1. The LUNA experiment at Gran Sasso Laboratory: studying stars by going underground Alessandra Guglielmetti Università degli Studi di Milano and INFN, Milano, ITALY Laboratory Underground Nuclear Astrophysics Outline: -Nuclear fusion reactions in stars: why measuring their cross section? -Why going underground to perform these experiments? -The LUNA experiment: most important and recent results - On-going measurements and future perspective: the LUNA-MV project

2. Why studying nuclear fusion reaction cross sections? -Stars are powered by nuclear reactions -Among the key parameters (chemical composition, opacity, etc.) to model stars, reactions cross sections play an important role - They determine the origin of elements in the cosmos, stellar evolution and dynamic - Many reactions ask for high precision data.

3. p + p  2 H + e + + p + e - + p  2 H + e + + 3 He + p  4 He + e + + 7 Be + e -  7 Li + 8 B  8 Be + e + + 13 N  13 C + e + + 15 O  15 N + e + + 17 F  17 O + e + + p-p chain CNO cycle Neutrino production in stars Solar neutrino puzzle: solved! Neutrino flux from the Sun can be used to study: Solar interior composition Neutrino properties ONLY if the cross sections of the involved reactions are known with enough accuracy

4. Big Bang nucleosynthesis Production of the lightest elements (D, 3 He, 4 He, 7 Li, 6 Li) in the first minutes after the Big Bang The general concordance between predicted (BBN theory) and observed (stellar spectra) abundances gives a direct probe of the Universal baryon density CMB anisotropy measurements (WMAP/Plank satellites) give an independent measurement of the Universal baryon density The agreement of the two results has to be understood in terms of uncertainties in the BBN predictions

5. He 3 4 Be 7 Li 7 H D p n 2 1 34 2 8 9 6 7 11 12 10 3 5 Li 6 13 1. n  p + e - + 2.p + n  D +  3.D + p  3 He +  4.D + D  3 He + n 5.D + D  3 H + p 6. 3 H + D  4 He + n 7. 3 H + 4 H  7 Li +  8. 3 He + n  3 H + p 9. 3 He + D  4 He + p 10. 3 He + 4 He  7 Be +  11. 7 Li + p  4 He + 4 He 12. 7 Be + n  7 Li + p 13. 4 He + D  6 Li +  BBN reaction network Apart from 4 He, uncertainties are dominated by systematic errors in the nuclear cross sections

6. Hydrogen burning 4p  4 He + 2e + + 2 e + 26.73 MeV Ne-Na cycle Mg-Al cycle CNO cycle

7. The importance of going underground… Sun: kT = 1 keV E C ≈ 0.5-2 MeV E 0 ≈ 5-30 keV for reactions of H burning kT but also E 0 << E C !! Cross sections in the range of pb-fb at stellar energies with typical laboratory conditions reaction rate R can be as low as few events per month

8. R has to be compared with background B B beam induced : reactions with impurities in the target, collimators,… secondary processes B env : natural radioactivity mainly from U and Th chains B cosmic : mainly muons Rate and background

9. LUNA site LUNA 1 (1992-2001) 50 kV LUNA 2 (2000  …) 400 kV Laboratory for Underground Nuclear Astrophysics RadiationLNGS/surface Muons Neutrons 10 -6 10 -3 LNGS (1400 m rock shielding  4000 m w.e.) LUNA MV (2015->...)

10. Cross section measurement requirements 3MeV < E  < 8MeV: 0.5 Counts/s 0.5 Counts/s 3MeV < E  < 8MeV 0.0002 Counts/s GOING UNDERGROUND HpGe Pb Cu underground passive shielding is more effective since μ flux,that create secondary γ’s in the shield, is suppressed  E  <3MeV  passive shielding for environmental background radiation

11. Background reduction - Si detectors - alpha Underground+Pb vs. Overground: factor ~10-15 reduction in the range 200 keV - 2.5 MeV

12. 3 He( 3 He,2p) 4 He Fundamental reaction of the p-p cycle Measured down to the lower edge of the solar Gamow peak No resonances  no nuclear explanation for the solar neutrino puzzle

13. E beam  50 – 400 keV I max  500  A protons I max  250  A alphas Energy spread  70 eV Long term stability  5eV/h LUNA 400kV accelerator

