Effect of Hydrogel Matrices on Calcite Crystal ... - ACS Publications

Apr 1, 2015 - The organic/inorganic hybrid nanocomposite structure of biominerals confers the enhanced mechanical properties in comparison to inorgani...
0 downloads 0 Views 5MB Size
Subscriber access provided by SUNY DOWNSTATE

Article

The effect of hydrogel matrices on calcite crystal growth morphology, aggregate formation, and co-orientation in biomimetic experiments and biomineralization environments Fitriana Nindiyasari, Andreas Ziegler, Erika Griesshaber, Lurdes Fernández Díaz, Julia Huber, Paul Walter, and Wolfgang w. Schmahl Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/cg5018483 • Publication Date (Web): 01 Apr 2015 Downloaded from http://pubs.acs.org on April 16, 2015

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 48

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 hydrogel matrices on calcite crystal growth morphology, aggregate formation, and co-orientation in biomimetic experiments and biomineralization environments Fitriana Nindiyasaria*, Andreas Zieglerb, Erika Griesshabera, Lurdes Fernández-Díazc,d*, Julia Huberb, Paul Waltherb Wolfgang W. Schmahla a

Department für Geo- und Umweltwissenschaften and GeoBioCenter, Ludwig-Maximilians-Universität, 80333 Munich, Germany. b Central Facility for Electron Microscopy, University of Ulm, Ulm, Germany. c Departamento de Cristalografía y Mineralogía, Universidad Complutense de Madrid, 28040 Madrid, Spain. d Instituto de Geociencias (UCM, CSIC). C/ José Antonio Novais 2, 28040 Madrid, Spain.

ABSTRACT We investigate the effect of gelatin, agarose and silica hydrogel with and without magnesium in the growth medium, on calcite single crystal growth and aggregate formation. We characterize the hydrogel and the mineral by cryo-SEM, high-resolution SEM and electron backscatter diffraction (EBSD). We image the pristine hydrogel fabric and the fabric of hydrogel incorporated into the mineral. We visualize the hydrogel-mineral interface and investigate the effect of the hydrogels on calcite micro- and mesostructure in the gel/calcite composits. We compare hydrogel fabrics in biomimetic hybrid composits with proteinaceous matrices and networks in biological carbonate tissues of bivalves, gastropods, brachiopods and corraline red algae. In Mg-free environments, silica gel has very little effect on crystal morphology and arrangement; the gel/calcite composite that forms is a single gradient mesocrystal. Agarose and gelatin hydrogels influence mineral organization in gel/calcite aggregates, these consist of very few subunits separated by hydrogel membranes. With Mg added to the growth medium large and small angle boundaries highly increase in number: silica gel/calcite aggregates consist of partial spherulites with mesocrystalline subentities; agarose, gelatin gel/calcite aggregates are regular spherulites, their subentities are single crystals. Thus, calcite crystal organization is influenced by accumulative split growth provoked by incorporation of magnesium. Key-words: nacre, mollusk shell, gel-growth, organic matrix-mediated, mesocrystal, hybrid composite

* To whom correspondence should be addressed. Tel: +49-890-2180-4354. Department für Geo- und Umweltwissenschaften,

Ludwig-Maximilians-Universität,

80333

[email protected] or [email protected]. ACS Paragon Plus Environment

Munich,

Germany.

Email:

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 2 of 48

INTRODUCTION Calcite is one of the two major carbonate mineral components in biological mineralized hard tissues and it is produced by numerous organisms, such as foraminifera, echinoderms, mollusks, brachiopods, isopods and decapods.1,2 In most cases the calcite crystals show a pronounced crystallographic preferred orientation in the hard tissue, however, the individual calcite crystals have a larger internal mosaic spread than calcite precipitated from solution.3-11 Biogenic carbonates incorporate Mg into their structure in proportions that can vary widely depending on the organism and the specific part of the hard tissue.7, 12-14 Mg contents in hard tissues vary between 0 and 0.23 mol % MgCO3 in planktonic foraminifera and reach values close to 10 mol % MgCO3 in brachiopods, while concentrations as high as 45 mol % MgCO3 are found in the polycrystalline matrix filling the space between the needles and the plates (10-15 mol% MgCO3) in sea urchin teeth.7, 10, 15-17 Hydrogels share characteristics with proteinaceous environments where biogenic minerals are formed.18 Important properties of hydrogels for crystal growth experiments are the possibility of fine tuning19-22 some of their characteristics such as the solid content, porosity, composition of aqueous phase in the pores23-26 and functionalization with specific active groups.27-29 This renders hydrogels as relevant model systems that mimic to some degree formation processes in biological regimes.30-31 Agarose is a linear polysaccharide extracted from marine red algae. It consists of β-1,3 linked Dgalactose and α-1,4 linked 3,6-anhydro-αL-galactose residues. The polymer chains form single and double helices that bundle together to a three-dimensional, porous network with fibrous characteristics when it is in gel form.31,32 Gelatin is a poly-peptide material derived from natural collagen through hydrolytic degradation, which breaks the triple-helix structure of collagen into single-strand molecules.33,34 Gelatin contains both acidic and basic amino acids.31,35 A variety of gelatins can be obtained depending on the origin of collagen and the production method.35 Aqueous solutions of agarose (~ 0.1% w/v) and gelatin (~ 5% w/v) form hydrogels on cooling below 27 ºC.35 The gelling process involves a helix-coil transition. Both agarose and gelatin hydrogels can be described as “reversible” or “physical” gels.36 This means that the gelling process is reversible and their networks are held together through physical interactions between biopolymer chains, including, among others, molecular entanglements, H-bonding and Coulombic forces.31,36 Silica hydrogel, on the other hand, is a “chemical” gel. It can be prepared by acidification of aqueous solutions of sodium metasilicate, the method used in this work, or by the hydrolysis of siloxanes.19 In the case of silica hydrogel, gelling is an irreversible process. It involves the formation of a three-dimensional molecular 2

ACS Paragon Plus Environment

Page 3 of 48

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

network of SiO links through the polymerization of, mainly, monosilicic acid (H4SiO4) and, to a lesser extent, other silicic species (H3SiO4- and H2SiO4).3,37 The structural network in silica hydrogels is established predominantly by covalent interactions.31,38 Both, silica and agarose hydrogels are chemically relatively inert crystal growth media due to the absence of strongly interacting chemical functionalities within their matrixes, which makes them highly suitable for the investigation of the influence of physical factors on crystal nucleation and growth.31 In contrast, it can be expected that gelatin hydrogels play a chemically active role in crystallization due to the presence of both acidic (carboxylic acid) and basic (amine) groups.31,39 Studies have revealed that agarose hydrogel-grown calcite crystals show most of the typical features of biogenic calcite crystals, such as occluded, fairly intact hydrogels.40-43 A recent study on the formation of calcite aggregates in gelatin hydrogels has shown that the aggregates incorporate gelatin in various amounts and proved that an increase in gelatin incorporation results in an increasing degree of internal structuring of the hydrogel-mineral aggregate.44 Different studies have emphasized the active role of peptides, which are constituents of gelatin, as promoters of Mg incorporation into calcite 39,45-47 In this study we complement our previous findings39,44 with the investigation of carbonate aggregate formation in all three major hydrogels: gelatin, agarose and silica. We explore calcite growth and coorientation in Mg-free and Mg-bearing hydrogel environments and investigate the influence of different hydrogel chemistries and hydrogel porosities on the internal constitution of the obtained mineral aggregates. Our aim is to record and juxtapose the influence of the different hydrogels on biomimetic calcite formation. We investigate the relation between hydrogel incorporation and calcite crystal organization from the mesoscale, over tens of micrometers-sized subunits to submillimeter crystal aggregates. We compare the influence of the different hydrogels on mineral nucleation and growth, and biopolymer distribution patterns. Finally we compare these biomimetic systems with biopolymer fabrics in modern marine carbonate biological hard tissues.

