Effect of pH on Amorphous Calcium Carbonate Structure and

Publication Date (Web): July 13, 2016. Copyright © 2016 American Chemical Society. *E-mail: [email protected]. Phone: +45 35 32 02 23 ...
1 downloads 0 Views 2MB Size
Subscriber access provided by the University of Exeter

Article

The effect of pH on amorphous calcium carbonate (ACC) structure and transformation Dominique J. Tobler, Juan Diego Rodriguez-Blanco, Henning O. Sørensen, Susan L. S. Stipp, and Knud Dideriksen Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.6b00630 • Publication Date (Web): 13 Jul 2016 Downloaded from http://pubs.acs.org on July 17, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Crystal Growth & Design is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

The effect of pH on amorphous calcium carbonate (ACC) structure and transformation Dominique J. Tobler*, Juan Diego Rodriguez Blanco, Henning O. Sørensen, Susan L.S. Stipp and Knud Dideriksen Nano-Science Center, Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen Ø, Denmark

Abstract A number of organisms produce crystalline calcium carbonate via a metastable precursor phase termed amorphous calcium carbonate (ACC). ACC also forms during production of CaCO3 for industrial purposes, e.g., paper manufacturing and synthesis of fillers for polymers. Previous studies suggest that the local structure of ACC controls crystallisation kinetics and pathways, i.e., the crystalline polymorph(s) that form(s) in the process. We used pair distribution function (PDF) analysis to provide evidence that the local structure of ACC gradually changes as the pH of the synthesis solutions is increased from 10.6 to 12.7, at ambient conditions. These changes correlate with the mole fraction of incorporated hydroxide ions, which varies gradually from negligible at pH 10.6 to 0.12 at pH 12.7. At lower pH (10.5), vaterite and calcite formed in less than 2 minutes but as pH increased, the lifetime of ACC increased and it transformed directly to calcite (i.e. no vaterite intermediate). Although higher pH led to a preferable transformation into calcite with decreasing crystal size, variations in ACC local structure cannot be linked to development of a calcite like motif, as has been suggested. Based on comparing the measured vaterite PDF to a range of structural models to determine Ca-Ca distances, we found no obvious structural similarity between vaterite and the ACC precursor either, although such analysis is complicated by the ambiguity of the vaterite structure. Presumably, the absence of vaterite and the prolonged ACC stability with increasing synthesis pH could indicate inhibition of crystal nucleation and growth by hydroxide ions. *Corresponding Author: D.J. Tobler, Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen Ø, Denmark. Email: [email protected], Phone: +45 35 32 02 23

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The effect of pH on amorphous calcium carbonate (ACC) structure and transformation Dominique J. Tobler*, Juan Diego Rodriguez Blanco, Henning O. Sørensen, Susan L.S. Stipp and Knud Dideriksen

Nano-Science Center, Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen Ø, Denmark (*[email protected])

Abstract A number of organisms produce crystalline calcium carbonate via a metastable precursor phase termed amorphous calcium carbonate (ACC). ACC also forms during production of CaCO3 for industrial purposes, e.g., paper manufacturing and synthesis of fillers for polymers. Previous studies suggest that the local structure of ACC controls crystallisation kinetics and pathways, i.e., the crystalline polymorph(s) that form(s) in the process. We used pair distribution function (PDF) analysis to provide evidence that the local structure of ACC gradually changes as the pH of the synthesis solutions is increased from 10.6 to 12.7, at ambient conditions. These changes correlate with the mole fraction of incorporated hydroxide ions, which varies gradually from negligible at pH 10.6 to 0.12 at pH 12.7. At lower pH (10.5), vaterite and calcite formed in less than 2 minutes but as pH increased, the lifetime of ACC increased and it transformed directly to calcite (i.e. no vaterite intermediate). Although higher pH led to a preferable transformation into calcite with decreasing crystal size, variations in ACC local structure cannot be linked to development of a calcite like motif, as has been suggested. Based on comparing the measured vaterite PDF to a range of structural models to determine Ca-Ca distances, we found no obvious structural similarity between vaterite and the ACC precursor either, although such analysis is complicated by the ambiguity of the vaterite structure. Presumably, the absence of vaterite and the prolonged ACC stability with increasing synthesis pH could indicate inhibition of crystal nucleation and growth by hydroxide ions.

1. Introduction Amorphous calcium carbonate (ACC) frequently forms in highly supersaturated solutions, as a precursor to crystalline CaCO3 minerals (e.g., vaterite, calcite, monohydrocalcite). If synthesised in the laboratory, by mixing CaCl2 and Na2CO3 solutions, this amorphous phase is often short lived and crystallises within minutes. However, the addition of (in)organic molecules can stabilize ACC through the inhibition of ACC dehydration and the formation of crystallization nuclei.1-6 This also explains the enhanced lifetime of biogenic ACC. For example, almost all biogenic ACC contains Mg2+ because Mg2+ readily replaces Ca2+.2, 3 Mg2+ has a higher dehydration enthalpy than Ca2+, thus the retention of water in Mg2+ bearing ACC is much

ACS Paragon Plus Environment

Page 2 of 20

Page 3 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

stronger than in pure ACC, thereby enhancing its thermal stability and its lifetime in solution.3, 7 Similarly, it has been shown that citrate incorporation considerably inhibits ACC dehydration in solution and during dry heating, while citrate also acts as an inhibitor for calcite growth.5, 8 Aside from the effect of additives, ACC lifetime is controlled by the type, structural arrangement and relative quantities of the various hydrous components present in its structure (i.e., mobile vs rigid water, hydroxide ions). This has been shown by thermogravimetric analyses on dry ACC.9,

10

For example, all the mobile

water present within ACC is easily lost at temperatures below 150 °C, whereas temperatures in excess of 250 °C are required to remove rigid, less accessible water and hydroxide ions. Concomitant with the dehydration process, a structural reorganisation of the ACC has been shown using NMR analyses9 and molecular modelling,11,

12

indicating the formation of a more ordered and stable atomic network with increasing

temperature. Identical observations were made for ACC crystallisation during sea urchin spicule formation,3, 13

where hydrated ACC was shown to transform to a less disordered, less hydrated ACC and then anhydrous