14. S 14,1 /5 S 14,1 x5 Standard CF88 14 N(p,γ) 15 O cross section influences CNO neutrino flux  Solar metallicity Globular cluster age

15. High resolution measurement Solid target + HPGe detector  single γ transitions  Energy range 119-367 keV  summing had to be considered Gas target+ BGO detector  high efficiency  total cross section  Energy range 70-230 keV High efficiency measurement CNO neutrino flux decreases of a factor  2 Globular Cluster age increases of 0.7 – 1 Gyr: new lower limit on the Age of the Universe T>14 Gy S 0 (LUNA) = 1.61 ± 0.08 keV b

16. He 3 4 Be 7 Li 7 H D p n 2 1 34 2 8 9 6 7 11 12 10 3 5 Li 6 13 1. n  p + e - + 2.p + n  D +  3.D + p  3 He +  4.D + D  3 He + n 5.D + D  3 H + p 6. 3 H + D  4 He + n 7. 3 H + 4 H  7 Li +  8. 3 He + n  3 H + p 9. 3 He + D  4 He + p 10. 3 He + 4 He  7 Be +  11. 7 Li + p  4 He + 4 He 12. 7 Be + n  7 Li + p 13. 4 He + D  6 Li +  BBN reaction network

17. The two Lithium problems 1)The BBN 7 Li predictions are a factor 2-4 higher than observations: a nuclear physics solution is highly improbable (e.g 3 He( 4 He,  ) 7 Be measurement at LUNA) 2) The amount of 6 Li predicted by the BBN is about 3 oom lower than the observed one in metal poor stars (debated but still «true» for a few metal poor stars) BBN predicts 6 Li/ 7 Li= 2 * 10 -5 much below the detected levels of about 6 Li/ 7 Li= 5 * 10 -2 Necessary to constrain nuclear physics input: 2 H( ,  ) 6 Li

18. 2 H( ,  ) 6 Li: available data No data in the BBN energy range! Upper limits (indirect meas)

19. 2 H( ,  ) 6 Li at LUNA: Experimental setup Strong beam induced background due to: 1)Rutherford scattering of 4 He beam on 2 H target 2) 2 H(d,n) 3 He reaction 3)Inelastic neutron scattering on different materials (Cu, Pb, Ge,…)   background in the 2 H( ,  ) 6 Li RoI The beam induced background weakly depends on the beam energy

20. About the experimental setup

22. 2 H( ,  ) 6 Li at LUNA: gamma spectra An irradiation at one given beam energy can be used as a background monitor for an irradiation at a different beam energy, if the two ROIs do not overlap Natural background subtracted 400 keV data (grey filled) 280 keV data (red empty) rescaled to take into account the weak energy dependence of the beam induced background

23. New LUNA data From the new data on the 2 H( ,  ) 6 Li reaction: 6 Li/ 7 Li = (1.5 ± 0.3) * 10 -5 2 H( ,  ) 6 Li at LUNA: results Standard BBN production as a possible explanation for the reported 6 Li detections is ruled out. “Non standard” physics solutions? M. Anders et al., Phys. Rev. Lett. 113 (2014) 042501

24. 17 O(p,  ) 14 N and 18 O(p,  ) 15 N reactions In AGB stars ( T=0.03-0.1 GK ) CNO cycle takes place in H burning shell Measured 17 O/ 16 O and 18 O/ 16 O abundances in pre-solar grain give information on AGB surface composition Information on mixing processes if cross sections are well known

25. 17 O(p,  ) 14 N and 18 O(p,  ) 15 N reactions Q = 1.2 MeV Two narrow resonances at 70 and 193 keV Main goal at LUNA: 70 keV resonance 17 O(p,  ) 14 N 18 O(p,  ) 15 N Q = 4 MeV Two narrow resonances at 95 and 152 keV Main goals at LUNA: rescan excitation function 95 keV resonance (strength and energy) Measure below 70 keV

26. 17 O(p,  ) 14 N and 18 O(p,  ) 15 N experimental setup proton beam from LUNA 400 kV enriched 17 O or 18 O solid targets 8 silicon detectors foils of Al Mylar to stop backscattered protons  low alpha particle energy (200- 250 keV for 17 O(p,  ) 14 N reaction) Beam entrance Silicon detectors (8 in total) Target position

