Article pubs.acs.org/cm
Opposite Particle Size Effect on Amorphous Calcium Carbonate Crystallization in Water and during Heating in Air Zhaoyong Zou,† Luca Bertinetti,† Yael Politi,† Anders C. S. Jensen,† Steve Weiner,‡ Lia Addadi,‡ Peter Fratzl,† and Wouter J. E. M. Habraken*,† †
Department of Biomaterials, Max Planck Institute of Colloids and Interfaces, Potsdam, Germany Department of Structural Biology, Weizmann Institute of Science, Rehovot, Israel
‡
S Supporting Information *
ABSTRACT: Calcium carbonate is a common constituent of many natural materials, such as shells and skeletons of marine animals. While it is well-documented that additives (organic and inorganic) modulate the crystallization of amorphous calcium carbonate (ACC), the effects of the intrinsic physicochemical characteristics of ACC, such as particle size, shape, and water content on the transformation to crystalline polymorphs, are still poorly understood. Here, we investigate the effect of particle size by preparing ACC nanoparticles with an average size ranging from ∼66 to ∼196 nm using a highresolution titration setup. Our results show that the particle size determined the polymorph selection in solution; an increasing proportion of vaterite to calcite was observed with decreasing particle size. The polymorph selection was ascribed to a higher apparent solubility of ACC with decreasing particle size, a parameter from which we could determine the surface energy of ACC to be ∼0.33 J/m2. Upon heating, particle size showed the opposite effect, as smaller particles favored a higher crystallization temperature from ACC into (only) calcite. When the particle size was large enough, crystallization occurred concomitantly with the removal of bulk water at lower temperatures, where the smallest particles transformed at ∼310 °C, only after losing the final (surface) water. Our results highlight the importance of particle size as well as the crystallization conditions on the stability and transformation mechanisms of ACC.
■
INTRODUCTION Amorphous calcium carbonate (ACC), thermodynamically the most unstable and water-soluble form of calcium carbonate (CaCO3), has been shown to play critical and diverse roles in biomineralization. The mature hard skeletal biominerals from several invertebrate phyla, which consist of mostly calcitic or aragonitic minerals, are formed through transient ACC phases,1,2 while other calcium carbonate biominerals are composed of permanently stabilized ACC.1,3 Because of its relatively high solubility, ACC is also used for storage as a readily available source of ions, for example, in crustaceans undergoing molt.4,5 Inspired by the significance of this material in nature, recent studies have demonstrated the controlled transformation of synthetic ACC into vaterite, aragonite, or calcite with a variety of morphologies.6−11 Studies on the dehydration and crystallization of ACC have revealed that depending on the synthesis method of ACC, the transformation process and the final crystalline product may vary substantially.12−17 Inorganic ions and organic additives such as phosphate, magnesium, phosphorylated proteins, and various polymers have been used to tune the stability of ACC.4,18,19 However, because of difficulties in controlling ACC formation and its characterization,20 the basis for varying stabilities of synthetic and biogenic ACC is still poorly © XXXX American Chemical Society
understood. Additionally, little direct experimental evidence is available on the transition mechanism of ACC into its crystalline polymorphs. An interesting feature of synthetic ACC as well as of biogenic ACC is that both consist of particles from tens of nanometers to almost a micrometer in size.1,13,21,22 However, the consequences of the particle size are not well understood. A number of studies have shown that the particle size could be an important parameter in determining the phase transformation behavior of nanoparticles.23−26 As the size decreases, the contribution of surface energy to the total free energy becomes increasingly important. Furthermore, the free energies of the three crystalline CaCO3 polymorphs are close enough that surface energy effects can cause crossovers in stability,27 leading to the formation of kinetically stabilized vaterite or aragonite polymorphs, instead of the thermodynamically favored calcite at ambient conditions. Therefore, introducing small changes in the ACC particle size, thereby tuning the surface to bulk ratio, may be crucial for polymorph selection during the crystallization of ACC. In addition, as shown by Bloch et al.,28 shortReceived: January 13, 2015 Revised: May 27, 2015
A
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials
solution to 50 mL of water at room temperature at a rate of 0.1 mL/ min and measuring the signal (in mV) as a function of calcium concentration. During the calibration of calcium, the pH of the solution was kept at 10.5 by titrating 0.01 M NaOH. The pH electrode was calibrated every day using standard buffers at pH 4.00, pH 7.00, and pH 9.00 (Metrohm AG). Scanning Electron Microscopy and Transmission Electron Microscopy. The morphology was imaged using a field emission scanning electron microscope (JEOL, JSM-7500F) working at an acceleration energy of 10 keV. Samples were not coated prior to investigation. Transmission electron microscopy was performed using a ZEISS EM 912 transmission electron microscope working at 120 keV. Small- and Wide-Angle X-ray Scattering. Synchrotron smalland wide-angle scattering experiments were carried out at the μ-Spot beamline (BESSY II storage ring, Helmholtz-Zentrum Berlin) using a multilayer monochromator and spot size of 100 μm. Two-dimensional (2D) scattering patterns were collected using a MarMosaic CCD detector (Rayonix L.L.C., Evanston, IL, USA). Radial integration of the 2D scattering patterns with the software Fit2D (A. Hammersley, ESRF, Grenoble, France) gave the spherically averaged scattering intensity as a function of the modulus of the scattering vector q, with q = 4π sin(θ)/λ, where 2θ is the scattering angle and λ is the wavelength. The resulting profiles were corrected for dark current, primary intensity, and sample transmission. An additional background signal resulting from the Kapton foil windows of the in-situ heating setup (THMS 600, Linkam Scientific Instruments, Surrey, UK) was also subtracted. X-ray Diffraction and Rietveld Refinement. The X-ray diffraction (XRD) measurements of the samples were recorded on a Bruker D8-Advance X-ray powder diffractometer using Cu Kα radiation (λ = 1.5406 Å) with scattering angles (2θ) of 10−80° and a scan speed of 0.36°/min. Rietveld refinement was performed with Fullprof software31 using spherical harmonics to model size anisotropy of vaterite.32 The instrumental line broadening was determined using a Ceria standard (NIST). Raman and Infrared Spectroscopy. Raman spectra were collected using a confocal Raman microscope (α300; WITec, Ulm, Germany) equipped with a Nikon objective (10×) and a 532 nm laser. Spectra were acquired with a CCD camera (DV401-BV; Andor Technology Ltd., Belfast, UK) behind a spectrometer (UHTS 300; WITec) (40 accumulations, integration time 1 s), where ACC powders were dropped on a microscope slide before measurement. Infrared spectra were recorded using a Fourier transform infrared spectrometer (Bruker Optik GmbH, Ettlingen, Germany) equipped with a MCTdetector (500 scans, resolution 2 cm−1). As the ACC we prepared tends to crystallize upon the use of physical force including the pellet machine or grinding, for infrared measurements, ACC was first dispersed in chloroform and then deposited on a ZnSe window, before measuring it in transmission. Thermogravimetric Analysis/Differential Scanning Calorimetry. Simultaneous weight loss and heat flow were measured during programmed heating (25−500 °C at 1.5 °C/min) using thermogravimetric analysis coupled with differential scanning calorimetry (SENSYS evo TGA-DSC, SETARAM Instrumentation, Caluire, France). Approximately 15 mg of dry powdered sample was placed in an alumina crucible. Dry nitrogen was used as the purge gas. ICP-OES Analysis. Elementary analysis was performed by inductively coupled plasma-optical emission spectrometry (Optima 8000 ICP-OES Spectrometer, PerkinElmer, Waltham, MA). Sample solutions were prepared by dissolving certain amounts of as-prepared powders in diluted HNO3 solution.