ACC, before forming a crystalline CaCO3 phase. This shows that the ACC local structure has a critical impact on its lifetime. In addition to the structural impact on ACC stabilisation, ACC that is formed slowly at pH 9 to 10 is suggested to be more stable.14, 15 These materials are proposed to form by aggregation of small ion clusters (that have been termed "prenucleation clusters") leading to two stable types of ACC, where pH dependent local structures are inherited from the ion clusters and prime the material for transformation to either vaterite or calcite. Similarly, Günther et al. (2005)16 suggested that ACC synthesised at 0 °C, from CO2 and solutions saturated with Ca(OH)2, possessed a calcitic motif based on data from extended X-ray absorption fine structure spectroscopy and the observation of preferential transformation to calcite. They did not report the pH during synthesis but earlier studies using similar methods showed a rapid pH decrease from ~12.3 to ~9 during ACC precipitation by CO2 injection.17 Thus, this ACC could have components that resulted from growth both at higher and lower pH. Furthermore, depending on the solution pH when CO2 injection stopped, calcite could form exclusively (higher pH) or form along with vaterite (lower pH). This could indicate a link between pH, ACC local structure and the crystal structure of the end product. Here, we focussed on formation and crystallisation of rapidly formed ACC in the pH range from 10.6 to 12.7. For this, we monitored the composition, local structure and lifetime of rapidly formed ACC as a function of the concentration of the added base, [NaOH], at ambient conditions and determined the crystallisation pathway following ACC breakdown. In situ spectrophotometric analyses combined with Xray powder diffraction and high resolution electron microscopy provided quantitative data on the role of [NaOH] in ACC lifetime and its crystallisation kinetics. Thermogravimetric and elemental analyses gave information about changes in ACC hydrous state (i.e., type and distribution of hydrous components), while high energy X-ray scattering and pair distribution function analyses permitted identification of ACC

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

structural changes with increasing [NaOH]. The combined results highlight a strong impact of hydroxide on ACC formation and crystallisation in solutions with pH >10 at ambient conditions. This is highly relevant for biomineral formation and also important to consider for industrial processes, e.g., coatings and fillers, scaling of pipes and desalination membranes.

2. Methods 2.1 CaCO3 synthesis and crystallisation A 13 mM CaCl2 solution and a number of 13 mM Na2CO3 solutions with added NaOH (ranging from 0.01 to 0.16 M) were prepared using reagent grade chemicals and ultrapure deionised water (MilliQ, resistivity > 18 MΩ cm-1). The pH of the carbonate solution was 11.2 while the NaOH containing carbonate solutions had a pH of 11.8, 12.5, 12.75 and 13 (±0.03) following the addition of 0.01, 0.04, 0.08 and 0.16 M NaOH. ACC synthesis and all crystallisation experiments were conducted at 25 °C. The kinetics of ACC formation and crystallisation in solution at different initial pH, i.e., with varying [NaOH], was investigated by monitoring the time dependent change in the solution turbidity using UV-Vis spectrophotometry following the method described in previous work.5, 18, 19 This was done by adding the CaCl2 solution to the Na2CO3 (± NaOH) solution inside a plastic cuvette placed in the spectrophotometer (STS-UV, OceanOptics). During the measurement, the solution was continuously stirred and the absorbance (at 450 nm) was monitored at 1 second intervals. In replica experiments, cuvettes were removed from the setup at regular time steps to determine the nature of the precipitate by using powder X-ray diffraction (XRD) and scanning electron microscopy (SEM). To collect the solid portion, the samples were immediately filtered (0.2 um polycarbonate filters) using vacuum filtration, the solid rinsed with isopropanol to remove remaining water and then quickly dried by blowing air over the solid.20 SEM and XRD data collected from these solids were compared with the absorbance profiles, to link changes in solution turbidity to ACC lifetime and the onset of crystallization. Dry ACC samples for composition and structure characterisation (with techniques detailed in Section 2.3) were prepared by mixing larger volumes of the calcium and carbonate solution (± NaOH) followed by vacuum filtration and fast drying as described above. For all tested solutions, this filtration process was completed within 30 s of mixing. For comparative structural analyses, two ACC samples were also synthesised following Methods I and II, described by Michel et al. (2008)21 and a sample of Ca(OH)2 was produced by mixing equal volumes of a 50 mM CaCl2 and 50 mM NaOH solution, followed by vacuum filtration and rinsing with isopropanol. The saturation indices for ACC and calcite were determined using the hydrogeochemical code PHREEQC22 using the minteq.v4 database23,

24

and the solubility products for pure ACC25, vaterite and calcite.26 The

ACS Paragon Plus Environment

Page 4 of 20

Page 5 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

saturation index, SI, is defined here as SI= log(IAP/Ksp), with IAP and Ksp being the ion activity product and thermodynamic solubility product.

2.2 CaCO3 characterisation Pair distribution function (PDF) analyses: The local atomic structure of ACC, that is formed in the presence of a range of [NaOH], was probed using synchrotron PDF analyses. PDF analyses were conducted at beamline 11-D-B (58.6 keV, λ = 0.2114 Å) at the Advanced Photon Source, Argonne National Laboratory, USA. Dry ACC samples were measured in Kapton® capillaries for 5 minutes and an empty capillary, to serve for background correction, was measured prior to the sample. 2D data measured using a ~40 x 40 cm amorphous Si 2D detector (Perkin Elmer) were azimuthal integrated and polarisation corrected to 1D scattering patterns using the software Fit-2D.27, PDFGETX2.

29, 30

28

PDFs were obtained from the scattering patterns using

Standard data processing in PDFGETX2, which includes background subtraction,

normalization and corrections for angular dependent nonlinear detector efficiency and incoherent scattering, was conducted as described previously in Tobler et al. (2015).5 Because the reduced scattering structure function, F(Q), i.e. Q[S(Q) − 1], contained spikes at Q > 21.5 Å-1, the Fourier transform included data only to Qmax = 21.3 Å-1. This value is similar to that used in other PDF studies of ACC.10, 31, 32 Fitting of the PDFs was conducted with PDFgui33 using structural data for calcite34 and Ca(OH)2.35 For vaterite, several structural models were fitted to the data (details in Supplementary Information). The instrument dampening factor was determined to be 0.047 by refining PDFs measured on synthetic calcite with crystal size of 1-10 um, precipitated from Na2CO3 and CaCl2 and aged in solution for 1 day. The structural parameters were fitted sequentially: 1) scaling, size of coherently scattering domains (assumed spherical) and correlated atomic movement (δ2), 2) unit cell dimensions and 3) atomic isotropic displacement parameters. Optimisation of the atomic positions was attempted but it often resulted in unrealistic distance for C-O of the carbonate group. Thermogravimetric analyses (TGA) and differential scanning calorimetry (DSC): The water content of ACC was determined using TGA (Netzsch TG 209 F1 Libra), where samples were heated at a rate of 10 °C min−1 from 30 to 800 °C in a N2 atmosphere. ACC samples were also analysed using DSC (Netzsch DSC 214 Polyma) to determine the crystallization temperature for ACC transformation to calcite. For this, ACC samples were heated from 30 to 500 °C in a N2 atmosphere at the same rate as for TGA measurements (10 °C min−1). X-ray diffraction (XRD) and scanning electron microscopy (SEM): CaCO3 polymorphs were identified by XRD (Bruker D8 Advance Da Vinci, Co Κα radiation, 0.02° step-1 from 10 to 70° in 2θ at 1° min−1) and imaged by SEM (FEI Quanta 3D). The dry samples were prepared as described above. To get an estimate of

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the calcite crystallite size at the end of ACC crystallisation, the full width at half maximum (FWHM) of three crystallographic directions (012), (104) and (110) was determined in 2θ and averaged. For SEM, the sample was placed on an aluminium SEM stub with sticky carbon tape and imaged in high vacuum with a 2 - 10 kV electron beam.

