The Stellar Initial Mass Function

TL;DR

This paper reviews the observational and theoretical understanding of the stellar initial mass function (IMF), highlighting its universal features and proposing a model based on fragmentation and accretion processes in star-forming regions.

Contribution

It introduces a simple model of successive mergers and accretion that reproduces the Salpeter law for the upper IMF, linking physical processes to observed IMF features.

Findings

Characteristic stellar mass linked to Jeans scale in molecular clouds.

Power-law decline at high masses consistent with scale-free accretion.

Proposed merger-accretion model reproduces Salpeter law.

Abstract

The current status of both the observational evidence and the theory of the stellar initial mass function (IMF) is reviewed, with particular attention to the two basic, apparently universal features shown by all observations of nearby stellar systems: (1) a characteristic stellar mass of the order of one solar mass, and (2) a power-law decline of the IMF at large masses similar to the original Salpeter law. Considerable evidence and theoretical work supports the hypothesis that the characteristic stellar mass derives from a characteristic scale of fragmentation in star-forming clouds which is essentially the Jeans scale as calculated from the typical temperature and pressure in molecular clouds. The power-law decline of the IMF at large masses suggests that the most massive stars are built up by scale-free accretion or accumulation processes, and the observed formation of these stars in dense clusters and close multiple systems suggests that interactions between dense prestellar clumps or protostars in forming clusters will play a role. A simple model postulating successive mergers of subsystems in a forming cluster accompanied by the accretion of a fraction of the residual gas by the most massive protostar during each merger predicts an upper IMF of power-law form and reproduces the Salpeter law with a plausible assumed accretion efficiency.