range order in amorphous materials, such as amorphous alumina, can also be size-dependent. In this study we investigated the influence of particle size on the stability and transformation mechanism of ACC. For this we used a high-resolution titration setup, which was recently applied for the time-resolved analysis of prenucleation species in solution.29 This setup enabled us to prepare batches of chemically and morphologically well-defined ACC nanoparticles with different particle sizes by adding CaCl2 at a fixed addition rate to a Na 2 CO 3 solution at varying concentrations, at a fixed Ca2+:CO32− ratio of 1:1. To investigate the effect of different destabilization conditions, we followed the transformation of ACC inside the native reaction solution and, after extraction, under continuous heating. The formation and transformation of ACC inside the reaction solution was monitored using online pH-electrode measurements and Ca2+-selective electrode [ion-selective electrode (ISE)] measurements.30 This also enabled us to derive the apparent solubility of the as-prepared ACC.20 The destabilization of extracted ACC under continuous heating was investigated through combined thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC), and in situ synchrotron small- and wide-angle scattering analysis (SAXS/WAXS). In both experiments, chemical, structural, and morphological analyses of extracted material at different time points were performed by infrared spectroscopy (IR), Raman spectroscopy, inductively coupled plasma-optical emission spectrometry (ICP-OES), X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM).
■
EXPERIMENTAL SECTION
Materials and General Preparative Methods. Analytical-grade calcium chloride dihydrate (CaCl2·2H2O) and sodium carbonate decahydrate (Na2CO3·10H2O) were purchased from Sigma-Adrich and used as-received. The calcium solution and carbonate solution were prepared by dissolving CaCl2·2H2O and Na2CO3·10H2O in ultrapure water. A computer-controlled titration system (905 Titrando, Metrohm AG, Herisau, Switzerland) was utilized for the experiments. The setup consists of a titration device controlling three dosing units (800 Dosino). The dosing unit incorporates a 5 mL cylinder with a minimal dose-able volume of 0.2 μL. ACC Synthesis. The experiments were performed at (24 ± 1) °C in a 100 mL vessel filled with a certain amount of carbonate solution (45−49.75 mL) under stirring. One molar calcium solution (0.25−5 mL) was dosed through the dosing unit into the reaction vessel at a rate of 10 mL/min to ensure a rapid mixing of the two solutions. The total volume of all solutions after dosing was 50 mL and the concentration ratio of calcium/carbonate after mixing was 1:1. In the experiments, samples were prepared at five different concentrations: 5, 10, 20, 40, and 100 mM (see Table S1 of the Supporting Information for detailed preparation procedures). Here, the different concentration levels reflect the concentration of free Ca2+ and CO32−/HCO3− species, in the hypothetical case that only free ions are present after complete mixing of both reactants. The precipitates were collected by fast vacuum filtering of the reaction solution after mixing, and rinsed with ethanol. The dry powders after filtration were then stored in a vacuum desiccator for further characterization within a few days. Monitoring pH and Calcium Concentration in Solution. The pH and calcium concentration of the solution during reaction were monitored using a pH electrode and a calcium-selective electrode (CaISE, Metrohm AG) connected to the titration system. Electrodes, vessel, and buret tips were cleaned with dilute hydrochloric acid and carefully rinsed with distilled water after every experiment. The Ca-ISE was calibrated before every experiment by dosing the 1 M calcium
■
RESULTS Synthesis of ACC with Different Particle Sizes. The morphology of the as-synthesized powders was investigated by SEM. As shown in Figure 1, as previously reported for ACC,14,33 well-distributed spherical structures with smooth surfaces were observed at all concentrations, however, with B
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials
approximate average size of 200, 120, 87, 73, and 66 nm, respectively. We note that ACC filtration and dehydration procedure did not influence the particle size, as was confirmed by particle size analysis using cryo-SEM images (Figure S1 of the Supporting Information). WAXS patterns of all samples (Figure S2a) showed two broad peaks at 21.9 and 31.6 nm−1. These positions are in accordance with earlier reports for ACC,17 and the absence of Bragg peaks confirm the amorphous nature of the samples. Raman spectra of all samples (Figure S2b) were identical, showing an intense broad peak at 1084 cm−1 [full width at halfmaximum (fwhm) width ∼29 cm−1], corresponding to the ν1 symmetric stretching mode of CO32−, a small peak at 727 cm−1 (fwhm width ∼59 cm−1), a broad peak at approximately 214 cm−1,1,34 and a broad peak at about 3374 cm−1, corresponding to the O−H stretching mode of water. The observed Ramanforbidden signals in the region 1300−1500 cm−1 are typical for an unstructured “glassy” material. Also, infrared analysis did not show any difference among the samples (Figure S2c,d). With infrared analysis, a splitting in the asymmetric stretch of the carbonate ions (ν3) at about 1480 and 1388 cm−1 was clearly seen, which together with the broad peak at around 696 cm−1, is characteristic of ACC. The peaks at 1072 and 861 cm−1 correspond to the symmetric stretch (ν1) and out-of-plane bending (ν2) bands of CO32−.35 The broad band at ∼3304 cm−1 and a peak at 1649 cm−1 are due to vibrations of structural water in ACC. No peaks of hydroxide (OH−) were observed in either Raman or infrared spectra. All these data showed that the as-synthesized powders were composed of hydrated ACC with no detectable structural and chemical differences. To investigate the chemical composition of ACC in greater depth for the presence of possible HCO3− or counterions such as Na+, we performed additional elemental analysis. The amount of Na+ and Ca2+ generated after a known amount of ACC dissolved in dilute nitric acid was measured by ICP-OES. Though Cl− cannot be measured by ICP-OES, it was assumed to be in the same concentration range as that measured for Na+. The water content was estimated by measuring the weight loss by TGA after heating ACC to 400 °C. With use of this data, the percentages of carbonate or bicarbonate ions were deduced using mass and charge balance equations. The results are presented in Table 1. Here, the maximum Na+:Ca2+ molar ratio was found to be 0.032, indicating that there were only small amounts of likely adsorbed Na+ or Cl− in the samples, in agreement with the increasing amounts of Na+ upon decreasing particle size. Furthermore, we deduced that the molar ratio of Ca2+ to H2O for all samples was ∼1.1, and that the content of HCO3− in the prepared ACCs was negligible.36,37 Relationship between Apparent Solubility of ACC and Particle Size. To determine the time range during which ACC was stable in the solution, and then follow its transformation, the changes of pH and calcium activity during the reaction were
Figure 1. SEM micrographs and corresponding size distribution of ACC particles prepared at different concentrations: ∼200 nm for (a) at 5 mM, ∼120 nm for (b) at 10 mM, ∼87 nm for (c) at 20 mM, ∼73 nm for (d) at 40 mM, and ∼66 nm for (e) at 100 mM.
