The LUNA experiment at the Gran Sasso Laboratory

X C I X C o n g r e s s o N a z i o n a l e S I F - T r i e s t e 2 3 - 2 7 s e p t e m b e r 2 0 1 3

Roberto Menegazzofor the LUNA Collaboration

outlook

LUNA: Why going underground to measure nuclear fusion reactions in a laboratory ?

The Sun: p-p chain, CNO cycle and solar neutrinosNucleosynthesis at work: 26AlHot environment: BBN and Novae

The LUNA-MV project: a big step forward

H burning → He He burning → C, O, NeC/O ... Si burning → Fe explosive burning

Solar Neutrinos and element abundances in

stars and BBN

TSUN = 0.015 GK ~ 2 keVTRGB = 0.1 GK ~ 80 keVTnovae = 0.3 GK ~ 140 keV

Nucleosynthesis

Reaction Rate for Charged Particles

σ (E ) = S(E)

Eexp −31.29 ⋅Z1 ⋅Z2 ⋅

µ

E

⎛

⎝⎜⎜

⎞

⎠⎟⎟

Gamow factor Astrophysical factor

Gamow Energy for H-burning reactions:few to several tens keV

Nuclear reactions that generate energy and synthesize elements take place inside the stars in a relatively narrow energy window: the Gamow peak

in the Sun: T = 1.5 107 KKT = 1 keV << Ecoul (0.5-2MeV)

pbarn < σ < nbarn

narrow resonance

Measurements

S(E)

non-resonant process

Sub-thr. resonance

tail of broad resonance

Extrapolations

sometimes extrapolation fails!

Extrapolation risks

Extrapolation down to astrophysical energies is

needed but ...

How to improvesignal to noise ratio

The cross section varies strongly with energyprecise beam energy resolutionhigh purity and stable targets

and it’s very small at low energies

setup efficiency

natural background(cosmic radiation,

radioactive isotopes)

beam induced background

Direct cross section measurements feasible with reduced cosmic-ray induced background

Experimental requirements

Underground measurements

Surface Gran Sasso

Why Going Underground ?

neutrons: 4 x 10-6 cm-2s-1 with fission and (α,n)muons: 1 /(m2.h), E > 1 TeV@ LNGS

Laboratori Nazionali del Gran Sasso

LUNAII

LUNAI

1400 m rock overburdenFlux attenuation: n 10-3

μ 10-6

underground area 18000 m2

support facilities on the surface

Laboratori Nazionali del Gran Sasso

LUNAII

LUNAI

1400 m rock overburdenFlux attenuation: n 10-3

μ 10-6

underground area 18000 m2

support facilities on the surface

LUNA Ibeams = p, αCurrent max = 1 mAVoltage range = 1 - 50 kVBeam energy spread: 20 eVLong term stability (8 h): 10-4 eV

LUNA IICockcroft-Walton acceleratorbeams = p, αCurrent max = 500 μA (protons)

250 μA (alphas)Voltage range = 50 - 400 kVAbsolute energy error: ±300 eVBeam energy spread < 100 eVLong term stability (1h): 5 eV

3He+3He → 4He + 2ppossible solution of the Solar Neutrino Problemcross section measured directly at Gamow energies

no extrapolation needed !

p+d → 3He + γ

C. Casella et al., NPA 706 (2002) 203

Measurements at LUNA I

count rate @ lowest energy: 2 cts/monthlowest cross section: 0.02 pbarnbackground < 4*10-2 cts/d in ROI

S(0) = 5.32(8) MeVb

R. Bonetti et al., PRL 82 (1999) 26

3He+4He → 7Be + γkey reaction in the p-p chain for 7Be e 8B neutrinos in the Sunfundamental for 7Li in BBN

gamma-prompt and activation method

uncertainties on the neutrino fluxes

8B -> from 12% to 10%7Be -> from 9.4% to 5.5% S3,4(0) = 0.560(17) keVb

A. Caciolli et al., EPJA 39 (2009) 179

Measurements at LUNA II

Measured background attenuation factor for the 3He(α,γ)7Be setup is ~ 10-5 !!!(i.e. 1.9 and 0.8 counts/day with ΔE = 20 keV)

F. Confortola et al., PRC 75 (2007) 065803

14N(p,γ)15O Bottleneck of the CN cyclestudied both with solid and

gas targetCNO neutrino fluxes reduced by a factor of 2Globular Cluster Age increased by 0.7 - 1.0 Gyreduced uncertainties below 8%

15N(p,γ)16O

Link the first and second CNO cycles

totally covered the Nova Gamow peakreduced the S-factor by a factor of 2 reduction 16O produced by novae explosions

