AmericanChemicalSociety.com

Organic molecules are ubiquitous in sedimentary basins and other natural environments, where they exert a strong influence on the stability and growth/dissolution kinetics of geochemical significant “sparingly soluble” minerals such as calcite (CaCO₃), barite (BaSO₄) and gypsum (CaSO₄.2H₂O) that comprise petroleum reservoirs, evaporites, and pore-filling precipitates, thus affecting petroleum, natural gas and groundwater formation, migration and flow, and contributing to the global carbon cycle (in the case of calcite). Minerals like barite are formed in petroleum recovery pipes. Organic molecules such as organic acids, amino acids, peptides and proteins are known to influence the thermodynamics as well as kinetics of mineral nucleation and crystal growth, thus, affecting the amount and type of crystals precipitated in a petroleum reservoir or other geochemical environment.

We investigated the fundamental molecular-level mechanisms by which organic acids influence calcite and hydroxyapatite (Ca₅(PO₄)₃OH) growth/dissolution rates. It has long been hypothesized in the literature that the negatively-charged functional groups (e.g. –carboxylate, RCOO^- or phosphorylate, -R-PO₄^2-) of organic acids, amino acids, or peptides provide a template for mineral nucleation and crystal growth. It has been suggested that these organics adsorb preferentially in certain crystallographic directions or specific step directions at growth spirals/etch pits, thus either stabilizing the crystal face/steps thermodynamically or hindering crystal face/step movement kinetically by blocking sites.  The goal of our study was to probe these proposed mechanisms for organic-mediated crystal nucleation and growth using computational chemistry methods. An alternative hypothesized mechanism in the literature proposes that the organic ligand adsorbs by disturbing the interfacial solvation preferential crystallographic directions.

We used Molecular Mechanics/Molecular Dynamics (MM/MD) simulations up to 25 ns to examine the interactions of a negatively-charged peptide with carboxyl and phosphoserine groups at hydroxyapatite crystal faces (100), (110) and (001), to investigate nucleation and growth mechanisms. We also used MM/MD simulations to study organic acid adsorption at calcite surfaces and compare our results to previously published experimental Atomic Force Microscopy (AFM) studies of the same systems to further probe crystal growth mechanisms.

For hydroxyapatite, we used MD simulations as well as a novel Bioinformatics computational chemistry approach to determine whether and how the structure of a 10-amino-acid long peptide influences the nucleation and growth pathway of hydroxyapatite (Ca₅(PO₄)₃OH) compared to the inorganic system (Yang et al., 2010).   In particular, we examined the alpha-helix versus random coil structures of the peptide, and used the Bioinformatics approach to determine if the peptide influences crystal morphology by modifying preferential growth directions (Yang et al., 2011).

Results for the nucleation of hydroxyapatite indicated that the random coil peptide promoted formation of an amorphous Ca-PO₄ cluster, which ultimately transforms to the crystalline mineral form. This pathway was more favorable than the formation of a crystalline, templated Ca-PO₄ cluster by the alpha-helix peptide, or any Ca-PO₄ cluster in the inorganic solution. These results indicate that the organic-templated crystal nucleation mechanism proposed in the literature does not hold, at least for Ca-PO₄ phases. These results are consistent with recent experimental evidence for nucleation of amorphous CaCO₃ precursors in Ca-CO₃ systems. 

The Bioinformatics results suggested a lack of geometrical templating between the peptide residues and all HAP surface sites, indicating that adsorption and subsequent crystal growth modulation may be structurally nonspecific. This result also contradicts the two commonly assumed mechanisms that peptides and proteins can adsorb preferentially due to stereochemical matching or by disturbing the interfacial solvation layer, and thus modify crystal growth and morphology.

For calcite, we examined adsorption of succinate, a dianion at neutral pH, along with two Na⁺ counter-ions at steps in various directions on the calcite {104} cleavage face, in a fully hydrated system (Mkhonto et al., in prep.). Adsorption strength was estimated in terms of the reaction energy (-DE_r, _{suc + 2Na+, hyd}) for succinate adsorption at each step, and stability of steps with succinate adsorbed was represented by surface energy of the step (G _{suc+2Na+, hyd}). Succinate was also unable to displace surface hydration waters over the simulation period. When removed manually, succinate adsorbed preferentially in specific step directions, thus, controlling crystal growth. Here, again, stereochemical templating was not the controlling factor. Rather, electrostatics, solvation, and other factors contributed to the final adsorbed geometry and energy. By comparing the present results to experimental AFM studies, we established that the mechanism for succinate-mediated etch pit dissolution/growth spiral morphology is controlled by kinetics (preferential step movement directions) rather than thermodynamics (increased face stability).  

In summary, we found no stereochemical templating effect or interfacial solvation-disturbance by the 10-amino-acid peptide on hydroxyapatite nucleation or crystal growth, consistent with recent experimental work from other groups on Ca-CO₃ nucleation. For calcite also, succinate did not displace surface water, however, we found an apparent effect of the small organic acid, succinate, on the crystal growth kinetics when surface water was manually removed. Thus, the second previously proposed mechanism that the ligand disturbs the interfacial solvation was also not supported by our results within the simulation period.

The results of our studies indicate that the computational approaches developed here may be extended to understand the formation of other sparingly soluble minerals such as barite that are formed in petroleum recovery pipes. Beyond petroleum chemistry and recovery, our study also provides a fundamental basis for understanding the mechanistic controls on the formation of unusual crystal morphologies of biominerals that may serve as biosignatures on Earth and, potentially, on other worlds, for designing biomimetic materials synthesis pathways and for technological processes such as scale inhibition.

References

1. Yang Y, Cui Q, and Sahai N (2010) How does bone sialoprotein promote the nucleation of hydroxyapatite? A molecular dynamics study using model peptides of different conformations. Langmuir 26: 9848-9859.
2. Yang Y, Mkhonto D, Cui Q, and Sahai N (2011) Theoretical study of bone sialoprotein in hydroxyapatite biomineralization. Cells Tissue Organs 194 182-197.
3. Xu, Z., Mkhonto D., Sahai, N. and Teng, H. (in prep.).Dynamics Simulations of Succinate Adsorption at Steps on Calcite (104): Adsorption Energies and Crystal Morphology Evolution. Fro submission to Geochim. Cosmochim. Acta.