varying particle sizes. Here, the broader size distribution, especially for the smaller sized particles, is likely due to poor mixing for the highly concentrated samples. The numberaverage particle size calculated according to the SEM micrographs was found to decrease from about 200 to 66 nm with increased concentration from 5 mM (a) to 100 mM (e) (Figure 1). We thus labeled the samples according to their Table 1. Summary of the Chemical Composition of ACC approximate particle size 200 nm 120 nm 87 nm 73 nm 66 nm
Ca2+ (wt %) 33.5 33.3 33.3 33.5 32.3
± ± ± ± ±
0.3 0.2 0.2 0.5 0.9
Na+ (wt %) 0.05 0.11 0.12 0.23 0.6
± ± ± ± ±
0.1 0.08 0.04 0.03 0.06
H2O (wt %) 16.9 16.0 17.1 16.7 16.4
± ± ± ± ±
1.2 1.0 1.6 1.1 0.6 C
CO32− (wt %) 50.7 49.3 50.4 51 47
± ± ± ± ±
1.7 1.3 1.7 2.2 3.6
HCO3− (wt %) −1.1 1.1 −1.1 −1.7 2.8
± ± ± ± ±
2.8 2.3 3.3 3.3 4.6
H2O/Ca2+ 1.12 1.07 1.14 1.11 1.13
± ± ± ± ±
0.04 0.03 0.05 0.03 0.02
CO32−/Ca2+ 1.01 0.99 1.01 1.02 0.97
± ± ± ± ±
0.05 0.04 0.05 0.07 0.12
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials
Figure 2. pH (a) and calcium activiy (b) evolution of the solution for experiments with ACC particle sizes of 200−73 nm. Mixing starts at t = 120 s. Here, (1) refers to the ACC plateau at which the samples were extracted and (2) to the transformation of ACC to crystalline CaCO3 polymorphs.
Figure 3. Average apparent solubility of ACC with different particle sizes (a).The relationship between ln(Ks) and inverse size 1/d of extracted ACC according to eq 1 (b).
to the extended Debye−Hückel equation.20 Here, apparent solubility refers to an empirically derived bulk solubility value under the assumption of a local equilibrium between the ACC and solution species. We observed that Ks increased from 7.5 × 10−7 for the 200 nm sized particles to 1.3 × 10−6 for the 73 nm sized particles. As a comparison, Clarkson et al.38 determined the Ks of ACC through the addition of CaCl2 solution into a 15 mM Na2CO3 solution (in 0.1 mol/L NaCl) in the presence of a small amount of triphosphate at different temperatures. The value of Ks was determined to be 9.1 × 10−7 at 25 °C, which lies exactly between 8.3 × 10−7 (120 nm, 10 mM) and 9.5 × 10−7 (87 nm, 20 mM), Ks values obtained in the present experiment. Many investigations have shown that the particle size influences the solubility of nanosized materials such as calcium and barium sulfates and amorphous silica.40,41 For this, the widely used Ostwald−Freundlich equation40 describing the size dependence of the solubility of solid particles in liquid solutions was derived, based on Gibbs thermodynamics and later optimized by Kaptay.42 For spherical particles the relationship can be written as
monitored (Figure 2). (For reproducibility of the measurements see Figure S3). In the experiments corresponding to particle sizes of 200−73 nm, the presence of ACC was indicated by a plateau in both pH and Ca2+ activity directly after mixing. The plateau was stable for ∼20−100 s, after that showing a slight decline especially for the smaller particle sizes until ∼300 s (point 2 in Figure 2a). Then an abrupt change to a final plateau was observed where all ACC has transformed into the mature (crystalline) polymorphs.30,38 Though until the final drop the extracted material was still ACC without any measurable changes in morphology or chemistry, for the above analysis, ACC samples were taken within the first 20 s after mixing. As large amounts of ACC precipitated with smaller particles sizes, the experiment corresponding to a particle size of 66 nm proved to be inappropriate for the online measurement of pH and Ca2+ activity during the ACC formation stage (Figure S4). However, trustworthy values for pH and Ca2+ activity were obtained for the ACC crystallization process. Interestingly, the values for Ca2+ activity and pH that corresponded to the ACC plateau were dependent on the particle size. On the basis of these values and the conditions before mixing, the apparent solubility of the ACC (Ks) for each experiment was calculated (Figure 3a, see Habraken et al.20,39). Briefly, as Ks is an activity product, the activity of CO32− was derived by subtracting the amount of CO32− in the precipitate (=Ca2+ bound) from the total amount of carbonate added and correcting for the presence of HCO3− in solution, applying the HCO3−/CO32− equilibrium constant and multiplying the obtained concentration by the activity coefficients according
RT 6σ 1 ln K s = + constant M ρ d
(1)
where R is the gas constant, T the absolute temperature, M the molecular weight of the solid in solution, ρ the density of the solid, σ the surface energy per unit surface between the solid and its solution, and Ks the solubility of a sphere of diameter d, respectively. On the basis of the values of apparent solubility and corresponding average particle size for ACC, a linear D
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials
Figure 4. SEM micrographs of the intermediate products observed at the end of the ACC plateaus for experiment with larger particle sizes (200 nm) (a−c) and for experiment with smaller particle sizes (66 nm) (d−f).
