Ab Initio Molecular Dynamics Simulations of Amorphous Calcium Carbonate: Interpretation of Pair Distribution Function and X‑ray Absorption Spectroscopy Data

• Published in: Crystal growth & design Vol. 21; no. 4; pp. 2212 - 2221
• Main Authors: Prange, Micah P, Mergelsberg, Sebastian T, Kerisit, Sebastien N
• Format: Journal Article
• Language: English
• Published: France American Chemical Society 07-04-2021
• ISSN: 1528-7483; 1528-7505
• DOI: 10.1021/acs.cgd.0c01655

The structure and transformation of hydrous amorphous calcium carbonate (ACC) are the key to understanding mineralization pathways and their relationship with the properties of the final crystalline materials. Quantitative interpretation of scattering experiments aimed at elucidating the structure of ACC is challenging due to the amorphous nature of this material and, therefore, requires models for the structure and scattering physics. Here, we generate physically realistic ensembles of hydrated ACC structures and their vibrational disorder from ab initio molecular dynamics (AIMD) simulations with an emphasis on enabling the consistent interpretation of the finer details of three complementary structural probes: neutron and X-ray pair distribution function (PDF) experiments and X-ray absorption spectroscopy (XAS). In each case, we simulate the signal directly in reciprocal space and then manipulate it into the real-space PDF or XAS spectrum using the same procedures for the experimental and theoretical data. Good agreement with the experimental data was obtained across the three techniques with the simulations accounting well for all features in the spectra. The remaining small discrepancies pointed to differences between real samples and the idealized simulated systems such as deviations from the nominal CaCO3·nH2O stoichiometry. Additionally, the simulations offered a more accurate description of the local coordination environment of calcium than previous shell-by-shell fits to spectra of synthetic ACC and classical molecular dynamics simulations. This work demonstrates that AIMD is a powerful approach for extracting detailed structural information from neutron PDF, X-ray PDF, and XAS characterization of amorphous carbonate phases.