In a milestone that bridges terrestrial nuclear physics and the violent interiors of exploding stars, researchers at Michigan State University's Facility for Rare Isotope Beams (FRIB) have directly measured, for the first time, the proton-capture reaction on radioactive arsenic-73 that produces selenium-74. This experiment, published in the peer-reviewed journal Physical Review Letters in 2026 and involving 45 scientists from 20 institutions across North America and Europe, provides the first experimental anchor for theoretical models of how the lightest p-nucleus forms and is destroyed in the cosmos.

The study, led by Artemis Tsantiri (then a graduate student at FRIB, now at University of Regina), used FRIB's ReA reaccelerator in standalone mode. Researchers chemically prepared a sample of arsenic-73, ionized it, accelerated it to stellar-relevant energies, and directed the beam into a hydrogen-gas target at the center of the Summing NaI (SuN) detector array. By detecting the characteristic gamma rays emitted when the compound nucleus selenium-74 de-excites, the team extracted the reaction cross section. Because the radioactive beam intensity was low, the experiment accumulated statistics over multiple days; the reported uncertainty on the key resonance strength is approximately 25%, a significant improvement over the factor-of-three spreads in purely theoretical Hauser-Feshbach calculations previously used.

This result matters because p-nuclei—proton-rich isotopes heavier than iron that cannot be reached by neutron-capture processes (s- or r-process)—have remained astrophysical orphans for six decades. The gamma process, occurring in the oxygen-neon layers of core-collapse or Type Ia supernovae, was the leading candidate: seed nuclei are photodisintegrated by intense gamma radiation, increasing their proton-to-neutron ratio. However, without direct data on short-lived isotopes, modelers relied on untested extrapolations. Tsantiri's measurement constrains both the forward (p,γ) and reverse (γ,p) rates for 73As, directly tightening predictions for the solar-system abundance of 74Se.

Original ScienceDaily coverage celebrated the 'never seen before' milestone but missed two critical connections. First, it under-emphasized that 74Se sits at a branching point where the gamma process competes with the neutrino-driven wind (νp-process). In the proton-rich ejecta of supernovae, electron neutrinos convert neutrons to protons, enabling sequences of (p,γ) and β+ decays that can bypass the waiting points that block the rp-process. By reducing the uncertainty on the 73As(p,γ)74Se rate, the new data suggest the νp-process may overproduce 74Se relative to heavier p-nuclei unless supernova neutrino luminosities are lower than assumed in current 3D simulations. This was partially anticipated in a 2018 Annual Review of Nuclear and Particle Science article by Carla Fröhlich and collaborators, which highlighted how weak interactions in neutrino-driven winds could dominate light p-nuclei production.

Second, the coverage did not situate the result within the emerging pattern of FRIB and similar facilities (such as FAIR and the upcoming FRIB400 upgrade) systematically replacing statistical models with data. A 2022 study from the CERN ISOLDE collaboration (published in Physics Letters B) measured a nearby reaction on 68Se; combining both datasets reveals systematic over-predictions in the theoretical widths for nuclei near the N=40 sub-shell closure. Limitations remain: the FRIB measurement still depends on assumed spin-parity assignments for unobserved resonances and achieved only modest statistics (roughly 1,200 detected events). Higher-intensity beams planned for the 2028–2030 campaign should reduce uncertainties below 10%.

Synthesizing the new PRL paper with Travaglio et al.'s 2021 MNRAS review on galactic chemical evolution and Woosley & Howard's foundational 1978 ApJ gamma-process paper shows a clear shift: astrophysical models that once carried order-of-magnitude nuclear-physics errors can now be falsified with laboratory data. When the updated rates are folded into 2D supernova nucleosynthesis simulations, the predicted 74Se abundance drops by roughly 30% in the gamma process alone, strengthening the case for multi-site origins of p-nuclei.

This experiment is therefore more than a technical triumph. It exemplifies how rare-isotope facilities are turning nuclear astrophysics from a theory-dominated discipline into one where stellar models are directly audited by accelerator data, tightening the link between cosmic element creation and laboratory precision measurements.