Local structure of liquid carbon controls diamond nucleation

Abstract

Diamonds melt at temperatures above 4000 K. There are no measurements of the steady-state rate of the reverse process: diamond nucleation from the melt, because experiments are difficult at these extreme temperatures and pressures. Using numerical simulations, we estimate the diamond nucleation rate and find that it increases by many orders of magnitude when the pressure is increased at constant supersaturation. The reason is that an increase in pressure changes the local coordination of carbon atoms from three-fold to four-fold. It turns out to be much easier to nucleate diamond in a four-fold coordinated liquid than in a liquid with three-fold coordination, because in the latter case the free-energy cost to create a diamond-liquid interface is higher. We speculate that this mechanism for nucleation control is relevant for crystallization in many network-forming liquids. On the basis of our calculations, we conclude that homogeneous diamond nucleation is likely in carbon-rich stars and unlikely in gaseous planets.

Cited by in corpus (13)

• Forward Flux Sampling for rare event simulations
• Liquid-liquid critical point in supercooled silicon
• Local order parameters for use in driving homogeneous ice nucleation with all-atom models of water
• A parameter-free, solid-angle based, nearest-neighbor algorithm
• State-of-the-art models for the phase diagram of carbon and diamond nucleation
• Diamond formation from hydrocarbon mixtures in planets
• Phase transitions in fluctuations and their role in two-step nucleation
• Nucleation barriers in tetrahedral liquids spanning glassy and crystallizing regimes
• Thermal Transport in Amorphous Graphene with Varying Structural Quality
• Thermodynamics and kinetics of crystallisation in deeply supercooled Stillinger-Weber silicon
• The structure in warm dense carbon
• An Efficient Self-optimized Sampling Method for Rare Events in Nonequilibrium Systems
• Deriving Reliable Nucleation Rates from Metadynamics Simulations: Application to Yukawa Fluids