The CNO cycle is the dominant mechanism for energy production in massive main-sequence stars during the hydrogen burning phase.

The ¹⁴N(p, γ)¹⁵O reaction is the slowest reaction of the CNO cycle and controls the rate of energy generation of this process. As a consequence, it directly determines the lifetime of massive stars. In our Sun ∼1% of energy is produced through CNO, contributing also to 1.6% of the solar neutrino flux. As recently assessed by the Borexino collaboration [9], the ¹⁴N(p, γ)¹⁵O remains the second largest contribution to the uncertainty budget in the estimation of the C and N abundances in the Sun after the CNO neutrino flux itself.

This reaction has been extensively studied by the LUNA collaboration [10–13] and other groups [14, 15] down to a center of mass energy of 70 keV. In fact, thanks to one of the first results obtained at the LUNA-400 accelerator, its rate was found to be a factor of two slower than expected [12]. This result had several implications, such as increasing the age of globular clusters by about 1 Gy [10] and reducing the expected CNO solar neutrino rates by a factor of two [11]. Recently, a new underground measurement was performed by Frentz et al. [16] at the CASPAR accelerator (SURF).

Nonetheless experimental data still lies far from the solar Gamow window at E = 27 keV. Low-energy reaction rate data therefore relies on extrapolations done with R-matrix analysis [17]. Up to the current days, the R-matrix analysis of the reaction failed to provide a consistent view of both the low and high energy data for the reaction. The presence of consistent gaps in the data sets and the fact that several weaker transitions have not been measured since Schröder et al. [18] therefore reinforce the need for a precision measurement of the ¹⁴N(p, γ)¹⁵O over a wide energy range with an angular distribution experiment. The high-current 3.5 MV Singletron^® of the Bellotti IBF represents the perfect framework where such an experiment can be conducted given the relatively weak population of some of the transitions and the reduction of the background in the γ-ray spectra given by the deep-underground location of the facility.

In this context, an improvement in the reaction rate, alongside the recently improved CNO flux measurement and further improvements from future such measurements, could provide important elements to solve the tension between solar models obtained using different metallicity levels [19].

Helium burning, and more generally, processes involving α-capture reactions, contribute greatly to the isotopic abundances observed in the Universe. After the bridging of the A = 5 and A = 8 mass gaps through the triple-α reaction [20, 21], helium burning begins at at T ≈ 0.2 GK and proceeds through the reactions ¹²C(α, γ)¹⁶O and ¹⁶O(α, γ)²⁰Ne. The latter reaction is sufficiently slow to provide only for a partial conversion of carbon into oxygen, leading to the currently observed ¹²C/¹⁶O abundance ratio of ca. 0.4 [22].

The cross section of ¹²C(α, γ)¹⁶O needs to be known at E _c.m. = 300 keV as an input for stellar models. At these energies, direct measurements are highly challenging due to the extremely low experimental count rates [53]. One approach to direct cross section measurements that has proven quite successful in the past is based on recoil separators [23–25]; γ-ray measurements in the deep underground will provide high-quality complementary cross section data.

Other α-capture induced reactions of great importance are (α, n) reactions on ¹³C and ²²Ne during thermal pulses in AGB stars and, in the case of ²²Ne(α, n)²⁵Mg, core He and shell C burning in massive stars [26–28]. They constitute the neutron sources for the main and weak s processes, responsible for the production of about half of all heavy (A > 56) elements in the Universe [29].

Also the measurements of the neutron sources suffer from very low cross sections that have prevented progress towards direct measurements at low energies in surface laboratories [30–34]. The ¹³C(α, n)¹⁶O reaction has recently been measured at previously unreachable energies deep underground by the LUNA and the JUNA collaborations [35, 36]. The Bellotti IBF will allow to connect the low-energy data from LUNA-400 to the high-energy region, providing an insight into normalization issues seen between the many high-energy data sets. In addition, the availability of ¹³C as a beam could allow measurements in inverse kinematics, penetrating deeper into the Gamow peak with a more advanced setup.

²²Ne(α, n)²⁵Mg has not been directly measured in over 20 years, and the ultra-low neutron background at the Bellotti IBF combined with the high-intensity α beam and an innovative setup constructed in the framework of the SHADES ERC project aims at providing for the first time data in the astrophysical energy range.

Carbon burning is a key stage of stellar evolution determining the final destiny of massive stars and of low mass stars in close binary systems, whose understanding is of paramount importance for the comprehension of SuperNova (SN) outcomes. SNe play a pivotal role in astrophysics, providing a major contribution to the chemical and physical evolution of galaxies, and generating the most compact objects in nature, such as neutron stars and black holes (type II, core collapse SNe). SNe are also used to determine distances on cosmological scale, to probe the history of the Universe (type Ia, carbon explosive burning) [37]. The efficiency of carbon burning in massive stars also determines the compactness of the stellar core at the onset of the final core collapse. The compactness of the core is a critical parameter that determines the final fate of these stars, whether they end their life as a supernova or directly collapse into a black hole [38, 39]. Moreover, carbon burning can trigger superburst events, which are long, energetic, and rare thermonuclear flashes on accreting neutron stars in low-mass X-ray binaries. These bursts are considered to be triggered by the unstable ¹²C+¹²C burning in the ash left over from the rp-process on the surface of a neutron star [40].

