Is \gamma-ray emission from novae affected by interference effects in the 18F(p,\alpha)15O reaction?

TL;DR

This study investigates the 18F(p,α)15O reaction in novae, revealing multiple resonances near the proton threshold and reducing the uncertainty in the reaction rate, which impacts predictions of gamma-ray emissions.

Contribution

The paper provides new measurements of resonances in 19Ne, challenging previous assumptions about interference effects and identifying the proton-width of a key resonance as the main source of uncertainty.

Findings

Identified at least 3 resonances between the proton threshold and 50 keV.

None of these resonances are consistent with Jπ=3/2+ angular distributions.

The unknown proton-width of the 48 keV resonance influences 18F production by at least a factor of two.

Abstract

The 18F(p,\alpha)15O reaction rate is crucial for constraining model predictions of the \gamma-ray observable radioisotope 18F produced in novae. The determination of this rate is challenging due to particular features of the level scheme of the compound nucleus, 19Ne, which result in interference effects potentially playing a significant role. The dominant uncertainty in this rate arises from interference between J\pi=3/2+ states near the proton threshold (Sp = 6.411 MeV) and a broad J\pi=3/2+ state at 665 keV above threshold. This unknown interference term results in up to a factor of 40 uncertainty in the astrophysical S-factor at nova temperatures. Here we report a new measurement of states in this energy region using the 19F(3He,t)19Ne reaction. In stark contrast with previous assumptions we find at least 3 resonances between the proton threshold and Ecm=50 keV, all with different angular distributions. None of these are consistent with J\pi= 3/2+ angular distributions. We find that the main uncertainty now arises from the unknown proton-width of the 48 keV resonance, not from possible interference effects. Hydrodynamic nova model calculations performed indicate that this unknown width affects 18F production by at least a factor of two in the model considered.