Underground measurement at LUNA found no evidence for a low-energy resonance in the 6 Li(p, γ ) 7 Be reaction

. The 6 Li(p, γ ) 7 Be reaction is involved in all three main nucleosynthesis scenarios: Big Bang Nucleosynthesis, the interaction of cosmic rays with interstellar matter, and stellar nucleosynthesis. Conﬂicting experimental results have been reported in literature for the 6 Li(p, γ ) 7 Be reaction cross section trend at astrophysical energies. A recent direct measurement found a resonance-like structure at E c.m. = 195 keV, corresponding to an excited state at E x ∼ 5800 keV in 7 Be which, however, has not been conﬁrmed by either theoretical calculations or other direct measurements. In order to clarify the existence of this resonance, a new experiment was performed at the Laboratory for Underground Nuclear Astrophysics, located deep underground at Laboratori Nazionali del Gran Sasso (Italy). The 6 Li(p, γ ) 7 Be cross section was measured in the energy range E c.m. = 60-350 keV with un-precedented sensitivity and no evidence for the alleged resonance was found.


Introduction
According to simulations of the Galaxy chemical evolution most of the solar system lithium was provided by low-mass stars [1] while less than half of it was produced by Big Bang Nucleosynthesis (BBN) [2,3] or Galactic cosmic rays interacting with interstellar matter.
The predicted BBN 6 Li/ 7 Li isotopic ratio is ∼ 10 −5 [4], significantly lower than the solar system value of 0.08 [5]. Very low 6 Li/ 7 Li values are expected for neutrino nucleosynthesis [6] and for stellar sources as well. In contrast, in case of Galactic or structure formation cosmic rays the 6 Li/ 7 Li production ratio is close to unity [7]. The 6 Li/ 7 Li isotopic ratio has been indeed proposed as a tool to constrain non-standard lithium production mechanisms [8] and pollution of stellar atmospheres [9] in the context of the cosmological lithium problem, The 6 Li(p,γ) 7 Be reaction plays a key role in determining the stellar 6 Li/ 7 Li. The 6 Li(p,γ) 7 Be reaction may indeed not only deplete 6 Li but also convert some of it to 7 Li, through 7 Be radioactive decay.
Measurements of the 6 Li(p,γ) 7 Be reaction cross section at low energies have reported inconsistent results on the slope of the astrophysical S -factor [10,11]. Moreover, the positive slope reported by [12] was interpreted as a new resonance at E c.m. = 195 keV, corresponding to an excited level at E x ≈ 5800 keV with J π = (1/2 + , 3/2 + ) and Γ p ≈ 50 keV. No evidence for such a resonance was found in the 3 He( 4 He,γ) 7 Be reaction at E c.m. = 4210 keV as reported in recent comprehensive study [13].
None of the theoretical calculations of the 6 Li(p,γ) 7 Be S -factor are able to reproduce the newly-reported resonance [14,15, and references therein], unless this is added ad-hoc to reproduce the experimental data [16].

Experimental Setup
We performed a new experiment [17] at the Laboratory for Underground Nuclear Astrophysics (LUNA), located deep underground at Laboratori Nazionali del Gran Sasso (Italy) [18].
A schematic view of the experimental setup is shown in Fig.1 Figure 1: Sketch of the experimental setup used for the measurement of the 6 Li(p,γ) 7 Be cross section at LUNA [17].
[19] high-intensity proton beam was collimated by a 3 mm diameter aperture and delivered through a copper pipe to the target, mounted at 55 • with respect to the beam direction. The Cu tube was used both as a cold trap and for secondary electron suppression. Three evaporated targets (thicknesses 100 − 200 µg/cm 2 ) were made from 6 Li 2 WO 4 powder and one (thickness 20 µg/cm 2 ) was made using 6 Li 2 O powder. The 6 Li isotopic enrichment level was 95% for all targets, which were water cooled during irradiation in order to limit target degradation [17]. A High-Purity Germanium (HPGe) detector positioned in close geometry to the target and at 55 • with respect to the beam direction was used to detect 6 Li(p,γ) 7 Be reaction γrays. To detect the α and 3 He particles from the 6 Li(p,α) 3 He reaction concurrently with the gamma rays from the 6 Li(p,γ) 7 Be reaction, a silicon detector was installed at 125 • from the beam direction. Efficiencies for both detectors were obtained using GEANT simulations, fine tuned through the comparison with experimental results [17].

Analysis and Results
To make consistency checks and verify results are unaffected by systematic effects, a measurement of the 6 Li(p,γ) 7 Be and 6 Li(p,α) 3 He excitation functions was performed for each target in the whole dynamic range of the LUNA-400 accelerator [17].
The 6 Li(p,γ) 7 Be experimental yield was calculated from the sum of the contributions from the direct capture to the ground state (γ 0 ) and to the 429 keV excited state of 7 Be (γ 1 ).
For the calculation of the 6 Li(p,γ) 7 Be reaction S -factor, we adopted a relative approach [17]: the (p,γ) yield was normalized at each energy to the (p,α) yield. This ratio can be expressed in terms of the (p,γ) and (p,α) S -factors. We adopted for the 6 Li(p,α) 3 He reaction the S -factor parametrization from [20]. For the (p,α) channel, the angular distribution coefficients A k and related uncertainties were taken from [21, and references therein]. For the (p,γ) channel we adopted the theoretical angular distribution described in [14]. Finally the measured S -factor was corrected for electron screening using the adiabatic approximation [22] with screening potential U e = 273 eV [20].
Our S -factor data have a monotonic dependence on the energy and show no evidence of the resonance reported by [12], see Fig.2. The 6 Li(p,γ) 7 Be reaction cross section was measured in the energy range 60 − 350 keV with ≤2% statistical and 12% systematic uncertainty. An R-matrix fit of our data and the data from [23] was performed and used to calculate a new 6 Li(p,γ) 7 Be reaction rate. The proposed reaction rate is 9% lower than NACRE [24] and 33% higher than reported in NACREII [16] at 2 MK, relevant for 6 Li depletion in premain sequence stars, and the reaction rate uncertainty has been significantly reduced [17], see Fig.3.
The result of a recent indirect study supports LUNA extrapolation for the 6 Li(p,γ) 7 Be S -factor [25]. R-matrix fit Figure 2: Astrophysical S-factor for the 6 Li(p,γ) 7 Be reaction as obtained by LUNA in red [17]. Previous experimental data and theoretical evaluations are also shown for comparison. The solid black line represents an R-matrix fit of LUNA data and data from [23].  Figure 3: Reaction rate for the 6 Li(p,γ) 7 Be reaction, normalized to the NACRE rate [24]. The NACRE II rate [16] is also shown for comparison. Dashed lines represent the uncertainty on the NACRE rate, while shaded areas represent the uncertainties from LUNA experiment (red) and from NACRE II (grey).