The large uncertainty affecting the low energy cross section of the ¹²C+¹²C reaction hampers the knowledge of the final the fate of stellar structures and, in turn, our understanding of SN phenomena. The ¹²C+¹²C fusion in stars proceeds primarily through the ¹²C(¹²C, α)²⁰Ne and the ¹²C(¹²C, p)²³Na reactions, at higher energy the neutron channel ¹²C(¹²C, n)²³Mg can be open. The cross sections of these processes are extremely low, in the sub-femto-barn range. Successful modelling of supernovae requires the cross sections to be known down to E around 1.5 MeV. Due to the extremely small cross sections, direct experiments are challenging already at energies above 2.2 MeV [41–46]. To overcome the experimental limitations, an indirect measurement was performed using the THM [47], covering the entire astrophysical region of interest from E = 2.7 MeV down to 0.8 MeV and revealing well resolved resonance structures. Further theory calculations [48] resulted in large corrections to the initially reported S-factors. Therefore, a direct measurement aiming to reach the astrophysical relevant energy is of crucial importance to reduce the present uncertainty. Measurements may be done detecting the γ-rays generated by the decay of the ²³Na and ²⁰Ne excited states, or by searching for the charged particles, p and α, emitted in the two reactions.

A prior experiment [43] used a HPGe detector equipped with 15 cm thick lead shield and an active muon veto in a surface laboratory to study the ¹²C(¹²C, p)²³Na and the ¹²C(¹²C, α)²⁰Ne reactions. The reaction yield has been measured observing the transitions from the first excited state to the ground state of the daughter nuclei, involving the emission of photons with energies of 440 and 1634 keV, respectively. This study demonstrates the potential of this technique and supports future experiments in an underground location to take advantage of a lower background and extend the measurements to lower energy. The LUNA collaboration already showed that the combination of the 3800 m.w.e. rock shield provided by the Gran Sasso massif, a 25 cm thick lead shield, a copper liner and nitrogen flushing is highly effective in reducing the gamma background in HPGe detectors. The reduction factor ranges between 10³ and 10⁴ both for the energy region above natural radioactivity and below it [49] (see Section 1). In particular, based on preliminary background measurements at LNGS, in the regions of the spectrum close to 440 keV and 1634 keV the Bellotti IBF will allow to achieve a gamma background about two orders of magnitude lower with respect to [43], enabling the measurement of the cross section in the energy range relevant to astrophysics.

Unlike other MeV ion beam facilities that operate in standard laboratories located at the surface, the severe limitation of prompt radiation production, especially neutrons, imposes a compelling severe limitation to the beam intensities and energies and to the materials that may be subject to irradiation in the underground laboratory. Despite the limitations, Ion Beam Analysis (IBA) niche applications (in particular depth profiling using narrow resonant (p, γ) and (α, γ) reactions for sub ppm sensitivity analysis and nm depth resolution) may be considered potential high value and cutting edge applications of the 3.5 MV underground accelerator. IBA is generally accomplished with 100 pnA to few pµA proton or helium currents that can be too small for the present configuration of the accelerator, but could be part of a future development if extensive programs of key material micro-analyses will be proposed. The Nuclear Resonance Reaction Analysis (NRRA) technique is far the most interesting niche application at IBF. While Elastic Backscattering (EBS) and Particle Induced X-ray Emission (PIXE) are sensitive to medium to heavy elements, Particle Induced Gamma-ray Emission (PIGE) is best suited for light elements and isotopes analysis and an intense program of material characterization involving the study of light impurities at (sub)ppm level may be specific of the underground laboratory taking the full advantage of the high sensitivity gamma spectrometers and the extremely low gamma background coupled to the unprecedented ion beam energy stability and low energy spread of the 3.5 MV Singletron^® accelerator of the Bellotti Facility. In parallel, a scientific plan aiming at the accurate measurement of unexplored (p, γ) and (α, γ) reactions suitable for IBA provides another field of investigation which will fully profit of the characteristics of the underground facility. In addition, it must be mentioned that in quantum technology the possibility to accurately implant ^12,13C ions in a fairly wide energy range (0.3–7 MeV) coupled to the possibility to use the proton and He beams for in-situ pre(post) irradiation and analysis might offer a valuable possibility to create new color centers in semiconductors for the development of single photon sources in advanced quantum photonics applications.