Exploring the possibility of Peter Pan discs across stellar mass

TL;DR

This work investigates whether primordial protoplanetary discs can survive to $\sim$50 Myr around stars up to $\sim$2 $M_\odot$ by modeling viscous disc evolution with external photoevaporation, internal photoevaporation, and accretion. It constructs a grid of simulations spanning host masses from $0.08$ to $1.9\,M_\odot$, two viscosities, and three external-UV field strengths, using mass-radius relations to set initial disc masses and radii and incorporating time-dependent X-ray luminosity. The key finding is that lifetimes can reach $\sim$50 Myr for $M_* \lesssim 0.6\,M_\odot$ in low-radiation environments, but decline sharply for higher-mass hosts (to $<50$ Myr for $M_* \gtrsim 0.8\,M_\odot$) as IPE and accretion become dominant; overall, lifetimes up to $\sim$15 Myr persist for $M_* \lesssim 2\,M_\odot$, and accretion cessation often precedes dispersal, forming inner gaps. The results strongly suggest Peter Pan discs are confined to M-dwarfs, with K-type and earlier stars unlikely hosts of such long-lived primordial discs, thereby constraining the physics of disc dispersal and informing the interpretation of old accreting discs observed in star-forming regions.

Abstract

Recently, several accreting M dwarf stars have been discovered with ages far exceeding the typical protoplanetary disc lifetime. These `Peter Pan discs' can be explained as primordial discs that evolve in a low-radiation environment. The persistently low masses of the host stars raise the question whether primordial discs can survive up to these ages around stars of higher mass. In this work we explore the way in which different mass loss processes in protoplanetary discs limit their maximum lifetimes, and how this depends on host star mass. We find that stars with masses $\lesssim$ 0.6 M$_\odot$ can retain primordial discs for $\sim$50 Myr. At stellar masses $\gtrsim$ 0.8 M$_\odot$, the maximum disc lifetime decreases strongly to below 50 Myr due to relatively more efficient accretion and photoevaporation by the host star. Lifetimes up to 15 Myr are still possible for all host star masses up to $\sim$2 M$_\odot$. For host star masses between 0.6 and 0.8 M$_\odot$, accretion ceases and an inner gap forms before 50 Myr in our models. Observations suggest that such a configuration is rapidly dispersed. We conclude that Peter Pan discs can only occur around M dwarf stars.

Figures (5)

• Figure 1: Disc mass loss rates of a number of processes as a function of host star mass. The processes are external photoevaporation (EPE), internal photoevaporation (IPE), and accretion. The dashed lines show the EPE rate as obtained from interpolating on the FRIED grid Haworth2018, for a range of disc masses. These are spaced logarithmically between $1.57\cdot 10^{-2}$ M$_{\textrm{Jup}}$ to 2.62 M$_{\textrm{Jup}}$ (from bottom to top). The disc radius is 100 au, and the radiation field is 10 G$_0$. The dotted blue and cyan lines show the IPE rates from Owen2012 and Picogna2019, respectively. The solid green line shows the fitted accretion rate dependence on stellar mass from Alcala2014.
• Figure 2: The disc lifetimes (points) and moments of cessation of accretion (triangles; taken to be equal to the moment of dispersal if it has not happened before dispersal) as a function of host star mass for a viscosity parameter of $\alpha=10^{-3}$ (left) and $\alpha=10^{-4}$ (right). Each point corresponds to a simulation, connecting lines are added to help guide the eye. Solid lines correspond to the dispersal of discs under a 'fixed' EPE rate (these reference rates are scaled linearly with the disc outer radius as the disc evolves). The dotted line corresponds to EPE rates as computed from the FRIED grid, with a radiation field of 10 G$_0$. The dashed lines correspond to the cessation of accretion (which does not happen before dispersal for EPE rates of $10^{-8}$ M$_\odot$ yr$^{-1}$ and a radiation field of 10 G$_0$). The horizontal black solid line corresponds to 50 Myr, the characteristic age of Peter Pan discs.
• Figure 3: The accretion rates as a function of time for all discs that live to at least 50 Myr under an EPE rate of $10^{-10}$ M$_\odot$ yr$^{-1}$, for a viscosity parameter of $\alpha=10^{-3}$ (left) and $\alpha=10^{-4}$ (right). The dotted lines are the prescribed accretion rates, and the solid lines are the actual accretion rates. The colour indicates host star mass, with dark purple corresponding to 0.08 M$_\odot$. Note that the accretion rate is a monotically increasing function of host star mass. Also shown are the observed ages and accretion rates of the Peter Pan discs listed in Silverberg2020 and Lee2020 as black points with error bars (the latter did not provide error estimates, so we opted to conservatively display the largest error bars of the sample of Silverberg2020). The two clusters of three points have the same ages, but we have displaced them horizontally for visibility. The left-most point of the clusters corresponds to their true age.
• Figure 4: Gas surface density profiles of a protoplanetary disc subject to accretion, IPE, and EPE, at 30 Myr and 34 Myr. The disc has a host star mass of 0.71 M$_\odot$, an $\alpha$ viscosity parameter of $10^{-3}$, and is subject to a reference EPE rate of $10^{-10}$ M$_\odot$ yr$^{-1}$.
• Figure 5: The evolution of the radius and mass ratio of the discs with a viscosity of $10^{-3}$ (left) and $10^{-4}$ (right) under a radiation field of 10 G$_0$. The colour scale indicates initial host star mass, ranging from 0.08 M$_\odot$ (dark purple) to 1.9 M$_\odot$ (yellow). The black solid line indicates the initial disc radii and mass ratios. The blue points correspond to the points of the FRIED grid, and the blue shaded region to the parameter space (convex hull) of these points in log space. Within this region we can interpolate on the FRIED grid, outside we resort to nearest-neighbour extrapolation.