Extremely low background experiments to measure key nuclear reaction cross sections of astrophysical interest are conducted at the world’s deepest underground laboratory, the Jingping Underground laboratory for Nuclear Astrophysics (JUNA). High precision measurements provide reliable information to understand nucleosynthetic processes in celestial objects and resolve mysteries on the origin of atomic nuclei discovered in the first generations of Pop. III stars in the universe and meteoritic SiC grains in the solar system.

Nuclear astrophysics is a growing interdisciplinary research field of nuclear physics and astrophysics. It investigates the origin and evolution of hundreds of atomic nuclei in the universe to understand the evolution of stars and galaxies. In addition to the events of the century, SN1987A and GW170817, many complementary data have been accumulated from the high-resolution spectroscopic observations of old faint stars and precise meteoritic analysis of isotopic anomalies and correlations among them.

The origin of atomic nuclei is classified into three astrophysical sites and epochs in cosmic evolution [1]: Big-Bang nucleosynthesis in the first 3–10 min of the early universe for the production of light mass nuclei, thermonuclear and explosive nucleosyntheses in stars for the production of intermediate-to-heavy mass nuclei after galaxies formed, and Galactic cosmic-ray interactions with interstellar medium for the production of rare abundant nuclei. In nuclear astrophysics [2], identifying key nuclear reactions and precisely determining their rates [3] have been a long-standing challenge for several nucleosynthetic processes, as shown in Fig. 1.

Several nucleosynthetic paths, such as the rapid-neutron capture process (r-process), involve many radioactive unstable nuclei, which are challenging to study experimentally because of their short lifetimes and low beam intensities. This difficulty remains even for the measurements of the cross sections of stable nuclei because most cosmic or stellar nucleosynthesis occurs at temperatures of \(T = 10^8-10^9\) K under extreme conditions. Although these temperatures are relatively lower than those of other astronomical phenomena, their energy scale is as low as \(E = 10 - 100\) keV. This energy scale is too low to directly measure the reaction cross sections of charged particles in laboratories because extremely small Coulomb penetrability significantly reduces the cross section [4].

In addition, the cosmic-ray background prevents the precise measurement of small reaction cross sections. Hence, eliminating the cosmic-ray background is critical for the detection of neutral particles like photons and neutrons to measure the reliable reaction rates of radiative-capture \((\alpha ,\gamma )\), (p, \(\gamma\)) and neutron emission (\(\alpha\), n) reactions, which are key to the nucleosynthetic processes in Fig. 1.

Fig. 1
figure 1

(Color online) Nuclear chart and known nucleosynthetic processes in the universe. Filled and open black boxes are the stable and radioactive isotopes, respectively, whose lifetimes were measured. Red and blue boxes are the newly synthesized isotopes and those with known masses, respectively. Isotopes in the yellow region are predicted theoretically. See Ref. [3] for more details

Full size image

The JUNA Collaboration is directed by CIAE, jointly supported by NSFC, CNNC, and CAS, and has established a unique underground facility equipped with a high-intensity accelerator for the direct measurements of extremely small cross sections at low astrophysical energies. JUNA is 2400 m underground at the China JinPing Laboratory complex (CJPL-II) established in 2014 and is the deepest underground laboratory in the world (Fig. 2). The laboratory is beneath rocks with a thickness of several kilometers; the rocks shield cosmic rays; and their corresponding background. Since the first beam was delivered in Dec. 2020, experiments on several key reactions have been carried out with a high-intensity accelerator in an ultra-low background environment.

Fig. 2
figure 2

(Color online) Measured residual muon fluxes in key underground facilities around the world, which are consistent with predicted values (gray line). The sizes of the circles correspond to the laboratory space by volume; red or blue denotes access by road tunnels or shafts, respectively

Full size image

From the onset, JUNA was dedicated to experiments at \(^{19}\)F(p, \(\alpha \gamma )^{16}\)O at very low energies of \(E_\text {c.m.}\) = 72.4–344 keV, covering the Gamow window [5, 6]. Fluorine is one of the most mysterious monoisotopic elements whose predicted abundance has a large uncertainty for the reaction rate. To have multiple origins, that is, asymptotic giant branch stars (AGB stars) in a relatively recent epoch with solar metallicity [7], and the core-collapse type II supernova (cc-SN II) in the early galaxy at low metallicity [8], the JUNA experiment reduces theoretical significantly, and the breakout condition of the carbon-nitrogen-oxygen cycle (CNO cycle) in AGB stars is more precisely constrained. The SN \(\nu\) nucleosynthesis of \(^{19}\)F is also revisited by considering flavor oscillation effects to infer nucleosynthetic constraints on an unknown mass hierarchy [9].

As part of the inception objectives of the JUNA collaboration, \(^{25}\)Mg(p,\(\gamma\))\(^{26}\)Al, which plays an important role in the production of \(^{26}\)Al (\(\tau _{1/2}=7.17\times 10^5\) y) in massive stars [10], was successfully studied. \(\gamma\) rays with an energy of 1.809 MeV emitted from \(^{26}\)Al were observed by the Gamma-ray Satellite INTEGRAL that provides the distribution and total accumulated mass of \(^{26}\)Al in the Milky Way. The estimated frequency of cc-SN II and Ib/c was 1.9 ± 1.1 events per century [11], providing a firm basis for the theoretical study of the Galactic chemical evolution of r-process elements [12]. The JUNA Collaboration team determined the reaction rate of \(^{25}\)Mg(p,\(\gamma\))\(^{26}\)Al with high accuracy by measuring resonance parameters at 92 keV and contributed to establishing a scheme of estimating the rate of SN events.

