Constraints on the pairing properties of main-sequence stars from observations of white dwarfs in binary systems

Observations of main-sequence stars conducted over the last several decades have clearly shown that something like 50 per cent of stars of spectral types G and F occur in multiple systems. For earlier spectral types, the incidence of multiplicity is even higher. Thus, a volume-limited sample of white dwarfs should reflect the percentage of binarity observed in stars of F to late B spectral types, which are their main-sequence progenitors. However, a study of the local volume-limited sample of white dwarfs (20 pc from the Sun) conducted by Holberg has shown that a white dwarf has a probability of only ∼32 per cent of occurring in a binary system, in stark contrast to the observations of multiplicity of main-sequence stars. Other studies have also led to the same conclusion. In this paper, we argue that the ‘hidden’ white dwarfs are either in double white dwarf systems or in Sirius-like systems. We also show that the white dwarf progenitors of the Sloan Digital Sky Survey (SDSS) white dwarf–M dwarf wide binaries are distributed according to Salpeter's initial mass function (IMF). However, they cannot be paired with secondary stars which are also drawn from this IMF, since such a pairing would produce a percentage of white dwarf–M dwarfs systems that is several times larger than observed.