EXPERIMENTAL PROCEDURES Crystal growth. Crystallization experiments were carried out using a double diffusion system with a hydrogel column length of 120 mm and diameter of 9 mm. The two vertical branches were filled with 5 mL of aqueous solution of 0.5M CaCl2 (Sigma Aldrich) and 5 mL of aqueous solution of 0.5M Na2CO3 (Sigma Aldrich) that were brought together by counter diffusion through the porous structure 3

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 4 of 48

of the hydrogel. Three different types of hydrogels were used. Gelatin and agarose hydrogels were prepared by respectively dissolving porcine gelatin (Sigma Aldrich; Type A, Bioreagent) and agarose (Sigma Aldrich; Purified; Plant Cell Culture), in two solvents at 60 ° C: (i) water and (ii) an aqueous solution of 0.1 M MgCl2. The concentration of gelatin and agarose was 10 wt % and 1 wt %, respectively. The gelation was carried out at 4 °C during an hour. Silica hydrogel was prepared by acidification of a sodium silicate solution (Merck, sp. gr.: 1.509 g/cm3; pH = 11.2) with 1N HCl to pH = 5.5. Magnesium-bearing silica hydrogel was prepared by adding an adequate volume of a MgCl2 solution to the silica sol prior to gelation. The silica hydrogel contains ~ 96.5wt% water filling its interconnecting pores. All hydrogels were set at 15 °C for 24 hours before the reagents were poured into the vertical branches of the double diffusion system. All solutions were prepared using high purity deionized (MiliQ) water (18.2 MΩ). Experiments were conducted at 15 ºC. When experiments start, in magnesium-bearing hydrogels the concentration of magnesium within the gel column is homogeneous. As time passes by, magnesium diffuses towards the reactant deposits, leading to the progressive depletion of gel column regions that are closer to the reactant deposits. The concentration of magnesium in the central region of the gel column remains basically constant during the duration of experiments. Sanchez-Pastor et al. (2011)48 investigated the diffusion of ions (Ca2+ and CO32-) from the reservoirs through the gel and the diffusion of impurities (CrO42-) added to the gel during preparation towards the reservoirs. Modelling with PHREEQC evidenced that, although the gel column became Cr(VI) depleted near the reservoirs, the concentration of Cr(VI) basically underwent no change in the middle region of the column. We expect a similar behaviour in the case of Mg2+. The slow diffusion of Mg2+ is even enhanced as it is surrounded by six water molecules that constitute its first hydration sphere, its diffusion is extremely slow. Crystallization of CaCO3 occurred within the hydrogels. In the case of gelatin and agarose, the crystals were extracted by dissolving the slice of hydrogel containing the precipitate in hot water (60 ºC). The precipitate was then filtered through a 1-µm pore size membrane, washed 3 times with hot Milli-Q water and dried at room temperature. The whole extraction procedure took ~ 15 minutes. The crystals grown in the silica hydrogel were extracted by dissolving the slice of hydrogel in a 1 M NaOH solution during 20 minutes. Subsequently, the precipitate was thoroughly rinsed and introduced into an ultrasonic bath during 10 minutes to get rid of possible rests of silica hydrogel. Subsequently, the crystals were again carefully cleaned with Milli-Q water and dried at room temperature. 4

ACS Paragon Plus Environment

Page 5 of 48

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

Morphological characterization of crystal aggregates. Crystal aggregates grown in the three different hydrogels were selected under a binocular stereomicroscope and picked using a fine painting brush. The samples were mounted on holders and coated with gold to be studied using scanning electron microscopy and X-ray spectrometry (JEOL JSM6400, 40 kV; JEOL JSM335F, 30 kV, equipped with an energy dispersive spectrometer LINK Ex1; Oxford Instruments 80 mm2 X-Max SDD). Due to the difficulty of removing all rests of hydrogel without producing any crystal surface alteration, the surface of crystals grown in silica hydrogel occasionally appeared unclean. It has affected the quality of the SEM micrographs and has required double coating of those samples with gold and carbon, Selective etching preparation for SEM imaging of crystal/hydrogel composites. The obtained crystals grown in the hydrogel were embedded in EPON resin. The samples were first cut using a Leica Ultracut ultramicrotome with glass knifes to obtain plane surfaces within the material that were then polished with a diamond knife (Diatome) by stepwise removal of material in a series of sections with successively decreasing thicknesses (90 nm, 70 nm, 40 nm, 20 nm, 10 nm and 5 nm, each step was repeated 15 times).49 The polished crystals were etched for 90 seconds using 0.1M HEPES (pH = 6.5) containing 2.5 % glutaraldehyde as a fixation solution. The etching procedure was followed by dehydration in 100 % isopropanol 3 times for 10 seconds each, before the specimens were critical point dried in a BAL-TEC CPD 030 (Liechtenstein). The dried samples were rotary coated with 3 nm platinum and imaged using a Hitachi S5200 FE-SEM at 4 kV. Preparation of the crystals for Electron backscatter (EBSD) analysis. The crystals were embedded into EPON resin and were treated with several sequential mechanical grinding and polishing steps down to a grain size of 1 µm. The final step was etch-polishing with colloidal alumina (particle size ~ 0.06 µm) in a vibratory polisher. Information obtained from EBSD measurements is presented as band contrast images, and as colour-coded crystal orientation maps with corresponding pole figures.The EBSD band contrast gives the signal strength of the EBSD-Kikuchi diffraction pattern. The strength of the EBSD signal is high when a crystal is detected, while it is weak or absent when a polymer is scanned. A mesocrystal is a crystalline object which consists of submicrometer sized distinct crystallites that are crystallographically co-aligned (see also Song and Cölfen 2010,50 Seto et al. 2012,51 Huber et al. 201552) to better than a threshold angle . As a working definition we suggest max=10°. A mosaic 5

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 6 of 48

crystal is a crystalline object consisting of subunits which are separated from each other by small angle boundaries. These subunits (of mosaic crystals) are also crystallographically co-aligned to better than max=10°. Investigation of hydrogel porosity using high-pressure freezing and cryo-SEM. High-pressure freezing is a method that prevents or minimizes damage of the structure of a material caused by ice crystal formation.53-54 The hydrogel samples, 10 wt % gelatin and 1 wt % agarose, were prepared by dropping a small portion of the hydrogel into the cavity of aluminum planchettes, having a total diameter of 3 mm, and a central cavity 2 mm wide and 300 µm deep.53 The silica hydrogel was prepared in a Petri dish following the procedure described above. After settling, a small slice of silica hydrogel was sampled and placed into the aluminum planchettes and both, the planchettes and the hydrogels, were immersed in hexadecane solution to fill cavities between the hydrogel and the planchettes. Subsequently the sample was covered with the flat side of another aluminum planchette. In order to assess damage to the gel fabric induced by potential ice crystal formation, hydrogels were either high-pressure frozen without further treatment or after treatment with 30 % isopropanol for 30 minutes. The samples did not shrink during isopropanol treatment. High-pressure freezing was carried out in a HPF Compact 01 (Wohlwend GmbH, Sennwald, Switzerland) at 230 MPa.55-56 The pressure and freezing temperature was reached within 30 ms. Planchettes with the frozen samples were then mounted on a holder in a cryo-chamber above liquid nitrogen and cryo-transferred to a BAF 300 freeze-etching device (Baltec, Balzers, Liechtenstein). Samples were warmed up to -93° C and the sample surface was freeze-dried for 30 min by sublimation of water. Then the frozen samples were coated with 3 nm of platinum at an angle of 45° and with 20 nm of carbon at an angle of 90°. The coated frozen samples were mounted onto the Gatan cryo-holder 626 (Gatan, Inc., Pleasanton, CA, U.S.A.) and cryo-transferred to the Hitachi S5200 FE-SEM.57 The samples were investigated at a temperature of -100 °C and an accelerating voltage of 10 kV. Imaging was performed by using analysis mode and the backscattered electron signal.