3. Results and discussion 3.1 ACC crystallisation The addition of calcium solution to the carbonate solution containing NaOH (where concentration ranged from 0 to 160 mM) yielded an instantaneous drop in pH to values between 10.6 and 12.7, depending on initial [NaOH] (Table 1). The time dependent change in solution absorbance (i.e. turbidity) resulting from mixing the calcium and carbonate solutions is shown in Figure 1A. The nature of the precipitates was probed with XRD and SEM in replicate experiments that were terminated at the time of interest. Under all tested conditions, ACC formed immediately upon solution mixing, at t = 0 s, as shown by the rapid increase in turbidity (Fig. 1A). ACC formation was fastest in the solution without added NaOH (pure system), while increasing [NaOH] lowered the precipitation rate, illustrated by the decrease in total turbidity and a slight decrease in the rate of turbidity increase. This fits well with results from PHREEQC calculations, suggesting that ACC saturation indices decrease with increasing amounts of NaOH as a result of formation of Ca(OH)+ solution complexes (Table 1). The ACC Ksp used here25 was derived at pH 10 to 11 and likely does not account for any small structural and/or compositional changes that might occur in ACC formed in solutions with added NaOH. At 0 mM NaOH, the ACC lifetime (time until onset of ACC crystallisation, tcryst, Fig. 1A) was shortest and crystallization to vaterite and calcite started after 50 s. As [NaOH] increased, ACC lifetime steadily increased, reaching approximately 6 min in the experiment with 160 mM NaOH (Fig. 1B). In these systems, the onset of CaCO3 crystallisation was marked by a sudden increase in turbidity to a peak value similar to what has been observed in experiments with citrate.5 This maximum turbidity itself denotes the end of ACC transformation to crystalline CaCO3 (verified by SEM and XRD), and it was completed within 30 to 40 s under all tested conditions. Vaterite formed as an intermediate phase in the pure system (0 mM NaOH), while in systems with added [NaOH], this phase was undetectable and only calcite was observed following ACC dissolution. A decrease in turbidity followed the period of ACC transformation to crystalline CaCO3 (Fig. 1A), which is explained by the settling of larger crystals and/or aggregates (Fig. 1C) that could not stay in suspension at the chosen stirring rate. This turbidity dip is most significant for the pure system and then decreases in magnitude with increasing [NaOH]. As a result, the highest final turbidity (~1.2) was observed in systems with 160 mM NaOH (Fig. 1A). This agrees well with the observation that smaller and thereby lighter crystals formed at higher [NaOH] (Fig. 1C).

ACS Paragon Plus Environment

Page 6 of 20

Page 7 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Figure 1: A) Turbidity as a function of time after mixing two solutions of 13 mM CaCl2 and 13 mM Na2CO3, with increasing amounts of NaOH added to the Na2CO3 solution prior to mixing. The dashed lines mark the ACC lifetime, i.e., the onset of ACC transformation to crystalline CaCO3 phases (tcryst), as confirmed by XRD analyses. B) ACC lifetime as a function of initial [NaOH]. C) SEM images of calcite formed after 2 hours in the crystallization experiments. The scale bar is 1 um for all images.

Another trend we observed was that during the ACC stage (prior to tcryst, Fig. 1A), turbidity first increases to a maximum and then gradually decreases until crystallization is observed. While the timing of the maximum ACC turbidity is less affected by [NaOH], the time for the following decrease in turbidity linearly increases with increasing [NaOH]. In our previous study on citrate effects on ACC formation and crystallisation,5 we made similar observations and we suggested this decrease in turbidity could be the result of ACC dehydration prior the onset of crystalline CaCO3 formation. Settling processes could equally lead to a decrease in turbidity but this is unlikely in this setup, because the solutions are stirred well enough to keep the hydrous ACC nanoparticles in suspension. A process that we cannot entirely rule out is that ACC particles aggregate, thereby affecting turbidity profiles. This could also explain the increased fluctuations in turbidity readout seen prior to ACC crystallisation (Fig. 1A). Regardless of the process(es) that result in this decrease in turbidity prior to tcryst, increasing [NaOH] clearly delays ACC crystallisation to calcite.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 20

Table 1: Solution pH measurements and data from PHREEQC and XRD analyses. Mixed solution pH

Activity of aqueous species (x103 M)

NaOH ACC (mM) End of ACC formation crystallisation (t = 0 s) 0 10.56 10.46

CaCO3

Ca(OH)

3.3

10

11.60

11.61

40

12.15

80

12.45

160

12.74

+

Saturation index, SI

XRD (2θ)a

FWHM

sum

ACC

calcite

0.01

3.3

0.75

2.81

0.054 ± 0.017

3.3

0.13

3.4

0.75

2.81

0.061 ± 0.009

12.17

2.8

0.45

3.3

0.69

2.75

0.098 ± 0.008

12.47

2.4

0.80

3.2

0.61

2.67

0.137 ± 0.004

12.70

1.7

1.3

3.0

0.47

2.54

0.204 ± 0.025

a

FWHM measured for the three crystallographic directions (012), (104) and (110) and then averaged.