Figure 5. SEM micrographs of the crystallized products appearing as the reactions continued for 600 s: experiment starting from ACC particles with sizes of approximately 200 nm (a), 120 nm (b), 87 nm (c), 73 nm (d), and 66 nm (e) and corresponding weight fraction of crystalline phases determined using Rietveld refinement (f) (see Figure S6).
a stable second plateau was obtained (t = ∼720 s, Figure 2). For the reaction with particle sizes around 200 nm, after the initial ACC plateau, the pH and Ca activity increased and then decreased again, indicating dissolution of ACC during the transformation process. With increasingly smaller particles, such a dissolution−reprecipitation signature became less prominent. SEM was used to characterize the intermediate products taken during the abrupt decrease in the Ca2+/pH curves at 300−350 s. For the experiment with 200 nm sized particles, the surface of the large ACC particles was significantly etched, and at the same time, rhombohedral calcite crystals were present, corresponding to large-scale dissolution− reprecipitation of ACC (Figure 4a−c). The surface of the smaller particles remained rather smooth, and large vaterite spheres were formed in the experiment with 73 nm sized particles (Figure 4d−f). In both cases, the attachment of ACC particles to the calcite and vaterite crystals was observed, suggesting possibly a continued growth of these crystals from ACC. This observation is in agreement with previous results for
relationship between ln Ks and 1/d was obtained (Figure 3b). From this curve, the ACC surface energy was calculated to be 0.33 J/m2 when a value of 1.62 g/cm3 was used as the density of ACC particles.43 Though no surface energy calculations for ACC have been reported in the literature, on the basis of the ACC water content, it can be predicted that the surface energy of ACC should be smaller than that of an anhydrous mineral like calcite (1.48 and 1.87 J/m2 for hydrated and anhydrous surfaces).27,44 Furthermore, the obtained value is in the same range, but slightly larger than the surface energy recently derived for ACP.39,45 It must be noted that, for less soluble materials, such as all the crystalline forms of CaCO3, the particle size dependence of the solubility will only become significant at much smaller particle sizes (in the range of a few nanometers) than those reported here for ACC. In general, eq 1 impliesin agreement with the data in Figure 3that smaller particles are less stable in dissolution conditions. Transformation of ACC in Solution. To investigate the transformation of ACC in solution, we completed the titration and maintained stirring to allow the reactions to continue until E
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials
Figure 6. TGA-DSC curves of ACC (a), contour plot of WAXS scattering intensity versus temperature and q for ACC with particle sizes of 200, 73, and 66 nm under continuous heating (b), and the change of the intensity of calcite peak (104) in the WAXS patterns (c). Dashed lines in (a) indicate the positons of the first endothermic peak (1) and the first (2) and second (3) exothermic peak. Dashed lines in (b) represent the temperature at which the peak (104) of calcite crystal appears.
the inorganic system46,47 and similar/other systems stabilized with additives.11,48 Finally, the calcium carbonate that was present after the reaction proceeded for 600 s was collected and characterized with XRD, IR, and SEM. XRD patterns and IR spectra (Figure S5) showed that all products contained both calcite and vaterite, but in different proportions. Quantitative analysis of the X-ray data was performed using Rietveld refinement software (Figure 5f), showing that the weight fraction of calcite decreased from ∼65% to ∼14% as particle size decreased from about 200 to 66 nm. SEM images (Figure 5a−e) further showed that ACC transformed predominantly to calcite rhombohedra for large ACC particles (close to 200 nm), whereas the percentage of vaterite spheres (∼4 μm) significantly increased with smaller initial ACC particles. In summary, decreasing the ACC particle size resulted in a transformation of ACC toward a final crystalline phase which changes from predominantly calcite to vaterite. Although vaterite would eventually transform into calcite30 in solution, the final plateaus in pH and free Ca2+ activity demonstrated that this transformation process was negligible in the time scale of the experiment. Heat-Induced Transformation of ACC in Air. The stability and crystallization of ACC under continuous heating was investigated using TGA/DSC and in situ SAXS/WAXS. TGA/DSC of extracted ACC samples (Figure 6a) showed
three main consecutive stages of events on programmed heating (1.5 °C/min). The most prominent feature was that in contrast to transformation in watersmaller ACC particles were more stable against crystallization upon heating in air compared to large particles. In detail, from 25 °C to 130−150 °C a broad endotherm in the DSC curves was observed, involving a total weight loss of 11−15%. For all samples with ACC particles larger than 70 nm, this event was followed by a rapid weight loss of 1.5−3.5%, concomitant with the first exothermic peak in the DSC curves, whereas the sample with the smallest particles (66 nm) showed a slower, gradual decrease in weight, during which the first exothermic event occurred. More interestingly, it was observed that the temperature of the first exothermic peak increased with a decrease of particle size from ∼130 °C for the 200 nm sized particles to ∼205 °C for the 66 nm sized particles. The third stage of little additional weight loss was complete at ∼310 °C, where the second exothermic peak emerged. This was the main exothermic event for the 66 nm sized particles. The weight loss of all three stages were attributed to the loss of H2O since the decomposition of the dehydrated CaCO3 to CaO and CO2 happens at approximately 750 °C.17 In situ wide-angle X-ray scattering patterns for ACC with particle sizes of 200, 73, and 66 nm (Figure 6b) provided information about the crystallization process. No peaks of CaCO3 crystals could be seen during the first stage, indicating F
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials
Figure 7. TEM images of a typical ACC particle (a), sample with 200 nm ACC particles after 200 °C (b) and 500 °C (c) and sample with 66 nm ACC particles after 250 °C (d) and 500 °C (e,f).