RESULTS Hydrogel fabric and calcite – hydrogel aggregate formation. The pristine fabric and pore structure of the three investigated hydrogels were determined with cryo-scanning electron microscopy of highpressure frozen samples treated with and without isopropanol (Figures 1a, b, c, S1, S2). This 6

ACS Paragon Plus Environment

Page 7 of 48

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

analytical approach9,45,46-49,54 ensures that the intrinsic hydrogel structure is not destroyed by ice crystal formation, and, more importantly, the observed fabric is inherent to a specific hydrogel. In our previous study of hydrogel structure44, this prerequisite was not met. Mg-free gelatin and agarose hydrogels (Figures 1b, 1c) have a fibrous structure with highly varying pore dimensions. In comparison to agarose, the fiber structure of gelatin hydrogel is more compact, which may be related to the gel concentration (1 wt.% for agarose, 10 wt.% for gelatin). In the 2D section of the gelatin hydrogel (Figure 1c), a characteristic mesh size in the order of 60 nm is visible, with occasional pores in the 150 nm size range. Smaller mesh sizes are very frequent, but they also might result from nonequatorial sectioning of large pores. Pore dimensions in the agarose gel are 100-200 nm, i.e. significantly wider than in the gelatin gel, while the fiber thickness of agarose and gelatin hydrogels are similar . Silica hydrogel has an entirely different structure (Figures 1a, see also supplementary Figures S1 and S3). It is composed of minute (less than 20 nm) spherical particles that do not appear to form a fibrous network or a foam structure. Figures 1d to 1f show the hydrogel fabric after complete decalcification of the mineral-gel aggregate (see also supplementary Figure S4). As silica gel does not seem to alter the growth of the mineral considerably we do not observe a significant difference between the gel fabrics prior and subsequent to mineral/silica composite formation (Figures 1a, 1d). Agarose and gelatin hydrogels have an effect on mineral-gel aggregate growth. In Figures 1e and 1f we see the large voids where the CaCO3 mineral was located prior to decalcification of the composite. We observe the accumulation of fibers and formation of thin gel membranes which form where the polymer was pushed ahead by the growing CaCO3 crystals. For the solid contents used in this study, membrane formation is more pronounced for agarose hydrogel than for gelatin hydrogel. Figures 2 to 7 and supplementary Figure S5 highlight the interlinkage between the hydrogels and the calcite mineral. Figure 2 shows the rhombohedral morphology of calcite grown in Mg-free silica hydrogel. The nanometric particulate character of silica hydrogel is apparent in all SEM images (Figure 2, see also supplementary Figure S3 for higher resolution). In the Mg-free agarose/calcite composite (Figure 3) and the Mg-free gelatin/calcite composite (Figures 4, see also supplementary Figure S5) we observe a hydrogel network incorporated into the crystals as well as compact hydrogel membranes of 100 nm thickness at boundaries between crystal subunits (yellow arrows in Figure 3b and yellow stars in Figure 4b). In Mg-free agarose the intra7

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 48

crystalline hydrogel network is a very pronounced characteristic of the composite and is intimately interwoven with the mineral (Figure 3c). The incorporation of gelatin into calcite aggregates is not as uniform (Figures 4 and supplementary Figure S5). In the crystal subunits I and II in Figure 4 we observe patches of calcite connected by the delicate gelatin network (supplementary Figure S5b), while in subunit III of the same aggregate a highly pronounced network occluded between the mineral prevails (Figure S5a). A thick gelatin membrane with a thickness of around 100 nm is also present in the composite (white stars in Figure 4b) and forms the border between two subunits of the gelatin hydrogel/calcite aggregate. The presence of Mg in silica hydrogel does not influence on the fabric of Mg-bearing silica hydrogel/calcite aggregate significantly (Figure 5). On the other hand, the presence of Mg in agarose and gelatin hydrogels exerts a major influence on the structure of the composite materials (Figures 6 and 7, S6). In all investigated Mg-bearing gelatin/calcite and Mg-bearing agarose/calcite aggregates the compact hydrogel membranes are missing (Figures 6 and 7). In Mg-bearing agarose/calcite composites the hydrogel structure is delicate (Figure 6) while in Mg-bearing gelatin/calcite composites the hydrogel network is thick and more pronounced (Figure 7). Calcite that grows in Mgbearing gelatin has a blocky appearance (Figure 7b) in comparison to calcite that forms in gelatin hydrogels without the addition of Mg. In Mg-bearing agarose calcite entities have a columnar structure with rounded outer shapes (Figure 6b). Morphological characteristics of calcite crystal aggregates grown in Mg-free hydrogels. Calcite forms the main component in the precipitates of our experiments with Mg-free growth media, irrespective of the type of hydrogel that was used. However, depending on the hydrogel, we observed significant morphological differences between the calcite aggregates (Figure 8). Calcite crystals grown in Mg-free silica hydrogel are single crystals (Figure 8a, b). Their habit is dominated by strongly terraced {104} faces and variably developed rough, blocky surfaces without a well-defined orientation. Edges bounding {104} faces are sharp for those converging in the three-fold axis (Figure 8a), and are curved and rough for those not converging in the three-fold axis (Figure 8b). The high roughness of the latter edges is depicted in Figure 8c, where both, their jagged appearance and the high porosity of flat {104} surfaces become evident.

8

ACS Paragon Plus Environment

Page 9 of 48

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

Calcite that grows in Mg-free agarose hydrogel shows a wide variety of morphologies, ranging from rhombohedral single crystals (Figure 8d) to crystal aggregates (Figure 8e). Single crystals are bounded by flat {104} faces; where only those edges converging in the three-fold axis appear smooth and straight (Figure 8d). Crystal aggregates consist of numerous individuals in an apparently radial arrangement. These constituting individuals are bounded by flat {104} faces with straight edges converging in the c-axis (Figure 8e). A detailed inspection of the {104} faces reveals that the crystals are highly porous as indicated by originally hydrogel-filled pores (Figure 8f). Calcite that grows in Mg-free gelatin hydrogel appears as aggregates with very rough surfaces. The shape of these aggregates varies from approximately spherical (Figure 8g) to dumbbell-like structures (Figure 8h). They all consist of subunits with very slightly diverging crystallographic orientation. Detailed imaging of both aggregates and subunits reveals a surface structure formed of isolated submicrometer blocks or pillars (size < 1 micrometer), which are separated by incorporated hydrogel. These pillars appear to be only slightly misoriented relative to each other (Figure 8i). Morphological and compositional characteristics of calcite crystal aggregates grown in high-Mg hydrogels. The characteristics of calcite crystal aggregates that grew in Mg-bearing gelatin, agarose and silica hydrogel (Figure 9) are significantly different from aggregates that formed in Mg-free media. These differences involve both, morphological and compositional features. Calcite grown in Mg-bearing silica hydrogel appears as bundled sheaf-like aggregates. These aggregates show an equatorial cleft and either consist of flat-surfaced similarly arranged units or numerous radial, rough surfaced units (Figure 9c, d). In both cases, the units that compose the aggregates are bounded by rhombohedron faces (Figure 9b, d). Detailed inspection of the aggregates’ surface shows roughly defined sub-blocks that grow together and leave nanometric sized pores in between. EDX analyses collected on the surface of these aggregates yield Mg contents that range from 0 to 3 mol% MgCO3. Calcite crystal aggregates grown in Mg-bearing agarose hydrogel show two characteristic morphologies: sheaf-like and spherical aggregates. Figure 5e depicts a sheaf-like aggregate with a marked equatorial cleft. At the surface, this type of aggregate displays {104} terraced faces of its subunits which are sub-parallel (Figure 9f). These aggregates have sizes in the range from 50 to 100 micrometers. EDX analyses collected on the surface of these aggregates provide Mg contents that vary between 2 and 5 mol% MgCO3. Spherical aggregates show numerous sub-blocks in a radial arrangement (Figure 9g). These sub-blocks are bounded by clearly distinguishable {104} 9