SEM observation of the crystalline CaCO3 phases showed a profound influence of [NaOH] on crystal size and morphology (Fig. 1C). In the pure system (no added NaOH), both vaterite and calcite were observed immediately after the disappearance of ACC. In solutions left for 2 hours, only calcite remained and SEM images showed rhombohedral crystals as imbricated twin clusters (Fig. 1C). In systems where NaOH was added, no vaterite was detected, not even immediately after the loss of ACC, and only calcite formed. At 10 mM NaOH, the calcite crystals were rhombohedral and twinned, as in the pure system, but the number of individual crystals within a particle was higher and the crystals were generally smaller. With further increase in [NaOH], both the particle and nanocrystal dimension further decreased, with particles becoming more elongated (Fig. 1C). The crystallite sizes, approximated by the average FWHM of three different crystallographic XRD reflections (assuming unstrained crystals, Table 1), also decreased with increasing [NaOH]. This gradual change in crystal shape and size clearly shows that calcite growth was distinctly affected by the addition of [NaOH], consistent with results on calcite growth at high pH.36 Intergrown crystals, as observed here, are frequently described in studies of the crystallization of poorly ordered precursors.5, 37-39 They are explained to be the result of spherulitic growth, a nucleation controlled process, where new nuclei continuously form on the surface of the growing spherulite but with no structural relationship between the preexisting and the new particles 40, 41. This nucleation controlled growth process is driven by the high supersaturation levels that build up as the amorphous phase dissolves, i.e., SI>2 to 3 in the CaCO3 system.42,

43

This fosters a lower nucleation barrier44 and a high driving force for crystallisation,

promoting a continuous growth front nucleation process that controls spherulitic growth. With increasing [NaOH], the particle and crystallite sizes decreased, consistent with other reports of growth inhibitor activity (e.g., citrate) resulting in smaller crystal size as a result of more frequent nucleation compared to the rate of crystal growth.5, 45

ACS Paragon Plus Environment

Page 9 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

3.2 ACC composition Thermogravimetric analyses (TGA) of dry ACC samples showed little difference in their total weight loss as a result of heating to 800 ºC (Table 2). This indicates that even if the ACC composition changed as a result of increasing [NaOH], TGA cannot provide a quantitative measure of it. Yet, the TGA and DSC profiles revealed some clear differences in ACC thermal stability as a function of [NaOH] in the formation solution (Fig. 2). The initial rapid decrease of mobile water as temperature approached ~100 °C was identical but the rate of water loss during the second major weight loss differed, as did the crystallization temperature, Tc, which increased as [NaOH] increased (arrows, Fig. 2). The fastest weight loss was observed in the pure system, where ACC transformation to calcite was complete at 162 ºC. In contrast, for ACC formed in systems with added NaOH, the weight loss was more gradual, e.g., continuing to 336 ºC for the solution containing 160 mM NaOH, before transforming to calcite. This is also shown by the exothermic peaks in the DSC profile (arrows in Fig. 2). For ACC formed in 0, 10 and 160 mM NaOH, the DSC profile shows a single peak for crystallisation, suggesting that these materials are a homogeneous solid solution. Conversely, at least two peaks develop in the DSC for ACC formed at 40 and 80 mM NaOH (pH 12.15 and 12.45, Table 1), similar to what was observed by Schmidt et al. (2014)10 for samples synthesised at pH 12.2 and 12.4. This indicates that ACC particles precipitated at intermediate NaOH concentrations have heterogeneities that result in variable crystallisation temperature. It is interesting that the low temperature peak(s) move(s) gradually to higher temperature as [NaOH] changes from 0 to 160 mM, whereas the higher temperature peak for crystallisation appears at an almost identical temperature, 335 ºC, for ACC formed with 40 to 160 mM NaOH. This temperature is very similar to that observed in several studies for ACC precipitated at higher pH.10, 31, 46, 47 The reasons for the heterogeneities indicated by DSC for 40 and 80 mM ACC are unknown but it is striking that it occurs over a relatively narrow pH interval, 12.15 to 12.45, both here and in Schmidt et al. (2014).10 The higher thermal stability of ACC formed in solutions with higher [NaOH] has been reported in previous studies10,

46, 48

and ACC crystallization becomes increasingly exothermic with increasing [NaOH] in the

synthesis solution. Schmidt et al. (2014)10 argued that this different behaviour likely results from differing populations of hydrous components, i.e., mobile water vs rigid water vs hydroxide, which then affects ACC dehydration rates and crystallization temperatures. They also showed that there is hydroxide and likely some trapped H2O (inaccessible to the atmosphere) in ACC, which persist until higher temperatures, i.e., 270 to 400 ºC. Schmidt et al. (2014)10 pointed out that such composition changes must be linked to structural differences to cause such variations in thermal stability but they could not find any evidence for this in their dataset.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

An alternative method to determine changes in ACC composition is to combust dry ACC samples (at 900 °C) and quantify the mass of the combustion products, i.e., C and H (Table 2). For the pure system, 10.0 wt% C loss corresponds to inorganic C from carbonate and 1.9 wt% H loss, to water and potentially some hydroxyl. With an increase in [NaOH] in the synthesis solution, a decrease in C (wt%) loss was observed, while the H wt% loss remained at 2.0 - 2.1, within experimental error (Table 2). We argue that such changes are consistent with a gradual replacement of carbonate by hydroxyl in ACC, with an increase in [NaOH]. Assuming the ACC molecular formula to be Ca(CO3)(1-x)(OH)2x.yH2O, we can calculate the molar units for water and hydroxyl (Table 2). The water content is less affected by the addition of NaOH. In contrast, the hydroxyl component clearly increases with [NaOH], reaching a value of about 0.12 at [NaOH] = 160 mM, which corresponds reasonably well with the H fraction associated with hydroxyls of 7 ± 3 % reported in Michel et al. (2008).21 This supports the increased thermal stability of ACC observed with TGA and DSC. Although Kojima et al. (1993)48 observed the replacement of carbonate by hydroxide ions in ACC, this was for material synthesized at pH 13.5 -14. We argue here that carbonate replacement by hydroxide already plays an important role at lower pH, i.e., lower [NaOH], leading to higher stability during heating. The negative value calculated for the OH- content in the 0 mM NaOH experiment could indicate the presence of bicarbonate in the solid. However, bicarbonate is thought to take part only in calcium carbonate nucleation reactions at near neutral pH.49 For example, titration experiments by Gebauer et al.15 showed that Ca2+ associates with the carbonate ions at pH 9-10. This is also consistent with PHREEQC calculations that show preferential bonding of Ca2+ and CO32-, rather than to HCO3-, for the 0 mM NaOH experiment (pH 10.6), i.e., the molar ratio of [CaCO3(aq)]/[CaHCO3+ (aq)] ~130. Consequently, we ascribe the negative value to analytical uncertainty.

ACS Paragon Plus Environment

Page 10 of 20

Page 11 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Figure 2: TGA and DSC data for ACC particles formed in the presence of 0 to 160 mM NaOH. TGA profiles (dotted lines) show gradual weight loss as a function of increasing temperature. DSC profiles (solid lines) show heat flow, with arrows marking the temperature where transformation to calcite occurred, i.e. the crystallization temperature (Tc).