M. Marta et al., PRC 83(2011)045804

A. Caciolli et al., A&A 533(2011)A66

CNO cycle

25Mg(p,γ)26Al

-26

0 5+

228 0+

6496 5+(4+) 190

6436

6414

6399

6364

6343

6280

4-

0+

2-

3+

4-

3+

37

58

93

108

130

26Al N

ov

ae exp

losiv

e

Bu

rnin

g (T

9 >0

.1)

AG

B o

r W-R

Stars (T

9 ~0

.05

)

Ex

(keV)

J

ECM

(keV)

1809

0

417 3+

6610 3- 304

26Mg

isomeric

state

Reactions per

day (200uA)

0.5·107

1·105

25

25

26Al 27Al

28Si 29Si 30Si

25Mg 24Mg 26Mg

23Na (n,�)

(p,�)

(n,p)

B. Limata et al., PRC 82 (2010) 015801

5

Figure 4. Undergroumd γ-spectrum of the 25Mg(p, γ)26Al reaction recorded atresonance energy Er = 189 keV.

3. Acknowledgments

”We would like to thank Helmut Baumeister, University of Munster (Germany) andMassimo Loriggiola, INFN Laboratori Nazionali di Legnaro (Itlay) for the Magnesiumtarget production. This work was supported by INFN and in part by the EuropeanUnion (TARI RII3-CT-2004-506222), the Hungarian Scientific Research Fund (K68801and T49245) and the Deutsche Forschung Gemeinschaft (DFG) (Ro429/41).”

-

References

[1] R.Diehl et al.,A&A, 298 (1995) 445[2] J.Knodlseder et al.,A&A, 345 (1999) 813[3] R.Diehl et al.,A&A, 411 (2003) L451L455[4] A. E. Champagne et al., Nucl.Phys.A 402 (1983) 179[5] P.M.Endt et al., Nucl.Phys.A 459 (1986) 61[6] Ch.Iliadis et al., Nucl.Phys.A 512 (1990) 509[7] D.C.Powell et al., Nucl.Phys.A 644 (1998) 263[8] A.Arazi et al., PRC 74 (2006) 025802[9] C.Angulo et al. Nucl.Phys.A 656(1999)3[10] A.Best Diploma Thesis, Ruhr-Universitat-Bochum, January 2007[11] M.Mehrhoff et al., Nucl.Instr.Meth.A 132 (1997) 671[12] Geant4 Collaboration, Nucl.Instr.Meth.A 506 (2003) 250, http://geant4.cern.ch[13] P.M.Endt, P.De Wit, C.Alderliesten, Nucl.Phys.A 476 (1988) 333[14] A.Spyrou et al., Phys.Rev.C 76 (2007) 015802[15] A.Best et al., to be published[16] A.Formicola et al., Nucl.Instr.Meth.A 507 (2003) 609

Mg - Al cycle

190 keV resonanceCharge = 25 C

AGB and RGB stars massive stars classical novae

17O 18F 14N 13C

13N

12C 15N 16O 18O

15N 17F

LUNA

Recent activity: 17O(p,γ)18F

17O+p is of paramount importance for understanding hydrogen-burning in different stellar environments:

It is important for galactic synthesis of 17O, 18F, and predicted O isotopic ratios in presolar grains.

It affects directly the production of 17O,18O, 18F, and 19F in Classical Novae

target preparation

Ta2O5 targets17O enrichment up to 69% (with 5% 18O)backing treated with citric acid and cooled to 25 °C during the anodization process

stoichiometry and isotopic ratio from 151 keV 18O(p,γ)19F resonance scans, RBS and SIMS measurements Targets stable up to ~ 20 CCaciolli et al. EPJA48(2012)144

28 keV @ 151 keV

17O(p,γ)18F: Prompt Gamma Spectroscopy

efficiency measured at three different distances to correct for summing effectlead shielding to reduce the natural background at low γ energies by a factor of 2new branchings observed

18F$

17O+p$

Ep=$193$keV$5789$

1080$

937$

3839$3791$

3358$

3134$2523$2101$

1041$

Ex(keV)$

also measured with activation technique

S-Factor

S-factor measured for the First time down to 200 keVUncertainty reduced by a factor of 4 with respect to previous worksNovae Gamow Peak covered totally for the first time with experimental data

D. Scott et al. PRL 109 (2012) 208001 Editors' Suggestion

NS61CH03-Fields ARI 14 September 2011 11:45

log 1

0 (Li

/H)

[Fe/H] = log10 [(Fe/H)/(Fe/H)◉]