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 10 of 48

rombohedron faces (Figure 9h). The size of these aggregates is in the 50-150 micrometer range. The Mg content detected by EDX on the surface of these aggregates varies between 4 and 10 mol% MgCO3. Calcite formed in Mg-bearing gelatin hydrogel grows to sphere-like aggregates with small rombohedron-like sub-blocks at the surface. Two different types of aggregates can be distinguished regarding their size and composition. The first type consists of aggregates with sizes between 150 and 200 micrometers (Figure 9i). These show ~ 0.8-1 micrometer sized sub-blocks on the surface (Figure 9j). EDX analyses performed on the surface of these aggregates yield Mg contents that range from 3 to 10 mol% MgCO3. The second type of precipitates includes aggregates with sizes < 100 micrometers (Figure 9k). Their surface consists of rhombohedron-shaped sub-blocks (Figure 9l). Mg concentrations derived from EDX measurements conducted on the surface of these aggregates vary between 13 and 28 mol% MgCO3. Characteristics of calcite crystal co-orientation patterns in mineral/hydrogel composites and their subunits. Calcite crystal co-orientation results were determined with high-resolution Electron Backscatter Diffraction (EBSD). Orientation results are presented in Figures 10 to 14, and supplementary Figure S7. Crystal orientation is shown with color-coded EBSD maps and corresponding pole figures and pole density distributions of calcite c- and a*-axes. Microstructural characteristics of the aggregates and their subunits are visualized with EBSD band contrast images (supplementary Figure S7). EBSD band contrast gives the signal strength of the EBSD-Kikuchi diffraction pattern. This signal is high when a mineral is detected, while it is weak or absent when a polymer or an amorphous phase is scanned. Thus, the band contrast highlights well the distributions of the mineral and the hydrogel into the composite, respectively. As the color-coded EBSD maps show, the crystals that formed in Mg-free silica hydrogel are single crystals with a slight orientational gradient (Figure 10a), while in Mg-free agarose hydrogels we get aggregates that comprise a few, in the case of the aggregate shown in Figure 11, five, more or less similarly oriented subunits (Figure 11a) that are separated from each other by pronounced hydrogel membranes (white arrows that point to the membrane shown in Figure 3a). Misorientation of calcite in crystals and aggregates grown in Mg-free hydrogels is presented in Figures 10, 11, and 12, respectively. In general, the increased degree of internal misorientation in gel grown crystals and aggregates shows that all occluded a certain amount of hydrogel. In calcite that is precipitated from 10

ACS Paragon Plus Environment

Page 11 of 48

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

solution (without any occluded hydrogel) we obtain an internal misorientation spread (mosaic spread) of about 0.5 degrees (Figure 10e), which more or less corresponds to our experimental resolution of +/- 0.3°. Misorientation spread in the Mg-free silica hydrogel crystal (Figures 10c, d) and in the aggregate grown in Mg-free agarose (Figure 11c, d) is up to 0.8 and 1.5 degrees, respectively. As the comparison between Figures 10f, g and 11g, h shows, misorientation spread in individual subunits of the aggregate grown in Mg-free agarose is lower than misorientation spread in the crystal grown in Mg-free silica hydrogel. The misorientation spread in the crystal grown in Mg-free silica hydrogel goes up to two degrees; we even observe a stepping in the misorientation versus distance curves (Figures 10f, g). This crystal displays a small orientational gradient as it contains submicrometer subentities that are minutely misoriented to each other and produce an almost continuous change in crystallographic orientation a crystal, which thus is a mesocrystal (see the definition of a mesocrystal in the Experimental Procedures section). In each of the two large main subunits of the aggregate that grew in Mg-free agarose hydrogel (Figures 11) calcite misorientation spread is about 0.5 degrees. We regard the aggregate that grew in Mg-free agarose hydrogel as a mosaic radial aggregate comprising a few large and highly co-oriented single crystalline subunits. The calcite aggregate that formed in Mg-free gelatin hydrogel is quite different from the crystals or aggregates that grew in silica and agarose hydrogels, respectively. This aggregate is a radial aggregate and is composed of a few distinctly sized and differently substructured major subunits (Figure 12a). The major subunits labelled 1 and 4 are typical mosaic crystals consisting of slightly misoriented mosaic blocks (see the definition of a mosaic crystal in the Experimental Procedures section). For the major subunits labelled 2 in Figure 12a the mosaic-blocks are in a fan-like arrangement with misorientations between 5 and 10 degrees (Figure 12a, f); this mosaic clearly forms a partial spherulite. The subentities of the partial spherulite are further substructured on a very fine scale, these are mesocrystals. The major subunits of the gelatin grown aggregate are separated from each other by gelatin membranes (white stars in Figure 4b). In addition to these membranes, the subentities of the major subunits contain occlusions of gelatin hydrogel, predominantly as a network (e.g. between and within the subentities of subunit 2). The density of mineralization within the gelatin gel appears to vary on the microscale. Some subunits are more mineralized (subunits I and II in Figure 4b, S5b), while others contain more of the dense, irregularly distributed gelatin hydrogel network (subunit III in Figure 4b, S5a). 11

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 12 of 48

The presence of Mg in the hydrogel strongly influences the morphology of the aggregate (Figure S6), its microstructure on several scale levels (Figure S7) as well as the mode of calcite co-orientation in the aggregate and in its subunits (Figures 13, 14). The aggregate that formed in Mg-bearing silica hydrogel is composed of three major subunits which are irregularly oriented relative to each other (Figures 13a, c, e). All these major subunits have a mosaic structure rendering them partial spherulites (Figure 13a, g, h). The subentities of these partial spherulites are mesocrystals (red arrows pointing to the slight scatter in the misorientation versus distance graph of a co-oriented unit in Figures 13g, h). Aggregates that form in Mg-bearing agarose hydrogels are cylindrically textured regular (complete) spherulites (Figures 14a, c, i). The subentities of the spherulite are assembled from highly co-oriented calcite rhombohedra and are single crystals (Figure 14g, h). The aggregate shown in Figure 14a and c has been sectioned in two positions: within its central part (a) and at the top of the aggregate (c). Depending on the position where the aggregate is cut, calcite c- and a*- axes are differently inclined to the plane of view. Where the aggregate cut at its top (Figure 14c), the calcite c-axes are about 8090 degrees out of the plane of view (Figure 14d), and the a*-axes are more or less within the plane of view (Figure 14d). Where the aggregate cut as a median section (Figure 14a) the orientation of the caxes is 50 to 90 degrees out of plane (Figure 14b) and the a*-axes are 40 to 90 degrees out of the plane of view (Figure 14b), respectively. In comparison to the aggregate grown in Mg-bearing silica hydrogel we do not see any sign of mesocrystallinity within the subentities that compose the calcite spherulites that we obtained in Mg-bearing agarose (compare Figure 13h to Figure 14f). The aggregate that formed in Mg-bearing gelatin hydrogel (Figures 14i, j, k) is also a complete spherulite with, in this cut, calcite c-axes being highly co-oriented and pointing 70 to 90 degrees out of the plane of view (Figure 14k). The a*-axes also show some systematic co-orientation; we observe a twin arrangement. As the misorientation versus distance curve shows, a mesocrystallinity in the subunits of the spherulite cannot be detected, the large angle boundaries between the subentities of the spherulite indicate a mosaic structure (Figure 14j).

DISCUSSION Hydrogel incorporation into calcite. Incorporation of silica hydrogel into calcite was reported as early as 1969.58 Recently the incorporation of both, gelatin and agarose hydrogels in amounts as high 12

ACS Paragon Plus Environment

Page 13 of 48

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

as ~4 wt% for crystals grown in gelatin (10 wt% gelatin solid content) and ~0.9 wt% for crystals grown in agarose, respectively (1 wt% agarose solid content) has been confirmed.27,31,40-44,59 Estroff’s group, following the force completion model for crystallization,60-63 stated that growth rate and hydrogel strength are the main factors that control hydrogel incorporation into crystal aggregates during growth

31,40-42

and demonstrated that high growth rates lead to high amounts of incorporated

hydrogel.41 In addition, a positive relationship has been observed between growth rate and crystallization pressure, the pressure that is exerted by a growing crystal against the hydrogel network. The ability of this network to resist the crystallization pressure without breaking or being pushed away is defined as the hydrogel strength.31,41 Strong hydrogels that resist high crystallization pressures are easily incorporated into the growing crystal while weak hydrogels are pushed away.41,6466