Table 2: Data from elemental, TGA and DSC analyses (wt% = percentage weight loss). NaOH (mM)

Elemental analysis (wt%)a C

H

Molar unitsb H2O

wt%b

-

OH

c

-

H2O

OH

Tc (°C)

TGA (wt%) 30 – 450 °C

0

10.0 ± 0.38

1.88 ± 0.28

1.14

-0.011

17.1

-

162

19.2

10

9.73 ± 0.19

1.99 ± 0.13

1.18

0.020

17.6

0.6

185

18.1

40

9.35 ± 0.09

2.14 ± 0.04

1.23

0.060

18.4

1.7

231 - 333

18.5

80

9.21 ± 0.03

2.12 ± 0.04

1.17

0.088

17.8

2.5

245 - 336

18.5

160

9.05 ± 0.04

2.11 ± 0.03

1.12

0.117

17.2

3.4

335

18.1

a

Uncertainty corresponds to one standard deviation from analyses of 3 separate ACC samples, synthesised on different days. b Calculated from C and H wt% loss obtained from element analyses. c Negative value likely resulting from large uncertainty in C & H analyses for the 0 mM NaOH experiment

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3.3 ACC atomic structure Synchrotron X-ray scattering of ACC formed at various [NaOH] produced very similar scattering F(Q) (=Q[S(Q) – 1]) and PDF patterns (Fig. 3A and B) and they match those reported in previous studies on ACC.5, 10, 21, 31, 32 In the ACC PDFs, the well defined peaks at 1.3 and 2.4 Å stem from the C-O distance in carbonate and the first neighbour Ca-O atom pairs, while the broader peaks at ~4 and 6 Å largely reflect atom pairs containing the more electron dense Ca, i.e., in Ca-Ca, Ca-O and Ca-C.21 However, the patterns show some variation with [NaOH] (all profiles were normalized to have matching intensity for the Ca-O peak at 2.4 Å). The most pronounced variation is at ~3.7 Å, where a peak forms and grows as a result of increasing [NaOH] (henceforth termed the ∆ ACC peak), whereas the intensity of the peak at r = 4.1 Å decreases slightly (Fig. 3C). In agreement with variations in the PDFs, small, gradual changes are also observed in the F(Q) pattern where the intensity of the peak at Q = 5.5 Å-1 varies. Detailed inspection of the data in Schmidt et al. (2014) also reveals pH dependent variation in the PDF at ~3.7 Å and the F(Q) at Q = 5.5 Å-1. Furthermore, the PDFs measured by Michel et al.21 and for our ACC synthesised with 0.3 M NaOH also features a peak at ~3.7 Å. To quantify the intensity variations between the scattering patterns resulting from varying [NaOH], differential PDFs were calculated by subtracting the PDF of the sample produced without NaOH from the ones obtained on ACC synthesised in the presence of NaOH, after intensity normalisation of the patterns according to the intensity for the 2.4 Å peak. This normalisation assumes that the coordination number for Ca does not differ between the samples in any significant way. This is a reasonable assumption given that the position and shape of the major Ca-O peak is virtually unaffected by changes in [NaOH]. Subsequently, the

∆ ACC peaks in the differential PDFs were fitted with Gaussian functions to obtain peak positions and intensities, heights and widths (Fig. S2). The derived peak heights increase with the OH content from the elemental analysis (Fig. 3D), consistent with structural changes that depend on the hydroxyl content of the ACC. Furthermore, the full width at half maximum is comparable with that of the Ca-O peak in calcite, indicating that the ∆ ACC peak represents a well defined interatomic distance. To probe if the magnitude of the ∆ ACC peak depended on the Q value chosen for truncation during Fourier transform, the PDFs were also generated using a maximum Q of 24 Å-1. The resulting PDFs were more noisy because of the spikes in the F(Q), but the same trend, i.e. increasing ∆ ACC peak with increasing [NaOH], is observed (Fig. 3D; empty symbols). Inspection of the interatomic distances predicted by structural models of calcium carbonates shows that the ∆ ACC peak is positioned within the range of r values, where the crystalline solids, calcite, vaterite and monohydrocalcite, have electron dense pairs of first neighbour Ca-Ca, as well as second neighbour Ca-O from correlated positions of Ca and the more distant O atoms of carbonate. This means that the ∆ ACC peak

ACS Paragon Plus Environment

Page 12 of 20

Page 13 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

could have contributions from both types of pairs. To produce well defined, correlated positions of second neighbour Ca-O however, the O atoms would have to be held in place by a structure that also includes two or more rigidly positioned, electron dense, Ca atoms. Because no other peaks appear in the ACC as [NaOH] increases, we conclude that the intensity of the ∆ ACC peak must have significant contributions from Ca-Ca pairs. One potential explanation for the observed changes in the PDFs with increasing [NaOH] could be that smaller amounts of Ca hydroxide form, either as a material with a crystal size of only 2-3 nanometres or as an amorphous material. However, PHREEQC calculations indicate that Ca(OH)2 (portlandite) is undersaturated prior to ACC formation at all mixing ratios of the calcium- and carbonate-bearing solution, even at the highest [NaOH] (SI=-0.09 to -1.1 for ACC showing the ∆ ACC peak; Fig. 4S). The calculated solubility in PHREEQC is for a crystalline solid and we would expect the solubility for a potentially nanocrystallline or amorphous Ca(OH)2 to be lower. Furthermore, the second most intense Bragg reflection for portlandite, positioned at Q = 1.28 Å-1, is absent in the F(Q) for the ACC. Finally, the position of the ∆ ACC peak at ~3.7 Å is slightly shifted compared to the 3.6 Å Ca-Ca distance in the synthesised portlandite, derived from structural fitting of the measured PDF (Fig. 3B). The portlandite PDF also shows peaks at high r values, and these we do not see in the ACC PDFs. Thus, we conclude that the ∆ ACC peak cannot be explained by the presence of Ca-hydroxide, but rather by an increase in hydroxyl groups incorporated into the ACC, in agreement with elemental analyses and TGA. Given that the r value of the ∆ ACC peak is similar to the Ca-Ca distance in portlandite, we propose that it reflects the interatomic distance of two edge sharing Ca atoms bridged by at least one hydroxyl, which would resemble the structural motif in portlandite. Based on the structure by Swainson,50 a similar motif is also present in monohydrocalcite, where two Ca atoms sharing an oxygen in a carbonate ion and in water give rise to Ca-Ca distances of 3.85 Å. In monohydrocalcite, the carbonate group coordinates bidentately to Ca, whereas nuclear magnetic resonance spectroscopy indicates that such coordination is uncommon in ACC formed at higher pH.21 Thus, we propose that the structural motif that gives rise to the ∆ ACC peak reflects edge sharing Ca atoms bridged by hydroxyl and monodentately coordinated carbonate.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3: A) F(Q) = Q[S(Q) − 1] and B) PDFs measured for ACC formed at various [NaOH] and using the two high pH synthesis methods reported by Michel et al. (2008),21 along with patterns for synthesised crystalline Ca(OH)2, calcite and vaterite. C) Zoom of the PDFs for ACC formed at various [NaOH] at r values from 3.3 to 4.8 Å, where similar trends with [NaOH] are observed in two sets of replica experiments. In B and C, the dotted line represents the r value, where the ∆ ACC peak develops at higher pH; the solid, fine lines denote the Ca-Ca interatomic distances in calcite and the shaded area gives the dominant Ca-Ca distances in vaterite. D) Intensity changes of the ∆ ACC peak for the two datasets in C as a function of the hydroxyl content from elemental analysis. Changes in peak intensity were determined from fitting a Gaussian function to the peak in the differential PDFs (Fig. 5S). Full symbols represent data with maximum Q of 21.3 Å-1 and empty symbols, maximum Q of 24 Å-1.