that the samples were amorphous and weight loss was attributed to loosely bound water. In all samples, the first exothermic peak in the DSC curves corresponded to the crystallization of ACC to calcite, as indicated by the appearance of the (104) peak of calcite in the WAXS patterns. As this also corresponded to the sudden increase in weight loss observed with TGA, this shows that the loss of water (an endothermic event) and the exothermic crystallization of the ACC occurred simultaneously. The crystallization rate (Figure 6c) was higher when the sample crystallized at a lower temperature, which suggested that the residual water in ACC accelerated the crystallization of ACC. The second exothermic peak represented a further crystallization of the remaining ACC into calcite; because of the lack of residual water, this likely proceeded via a solid-state transformation. Under these conditions, all ACCs transformed to calcite. As we were not able to detect any chemical or structural differences between different samples, the difference in thermal stability was attributed solely to the difference in particle size, where the 200 nm sized particles transformed at a much lower temperature and apparently by a different mechanism than the 66 nm sized particles. Here, the large distribution in particle sizes is likely to account for the presence of multiple crystallization events. To further clarify this difference in transformation mechanism, the morphology change of the samples after heating was studied. SEM images of the samples taken before the second crystallization peak (Figure S7) exhibited no morphological changes as compared to the amorphous samples. Further heating of the large particles until after the second crystallization peak (500 °C) (Figure S7) resulted in small morphological changes for the larger samples, such as changes of the spherical shape and smooth surface, while the smaller ACC particles agglomerated, and formed larger particles. However, TEM images (Figure 7) of the particles after the first crystallization peak showed the presence of 10−20 nm sized pores randomly distributed inside the particles, indicating internal structural rearrangements because of the loss of bulk
water. This change was also reflected in the in situ SAXS analysis of 200 nm ACC particles under continuous heating. As shown in Figure S8a, a peak at q ∼1.8 nm−1 appeared and became more intense upon heating until 100 °C, indicating structural rearrangements on the nanometer scale. During crystallization, another broad peak appeared at q ∼0.7 nm−1 (140 °C), which became more intense until 160 °C (Figure S8b). Corresponding to the observed pores in TEM, this peak could be fitted by a form factor of 10−20 nm sized spherical particles. Finally, by analysis of the fwhm of the first peak (104) arising in the WAXS patterns, the coherence length of calcite crystallites for all samples was determined to be 39−45 nm, a value which should be smaller than the size of calcite crystallites, and in our experiments was only slightly smaller than the average particle size of the 66 nm sample. These results suggest that (heat-induced) sintering of smaller particles was necessary to obtain the calcitic long-range order, whereas the larger particles transformed individually.
■
DISCUSSION Although the crystallization of ACC has been extensively studied by a number of approaches,7,16,49 the mechanism(s) behind this transformation pathway is not yet clear. From the literature it can be deduced that the transformation mechanism depends on the way ACC is prepared,15 as well as the way it is destabilized.50 The results presented in this paper showed that particle size may be the link between the synthesis conditions and the final stability/polymorph selection of the ACC since different synthesis conditions yielded different particle sizes, and the resulting crystallization kinetics depended strongly on particle size. Furthermore, by performing the destabilization in two different modes (in solution and upon heating), we monitored how aqueous and dry environments (i.e., the presence of water) influenced the effect of ACC particle size on final polymorph selection and stability. The observed particle size dependence on polymorph selection in solution provides an explanation for the presence of vaterite, calcite, or a mixture of both, as previously reported. G
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials For example, Clarkson et al.38 observed the formation of only calcite after the crystallization of ACC at low concentration (2.5 mM CaCl2 and 15 mM Na2CO3), indicating large particles according to our data. Similar to our experiment with 200 nm sized particles, an additional dissolution−reprecipitation event was observed by them. Ogino et al.30 showed that the polymorph selection directly after the formation of ACC changes from a calcite/vaterite mixture to only vaterite upon increasing the initial supersaturation starting from a mixture of 19 mM Ca2+ and 9 mM CO32−. In agreement with our data, at these conditions they obtained particles in the 100 nm range (120 and 87 nm samples). The reactions were also performed for longer time periods, and it was observed that the transformation of vaterite to calcite is much slower than the crystallization of ACC. The same trends were also observed in a study by Kim et al.51 Recently, Rodriguez-Blanco et al.46 showed that ACC transformed to calcite via a vaterite intermediate. As all the ACC particles had a small particle size (∼20−45 nm), which were smaller than our smallest particles (66 nm), this transformation mechanism is consistent with our observations. It should be noted that ionic strength (IS) and pH vary between the in situ transformation experiments. The observed relationship between particle size and solubility, however, is independent of the IS as Ksp is an activity product, which is corrected for changes in ionic strength. The Ksp is a material property that is not affected by pH. Thus, the slight variation in pH at the ACC plateau does not influence the relationship between particle size and solubility. The same is true for the polymorph selection from ACC as, in the case of a dissolution− reprecipitation mechanism, the driving force for vaterite/calcite formation is the ratio of the solubilities of ACC and vaterite/ calcite (Ksp(ACC)/Ksp(Vat/Cal)). Both terms are independent of ionic strength and pH. However, the higher IS for the smaller particles may induce aggregation. If this indeed occurs, it could lead to higher local supersaturations favoring the formation of vaterite. Such an event, however, would be indicated by a sudden change in polymorph selection, a phenomenon for which we do not find any evidence in our data set. For the heat-induced transformation of ACC, the effect of particle size can explain previous data that were not well understood. Exceptionally large ACC particles with an average size ranging from 475 to 759 nm were previously synthesized by hydrolysis of dimethyl carbonate as a source of CO2 in an aqueous solution of CaCl 2.13 The exothermic process representing the crystallization of these large ACC particles was found to occur at 105 °C, followed by an endothermic process due to water loss at 149 °C. ACC with smaller particle sizes was also shown to be much more stable. For example, as described by Koga et al.,52 ACC particles with sizes less than 50 nm were synthesized by rapidly adding CaCl2 solution (0.1 M) to a mixed solution of Na2CO3 (0.1 M) and NaOH (in the range of 0−2 M) at 5 °C. The as-synthesized ACC particles remained amorphous until crystallization at 300 °C, which corresponds very well to our observed trend of size effect of ACC. Recently, Schmidt et al.14 synthesized ACC based on procedures of Koga et al.52 where they observed a linear correlation between the crystallization temperature and pH of the starting synthesis solution. According to our observations, the average size of ACC particles decreased when the pH of the reaction solution increased. The influence of pH in our experiments was limited to the change in the ratio of free CO32− to HCO3−, according to the second equilibrium
constant (pKa = 10.33), thereby increasing CaCO3 supersaturation upon an increase in pH. In the work presented here, no significant chemical or structural differences were observed in any of the samples, thereby excluding pH-related modifications. The most striking observation remains the opposite effect of particle size on crystallization from solution or induced by heat. Indeed, small particles are less stable (more soluble) in solution, while upon heating they are more stable against crystallization than larger particles. These two processes are summarized in Figure 8. In solution, large particles transformed
Figure 8. Effect of particle size on ACC transformation in solution and upon heating. In solution, by decreasing the particle size the polymorph selection changes from predominantly calcite to a pure vaterite phase (blue arrows). Upon heating, a decrease in particle size shifts the formation of calcite to higher temperatures (red arrows). Large particles crystallize concomitantly with the release of water, forming porosity inside the spheres. The smallest particles (66 nm sample) crystallize only after all of the water evaporated, which was accompanied by a visible agglomeration. The darker layer around the ACC spheres represents the more ordered surface layer, forming a core−shell-like structured particle.
