Abstract
Portlandite, as a most soluble cement hydration reaction product, affects mechanical and durability properties of cementitious materials. In the present work, an atomistic kinetic Monte Carlo (KMC) upscaling approach is implemented in MATLAB code in order to investigate the dissolution time and morphology changes of a hexagonal platelet portlandite crystal. First, the atomistic rate constants of individual Ca dissolution events are computed by a transition state theory equation based on inputs of the computed activation energies (ΔG*) obtained through the metadynamics computational method (Part 1 of paper). Four different facets (100 or , 010 or 00, or , and 001 or 00) are considered, resulting in a total of 16 different atomistic event scenarios. Results of the upscaled KMC simulations demonstrate that dissolution process initially takes place from edges, sides, and facets of 010 or 00 of the crystal morphology. The steady-state dissolution rate for the most reactive facets (010 or 0) was computed to be 1.0443 mol/(s cm2); however, 0.0032 mol/(s cm2) for or , 2.672 × 10−7 mol/(s cm2) for 001 or 00, and 0.31 × 10−16 mol/(s cm2) for 100 or were represented in a decreasing order for less reactive facets. Obtained upscaled dissolution rates between each facet resulted in a huge (16 orders of magnitude) difference, reflecting the importance of crystallographic orientation of the exposed facets.
Highlights
To upscale atomistic simulations of mineral phases dissolution/precipitation towards much larger timescales and microscopic crystal sizes, kinetic Monte Carlo (KMC) simulations are applied using Molecular dynamics (MD) results from the Part 1 portlandite case study [1]
According to the activation energy (∆G*) obtained through a metaD computational method by Salah Uddin et al (Part 1 of companion paper) regarding calcium atoms (Ca) dissolution for all different scenarios, the dissolution rate constants were initially computed for all possible scenarios using Equation (1)
The difference between the activation energy (∆G*) computed through metaD and enthalpy (∆Ha) through density functional theory (DFT) methods demonstrates the contribution of entropy for total activation energy calculation, which might be significant and should be considered for atomistic rate constant computations
Summary
To upscale atomistic simulations of mineral phases dissolution/precipitation towards much larger timescales and microscopic crystal sizes, kinetic Monte Carlo (KMC) simulations are applied using Molecular dynamics (MD) results from the Part 1 portlandite case study [1]. The interfacial properties are reflecting the chemical composition, type of bonds, crystallographic orientation of the exposed facets, impurities incorporated in the crystal, and lattice defects. Piana et al [4] carried out a 3D microscopic KMC simulation of a growing urea crystal in which the rate constants for corners and edge crystal sites were approximated by data from islands/steps on facets. Chen et al [5] demonstrated the approach on NaCl crystal, by calculating the dissolution rate constants from ab initio MD simulations. NaCl (100) facets with different site types (e.g., edges and corners) were sufficient to perform a KMC simulation for the whole crystal due to crystal symmetry, high aqueous solubility and the absence of intramolecular degrees of freedom within the lattice. A simplified approach to efficiently determine the dissolution rates on the basis of crystal structure is proposed by Elts et al [3]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