–9.0

–9.5

–10.0

–10.5

–11.0

–11.5

–12.0

–3.0 –2.5 –2.0 –1.5 –1.0 –0.5 0

CMB + BBN

7Li6Li limits

6Li6Li◉

7Li

6Li

Figure 3Lithium abundances in selected metal-poor Galactic halo stars. For each star, both lithium isotopes areplotted versus the star’s metallicity: [Fe/H] = log10[(Fe/H)obs/(Fe/H)!]. Upper points show 7Li. Theflatness of 7Li versus iron is known as the Spite plateau; it indicates that the bulk of the lithium is unrelatedto Galactic nucleosynthesis processes and thus is primordial. The horizontal band gives the CMB+WMAPprediction; the gap between this prediction and the plateau illustrates the 7Li problem. Points below theSpite plateau show 6Li abundances; the apparent flatness of these points constitutes the 6Li problem. Curvesshow predictions of a Galactic cosmic-ray nucleosynthesis model. Points have been corrected for pre-main-sequence depletion. Abbreviation: CMB, cosmic microwave background. Reproduced from Reference 46with permission.

Moreover, the Spite plateau level measures the primordial abundance. Thanks to the sustainedeffort of several groups (46, 48–56), a large sample of halo stars have measured lithium abundances.The dominant errors are systematic. A careful attempt to account for the full lithium error budgetfound (57)

LiH

= (1.23+0.68−0.32) × 10−10, 9.

where the 95%-CL error budget is dominated by systematics (see also Section 3.1).Finally, lithium has now been observed in stars in an accreted metal-poor dwarf galaxy. The

Li/H abundances are consistent with the Spite plateau, indicating the plateau’s universality (58).

2.2.3. 6Li. Due to the isotope shift in atomic lines, 6Li and 7Li are in principle distinguish-able spectroscopically. In practice, the isotopic splitting is several times smaller than the thermalbroadening of stellar lithium lines. Nevertheless, the isotopic abundance remains encoded in thedetailed shape of the lithium absorption profile.

High–spectral resolution lithium measurements in halo stars attain the precision needed toobserve isotope signatures. Some researchers have claimed to detect 6Li with isotopic ratios in the

www.annualreviews.org • Primordial Lithium Problem 53

Ann

u. R

ev. N

ucl.

Part.

Sci

. 201

1.61

:47-

68. D

ownl

oade

d fr

om w

ww

.ann

ualre

view

s.org

by U

nive

rsita

deg

li St

udi d

i Pad

ova

- Agr

ipol

is L

egna

ro o

n 05

/29/

12. F

or p

erso

nal u

se o

nly.

7Li 6Li

Unfolding primordial abundancesObservation of a set of primitive objects, born when the Universe was young, and extrapolate to zero metallicity: Fe/H, O/H, Si/H → 0

7Li abundance: observation of the absorption line at the surface of metal-poor stars in the halo of our Galaxy6Li abundance: no direct observation. From the asymmetry of the 7Li absorption line

7Li abundance from line intensity6Li/7Li from line shape

M.Asplund et al. Astrophys. J. 644 (2006)

6Li and 7Li nucleosynthesis

-14.0

7Li

6Li

The Li problems

Observational Results:

7Li abundance is 3-4 times lower than foreseen (Spite Plateau): well established 7Li problem

6Li abundance is orders of magnitude higher than expected (Asplund 2006)

However the Second Lithium problem is debated, because convective motions on the stellar surface can give an asymmetry of the absorption line, mimicking the presence of 6Li

Uncertainty from 2H(α,γ)6Li

d(α,γ)6Li reaction Why is it important ? How much do we know about it ?

- d(α,γ)6Li is the main reaction for 6Li production- In BBN, this reaction occurs at energies in the range 50 < Ecm < 400 keV- No direct measurement exists at Ecm < 650 keV (Elab < 1950 keV)- Theoretical calculations for the astrophysical S-factor differ by more

than one order of magnitude

BBN

LUNA

F.Hammache et al. PHYS. REV. C 82 (2010)

LUNA direct measurement at Ecm ≤ 133 keV

(n,n’γ) reactions on the surrounding materials (Pb, Ge, Cu)

γ-ray background in the ROI for the D(α,γ)6Li DC transition (∼1.6 MeV)

LNGS constraints on available beam time and neutron production

Beam Induced Background

well understood beam induced background

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

0.01

0.1

1

10

100

Yie

ld [c

ount

s/C

] 1500 1550 1600 1650 1700

1

2

3

4

5

63Cu

63Cu

63Cu

56Fe

65Cu 52Cr

71mGe

56Fe63Cu

63Cu

40K

208Tl

214Bi

206Pb

Eα = 400 keV, P(D) = 0.3 mbar, Q = 263 C

Spectrum from D(α,γ)6Li

M. Anders et al., EPJA 49 (2013) 28

Deuterium inlet

Radon Box (N2 flushing)