directly into calcite while smaller ones led first to the appearance of vaterite. We attribute this to the size-dependent dissolution kinetics. Smaller ACC particles dissolve faster and, consequently, lead to higher local supersaturations, which may favor the nucleation of vaterite rather than calcite from solution. Indeed, when the solution is supersaturated with respect to both vaterite and calcite formation, the lower kinetic barrier may favor vaterite nucleation. Upon dissolution of large ACC particles, the supersaturation increases so slowly that vaterite formation is inhibited, leading directly to the nucleation of calcite. Since no structural differences were found between small and large ACC particles by means of IR and Raman spectroscopy, the higher stability of smaller ACC particles upon heating (see Figure 8) should be related to the ratio of surface to bulk, which is obviously larger for smaller particles. Considering that the total energy of an ACC particle is the sum of the surface and the bulk contributions, we speculate that the surface H
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials
selected SAXS patterns. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemmater.5b00145.
contribution in air could be at the origin of this effect. In an amorphous solid, especially if partially hydrated, the mobility of the atoms can be much higher than that in a crystalline solid, and in particular, the molecular moieties in regions near the surface could more easily rearrange in an amorphous particle to fit the interface with air than in a crystalline solid where the surface moieties are constrained by the underlying lattice. This would mean that the surface energy of ACC is much lower than that of calcite (which is in agreement with our observation of 0.33 J/m2 for hydrated ACC versus 1.48 J/m2 for calcite), while the situation is of course opposite for the bulk energies. Hence, the total energy could become lower for an ACC particle than for a calcite particle of the same size, provided the particles are small enough so that the contribution of near-surface regions compensates for the difference in bulk energies. For the biogenic system, where the presence of additives can inhibit a dissolution−reprecipitation mechanism,10,11 such a particle size dependence could play an important role in stabilizing the ACC, especially as reported particle sizes in biogenic systems are small (20−50 nm).2,11 The above thermodynamic considerations, however, do not explain the large increase in crystallization temperature between samples where the average particle size difference was not large, namely, 73 nm (with a crystallization temperature of 170 °C) and 66 nm (310 °C). In this context, it is interesting to note that most of the (mobile) water has left the calcium carbonate preparations at a temperature of 150 °C for all particle sizes (see Figure 6a). This means that a kinetic effect could dramatically amplify the thermodynamic one discussed in the previous paragraph. If a small particle is sufficiently stabilized by its surface so as not to crystallize at temperatures below 150 °C, it will be further stabilized by the fact that without water the mobility of ions is dramatically reduced. This could explain why the crystallization temperature jumps from 170 to 310 °C, when the particle size decreases only from 73 to 66 nm. It is interesting to note that similar effects were observed in another mineral type, namely, the rutile−anatase system, with two different polymorphs of TiO2.53 Where at large particle sizes (∼30 nm) the rutile is the most stable phase, as it has the lowest bulk energy, decreasing the particle size (∼10 nm) favors the formation of anatase, which has the lower surface energy. In conclusion, a clear effect of particle size of amorphous calcium carbonate on the stability and polymorph selection upon crystallization was observed. This may explain many discrepancies between the crystallization pathways of various synthetic ACCs described in the literature. Most interesting here is the opposite effect of particle size on ACC stability in solution and upon heating. To understand biological crystallization from an ACC precursor, it is therefore not only important to know the chemical and structural properties of the specific precursor but also the conditions in which the ACC precursor transforms into the final crystalline state.
■
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Authors thank Stefan Siegel and Chenghao Li for support during the SAXS +WAXS measurements at the μ-Spot beamline at BESSY II, Helmholtz-Zentrum Berlin, and Sidney J. Omelon for editing the paper. Zhaoyong Zou is thankful for the support provided by the China Scholarship Council (CSC). The research was supported by a German Research Foundation grant within the framework of the Deutsch−Israelische Projektkooperation DIP.