27. Clear peak in the green ROI (determined from 193 keV resonance) Shape looks reasonable No obvious structures in the off resonance The 17 O(p,  ) 14 N reaction: 70 keV resonance

28. Comparison with literature is not easy to make (several re- analysis, not all published) Our value is higher than previously reported  important astrophysical consequences

29. 151 keV resonance 217 keV resonance 334 keV resonance 95 keV resonance PRELIMINARY The 18 O(p,  ) 15 N reaction Data taking completed. Data analysis on going

30. The 22 Ne(p,  ) 23 Na reaction NeNa cycle of H burning: -massive stars -RGB and AGB stars -classical novae and supernovae IA Impact on the abundances of: Ne, Na, Mg and Al isotopes

31. The 22 Ne(p,  ) 23 Na reaction LUNA range Two large HPGe detectors (135% and 90% relative efficiency) 15-20 cm thick lead shielding 5 cm additional copper shield for 55° detector

32. The 22 Ne(p,  ) 23 Na reaction Experiment carried out within the “green” box “red” resonances directly observed for the first time for the “black” resonances at 71, 105 and 215 keV an upper limit was determined  BGO phase almost concluded (71 and 105 keV res + DC component) F. Cavanna et al., Phys. Rev. Lett. 115 (2015) 252501

33. LUNA 400 kV new program 2016-2019: a bridge toward LUNA MV 13 C( ,n) 16 O – neutron source for the s-process (formation of heavy elements) 12 C(p,  ) 13 N and 13 C(p,  ) 14 N – relative abundance of 12 C- 13 C in the deepest layers of H-rich envelopes of any star 2 H(p,  ) 3 He – 2 H production in BBN 22 Ne( ,  ) 26 Mg – competes with 22 Ne( ,n) 25 Mg neutron source for the s-process (formation of heavy elements) 6 Li(p,  ) 7 Be – improves the knowledge of 3 He( ,  ) 7 Be key reaction of p-p chain

34. LUNA MV- scientific program 13 C( ,n) 16 O and 22 Ne( ,n) 25 Mg : neutron sources for the s-process (formation of heavy elements) 12 C( ,  ) 16 O: key reaction of Helium burning: determine C/O ratio and stellar evolution 12 C+ 12 C: energy production and nucleosynthesis in Carbon burning. Global chemical evolution of the Universe Reactions occuring at higher temperature than those belonging to Hydrogen burning or BBN Higher energy machine needed!

35. Oxygen-16 Consequences  late stellar evolution  composition of C/O White dwarfs  Supernova type I explosion  Supernova type II nucleosynthesis relevant questions: Energy production and time scale of Helium burning: 4 He(2 ,  ) 12 C( ,  ) 16 O( ,  ) 20 Ne Neutron sources for s process: 14 N( ,  ) 18 F(  + ) 18 O( ,  ) 22 Ne( ,n) 22 Ne( ,  ) 12 C( ,  ) 16 O – Holy Grail of Nuclear Astrophysics Stellar Helium burning in Red Giant Stars the He burning is ignited on the 4 He and 14 N ashes of the preceding hydrogen burning phase (pp and CNO)

36. 1 H 4 He 12 C 16 O 20 Ne 40 Ca 56 Fe 118 Sn 138 Ba 195 Pt 208 Pb 232 Th 238 U a -elements Type II SN Fe-peak Type I SN N=82 r-process peak Type II SN N=126 r-process peak Type II SN 1 H 12 C 16 O 20 Ne 40 Ca 56 Fe 118 Sn 138 Ba 195 Pt 208 Pb 232 Th 238 U  -elements Type II SN Fe-peak Type I SN N=82 r-process peak Type II SN N=126 r-process peak Type II SN N=82 s-process peak AGB stars N=82 s-process peak AGB stars N=126 s-process peak AGB stars N=126 s-process peak AGB stars 19 F Big Bang H-burning & He-burning Nuclear Astrophysics ambitious task is to explain the origin and relative abundance of the elements in the Universe Element abundances in the solar system n source reactions

37. 13 C( ,n) 16 O experimental status of the art Big uncertainties in the R-matrix extrapolations. Presence of subthreshold resonances. A low background environment is mandatory for any new study Neutron energy 2.5-3 MeV (in the energy range foreseen) Heil 2008