The production mechanism of \(^{40}\)Ca observed in the oldest ultra-metal-poor star [13] remains an unresolved mystery in recent astronomical observations. In the standard stellar evolution model, most \(^{19}\)F produced by the hot CNO cycle are recycled back to \(^{16}\)O by the \(^{19}\)F(p, \(\alpha )^{16}\)O reaction; therefore, the flow cannot reach the production of \(^{40}\)Ca. However, this scenario is subject to the error bars of its competing reaction rate for \(^{19}\)F(p, \(\gamma )^{20}\)Ne that breaks the hot CNO cycle as the temperature increases. The JUNA Collaboration team directly measured this reaction down to \(E_\text {c.m.}\) = 186 keV and found that the resonance at 225 keV contributes to the enhancement in the thermonuclear reaction rate of \(^{19}\)F(p, \(\gamma )^{20}\)Ne [14]. The hydrostatic burning in Pop. III stars calculated using new rate results of \(^{40}\)Ca abundance is consistent with those observed in the oldest known ultra-metal-poor stars [14].

The JUNA Collaboration team studied the \(^{13}\)C(\(\alpha\), n)\(^{16}\)O reaction and successfully measured its cross section in the range of \(E_\text {c.m.}\) = 240 keV−1.9 MeV, removing uncertainty from previous experimental data [15]. Since this reaction and \(^{22}\)Ne(\(\alpha\), n)\(^{25}\)Mg are a major neutron source for the s-process in AGB stars, the precise determination has long been wanted to predict reliable nuclear abundances from iron to \(^{209}\)Bi theoretically. The JUNA experiment covers the Gamow window for the intermediate-neutron capture process (i-process) recently identified as a new process in metal-deficient AGB stars and collapsars, which are major r-process sites of massive stars collapsing to a black holes [16]. Comprehensive theoretical studies of the s-, i-, and r-processes in multiple astrophysical sites are being conducted with JUNA’s new precise data.

Recently, the JUNA Collaboration team reported an experimental result [17] on \(^{18}\)O\((\alpha ,~\gamma )^{22}\)Ne. For the first time, the team successfully determined the energy \(E_\text{lab} = 474.0 \pm 1.1\) keV, spin-parity \(J^{\pi } = 1^-\), and \(\gamma\)-width \(\omega _{\gamma } = 0.25 \pm 0.03\) \(\upmu\)eV of the most effective resonance dominating the total reaction rate at the Gamow window. \(^{22}\)Ne is produced via \(^{14}\)N\((\alpha ,~\gamma )^{18}\)F\((e^- \nu )^{18}\)O\((\alpha ,~\gamma )^{22}\)Ne, which is key to identifying the conversion efficiency of the CNO cycle, whose main product is \(^{14}\)N, to the advanced burning stage where the s-process occurs using neutrons produced via the reaction \(^{22}\)Ne(\(\alpha\), n)\(^{25}\)Mg in both AGB and supergiant stars. The improved JUNA data also contributes to probing the unknown origin of the isotopic abundance ratio of \(^{21}\)Ne/\(^{22}\)Ne observed in meteoritic stardust SiC grains. The so-called mainstream components of SiC grains arise from AGB stardust. Therefore, both \(^{18}\)O\((\alpha ,~\gamma )^{22}\)Ne and its competing reaction \(^{18}\)O\((\alpha\), n)\(^{21}\)Ne affect the resultant isotopic ratio. The improved JUNA data remove the uncertainty in this isotopic ratio. The observed \(^{21}\)Ne/\(^{22}\)Ne ratios in meteoritic stardust SiC grains are used to constrain the physical conditions of nuclear burning processes that depend on the initial mass of the parent AGB star and other astronomical parameters in modeling stellar evolution.

Phase I of the JUNA collaboration has almost completed the high precision measurements of several key reactions at the lowest cosmic-ray background. Phase II of the JUNA collaboration has begun and will study the following reactions: \(^{12}\)C(\(\alpha\), \(\gamma\))\(^{16}\text{O}\) to improve knowledge on stellar evolution and the explosion mechanism of massive stars and resolve the mystery of the massive black-hole mass-gap [18]; \(^{12}\)C(\(^{12}\)C, \(\alpha\))\(^{20}\)Ne and \(^{12}\)C(\(^{12}\)C, p)\(^{23}\)Na to identify the unknown heat source of X-ray Superburst and clarify the double-degenerate explosion mechanism of SN Ia [19]; \(^3\)He(\(\alpha ,\gamma\))\(^7\)Be and \(^3\)H(\(\alpha ,\gamma\))\(^7\)Li to understand the overproduction of Big-Bang lithium and precisely constrain neutrino-mixing parameters from missing solar neutrino flux [20]; \(^{11}\)B(\(\alpha\), n)\(^{14}\)N, \(^{15}\)N(\(\alpha ,~\gamma\))\(^{19}\)F, and \(^{17}\)O(n, \(\alpha\))\(^{14}\)C and its reverse reaction to elucidate the physical conditions of \(\alpha\)-rich freezeout and the r-process nucleosynthesis in magneto-hydrodynamic jets from cc-SN II, collapsars, and binary neutron star mergers [12, 21]. JUNA is expected to produce relevant nuclear data to enrich our knowledge of stellar and cosmic nucleosyntheses to solve many mysteries in astronomy and astrophysics.