For a given hydrogel its strength increases with increasing hydrogel solid content.19,32-36 In

agreement with Estroff’s and coworkers’ results on calcite crystals grown in agarose, we recently found a positive correlation between the amount of hydrogel incorporated into calcite crystals and the solid content of gelatin hydrogels.44 Growth rate is also proportional to supersaturation.63-67 Consequently, high amounts of hydrogel are expected to be incorporated when crystals grow under high supersaturation conditions. In calcite/agarose composites formed under high supersaturation conditions, it was found that the amount of incorporated hydrogel corresponded to ~100% of the agarose fibers that constitute the hydrogel network.31 Calcite-hydrogels composites formed in Mg-free hydrogel media. Irrespective of the type of hydrogels used and the presence or absence of Mg in the growth medium, all calcite crystals and crystal aggregates that were obtained in this study incorporated hydrogels. The morphologies of calcite crystals and crystal aggregates obtained in Mg-free hydrogels point to them forming under high supersaturation conditions. This interpretation is supported by the rough surfaces and hopper developments of calcite aggregates grown in Mg-free gelatin hydrogels and the radial crystal aggregates with curved edges and surfaces obtained from Mg-free agarose hydrogels. Under these conditions the high growth rate effect will dominate over possible effects contributed from the incorporated hydrogels and their strength.32,33,68 The high crystal co-orientation that is a characteristic feature of silica-calcite composites can be explained according to the model proposed by Estroff and collaborators.31,40-42 The hydrogel membranes between the subunits of calcite/agarose and calcite/gelatin hydrogels are created as the hydrogel networks are pushed aside by the growing crystal. No evidence of these membranes can be 13

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 14 of 48

found in SEM images of agarose or gelatin hydrogels prior to crystallization (Figures 1b, c). Therefore, we conclude that these membranes form during crystallization, as a result of the crystal growth process itself (Figures 1e, 1f). While the incorporation of Mg-free silica hydrogel into calcite hardly seems to have any effect on any characteristics of calcite crystals, Mg-free agarose and gelatin hydrogels appear to be responsible for the development of specific microstructural features of the mineral. In the presence of agarose and gelatin hydrogels (without Mg in the growth medium) calcite aggregates consist of subunits that are separated from each other by hydrogel membranes. These are thinner in Mg-free agarose hydrogel (Figure 3e) than those formed in Mg-free gelatin hydrogel (Figures 3h, i). In the case of calcite grown in Mg-free agarose hydrogel, the subunits are few in number, they are co-oriented, and they have a very low internal misorientation spread (0.5º). The incorporated hydrogel network is evenly distributed within the subunits. On the contrary, calcite aggregates grown in Mg-free gelatin hydrogel consist of numerous subunits which are further substructured. The subunits can be very different from each other in both internal misorientation (differences can be as high 10 degrees) and in the amount of incorporated hydrogel. Under the conditions established in this work the silica hydrogel porous network is built up by coalescence of silica nanoparticles. If the hydrogel is able to deform plastically, as it is the case for gelatin and agarose hydrogels, hydrogel fibers will displace and become squeezed together in between those crystals. The hydrogel polymers hence become concentrated in areas between crystals and rearrangements occur in the structure of the network: membranes form such as those observed between calcite subunits grown in gelatin and agarose hydrogels. It stands to reason that the thickness of these membranes will depend on the density of hydrogel fibers in a particular hydrogel. The very different thickness of hydrogel membranes in calcite/agarose and calcite/gelatin composites can, thus, be explained by the different density of these two hydrogels. While the high solid concentration in gelatin (10 wt% solid) leads to a dense hydrogel, the much lower solid content of agarose gel leads to a light (1 wt% solid content) hydrogel. Since the dipolar forces involved in the co-orientation of crystal sub-blocks strongly decrease with distance, it can be expected that the formation of thick hydrogel membranes in dense hydrogels leads to less co-oriented aggregates composed of a large number of subunits. These classifications agree well with the features that we find when we compare the calcite microstructure obtained in different hydrogels: few highly co-oriented subunits in aggregates grown in the light agarose hydrogel versus numerous less co-oriented subunits in aggregates grown in the dense gelatin hydrogel. Is a brittle hydrogel locally broken, membranes will 14

ACS Paragon Plus Environment

Page 15 of 48

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

not form and a porous network will remain. The microstructure of calcite-silica composites evidences this disruption of the silica hydrogel porous network where gaps in the calcite-silica hydrogel interface can clearly be observed (supplementary Figure S3). The formation of membranes has a two-fold effect on the characteristics of the hydrogel network: on the one hand, since hydrogel strength is concentration dependent, hydrogel membranes are stronger than the rest of the hydrogel network. On the other hand, since diffusivity through a porous medium is strongly affected by pore size and pore wall thickness, the formation of hydrogel membranes locally alters diffusivity, hindering mass transfer and leading to a local decrease in supersaturation and, consequently, in growth rates. The complex balance between these two effects, one, which favors hydrogel incorporation and the other that makes it less likely, can explain the uneven distribution of hydrogel in adjacent subunits that are separated by a hydrogel membrane. In turn, this uneven hydrogel incorporation determines internal structuring, with adjoining subunits being mesocrystals or mosaic crystals, depending on the amount of incorporated hydrogel. Calcite-hydrogel composites formed in Mg-bearing hydrogel media. Calcite-hydrogel aggregates formed in the presence of Mg show major differences with respect to their counterparts grown in Mgfree hydrogels. Irrespective the hydrogel considered these differences include more complicated mineral microstructures and morphologies; higher number of subunits constituting the aggregates and larger misorientation within and between subunits are the results. These differences are less marked in aggregates that formed in Mg-bearing silica hydrogel but become important in aggregates grown in Mg-bearing agarose hydrogel and, more so, in aggregates that grew in Mg-bearing gelatin hydrogels. It is interesting to observe, that the amount of Mg incorporated into calcite follows the same trend, with the lowest MgCO3 contents (up to 3 mol%) detected in aggregates formed in silica hydrogel and the highest MgCO3 contents (up to 28 mol%) found in aggregates grown in gelatin hydrogel. It has frequently been argued that high supersaturation39 promotes the incorporation of impurities into crystals structures. Although the characteristics of all the aggregates formed in this work point to them forming under high supersaturations, we have no evidence that would allow us to assign the differences in MgCO3 contents of calcite grown in different hydrogels to differences in supersaturation during crystal growth. In a recent paper, we proposed an active role of the acidic groups of gelatin in promoting Mg2+ desolvation and facilitating its incorporation into the calcite structure to explain the very high MgCO3 contents detected in calcite-gel composites.39 The role of Mg2+ in the formation of calcite-gelatin composites is two-fold. On the one hand, there is 15

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 16 of 48

overwhelming evidence that Mg2+ significantly inhibits calcite growth.69-73 On the other hand, Mg2+ incorporation into the calcite structure substituting Ca2+ requires a tilting of the planar carbonate groups, thereby inducing structural disorder.74-75 Moreover, due to their different geometries, Mg2+ incorporates differently into kink sites on parallel 441 growth steps on calcite {104} surfaces (obtuse “+”, more open, step edges and acute “-“, less accessible, step edges). This induces misfit strain into the system.76-79 Mg2+ inhibition of the calcite growth rate could lead to a reduction in the amount of hydrogel incorporated into the calcite-hydrogel composites, an effect not supported by our observations. On the contrary, the comparison between Figure 2c and Figure 5c evidences that a similar amount of network is incorporated in silica-calcite composites, irrespective whether Mg2+ was present in the growth medium or not. A similar conclusion can be derived for calcite-agarose and calcite-gelatin composites from the comparison of Figure 3c to Figure 6c and Figure 4c to Figure 7c, which, in fact, evidence that a higher amount of network was occluded when Mg was present in the growth medium. The absence of hydrogel membranes is striking and two Mg-related effects might cause this feature: (i) possible changes in rheological properties of both hydrogels (when Mg2+ is present) and (ii) changes in crystallization pressure due to the growth inhibiting effect of Mg2+. Although the second effect more likely plays a role, a modified mechanical response of Mg-bearing gelatin and agarose hydrogels to crystallization pressure cannot be entirely disregarded. Indeed, changes in gelatin fibers structuring related to the presence of specific ions in the growth medium have been previously proposed by Tlalik et al. (2006)80 to explain differences in the characteristics of gelatin-fluorapatite nanocomposites. The lattice strain that is induced by Mg2+ incorporation into the calcite structure can play a cooperative role in hydrogel incorporation and to some degree explains the higher misorientations within and between the subunits of the aggregates formed in any of the hydrogels in the presence of Mg in the growth medium. This interpretation is in good agreement with the degree of misorientation correlating the MgCO3 content of the different aggregates. Aggregates that grew in silica hydrogel have the lowest Mg contents and show a low degree of misorientation between subunits, while aggregates that grew in gelatin hydrogel incorporated the highest amount of Mg and show the highest degree of misorientation within the aggregate. To explain the frequent small angle boundaries in Mgcalcite, in a previous paper we proposed the formation of dislocations at regular intervals that release the misfit strain associated with the differential incorporation of Mg.39 This mechanism can lead to the formation of spherulites through accumulative split growth and can, to a certain extent, explain 16