ACS Paragon Plus Environment

Page 14 of 20

Page 15 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Recently, it has been proposed that several types of ACC exist and that some of these types have pH dependent local structure that determines the identity of their transformation product.31, 51 When aged in their synthesis solution, we observe that ACC with a clear ∆ ACC peak transforms to calcite without any sign of intermediate formation of vaterite (Section 3.1). Conversely, the ACC synthesised without added NaOH, which shows only a minor ∆ ACC peak, produced intermediate vaterite during crystallisation. These observations could indicate that the local structure of the ACC plays a role in determining the identity of the transformation product. If so, we would expect a structural similarity between the types of ACC and their transformation product, which would give rise to common features in their PDFs. In calcite, the first shell Ca to Ca distance is ~4 Å, reflecting Ca octahedra that share a corner O atom in one carbonate group and are bridged by another carbonate group. The second shell Ca is at ~5 Å and stems from Ca octahedra that are bridged by two carbonate groups of different orientation. Figure 3b shows the PDF measured for calcite aged hydrothermally at 200 °C for 27 days.52 The figure also shows the Ca-Ca distances derived from fitting the calcite structure34 to the data. Comparison of these Ca-Ca interatomic distances with the PDFs for ACC shows that the measured peak intensity decreases at the r value of ~4 Å with increasing [NaOH], whereas the one at ~5 Å is virtually unaffected by [NaOH]. Feasibly, the opposite intensity changes at r ≈ 3.7 Å and 4 Å indicate an increase in the proportion of edge sharing Ca polyhedra at the expense of corner sharing Ca. In conclusion, there is no indication for the appearance of a calcite like motif in the synthesised ACC as [NaOH] increases. The structure of the metastable CaCO3 polymorph, vaterite, is not as well understood as that of calcite, in spite of it being widely investigated.53, 54 It is generally accepted that vaterite is composed of layers of Ca separated by layers of carbonate groups oriented perpendicular to the Ca plane. Recently, it has been proposed, based on ab initio modelling, that vaterite is probably a group of structures with only minor variation in structure and stability.53 Furthermore, high resolution transmission electron microscopy (TEM) has shown that structural domains, with dimensions of a few nanometres can exist within a "single crystal" of vaterite.54 In the material, the organisation of the Ca layers of these domains matched, giving rise to a pseudoepitaxial relationship between the domains. This would modify the PDF compared with the pattern for a true single crystal or for a material composed of two separate crystalline phases. Thus, based on the TEM studies, we expect a fit of the structural models of vaterite to the measured PDF to be less than optimal, giving only a range of possible Ca-Ca distances in the material. The results of the fitting are presented in Table 3S for all published structures. Notably, the modelled structures by Wang and Becker (2009)55 and Demichelis et al. (2013, 2012)53, 56 gave markedly better fit to the data (Fig. 6S), although the weighted residual, Rw, remained rather high (0.243 and above) with the constraints imposed on the fitting. Common to all the models is that the Ca-Ca distance across layers is ~ 4.2

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Å, reflecting Ca polyhedra that share an oxygen and are bridged by a carbonate. Within the Ca layer, several bonding environments exist in the better fitting modelled structures: octahedra sharing edges have Ca-Ca distances of 3.7 to 4.1 Å, those that share an O and are bridged by carbonate have Ca-Ca distances of 4.0 to 4.3 Å, and those that are bridged by two carbonate groups are separated by 4.4 to 4.5 Å. Calculations of the contribution of the variably spaced Ca-Ca atomic pairs to the PDF (Fig. 6S) shows that pairs at ~3.9 and at ~4.2 are particularly abundant and that more distant shells of Ca-Ca are located at ~6 Å. Comparison of these distances with the ACC PDFs (grey shaded areas in Figure 3B and C) shows that most Ca-Ca distances in the modelled structures are placed at r values where little or no change occurs to the ACC PDF as synthesis pH changes. Thus, we conclude that the most common structural motifs in the fitted vaterite structures result in Ca-Ca distances that differ from the r values where we observe pH dependent changes in the ACC PDF. However, the less abundant motifs found in the model of Wang and Becker (2009)55 and the Cc and P3121 models of Demichelis et al. (2012, 2013)53,

56

have Ca-Ca distances of about 4.1 Å. These distances

correspond to a higher peak intensity in the PDF for the low pH, vaterite forming ACC. Thus, we cannot fully exclude that these ACC variants might be structurally primed for vaterite formation.

4. Conclusions The presence of NaOH during CaCO3 crystallisation in highly supersaturated solutions (Table 1) results in the formation of ACC with composition and structure that is different than ACC formed from a pure system at 10.6 pH. With increased [NaOH], i.e. pH, the ACC lifetime before transformation to a crystalline phase is extended (Fig. 1) and thermal stability during dry heating increases (Fig. 2). This is linked to an increase in hydroxide incorporation into ACC (Table 2), which is mirrored in atomic scale changes in ACC structure (Fig. 3). ACC transformation to crystalline CaCO3 polymorphs is similarly affected by increased NaOH. Vaterite forms as intermediate phase prior calcite crystallization when NaOH is absent, whereas ACC directly transforms to calcite in solutions with high NaOH concentration. Variations seen in ACC atomic scale structure cannot be linked directly to this preferred transformation to calcite. Instead, ACC variants formed in solutions with increasing [NaOH] are structurally more primed for vaterite formation than calcite. This suggests that ACC transformation pathways under the tested conditions are not controlled by its structural motif. Under all tested conditions, calcite crystallization occurred via spherulitic growth leading to imbricated crystal morphologies. A clear decrease in particle and crystal size correlates with increasing concentrations of NaOH (Fig. 1C), showing that increasing hydroxide presence distinctly affects calcite nucleation and growth. This in turn could be a reason why vaterite formation is inhibited. While these results are directly applicable to the understanding of early stage biomineralisation processes occurring in alkaline carbonate precipitating waters, e.g., cyanobacterial precipitation of calcite and

ACS Paragon Plus Environment

Page 16 of 20

Page 17 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

magnesite57 and microbialites,58 these findings contribute to optimising industrial CaCO3 synthesis in alkaline solutions (e.g., using dissolved Ca(OH)2).

