Simulations of star formation in Ophiuchus – II. Multiplicity

Abstract

Lomax et al. have constructed an ensemble of 60 prestellar cores having masses, sizes, projected shapes, temperatures and non-thermal radial velocity dispersions that match, statistically, the cores in Ophiuchus; and have simulated the evolution of these cores using SPH. Each core has been evolved once with no radiative feedback from stars, once with continuous radiative feedback, and once with episodic radiative feedback. Here we analyse the multiplicity statistics from these simulations. With episodic radiative feedback, (i) the multiplicity frequency is ∼ 60% higher than in the field; (ii) the multiplicity frequency and the mean semi-major axis both increase with primary mass; (iii) one third of multiple systems are hierarchical systems with more than two components; (iv) in these hierarchical systems the inner pairings typically have separations of a few au and mass ratios concentrated towards unity, whereas the outer pairings have separations of order 100 au and a flatter distribution of mass ratios. The binary statistics are compatible with observations of young embedded populations, and -if wider orbits are disrupted preferentially by external perturbations -with observations of mature field populations. With no radiative feedback, the results are similar to those from simulations with episodic feedback. With continuous radiative feedback, brown dwarfs are under-produced, the number of multiple systems is too low, and the statistical properties of multiple systems are at variance with observation. This suggests that star formation in Ophiuchus may only be representative of global star formation if accretion onto protostars, and hence radiative feedback, is episodic.

Using binary statistics in Taurus-Auriga to distinguish between brown dwarf formation processes

“…A pre-stellar core can condense out from a self-gravitating collapsing molecular cloud to form a star (André et al 2014;Lomax et al 2015) when it reaches the density-and temperature-dependent Jeans-mass of the ambient medium (Larson 1998;Bate & Bonnell 2005;Bonnell et al 2006;Klessen et al 2007). Fragmentation of this kind can continue until a core becomes opaque to its own cooling radiation so that the Jeans-mass cannot decrease further (Rees 1976;Low & Lynden-Bell 1976;Silk 1977).…”

ALMA Observations of Starless Core Substructure in Ophiuchus

“…Furthermore, many stars are found in binary or higher-order multiple systems (roughly half of solar-type stars, e.g., Duquennoy & Mayor 1991;Raghavan et al 2010;Duchêne & Kraus 2013), and there is strong evidence to suggest that multiplicity may be more common in younger protostellar systems (e.g., Looney et al 2000;Connelley et al 2008;Chen et al 2013;Tobin et al 2016b). A high protostellar multiplicity is also expected based on models comparing the core mass function (CMF) and initial stellar mass function (Goodwin et al 2008;Holman et al 2013;Lomax et al 2015). While the formation of binary and higher-order protostellar systems could arise through fragmentation of a protostellar disk (e.g., Kratter & Lodato 2016;Tobin et al 2016a), some of these multiple systems could also begin forming earlier, at the starless core stage.…”

A Theoretical Perspective on the Formation and Fragmentation of Protostellar Discs

“…Figure 1 is a grey-scale column-density image of a prestellar core generated in this way. The procedure for generating synthetic cores, and its application to the cores observed in Ophiuchus, is described in more detail in Lomax et al (2013Lomax et al ( , 2014. …”

“…Whether they are a good representation of the cores in Ophiuchus depends on whether the assumptions we have made in the preceding section are valid. The assumptions that turn out to be critical are (i) that the cores are randomly oriented (Lomax et al 2013), (ii) the treatment of radiative feedback (Lomax et al 2014(Lomax et al , 2015b, and (iii) the ratio of solenoidal to compressive turbulence (Lomax et al 2015a). Our assumptions about the density profile, the fractal dimension, and the density-and velocity-scaling exponents appear to be much less critical, in the sense that they do not greatly influence the properties of the stars that form in the simulations.…”

“…In the second prescription, labelled episodic feedback, we use the phenomenological "sub-sink" accretion model described in Stamatellos et al (2011) -and also used in Stamatellos et al (2012) and Lomax et al (2014). In this phenomenological model, which is based on the theory developed by Zhu et al (2009), highly luminous, short-lived accretion bursts are separated by long intervals of low accretion and low luminosity.…”