609068 Examining Nucleation Pathways

Monday, November 16, 2020
Thermodynamics and Transport Properties (01A) (PreRecorded+)
Aaron R. Finney and Matteo Salvalaglio, Chemical Engineering, University College London, London, United Kingdom

Alternative pathways for the production of materials from solutions have been proposed that, seemingly, deviate from classical notions of direct crystal nucleation and growth [1]. Metastable phases often precede the most stable crystalline phase and, while such pathways are often described as being ‘nonclassical’, their presence does not necessarily negate classical nucleation theory (CNT) to adequately describe the thermodynamics and kinetics of stepwise phase transformations. Indeed, two-step nucleation provides a conceptual framework to explain such observations [2]. This is, however, distinct from the idea that phases emerge through a process of particle attachment [1].

In silico experiments [3,4] suggest that, within the limit of solution stability, single-step CNT adequately describes NaCl formation from constituent ions in homogeneous, metastable aqueous solutions. Here we apply atomistic simulations with enhanced sampling techniques to study the thermodynamics and kinetics of ion association in the same system. Our results indicate that the crystal nucleation free energy landscape is different from one predicted using a straightforward interpretation of CNT. For instance, clusters adopt a crystalline structure consistent with rock-salt NaCl only beyond a certain size threshold that is reproducible in simulations. We compare the resulting nucleation pathways to those observed in a model system of colloidal particles in a continuum solvent, where particle interaction strengths and mass densities can be controlled to identify pathways in different regions of a phase diagram. We explain our findings within the context of a recently proposed classical approach to two-step nucleation, which adopts a two-dimensional reaction coordinate describing both the size and crystalline order of particles within embryonic clusters [5].

[1] J. J. De Yoreo et al. (2015), Science, 349, 6247. [2] P. G. Vekilov (2010), Nanoscale, 2, 2346-2357. [3] N. E. R. Zimmerman et al. (2015), J. Am. Chem. Soc., 137, 13352-13361; N. E. R. Zimmerman et al. (2018), J. Chem. Phys., 148, 222838. [4] H. Jiang et al. (2018), J. Chem. Phys., 148, 044505; H. Jiang et al. (2019), J. Chem. Phys., 150, 124502. [5] D. Kashchiev (2020), J. Cryst. Growth, 530, 125300.


Extended Abstract: File Not Uploaded
See more of this Session: Computational Studies of Self-Assembly
See more of this Group/Topical: Engineering Sciences and Fundamentals