Supporting Information The Supporting information is available free of charge on the ACS Publications website at DOI: Additional table (3S) that details the structures used for vaterite data fitting; 3 additional figures that show (4S) saturation indices for Ca(OH)2 as a function of NaOH content and CaCl2 and Na2CO3 mixing ratio, (5S) differential PDFs for ACC synthesised with/without NaOH, and (6S) observed and fitted PDFs for vaterite.

Acknowledgements We thank Heloisa N. Bordallo and Marianne Lund Jensen for access to and support with TGA/DSC measurements, Birgitta Kegel for elemental analyses and Dina A. Belova and Logan N. Schultz for providing synthetic calcite. We are very grateful for the assistance provided by Karina Chapman, Peter Chupas, Rick Spence and Kevin A. Beyer at APS beamline 11 ID-B. This work was supported by the Engineering and Physical Sciences Research Council [Grant number EP/I001514/1] through the program grant for the Materials Interface with Biology (MIB) Consortium. Use of the Advanced Photon Source was supported by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. Support for travel to the synchrotron facilities came from the Danish Council for Independent Research (via DANSCATT). JDRB and DJT also acknowledge financial support by the NanoCArB (PIEF-GA-2013-624016) and MIRO (PIEF-GA-2013-624619) Marie Curie Intra-European Fellowship (IEF), respectively.

References (1) Addadi, L.; Raz, S.; Weiner, S. Adv. Mater. 2003, 15, 959-970. (2) Loste, E.; Wilson, R. M.; Seshadri, R.; Meldrum, F. C. J. Cryst. Growth 2003, 254, 206-218. (3) Politi, Y.; Batchelor, D. R.; Zaslansky, P.; Chmelka, B. F.; Weaver, J. C.; Sagi, I.; Weiner, S.; Addadi, L. Chem. Mater. 2010, 22, 161-166. (4) Gebauer, D.; Coelfen, H.; Verch, A.; Antonietti, M. Adv. Mater. 2009, 21, 435-439. (5) Tobler, D. J.; Rodriguez-Blanco, J. D.; Dideriksen, K.; Bovet, N.; Sand, K. K.; Stipp, S. L. S. Adv. Funct. Mater. 2015, 25, 3081-3090. (6) Ihli, J.; Kim, Y.-Y.; Noel, E. H.; Meldrum, F. C. Adv. Funct. Mater. 2013, 23, 1575-1585. (7) Rodriguez-Blanco, J. D.; Shaw, S.; Bots, P.; Roncal-Herrero, T.; Benning, L. G. Geochim. Cosmochim. Acta 2014, 127, 204-220. (8) Akiva-Tal, A.; Kababya, S.; Balazs, Y. S.; Glazer, L.; Berman, A.; Sagi, A.; Schmidt, A. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 14763-14768. (9) Ihli, J.; Wong, W. C.; Noel, E. H.; Kim, Y.-Y.; Kulak, A. N.; Christenson, H. K.; Duer, M. J.; Meldrum, F. C. Nat. Commun. 2014, 5, 3169.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(10) Schmidt, M. P.; Ilott, A. J.; Phillips, B. L.; Reeder, R. J. Cryst. Growth Des. 2014, 14, 938-951. (11) Saharay, M.; Yazaydin, A. O.; Kirkpatrick, R. J. J. Phys. Chem. B 2013, 117, 3328-3336. (12) Reeder, R. J.; Goodwin, A. L.; Michel, F. M.; Phillips, B. L.; Keen, D. A.; Dove, M. T. Geochim. Cosmochim. Acta 2010, 74, A855-A855. (13) Politi, Y.; Metzler, R. A.; Abrecht, M.; Gilbert, B.; Wilt, F. H.; Sagi, I.; Addadi, L.; Weiner, S.; Gilbert, P. U. P. A. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 17362-17366. (14) Gebauer, D.; Gunawidjaja, P. N.; Ko, J. Y. P.; Bacsik, Z.; Aziz, B.; Liu, L.; Hu, Y.; Bergstrom, L.; Tai, C.-W.; Sham, T.-K.; Eden, M.; Hedin, N. Angew. Chem. Int. Ed. 2010, 49, 8889-8891. (15) Gebauer, D.; Volkel, A.; Colfen, H. Science 2008, 322, 1819-1822. (16) Günther, C.; Becker, A.; Wolf, G.; Epple, M. Z. anorg. allg. Chem. 2005, 631, 2830-2835. (17) Matsushita, I.; Hamada, Y.; Moriga, T.; Ashida, T.; Nakabayashi, I. J. Ceram. Soc. Jpn. 1996, 104, 1081-1084. (18) Rodriguez-Blanco, J. D.; Vallina, B.; Blanco, J. A.; Benning, L. G. Mineral. Mag. 2014, 78, 13731380. (19) Tobler, D. J.; Rodriguez-Blanco, J. D.; Dideriksen, K.; Sand, K. K.; Bovet, N.; Benning, L. G.; Stipp, S. L. S. Procedia Earth Planet. Sci. 2014, 10, 143-148. (20) Rodriguez-Blanco, J. D.; Shaw, S.; Benning, L. G. Mineral. Mag. 2008, 72, 283-286. (21) Michel, F. M.; MacDonald, J.; Feng, J.; Phillips, B. L.; Ehm, L.; Tarabrella, C.; Parise, J. B.; Reeder, R. J. Chem. Mater. 2008, 20, 4720-4728. (22) Parkhurst, D. L. User’s guide to PHREEQC–A computerprogram for speciation, reaction-path, advective-transport, andinverse geochemical calculations; US Geological Survey Water-Resources Investigations, 1995. (23) Herndon, V. MINTEQA2/PRODEFA2, A Geochemical Assessment Model for Environmental Systems: User Manual Supplement for Version 4.0; In ed.; HydroGeoLogic, Inc. & Allison Geoscience Consultants, Inc.: Athens, Georgia, USA, 1998. (24) Allison, J. D.; Brown, D. S.; and Novo-Gradac, K. J. MINTEQA2/PRODEFA2—A geochemical assessment model for environmental systems—Version 3.0 user’s manual. In ed.; Athens, Georgia, Environmental Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency, 1991. (25) Brecevic, L.; Nielsen, A. E. J. Cryst. Growth 1989, 98, 504-510. (26) Plummer, L. N.; Busenberg, E. Geochim. Cosmochim. Acta 1982, 46, 1011-1040. (27) Hammersley, A. P.; Svensson, S. O.; Thompson, A. Nucl. Instrum. Methods Phys. Res., Sect. A 1994, 346, 312-321. (28) Hammersley, A. P. FIT2D: An Introduction and Overview. In ed.; ESRF Internal Report, ESRF97HA02T: 1997. (29) Chupas, P. J.; Qiu, X. Y.; Hanson, J. C.; Lee, P. L.; Grey, C. P.; Billinge, S. J. L. J. Appl. Crystallogr. 2003, 36, 1342-1347. (30) Qiu, X.; Thompson, J. W.; Billinge, S. J. L. J. Appl. Crystallogr. 2004, 37, 678. (31) Radha, A. V.; Fernandez-Martinez, A.; Hu, Y. D.; Jun, Y. S.; Waychunas, G. A.; Navrotsky, A. Geochim. Cosmochim. Acta 2012, 90, 83-95. (32) Wallace, A. F.; Hedges, L. O.; Fernandez-Martinez, A.; Raiteri, P.; Gale, J. D.; Waychunas, G. A.; Whitelam, S.; Banfield, J. F.; De Yoreo, J. J. Science 2013, 341, 885-889. (33) Farrow, C. L.; Juhas, P.; Liu, J. W.; Bryndin, D.; Bozin, E. S.; Bloch, J.; Proffen, T.; Billinge, S. J. L. J. Phys.: Condens. Matter 2007, 19, (33). (34) Graf, D. L. Am. Mineral. 1961, 46, 1283-1316. (35) Henderson, D. M.; Gutowsky, H. S. Am. Mineral. 1962, 47, 1231-1251. (36) Ruiz-Agudo, E.; Putnis, C. V.; Rodriguez-Navarro, C.; Putnis, A. Geochim. Cosmochim. Acta 2011, 75, 284-296. (37) Vallina, B.; Rodriguez-Blanco, J. D.; Brown, A. P.; Blanco, J. A.; Benning, L. G. Nanoscale 2015, 7, 12166-12179. (38) Dideriksen, K.; Frandsen, C.; Bovet, N.; Wallace, A. F.; Sel, O.; Arbour, T.; Navrotsky, A.; De Yoreo, J. J.; Banfield, J. F. Geochim. Cosmochim. Acta 2015, 164, 94-109. (39) Zhong, C.; Chu, C. C. Cryst. Growth Des. 2010, 10, 5043-5049.