Alpha beam

Ge Detector

Lead shield

Deuterium exhaust

Silicon Detector

Pipe Calorimeter

Crown HPGe

Detector

- Germanium detector close to the beam line to increase the detection efficiency

- Pipe to reduce the path of scattered deuterium, to minimize the d(d,n)3He reaction yield

- Target length optimized- Copper removal - Silicon detector to monitor the neutron

production through the d(d,p)3H reaction- Lead, Radon Box to reduce and stabilize

Natural Background- Borated polyethylene envelope to reduce

neutron contamination

Experimental setup

1500 1550 1600 1650 1700Eγ [keV]

1

2

3

4

5

Yie

ld [c

ount

s/C]

Run 1: 400 keVRun 1: 280 keV

P(D2) = 0.3 mbar, Q(400) = 263 C, Q(280) = 278 C

What should we observe ? A single γ-ray withEγ = 1473.48 + Ecm ± Edoppler

400 keV280 keV

The shape of Beam Induced Background spectra weakly depends on the α-beam energy

Some results

Next measurements

18O(p,α)15N, 18O(p,γ)19F, and 23Na(p,γ)24Mg under study

17O(p,α)14N data taking for 65 keV and 183 keV resonances (end 2013)

22Ne(p,γ)23Na setup under construction and calibration (2014)

A new accelerator underground

Limits of a 400 kV accelerator

Solar fusion reactionsStellar Helium and Carbon burningNeutron sources for astrophysical s-processes

proposed solutions:

LUNA-MV at Gran Sasso National Laboratory (Italy)CANFRANC (Spain)Felsenkeller (Germany) <-- shallow undergroundDIANA (formerly part of DUSEL) (United States)ChinaSouth America

A new, higher energy underground accelerator is needed !

April 2007: a Letter of Intent (LoI) was presented to the LNGS Scientific Committee (SC) containing key reactions of the He burning and neutron sources for the s-process:

12C(α,γ)16O13C(α,n)16O22Ne(α,n)25Mg(α,γ) reactions on 14,15N and 18O

These reactions are relevant at higher temperatures (larger energies) than reactions belonging to the hydrogen-burning studied so far at LUNA

Single ended 3.5 MV positive ion accelerator

LUNA - MV project

•  In a very low background

environment such as LNGS, it is

mandatory not to increase the

neutron flux above its average value

Study of the LUNA-MV neutron shielding by

Monte Carlo simulations

OnC 1613 ),(αα beam intensity: 200 µA

Target: 13C, 2 1017at/cm2 (99% 13C enriched)

Beam energy(lab) ≤ 0.8 MeV

MgnNe 2522 ),(αα beam intensity: 200 µA

Target: 22Ne, 1 1018at/cm2 Beam energy(lab) ≤ 1.0 MeV

OC 1612 ),( γαOnC 1613 ),(α from α beam intensity: 200 µA

Target: 13C, 1 1018at/cm2 (13C/12C = 10-5) Beam energy(lab) ≤ 3.5 MeV

•  Maximum neutron production rate : 2000 n/s

•  Maximum neutron energy (lab) : 5.6 MeV

the estimated n-flux (Fluka & Geant 4 simulations) will increase less than 1% of the LNGS natural flux !

LUNA - MV project

“Progetto Premiale” LUNA - MV

Italian Research Ministry financed the LUNA-MV special project with 2.8 MEuro in

2012Time schedule: 2012 - 2013 Hall preparation, Tender for the accelerator and Shielding2014 Beam lines R&D, Infrastructures2015 Accelerator installation, Beam lines construction, Detectors installation2016 Calibration of the apparatus and first tests of beam on target

Location at LNGS underground laboratory still uncertain !

New funding from 2nd “Progetto Premiale” under discussion !

Grazie

Atomki (Zs. Fülöp, Gy. Gyurky, E. Somorjai, T. Szucs), Bochum (C. Rolfs, F. Strieder, H.P. Trautvetter),

Dresden (M. Anders, D. Bemmerer, Z.Elekes), Edinburgh (M. Aliotta, T. Davinson, D. A. Scott),

Genova (F. Cavanna, P. Corvisiero, P. Prati), LNGS (A. Formicola, M. Junker),

Milano (C. Bruno, A. Guglielmetti, D. Trezzi), Napoli (G. Imbriani, A. Di Leva),

Padova (C. Broggini, A. Caciolli, R. Depalo, R.M.), Roma (C. Gustavino),

Teramo (O. Straniero), Torino (G. Gervino)

New collaborators are welcome!