ACS Paragon Plus Environment

Page 17 of 48

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 morphological characteristics of the aggregates formed in agarose and gelatin hydrogels in the presence of Mg. Other authors have connected misorientation between crystal units in calcite aggregates grown from aqueous solution in the presence of Mg the consequence of a growth mechanism that involves the aggregation of nanocrystals bridged by small regions of magnesiumstabilized amorphous calcium carbonate (ACC).81 Confirming this mechanism as also playing a role in the morphological evolution of calcite aggregates formed in Mg-bearing hydrogels will require further investigation. The comparison with nature: Biopolymer membrane and network structures in natural skeletons. Biological hard tissues are complex hierarchical structures that consist of a large variety of design concepts that extend over several scale levels.82-84 On small structural scales biological hard tissues are composites of a compliant matrix that is reinforced by stiff components, often minerals. For example plant tissues consist of micrometer sized cellulose fibers embedded in an amorphous hemicellulose matrix,85 arthropods with a mineralized cuticle contain nanometer sized chitin fibers that are surrounded by a proteinaceous matrix86-88 and mineralized tissues such as shells and are assembled by stacks of mineral platelets or arrays of mineral fibers embedded in a proteinpolysaccharide matrix.89-92 The compliant biopolymer component of gastropod, bivalve and brachiopod shells is an insoluble silk-like protein that forms the basic matrix structure for the assemblage of soluble polyanionic proteins and calcium carbonate minerals.89-72 While the silk-like proteins form membranes and provide the scaffold, polyanionic proteins mainly control crystal nucleation, orientation and formation of carbonate crystal entities such as platelets, prisms and fibers.90-91 Figure 15 shows organic matrices in bivalve and gastropod shells that were made visible by complete decalcification of the carbonate coupled to concurrent biopolymer fixation. These membranes are a few nanometers thick and envelope aggregations of carbonate nanoparticles (Figures 16a, b). In biological hard tissues these interlamellar (interfiber) membranes are always porous (Figures 15b, d) since they have to facilitate the transfer of carbonate mineral and crystallographic orientation information from one compartment to the other.93-94 Within the matrix-membrane lined compartments, a biopolymer network of fibrils is present (Figure 15e, f, g, h). These infiltrate the space between the nanoparticles and cause the misorientation between them. For example, in a nacre tablet misorientation between crystallites is between 2 to 4 degrees.9 For carbonate biomimetic systems Nindiyasari et al. 2014 have shown that 17

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 18 of 48

gelatin is incorporated in the subunits of a carbonate-gelatin hydrogel aggregate.44 An increase in gelatin content of the hydrogel from 2.5 to 10 wt% leads to an increase in misorientation between crystallites in the subunit of the composite aggregate from 2 to 6 degrees. Organic matrix formation was investigated by Nakahara and Bevelander 196992, Weiner and Traub 198096, Fritz and Morse 199897, Addadi et al. 200698, Nudelmann et al. 200699, Checa et al. 200593, Cartwright and Checa 2007100, Checa et al. 201194, Gries et al. 201193. Cartwright and Checa 2007100 propose that organic matrix formation starts with the self-assembly of a liquid-crystalline core consisting of chitin fibers assembled through self-organization. As these fibers are not bonded closely together this fibrous structure is not strong enough to form a scaffold. According to Cartwright and Checa 2007100 proteins with an affinity for chitin cover the chitin fiber layer. A chitin fiber-polymer composite is obtained that is now strong enough to act as the supporting matrix for the carbonate hard tissue100. Hydrogels are regarded to model biogenic matrix environments due to their ability to confine space, to determine diffusion rates, and thus local concentrations or supersaturation of solutes. However, hydrogel fiber organization lacks any order, unlike cholesteric liquid phases, e.g. chitin. Figures 1, S1, S2, S4 show the fabric of gelatin and agarose hydrogels prior to composite aggregate formation and after aggregate decalcification. We observe some co-alignments of hydrogel fibers after aggregate decalcification, however, we do not find any membranes in the decalcified hydrogelcalcite composites resembling those present in carbonate hard tissues. We do not see any membrane pattern regularity in our biomimetic organic matrix (Figures 15a, c, e, f) nor a regular porosity (Figure 15b, d). Further, although we found that hydrogels induce a hierarchical organization of calcite which crystallizes in them, the obtained mineral microstructures lack the regularity of the hierarchical structures of biocalcite in biological hard tissues, and, perhaps most importantly, our biomimetic systems do not yet produce a compact solid which is able to support strong mechanical forces. Nevertheless, hydrogel growth media can be fine-tuned to control the way in which they react to crystallization pressure and to provide properties that support the formation of hybrid composites. Biological organic matrices show a much higher degree of chemical complexity than our model systems. Further research is needed using hydrogels where polypeptides and oligo- as well as polysaccharides are simultaneously present59, a situation that can be attained by combinations of different hydrogels and chitin. The organic/inorganic hybrid nanocomposite structure of biominerals confers the enhanced mechanical properties in comparison to inorganic calcium carbonate. 18

ACS Paragon Plus Environment

Page 19 of 48

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

Accordingly crystals can acquire non-intrinsic properties by hosting materials with specific functionalities and interfaces.101-102 The ability of calcite crystals to incorporate the polymeric network of hydrogels during growth has recently inspired novel biomimetic-based synthetic routes for the production of composites which comprise inorganic single crystals occluding a nanoparticlefunctionalized gel matrix.103-104

The main challenge of these synthetic routes is to achieve a

homogeneous distribution of both gel fibers and nanoparticles within the volume of single crystals. Understanding the factors that determine if the gel matrix is incorporated or pushed away during crystal growth is extremely important for the successful development of these novel synthetic approaches.

CONCLUSIONS The main feature of the mineralization environment of biological carbonate hard tissues is an extracellular-matrix that consists of a three-dimensional assembly of proteins, polysaccharides and glycoproteins.31 Hydrated networks of these biopolymers compartmentalize space and thus control morphology, functionalities, which are the template for carbonate crystal nucleation and induce misorientation between nanoscale units which assemble to biocrystals such as nacre tablets or calcite fibers. Hydrogels mimic to some extent biological extracellular matrices, since the local crystallization microenvironment in the hydrogel is distinguished from that in solution by confinement of solutes in the pores of the hydrogel network. In this study we investigated the influence of silica, agarose and gelatin hydrogels on calcite crystal assembly and co-orientation as well as mineral aggregate formation for a better understanding of both, biomimetic carbonate formation as well as the assembly of matrix-mediated carbonate biological hard tissues. An overview of the observed micro-, meso- and nanostructures of calcite that formed in the presence of silica, agarose and gelatin hydrogels, respectively, is provided in Table 1. From our study we deduce the following conclusions: 1.

While the hydrogel fabrics of agarose and gelatin hydrogels are comparable to each other, they are significantly different from that of silica hydrogel. In the case for the first two gels the fabric consists of a 3D network of fibers, whereas silica hydrogel is a three-dimensional agglomeration of nanometric spherical particles. 19

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

2.

Page 20 of 48

All three hydrogels are occluded into the growing mineral, however, when the growth medium is free of Mg silica hydrogel does not impose a marked effect on mineral co-orientation and aggregate formation. Agarose gel and especially gelatin hydrogel impose a strong influence on aggregate formation. In magnesium-free environments either single crystals or radial aggregates with very few misoriented subunits are formed.

3.

When magnesium is present in the growth medium the amount of internal boundaries in the crystal aggregates increases drastically and the number of subunits in an aggregate increases accordingly. In the case of calcite grown in silica hydrogel very few subunits form an aggregate, and the individual subunits are partial spherulites. The subcrystals of the partial spherulite are further structured and are mesocrystals.

