State-of-the-art models for the phase diagram of carbon and diamond nucleation

TL;DR

This paper reviews recent modeling approaches for carbon's phase diagram, emphasizing bond-order potentials like LCBOP, and discusses controversial hypotheses about liquid carbon phases and diamond formation in planetary interiors.

Contribution

It provides a detailed analysis of LCBOP models for carbon and evaluates their implications for phase behavior and nucleation hypotheses.

Findings

LCBOP models accurately reproduce high-pressure carbon properties

Liquid carbon may exhibit two distinct phases separated by a first-order transition

Diamonds could form via homogeneous nucleation in planetary interiors

Abstract

We review recent developments in the modelling of the phase diagram and the kinetics of crystallization of carbon. In particular, we show that a particular class of bond-order potentials (the so-called LCBOP models) account well for many of the known structural and thermodynamic properties of carbon at high pressures and temperatures. We discuss the LCBOP models in some detail. In addition, we briefly review the ``history'' of experimental and theoretical studies of the phase behaviour of carbon. Using a well-tested version of the LCBOP model (viz. LCBOPI+) we address some of the more controversial hypotheses concerning the phase behaviour of carbon, in particular: the suggestion that liquid carbon can exist in two phases separated by a first-order phase transition and the conjecture that diamonds could have formed by homogeneous nucleation in Uranus and Neptune.