Solutes don’t crystallize! Insights from phase diagrams demystify the “magic” of crystallization

Crystals are ubiquitous in nature and technology. Despite the importance of understanding mechanisms of crystallization, conventional theories have proved inadequate. Consideration of phase diagrams reveals that these theories’ failings result from thermodynamically invalid assumptions, including that solutes are the crystallizing phase and that K_sp describes solubility. Thermodynamics requires that the nutrient (solvent) be the crystallizing phase. This perspective provides tools to rigorously characterize a phase diagram to explicitly determine the components of the crystallizing system and the influence of the diluent on the liquidus of the nutrient. It is shown that diluents can change the liquidus of the nutrient but do not participate in the rate-determining step of crystal growth. The rate-determining step is found to be the propagation of the crystalline phase boundary through a melt-like intermediate. This transition-zone theory model accurately describes crystal growth rates for all concentration-temperature conditions and resolves “riddles” of crystal morphology and growth rates.