38. 22 Ne( ,n) 16 O experimental status of the art Precise measurement of the known resonances down to the one at E  = 831 keV to be performed at first, followed by a detailed search for unknown resonances down to E  ~ 600 keV. Neutron energy 50-450 keV (in the energy range foreseen) Jaeger 2001

39. LUNA MV project LUNA MV accelerator will be installed in the south part of Hall C of LNGS laboratory (OPERA location) Dimensions of the hall: 27x12.5x5.5 m 3 OPERA decommissioning started in Jan 2015. Should be finished by October 2016

40. LUNA MV project Accelerator: Intense H +, 4 He +, 12 C + e 12 C ++ beams in the energy range: 350 keV-3.5 MeV. Two beam lines with all necessary elements (magnets, pumps, valves,...). Total budget (about 3.9 Meuro) from LUNA MV «Premium projects» (total 5.3 Meuro) of the Italian Research Ministry Tender assigned to HVEE in December 2015. Timeline: Accelerator built and tested by HVEE by 11/2017. Accelerator delivered to LNGS by 01/2018 Accelerator installed and tested at LNGS by 07/2018. Then first experiments… Experimental program for the first five years of operation to be defined by July 2016

41. LUNA MV project Building & shielding: Reactions to be studied at LUNA MV will produce a small amount of neutrons…in a very low background laboratory… GEANT4 simulations with different materials in order to find the best compromise among performance as neutron shield, price, easiness of decommissioning, thickness (maximize internal space, … )

42. LUNA MV project The «simple» solution of a 80 cm thick concrete shielding has been selected (  n ) av < 1.5 10 -6 n cm -2 s -1 To be compared with  n (LNGS) = 3.0 10 -6 n cm -2 s -1 Validation through independent calculation by MCNP code presently underway Timeline: Engineering of shielding & building concluded by 06/2016

43. LUNA MV project-commissioning measurement 14 N(p,  ) 15 O Use of neutrino flux as a probe of solar interior composition (metallicity) CNO neutrino play a key role: Borexino can detect them Necessary to better constrain nuclear physics inputs i.e. 14 N(p,  ) 15 O

44. Measure in the range 200 keV - 1.5 MeV with both accelerators and same experimental setup to minimize systematic effects Study angular distribution with HPGe detectors in far geometry. Reduce the nuclear physics uncertainty from 7% to 5% and measure CNO neutrinos ( 15 O) with 10% uncertainty allow to determine solar metallicity with 17% accuracy (now >30%) LUNA MV project-commissioning measurement 14 N(p,  ) 15 O Already measured at LUNA 400kV down to 70 keV (110 keV with angular distribution). Target production known. R matrix extrapolation to Gamow peak energies affected by high uncertainties at higher energies (g.s. transition in the figure)

45. The LUNA collaboration A. Boeltzig*, G.F. Ciani*, A. Formicola, S. Gazzana, I. Kochanek, M. Junker | INFN LNGS /*GSSI, Italy D. Bemmerer, M. Takacs, T. Szucs | HZDR Dresden, Germany C. Broggini, A. Caciolli, R. Depalo, R. Menegazzo, D. Piatti | Università di Padova and INFN Padova, Italy C. Gustavino | INFN Roma1, Italy Z. Elekes, Zs. Fülöp, Gy. Gyurky| MTA-ATOMKI Debrecen, Hungary O. Straniero | INAF Osservatorio Astronomico di Collurania, Teramo, Italy F. Cavanna, P. Corvisiero, F. Ferraro, P. Prati, S. Zavatarelli | Università di Genova and INFN Genova, Italy A. Guglielmetti, D. Trezzi | Università di Milano and INFN Milano, Italy A. Best, A. Di Leva, G. Imbriani, | Università di Napoli and INFN Napoli, Italy G. Gervino | Università di Torino and INFN Torino, Italy M. Aliotta, C. Bruno, T. Davinson | University of Edinburgh, United Kingdom G. D’Erasmo, E.M. Fiore, V. Mossa, F. Pantaleo, V. Paticchio, R. Perrino*, L. Schiavulli, A. Valentini| Università di Bari and INFN Bari/*Lecce, Italy