On Mechanisms for Mesocrystal Formation: Magnesium Ions and Water Environments Regulate Crystallization of Amorphous Minerals

Nature produces hierarchical, functional materials by shaping amorphous mineral precursors under physiological conditions. Although biominerals inspire the architectures of synthetic counterparts, the biogenic phase transformations yielding precise crystalline forms, polymorphs and structures are unclear. Elucidating the transformation and structuration of amorphous minerals, herein we show distinct crystallization and structuration schemes synergistically controlled by environmental water contents and the Mg/Ca atomic ratio within amorphous mixed metal carbonates. Control of phase transformation, as well as resultant crystalline micro- and nano-structures, reflects the significance of the amorphous precursors of biominerals as disordered by design. Thereby, we complement the literature-known, suggested (bio)polymer-mediated ‘divide and protect’ mechanism of amorphous mineral stabilization by a Mg2+-based ‘unite and protect’ strategy. Altogether, this allows delineating a novel mechanism for mesocrystal formation based on the interface-coupled dissolution–re-precipitation of mesoscale amorphous precursors, which appear important in biomineralization. In the latter case, the recruitment of environmentally abundant Mg2+ species can also supplement the functions of biomolecules.

Vacuum dried ACC samples were incubated at relative humidity of 85% to achieve a complete transformation to vaterite.The ACC and vaterite samples were mixed in fixed proportions (as in the table below) and their IR spectra were immediately measured.The relative areas of ν2 peaks for ACC (861 cm -1 ) and vaterite (874 cm -1 ) are plotted against the ACC content in the sample.

No
The data points fit well to an exponential decay function described as: ACC (% wt.) = 18.24 × ln ( (Av -6.8909) /0.334) where Av represents the relative area (%) of the ν2 (861 cm -1 ) peak of ACC.The equation is used for estimating the ACC contents of mineral samples.
For kinetic modelling, the mass fractions of ACC and vaterite at distinct time points were converted to volume fractions by applying the density values of ACC and vaterite as 1.49 and 2.54 g/cm 3 , respectively 1 .

Figure S9.
The kinetics of the amorphous to crystalline phase transformation were evaluated by plotting the Avrami 2 parameter ln(−ln(1−X)] as a function of time ln(t), representing the linear equation: The Avrami exponent is provided by the slope n, which can elucidate the mechanism of phase transformation.The intercept provides for the kinetic rate constant k, which is related to the energy of activation.
For estimating k, a modification proposed by Khanna and Taylor 3 is applied in the form of: The  4 .This sigmoidal function is applied in the form of: where X is the transformed volume fraction at time t, kG is a rate coefficient and t0 represents the induction time i.e. the time period before an observable transformation process.Under the applied conditions, the values of the maximum (i.e.final) transformed volume fraction, X0 are close to unity.Under different humidity conditions, the Gompertz fits of the kinetic data for different amorphous samples (Figure 4) are summarized in the following

Figure S1 .
Figure S1.Representative IR spectra of synthetic amorphous calcium carbonate (ACC) samples before and after washing with ethanol.Arrows indicate band frequencies for impurities associated with the ACC phase.

Figure S2 .
Figure S2.Representative IR spectra of amorphous carbonate minerals with distinct occluded Mg 2+ contents.

Figure S3 .
Figure S3.(a) TGA-derived temperature-dependent mass losses and (b) simultaneously acquired DSC profiles for amorphous metal carbonate phases.

Figure S5 .
Figure S5.Representative TEM images of morphological changes in (a-d) Mg0.5ACC and (eg) Mg0.1ACC samples by prolonged exposure to the electron beam.

Figure S6 .
Figure S6.Representative IR spectra of MgXACC phases after thermal treatment at 500°C, indicating the formation of calcite independent of the Mg 2+ contents occluded in amorphous precursors.

Figure S7 .
Figure S7.Representative IR spectra of MgXACC phases after thermal treatment at 300°C, indicating the conservation of amorphous compositions independent of occluded Mg 2+ contents.Note the distinct intensities of the band frequency for water (inset) in relation to Mg 2+ ion contents incorporated in the amorphous mineral phases.

Figure S8 .
Figure S8.The calibration for IR data for elucidating the compositions of ACC-vaterite mixtures.

4 Figure S10 .
Figure S10.Considering the incompatibility of the Avrami model with the experimental observations of ACC and Mg0.1 ACC transformation, the kinetic data were fitted with the Gompertz function4 .This sigmoidal function is applied in the form of:

Figure S13 .
Figure S13.Representative electron diffraction patterns of crystallized mineral derived after incubating ACC at 85% RH.

Figure S14 .
Figure S14.Representative electron diffraction patterns of transformed ACC, crystallized by incubation at 93% RH.