The impact of metallicity on nova populations

Abstract

The metallicity of a star affects its evolution in a variety of ways, changing stellar radii, luminosities, lifetimes, and remnant properties. In this work, we use the population synthesis code binary_c to study how metallicity affects novae in the context of binary stellar evolution. We compute a 16-point grid of metallicities ranging from Z = 10−4 to 0.03, presenting distributions of nova white dwarf masses, accretion rates, delay-times, and initial system properties at the two extremes of our 16-point metallicity grid. We find a clear anticorrelation between metallicity and the number of novae produced, with the number of novae at Z = 0.03 roughly half that at Z = 10−4. The white dwarf mass distribution has a strong systematic variation with metallicity, while the shape of the accretion rate distribution is relatively insensitive. We compute a current nova rate of approximately 33 novae per year for the Milky Way, a result consistent with observational estimates relying on extra-Galactic novae but an under-prediction relative to observational estimates relying on Galactic novae. However, the shape of our predicted Galactic white dwarf mass distribution differs significantly to existing observationally derived distributions, likely due to our underlying physical assumptions. In M31, we compute a current nova rate of approximately 36 novae per year, under-predicting the most recent observational estimate of $65^{+15}_{-16}$. Finally, we conclude that when making predictions about currently observable nova rates in spiral galaxies, or stellar environments where star formation has ceased in the distant past, metallicity can likely be considered of secondary importance compared to uncertainties in binary stellar evolution.

Binary Interaction Dominates Mass Ejection in Classical Novae

“…As with the envelope mass, we take the envelope composition as a given because of uncertainties in the strengths of the various enrichment mechanisms and explore two compositions: solar composition (Asplund et al 2009), which is equivalent to the assumption that no mixing occurs with the underlying C/O core, and an enhanced C/O composition, with 80% solar composition by mass plus an additional 10% by mass of 12 C and 16 O each. The composition of the accreted material may not be solar, which can induce changes to the ignition mass (Piersanti et al 2000;Starrfield et al 2000;José & Hernanz 2007;Chen et al 2019;Hillman & Gerbi 2022;Kemp et al 2022), but since we parameterize the convective envelope masses by hand in this study, we bypass this complication.…”

Viability of Novae as Sources of Galactic Lithium

“…The authors wish to thank the anonymous referee for their helpful feedback. Kemp et al (2021) and Kemp et al (2022), which discuss which binary systems produce the most novae. However, the key features and distributions of nova properties when considering which systems produce the most nova ejecta requires a brief description.…”

“…A metallicity-dependent grid of these pre-computed ejecta-delay time distributions is taken as a direct input into our galactic chemical evolution models. The underlying binary population synthesis models are a subset (Z = 10 −4 , 10 −3 , 5 × 10 −3 , 0.01, 0.02) of those presented in Kemp et al (2022), and further details about the process of simulating each nova eruption in binary c (Izzard et al 2004(Izzard et al , 2006(Izzard et al , 2009(Izzard et al , 2018 are available in Kemp et al (2021).…”

“…In this work, we instead rely on the results of previously computed theoretical nova populations (Kemp et al 2022), where each nova eruption is treated individually based on the white dwarf mass M WD and accretion rate M  . Our chemical evolution model and treatment of novae are described in Section 2, our findings on novae as sources of Li in the Milky Way are presented in Section 3, and commentary on the significance and uncertainty in our findings is presented in Section 4.…”

The evolution of CNO elements in galaxies

“…Recent studies report rates of 43:7 þ19:5 À8:7 yr À1 (De et al 2021), 26 AE 5 yr À1 (Kawash et al 2022), and 28 þ5 À4 yr À1 (Rector et al 2022). The past Galactic nova rate is basically unknown, though some population synthesis studies suggest the existence of an anticorrelation between metallicity and the number of novae produced: for instance, according to Kemp et al (2022), the number of novae at Z ¼ 0:03 is roughly half that at Z ¼ 10 À4 . Clearly, there is a high degeneracy between the nova rate and the nova yields assumed by any GCE model.…”