■
REFERENCES
(1) Addadi, L.; Raz, S.; Weiner, S. Taking advantage of disorder: Amorphous calcium carbonate and its roles in biomineralization. Adv. Mater. 2003, 15, 959−970. (2) Politi, Y.; Arad, T.; Klein, E.; Weiner, S.; Addadi, L. Sea urchin spine calcite forms via a transient amorphous calcium carbonate phase. Science 2004, 306, 1161−1164. (3) Aizenberg, J.; Lambert, G.; Weiner, S.; Addadi, L. Factors Involved in the Formation of Amorphous and Crystalline Calcium Carbonate: A Study of an Ascidian Skeleton. J. Am. Chem. Soc. 2002, 124, 32−39. (4) Akiva-Tal, A.; Kababya, S.; Balazs, Y. S.; Glazer, L.; Berman, A.; Sagi, A.; Schmidt, A. In situ molecular NMR picture of bioavailable calcium stabilized as amorphous CaCO3 biomineral in crayfish gastroliths. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 14763−14768. (5) Shechter, A.; Berman, A.; Singer, A.; Freiman, A.; Grinstein, M.; Erez, J.; Aflalo, E. D.; Sagi, A. Reciprocal Changes in Calcification of the Gastrolith and Cuticle During the Molt Cycle of the Red Claw Crayfish Cherax quadricarinatus. Biol. Bull. 2008, 214, 122−134. (6) Lee, K.; Wagermaier, W.; Masic, A.; Kommareddy, K. P.; Bennet, M.; Manjubala, I.; Lee, S.-W.; Park, S. B.; Cölfen, H.; Fratzl, P. Selfassembly of amorphous calcium carbonate microlens arrays. Nat. Commun. 2012, 3, 725. (7) Kim, Y.-Y.; Hetherington, N. B. J.; Noel, E. H.; Kröger, R.; Charnock, J. M.; Christenson, H. K.; Meldrum, F. C. Capillarity Creates Single-Crystal Calcite Nanowires from Amorphous Calcium Carbonate. Angew. Chem., Int. Ed. 2011, 50, 12572−12577. (8) Cantaert, B.; Kim, Y.-Y.; Ludwig, H.; Nudelman, F.; Sommerdijk, N. A. J. M.; Meldrum, F. C. Think Positive: Phase Separation Enables a Positively Charged Additive to Induce Dramatic Changes in Calcium Carbonate Morphology. Adv. Funct. Mater. 2012, 22, 907−915. (9) Natalio, F.; Corrales, T. P.; Panthöfer, M.; Schollmeyer, D.; Lieberwirth, I.; Müller, W. E. G.; Kappl, M.; Butt, H.-J.; Tremel, W. Flexible Minerals: Self-Assembled Calcite Spicules with Extreme Bending Strength. Science 2013, 339, 1298−1302. (10) Gal, A.; Habraken, W.; Gur, D.; Fratzl, P.; Weiner, S.; Addadi, L. Calcite Crystal Growth by a Solid-State Transformation of Stabilized Amorphous Calcium Carbonate Nanospheres in a Hydrogel. Angew. Chem., Int. Ed. 2013, 52, 4867−4870. (11) Gal, A.; Kahil, K.; Vidavsky, N.; DeVol, R. T.; Gilbert, P. U. P. A.; Fratzl, P.; Weiner, S.; Addadi, L. Particle Accretion Mechanism Underlies Biological Crystal Growth from an Amorphous Precursor Phase. Adv. Funct. Mater. 2014, 24, 5420−5426. (12) Reeder, R. J.; Tang, Y. Z.; Schmidt, M. P.; Kubista, L. M.; Cowan, D. F.; Phillips, B. L. Characterization of Structure in Biogenic Amorphous Calcium Carbonate: Pair Distribution Function and
ASSOCIATED CONTENT
S Supporting Information *
Detailed Experimental conditions, Cryo-SEM image, WAXS, infrared and Raman results of ACC samples, reproducibility of the pH and Ca activity measurement, pH/Ca2+ measurement of 66 nm particles, XRD and infrared of crystallized samples after transformation in solution, Rietveld refinement fitting, SEM images of the sample taken after the first and second crystallization of ACC, and stacked temperature series of I
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
Article
Chemistry of Materials Nuclear Magnetic Resonance Studies of Lobster Gastrolith. Cryst. Growth Des. 2013, 13, 1905−1914. (13) Faatz, M.; Grohn, F.; Wegner, G. Amorphous calcium carbonate: Synthesis and potential intermediate in biomineralization. Adv. Mater. 2004, 16, 996−1000. (14) Schmidt, M. P.; Ilott, A. J.; Phillips, B. L.; Reeder, R. J. Structural Changes upon Dehydration of Amorphous Calcium Carbonate. Cryst. Growth Des. 2014, 14, 938−951. (15) Ihli, J.; Wong, W. C.; Noel, E. H.; Kim, Y.-Y.; Kulak, A. N.; Christenson, H. K.; Duer, M. J.; Meldrum, F. C. Dehydration and crystallization of amorphous calcium carbonate in solution and in air. Nat. Commun. 2014, 5, 3169. (16) Ihli, J.; Kim, Y. Y.; Noel, E. H.; Meldrum, F. C. The Effect of Additives on Amorphous Calcium Carbonate (ACC): Janus Behavior in Solution and the Solid State. Adv. Funct. Mater. 2013, 23, 1575− 1585. (17) Radha, A. V.; Forbes, T. Z.; Killian, C. E.; Gilbert, P. U. P. A.; Navrotsky, A. Transformation and crystallization energetics of synthetic and biogenic amorphous calcium carbonate. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 16438−16443. (18) Politi, Y.; Batchelor, D. R.; Zaslansky, P.; Chmelka, B. F.; Weaver, J. C.; Sagi, I.; Weiner, S.; Addadi, L. Role of Magnesium Ion in the Stabilization of Biogenic Amorphous Calcium Carbonate: A Structure−Function Investigation. Chem. Mater. 2009, 22, 161−166. (19) Bentov, S.; Weil, S.; Glazer, L.; Sagi, A.; Berman, A. Stabilization of amorphous calcium carbonate by phosphate rich organic matrix proteins and by single phosphoamino acids. J. Struct. Biol. 2010, 171, 207−215. (20) Habraken, W. J. E. M. The integration of ion potentiometric measurements with chemical, structural, and morphological analysis to follow mineralization reactions in solution. Methods Enzymol. 2013, 532, 25−44. (21) Nudelman, F.; Sonmezler, E.; Bomans, P. H. H.; de With, G.; Sommerdijk, N. Stabilization of amorphous calcium carbonate by controlling its particle size. Nanoscale 2010, 2, 2436−2439. (22) Foran, E.; Weiner, S.; Fine, M. Biogenic Fish-gut Calcium Carbonate is a Stable Amorphous Phase in the Gilt-head Seabream, Sparus aurata. Sci. Rep. 2013, 3, 1700. (23) Beekman, M.; Disch, S.; Rouvimov, S.; Kasinathan, D.; Koepernik, K.; Rosner, H.; Zschack, P.; Neumann, W. S.; Johnson, D. C. Controlling Size-Induced Phase Transformations Using Chemically Designed Nanolaminates. Angew. Chem., Int. Ed. 2013, 52, 13211−13214. (24) Bardhan, R.; Hedges, L. O.; Pint, C. L.; Javey, A.; Whitelam, S.; Urban, J. J. Uncovering the intrinsic size dependence of hydriding phase transformations in nanocrystals. Nat. Mater. 2013, 12, 905−912. (25) Zhang, J.; Xu, Q.; Li, M. J.; Feng, Z. C.; Li, C. UV Raman Spectroscopic Study on TiO2. II. Effect of Nanoparticle Size on the Outer/Inner Phase Transformations. J. Phys. Chem. C 2009, 113, 1698−1704. (26) Fang, J.; You, H.; Kong, P.; Ding, B.; Song, X. Size-dependent structure transformation from amorphous phase to crystal. Appl. Phys. Lett. 2008, 92, 143111. (27) Navrotsky, A. Energetic clues to pathways to biomineralization: Precursors, clusters, and nanoparticles. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 12096−12101. (28) Bloch, L.; Kauffmann, Y.; Pokroy, B. Size Effect on the Short Range Order and the Crystallization of Nanosized Amorphous Alumina. Cryst. Growth Des. 2014, 14, 3983−3989. (29) Gebauer, D.; Volkel, A.; Colfen, H. Stable Prenucleation Calcium Carbonate Clusters. Science 2008, 322, 1819−1822. (30) Ogino, T.; Suzuki, T.; Sawada, K. The formation and transformation mechanism of calcium carbonate in water. Geochim. Cosmochim. Acta 1987, 51, 2757−2767. (31) Rodríguez-Carvajal, J. Recent developments of the program FULLPROF. In Commission on Powder Diffraction; Newsletter No. 26; International Union of Crystallography: Chester, England, 2001; pp 12−19.
