4

130TeO2 (Čerenkov)

Isotope/crystals in preparation 13 likely 75

(contract)

MoU to be signed 6 likely 100-130(estimate)

CUORE experience Light Detectors 5 to be

proved10-20(estimate)

✓1 ton isotope✓Bkg < 1 count/keV/ton/y✓ΔE ~ 5 keV FWHM

Conclusion: which one after CUORE?

21

Backup

• natTeO2 bolometers (34% 130Te), 750g each (ΔE =5 keV FWHM)

• Past: Cuoricino ‣ 62 bolometers

11 kg (130Te)×2y, Bkg: 0.16 cpy/keV/kg

‣ T0ν1/2 > 2.8×1024 years (90% CL)〈mββ〉< 300~700 meV

• Future: Cuore (data taking in 2015)‣ Expected bkg: 0.01~0.04 cpy/keV/kg‣ Exp. T0ν1/2 > 1.6 ×1026 years 〈mββ〉< 40~94 meV

• Present: Cuore-0, a CUORE-like tower.‣ same mass of Cuoricino, 0.05 cpy/keV/kg.

Energy [keV]2480 2500 2520 2540 2560 2580

Rat

e [c

ount

s/ (1

keV

)]

05

101520

253035

404550

Best Fit68% C.L.90% C.L.

0νDBD

60Co γ + γ Cuoricino

130Te CUORE

23

CUORE - @LNGS

CUORE: 988 bolometers

750 kg TeO2

200 kg 130Te

• The spread in the values does not influence the isotope choice.

• Generates problems when comparing exclusions and evidences from experiments running different isotopes.

March 26, 2012 0:7 WSPC/INSTRUCTION FILE bb0

16

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

M0ν

QRPAEDFPHFBIBM-2LSSM

76Ge 82Se 96Zr 100Mo116Cd124Sn 130Te 136Xe48Ca 150Nd128Te

Fig. 3. Values of the NME calculated with the methods in Tab. 2 55.

and we do not discuss the advantages and disadvantages of each of them. We onlypresent the references to the original papers in Tab. 2 and the latest results in Fig. 3.

From Fig. 3 we reach the following conclusions:

(1) The LSSM value of each NME is typically smaller than the corresponding onecalculated with other approaches. Moreover, the LSSM value of each NMEdepends weakly on the nucleus, except for the double-magic nucleus 48Ca. If0νββ-decay of different nuclei will be observed in future experiments, this char-acteristic feature of the LSSM can be checked, because the LSSM predicts thefollowing ratio of half-lives of different nuclei:

T 0ν1/2(Z1, A1)

T 0ν1/2(Z2, A2)

!G0ν(Q2, Z2)

G0ν(Q1, Z1)(68)

(2) There is a large discrepancy between the values of NMEs calculated with dif-ferent approaches. The ratios of the maximal and minimal values of each NMEare 3.1 (48Ca), 2.4 (76Ge), 2.0 (82Se), 3.7 (96Zr), 1.8 (100Mo), 1.3 (116Cd), 1.8

Table 2. Methods of calculation of nuclear matrix elements of0νββ-decay.

Method References

Quasi-particle Random Phase Approximation (QRPA) 49,50

Energy Density Functional method (EDF) 51

Projected Hartree-Fock-Bogoliubov approach (PHFB) 52

Interacting Boson Model-2 (IBM-2) 53

Large-Scale Shell Model (LSSM) 54

NME .

24