4.

In the case of agarose and gelatin hydrogels the addition of Mg to the growth medium drastically induces the formation of small angle and large angle boundaries and we obtain in both cases complete spherulites. In a radial profile the small angle boundaries are below 10 degrees, while in tangential profiles through the spherulite grain boundary misorientations form major steps in the order of 8 to 15 degrees. We do not see mesocrystallinity in the individual subunits of these spherulites. We attribute this growth process to accumulative split growth related to Mg incorporation.

5.

In weak hydrogels such as Mg-free gelatine or agarose gels, growing crystals push gel matrix ahead of their growth front, such that gel accumulates to form membranes between crystals which nucleate in proximity and abut against each other after growth. In strong hydrogels such as Mgbearing gelatine or agarose the gel becomes more incorporated than pushed ahead of the growing crystals, and consequently the membranes between abutting crystals are less pronounced.

Acknowledgements. This research is part of the German-Spanish joint Acciones Integradas program (AIB2010 DE-0008), DAAD-50749739. F.N. is grateful for a stipendium by KAAD. E.G. is supported by Deutsche Forschungsgemeinschaft, DFG grant number GR-1235/9-1. This research was partially funded by project CGL2010-20134-C02-01 (MECC-Spain). We sincerely thank the staff of the National Microscopy Centre (ICTS) of the Complutense University (UCM) for technical support and assistance.

20

ACS Paragon Plus Environment

Page 21 of 48

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

Associated Content. Supporting Information. Cryo-FE-SEM images of high-pressure frozen gel samples showing the fabric of silica, agarose, gelatin. Cryo-FE-SEM images of fracture surfaces showing the fabrics of high-pressure frozen Mgfree hydrogels (gelatin, agarose and silica) that were not treated with isopropanol. The assemblage of silica nanospheres to a very loose 3D hydrogel network. The fabric of agarose and gelatin hydrogels after complete decalcification of the hydrogel - calcite aggregate. FE-SEM images showing magnifications of subunit III (a) and subunit I (b) presented in Figure 4b (a Mg-free gelatin - calcite aggregate). FE-SEM image of microtome cut, polished, etched and critical point dried surfaces of Mg-free and Mg-beaing hydrogel-calcite aggregates (silica, agarose and gelatin hydrogel). EBSD band contrast images that highlight internal features of the investigated single crystals and aggregates obtained from the three different hydrogels, in the presence and absence of Mg in the growth medium. This material is available free of charge via the Internet at http://pubs.acs.org.

REFERENCES 1.

Lowenstam, H.A. and Weiner, S. (Eds) (1989) On Biomineralization. Oxford University Press, Inc., New York, 324 pp 2. Mann, S. Biomineralization: Principles and concepts of in biorganic materials chemistry. Oxford University Press, New York, 2001. 3. Robach, J. S.; Stock, S. R.; Veis, A. J. Struct. Biol. 2005, 151, 18-29. 4. Robach, J. S.; Stock, S. R.; Veis, A. J. Struct. Biol. 2009, 168, 452-466. 5. Alvares, K.; Dixit, S.N., Lux, E.; Veis, A. J. Biol. Chem. 2009, 284,26149-26160. 6. Li, H.; Xin, H.L.; Kunitake, M.E.; Keene, E.C.; Muller, D.A.; Estroff, L.A. Adv. Funct. Mater. 2011, 21, 2028–2034. 7. Griesshaber, E.; Schmahl, W. W.; Neuser, R.; Pettke, T.; Blüm, M.; Mutterlose, J.; Brand, U. Am. Mineral. 2007, 92, 722-734. 8. Goetz, A. J.; Steinmetz, D. R.; Griesshaber, E.; Zaefferer, S.; Raabe, D.; Kelm, K.; Irsen, S.; Sehrbrock, A.; Schamhl, W. W. Acta Biomaterialia 2011, 7, 2237–2243. 9. Griesshaber, E.; Schmahl, W. W.; Ubhi, H. S.; Huber, J.; Nindiyasari, F.; Maier, B.; Ziegler, A. Acta Biomaterialia 2013, 9, 9492-9502. 10. Goetz, A. J.; Griesshaber, E.; Abel, R.; Fehr, Th.; Ruthensteiner, B.; Schmahl, W. W. Acta Biomaterialia 2014, 10, 3885-3898. 11. Kim, Y.-Y.; Schenk, A. S.; Ihli, J.; Kulak, A. N.; Hetherington, N. B. J.; Tang, C. C.; Schmahl, W.; Griesshaber, E.; Hyett, G.; Meldrum, F. C. Nat. Comms. 2014, 5, 1-14. 21

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

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.

Page 22 of 48

Raz, S.; Weiner, S.; Addadi, L. Adv. Mater. 2000, 12, 38-42. Erez J. Rev. Mineral. Geochem., 2003, 54, 115-149. Bentoz, S.; Erez, J. Geochem. Geophy. Geosy. 2006, 7, 1-11. Ma, Y.; Aichmayer, B.; Paris, O.; Fratzl, P.; Meibom, A.; Metzler, R. A.; Politi, Y.; Addadi, L.; Gilbert, P. U. P. A.; Weiner, S. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 6048–6053. Lenders, J. J. M.; Dey, A.; Bomans, P. H. H.; Spielmann, J.; Hendrix, M. M. R. M.; de With, G.; Meldrum F. C.; Harder, S.; Sommerdijk, N. A. J. M. J. Am. Chem. Soc. 2011, 134, 1367-1373. Carpenter, S .J.; Lohmann, K.C. Geochim. Cosmochim. Ac. 1992, 56, 1837-1849. Sancho-Thomas, M.; Fermani, S.; Durán-Olivencisa, M. A.; Otálora, F.; Gómez-Morales, J.; Falini, G.; García-Ruiz, J. M. Cryst. Growth Des. 2013, 13, 3884-3891. Henisch, H. K. Crystals in Gels and Liesegang Rings; Cambridge University Press: Cambridge, 1988. McCauley, J. W.; Roy, R. Am. Mineral. 1974, 59, 947-963. García-Ruíz, J. M.; Amorós, J. L. J. Cryst. Growth 1981, 55, 379-383. Prieto, M.; Putnis, A.; Arribas, J. ;Fernández-Díaz, L. Min. Mag. 1992, 56, 587-598. Fernández-Diaz, L.; Putnis, A.; Prieto, M.; Putnis, C. V. J. Sediment Res. 1996, 66, 482-491. Helbig, U. J. Cryst. Growth 2008, 310, 2863-2870. Falini, G.; Fermani, S.; Gazzano, M.; Ripamonti, A. Chem. Eur. J. 1997, 3, 1807-1814. Sancho-Thomas, M.; Fermani, S.; Goffredo, S.; Dubinsky, Z.; García-Ruiz, J. M.; GómezMorales, J.; Falini, G.; Cryst. Eng. Comm. 2014, 16, 12571267. Grassmann, O.; Muller, G.; Löbmann, P. Chem. Mater. 2002, 14, 4530-4535. Grassmann, O.; Neder, R. B.; Putnis, A.; Löbmann, P. Am. Mineral. 2003, 88, 647-652. Grassmann, O.; Löbmann, P. Chem. Eur. J. 2003, 9, 1310-1316. Dorvee, J.R.; Boskey, A.L.; Estroff, L.A. CrystEngComm., 2012,14, 5681-5700. Asenath-Smith, E.; Li, H.; Keene, E. C.; She, Z. W.; Estroff, L. A. Adv. Funct. Mater. 2012, 22, 2891-2914. Normand, V.; Lootens, D. L.; Amici, E.; Plucknett, K. P.; Aymard, P. Biomacromolecules 2000, 1, 730-738. Van den Buckle, A. I.; Bogdanov, B.; de Rooze, N.; Schacht, E. H.; Cornelissen, M.; Berghmans, H. Biomacromecules, 2000, 1, 31-38. Lee, K. Y.; Mooney, D. J. Chem. Rev. 2001, 101, 1869-1879. Dorvee, J. A.; Boskey, A. L.; Estroff, L.A. CrystEngComm. 2012, 14, 5681–5700. Hoffman, A. S. Advanced Drug Delivery Reviews 2002, 54, 3-12. Lefaucheux, F.; Robert, M. C. Handbook of Crystal Growth, Vol. 2, Ed: D. T. J. Hurle; Elsevier Science: Amsterdam 1994 . Warren, B. E. Chem. Rev. 1940, 26, 237-255. Nindiyasari, F.; Griesshaber, E.; Fernández-Díaz, L.; Astilleros, J. M; Sánchez-Pastor, N.; Ziegler, A.; Schmahl, W. W. Cryst. Growth Des. 2014, 14, 4790-4802. Li, H.; Estroff, L. A. CrystEngComm. 2007, 9, 1153-1155. Li, H.; Estroff, L. A. Adv. Mat. 2009, 21, 470-473. Li, H.; Fujiki, Y.; Sada, K.; Estroff, L. A. CrystEngComm. 2011, 13, 1060-1062. 22