(32) Jarvinen, M. Application of symmetrized harmonics expansion to correction of the preferred orientation effect. J. Appl. Crystallogr. 1993, 26, 525−531. (33) Rieger, J.; Frechen, T.; Cox, G.; Heckmann, W.; Schmidt, C.; Thieme, J. Precursor structures in the crystallization/precipitation processes of CaCO3 and control of particle formation by polyelectrolytes. Faraday Discuss. 2007, 136, 265−277. (34) Fernandez-Martinez, A.; Kalkan, B.; Clark, S. M.; Waychunas, G. A. Pressure-Induced Polyamorphism and Formation of ‘Aragonitic’ Amorphous Calcium Carbonate. Angew. Chem., Int. Ed. 2013, 52, 8354−8357. (35) Chen, S. F.; Colfen, H.; Antonietti, M.; Yu, S. H. Ethanol assisted synthesis of pure and stable amorphous calcium carbonate nanoparticles. Chem. Commun. 2013, 49, 9564−9566. (36) Weiner, S.; Levi-Kalisman, Y.; Raz, S.; Addadi, L. Biologically formed amorphous calcium carbonate. Connect. Tissue Res. 2003, 44, 214−218. (37) Gower, L. B. Biomimetic Model Systems for Investigating the Amorphous Precursor Pathway and Its Role in Biomineralization. Chem. Rev. 2008, 108, 4551−4627. (38) Clarkson, J. R.; Price, T. J.; Adams, C. J. Role of metastable phases in the spontaneous precipitation of calcium carbonate. J. Chem. Soc., Faraday Trans. 1992, 88, 243−249. (39) Habraken, W. J. E. M.; Tao, J.; Brylka, L. J.; Friedrich, H.; Bertinetti, L.; Schenk, A. S.; Verch, A.; Dmitrovic, V.; Bomans, P. H. H.; Frederik, P. M.; Laven, J.; van der Schoot, P.; Aichmayer, B.; de With, G.; DeYoreo, J. J.; Sommerdijk, N. A. J. M. Ion-association complexes unite classical and non-classical theories for the biomimetic nucleation of calcium phosphate. Nat. Commun. 2013, 4, 1507. (40) Dundon, M. L.; Mack, E. The solubility and surface energy of calcium sulfate. J. Am. Chem. Soc. 1923, 45, 2479−2485. (41) Alexader, G. B. The Effect of Particle Size on the Solubility of Amorphous Silica in Water. J. Phys. Chem. 1957, 61, 1563−1564. (42) Kaptay, G. On the size and shape dependence of the solubility of nano-particles in solutions. Int. J. Pharm. 2012, 430, 253−257. (43) Bolze, J.; Peng, B.; Dingenouts, N.; Panine, P.; Narayanan, T.; Ballauff, M. Formation and growth of amorphous colloidal CaCO3 precursor particles as detected by time-resolved SAXS. Langmuir 2002, 18, 8364−8369. (44) Forbes, T. Z.; Radha, A. V.; Navrotsky, A. The energetics of nanophase calcite. Geochim. Cosmochim. Acta 2011, 75, 7893−7905. (45) Nielsen, A. E.; Söhnel, O. Interfacial tensions electrolyte crystalaqueous solution, from nucleation data. J. Cryst. Growth 1971, 11, 233−242. (46) Rodriguez-Blanco, J. D.; Shaw, S.; Benning, L. G. The kinetics and mechanisms of amorphous calcium carbonate (ACC) crystallization to calcite, via vaterite. Nanoscale 2011, 3, 265−271. (47) Rodriguez-Blanco, J. D.; Shaw, S.; Bots, P.; Roncal-Herrero, T.; Benning, L. G. The role of pH and Mg on the stability and crystallization of amorphous calcium carbonate. J. Alloys Compd. 2012, 536 (Suppl 1), S477−S479. (48) Politi, Y.; Mahamid, J.; Goldberg, H.; Weiner, S.; Addadi, L. Asprich mollusk shell protein: in vitro experiments aimed at elucidating function in CaCO3 crystallization. CrystEngComm 2007, 9, 1171−1177. (49) Han, T. Y.-J.; Aizenberg, J. Calcium Carbonate Storage in Amorphous Form and Its Template-Induced Crystallization†. Chem. Mater. 2007, 20, 1064−1068. (50) Noel, E. H.; Kim, Y. Y.; Charnock, J. M.; Meldrum, F. C. Solid state crystallization of amorphous calcium carbonate nanoparticles leads to polymorph selectivity. CrystEngComm 2013, 15, 697−705. (51) Kim, W.-S.; Hirasawa, I.; Kim, W.-S. Polymorphic Change of Calcium Carbonate during Reaction Crystallization in a Batch Reactor. Ind. Eng. Chem. Res. 2004, 43, 2650−2657. (52) Koga, N.; Nakagoe, Y. Z.; Tanaka, H. Crystallization of amorphous calcium carbonate. Thermochim. Acta 1998, 318, 239−244. (53) Gribb, A. A.; Banfield, J. F. Particle size effects on transformation kinetics and phase stability in nanocrystalline TiO2. Am. Mineral. 1997, 82, 717−728. J
DOI: 10.1021/acs.chemmater.5b00145 Chem. Mater. XXXX, XXX, XXX−XXX