ACS Paragon Plus Environment

Page 18 of 20

Page 19 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

(40) Andreassen, J. P. J. Cryst. Growth 2005, 274, 256-264. (41) Granasy, L.; Pusztai, T.; Tegze, G.; Warren, J. A.; Douglas, J. F. Phys. Rev. E 2005, 72. (42) Andreassen, J. P.; Flaten, E. M.; Beck, R.; Lewis, A. E. Chem. Eng. Res. Des. 2010, 88, 1163-1168. (43) Beck, R.; Andreassen, J. P. Cryst. Growth Des. 2010, 10, 2934-2947. (44) De Yoreo, J. J.; Gilbert, P. U. P. A.; Sommerdijk, N. A. J. M.; Penn, R. L.; Whitelam, S.; Joester, D.; Zhang, H.; Rimer, J. D.; Navrotsky, A.; Banfield, J. F.; Wallace, A. F.; Michel, F. M.; Meldrum, F. C.; Coelfen, H.; Dove, P. M. Science 2015, 349, 498. (45) Jensen, A.; Ibsen, C.; Sutherland, D.; Birkedal, H. Cryst. Growth Des. 2014, 14, 6343-6349. (46) Koga, N.; Nakagoe, Y. Z.; Tanaka, H. Thermochim. Acta 1998, 318, (1-2), 239-244. (47) Radha, A. V.; Forbes, T. Z.; Killian, C. E.; Gilbert, P. U. P. A.; Navrotsky, A. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, (38), 16438-16443. (48) Kojima, Y.; Kawanobe, A.; Yasue, T.; Arai, Y., Nippon Seramikkusu Kyokai Gakujutsu RonbunshiJ. Ceram. Soc. Jpn. 1993, 101, 1145-1152. (49) Bewernitz, M. A.; Gebauer, D.; Long, J.; Coelfen, H.; Gower, L. B. Faraday Discuss. 2012, 159, 291-312. (50) Swainson, I. P. Am. Mineral. 2008, 93, 1014-1018. (51) Gebauer, D.; Kellermeier, M.; Gale, J. D.; Bergstrom, L.; Coelfen, H. Chem. Soc. Rev. 2014, 43, 2348-2371. (52) Schultz, L. N.; Dideriksen, K.; Lakshtanov, L.; Hakim, S. S.; Muter, D.; Hausser, F.; Bechgaard, K.; Stipp, S. L. S. Cryst. Growth Des. 2014, 14, 552-558. (53) Demichelis, R.; Raiteri, P.; Gale, J. D.; Dovesi, R. Cryst. Growth Des. 2013, 13, 2247-2251. (54) Kabalah-Amitai, L.; Mayzel, B.; Kauffmann, Y.; Fitch, A. N.; Bloch, L.; Gilbert, P.; Pokroy, B. Science 2013, 340, 454-457. (55) Wang, J. W.; Becker, U. Am. Mineral. 2009, 94, 380-386. (56) Demichelis, R.; Raiteri, P.; Gale, J. D.; Dovesi, R. Crystengcomm 2012, 14, 44-47. (57) Thompson, J. B.; Ferris, F. G. Geol. 1990, 18, 995-998. (58) López-García, P.; Kazmierczak, J.; Benzerara, K.; Kempe, S.; Guyot, F.; Moreira, D. Extremophiles 2005, 9, 263-274.

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Table of Contents Use Only.

Title: The effect of pH on amorphous calcium carbonate (ACC) structure and transformation Authors: Dominique J. Tobler*, Juan Diego Rodriguez Blanco, Henning O. Sørensen, Susan L.S. Stipp and Knud Dideriksen

Synopsis: With increasing NaOH concentrations (i.e., pH), the hydroxide content in precipitated amorphous calcium carbonate (ACC) increases, thereby increasing its lifetime and thermal stability, and it promotes direct ACC crystallization to calcite. Furthermore, observed changes in ACC local atomic structure with increasing hydroxide content indicate that ACC is not structurally primed for crystallization to one or the other crystalline CaCO3 phase.

ACS Paragon Plus Environment

Page 20 of 20