ACS Paragon Plus Environment

Page 23 of 48

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

43. Simon, P.; Carrillo-Cabrera, W.; Huang, Y. X.; Buder, J.; Borrmann, H.; Cardoso-Gil, R., Roseeva, E.; Yarin, Y.; Zahnerty, T.; Kniep, R. Eur. J. Inorg. Chem. 2011, 35, 5370-5377. 44. Nindiyasari, F.; Fernández-Díaz, L.; Griesshaber, E.; Astilleros, J. M; Sánchez-Pastor, N.; Schmahl, W. W. Cryst. Growth Des. 2014, 14, 1531-1542. 45. Elhadj, S.; De Yoreo, J. J.; Hoyer, J. R.; Dove, P. M. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 19237. 46. Boedtker, H.; Doty, P. J. Phys. Chem. 1954, 58, 968−983. 47. Zhang, Z. K.; Li, G. Y.; Shi, B. J. Soc. Leather Technol. Chem. 2006, 90, 23−28. 48. Sánchez-Pastor, N.; Gigler, A.; Cruz, J.; Park, S.; Jordan, G.; Férnandez-Díaz, L. Cryst. Growth Des. 2011, 11, 348-389. 49. Fabritius, H.; Walther, P.; Ziegler, A. J. Struct. Biol. 2005, 2, 190-199. 50. Seto, J.; Ma, Y.; Davis, S.A.; Meldrum, F.; Gourrier, A.; Kim, Y.-Y.; Schilde, U.; Sztucki, M.; Burghammer, M.; Maltsev, S.; Jäger, C.; Cölfen, H. PNAS 2012, 109, 3699-3704. 51. Song, R.-Q.; Cölfen, H. Adv. Mater. 2010, 22, 1301-1330. 52. Huber, J.; Griesshaber, E.; Nindiyasari, F.; Schmahl, W. W.; Ziegler, A. J. Struct. Biol., in press 53. Moor, H. Riehle, U. Proceedings 4th European Reg. Conf. Electron Microsc. 1968, 2, 33-34. 54. Müller, M. Moor, H. Proceedings of the 42nd Ann. Meet. Electron Microsc. Soc. Am. (ed. By G. W: Bailey, 1984, 6-9. 55. Walther, O.; Ziegler, A. J. Microsc. 2002, 208, 3-10. 56. Walther, P. J. Microsc. 2003, 212, 34-43. 57. Walther, P. J. Microsc. 2008, 232, 379-385. 58. Nickl, H. J.; Henisch, H. K. J. Electrochem. Soc. 1969, 116, 1258-1260. 59. Huang, Y.; Buder, J.; Cardoso-Gil, R.; Prots, Y.; Carrillo-Cabrera, W.; Simon, P.; Kniep, R. Angew. Chem. Int. Ed. 2008, 47, 8280-8284. 60. Chernov, A. A.; Temkin, D. E. in Current Topics in Materials Science , Vol. 2 (Ed: E. Kaldis ), North-Holland, New York 1977, 3-77. 61. Chernov, A. A., Temkin, D.E., Melnikova, A. M. Sov. Phys. Crystallogr. 1976, 21, 369-74. 62. Chernov, A. A., Temkin, D. E., Melnikova, A. M. Sov. Phys. Crystallogr. 1977, 22, 656-58. 63. Chernov, A. A. Modern Crystallography III: Crystal Growth. Springer: Berlin, Heidelberg, New York, Tokyo, 1984. 64. Pötschke, J.; Rogge, V. J. Cryst. Growth 1989, 94, 726–738. 65. Gavira, J. A.; García-Ruiz, J. M. Acta Crystallogr. Sect. D-Biol. Crystallogr. 2002, 58, 16531656. 66. Khaimov-Mal’kov, V. I. Sov. Phys.: Crystallogr. 1958, 3, 487-493. 67. De Yoreo, J. J.; Vekilov, P. G. Rev. Mineral. Geochem. 2003, 54, 57-93. 68. Ross-Murphy S.B. Polymer 1992, 33, 2622-2627. 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

Page 24 of 48

69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80.

Davis, K. J., Dove, P. M., De Yoreo J. J. Mat. Res. Soc. Symp. 2000, 620, M9.5.1-M9.5.7. Davis, K. J., Dove, P. M., De Yoreo, J. J. Science 2000, 290, 1134–1137. Astilleros, J. M., Pina, C. M., Fernández-Díaz, L. Putnis, A. Surf. Sci. 2003, 545, 767-773. Astilleros, J. M., Fernández-Díaz, L., Putnis, A. Chem. Geol. 2010, 271, 52–58. Nielsen, L.C., De Yoreo, J. J., De Paolo, D. J. Geochim Cosmochim Ac. 2013, 115, 100-114. Bischoff, W.D., Sharma, S.K ., Mackenzie, F.T. Am. Min. 1985, 70, 581-589. Urmos, J., Sharma, S. K., Mackenzie F. T. Am. Min. 1991, 76, 641-646. Staudt, W.J.; Reeder, R.J.; Schoonen, W. A. A. Geochim. Cosmochim. Ac. 1994, 58, 2087-2098. Paquette, J., Reeder, R. J. Geochim. Cosmochim. Ac. 1995, 59, 735-749. Jordan, G., Rammensee, W. Geochim. Cosmochim. Ac. 1998, 62, 941-947. Davis, K. J.; Dove, P. M.; Wasylenki, L.E.; De Yoreo, J. J. Am.Min. 2004, 89, 714–720. Tlatlik, H.; Simon, P.; Kawska, A.; Zhan, D.; Kniep, R. Angew. Chem. Int. Ed. 2006, 45, 1905 – 1910. 81. Nishino, Y.; Oaki, Y.; Imai, H. Cryst. Growth Des. 2009, 9, 223-226. 82. Fratzl, P.; Weinkamer, R. Prog. Mater. Sci. 2007, 52, 1263-1334. 83. Dunlop, J. W. C.; Fratzl, P. Scripta Mater. 2013, 68, 8-12. 84. Bar-On, B.; Wagner, H. D. Ac. Biomater. 2013, 9, 8099-8109. 85. Burgert, I.; Fratzl, P. Phil. Trans. R. Soc. A. 2009, 367, 1541-1557. 86. Hild, S.; Neues, F.; Znidarsic, N.; Struts, J.; Epple, M.; Marti, O.; Ziegler, A. J. Struct. Biol. 2009, 168, 426-436. 87. Seidl, B.; Huemer, K.; Neues, F.; Hild, S.; Matthias, E.; Ziegler, A. J. Struct. Biol. 2011, 3, 512526. 88. Huber, J.; Fabritius, H.-O.; Griesshaber, E.; Ziegler, A. J. Struct. Biol. 2014, 188, 1-15. 89. Blank, S.; Arnoldi, M.; Khoshnavaz, S.; Treccani, L.; Kuntz, M.; Mann, K.; Grathwohl, G.; Fritz, M. J. Microscopy 2003, 112, 280-291. 90. Marin, F.; Luquet, G. C. R. Palevol. 2014, 3, 469-492. 91. Marin, F.; Luquet, F.; Marie, B.; Medakovic, D. Curr. Top. Dev. Biol. 2008, 80, 209-276. 92. Gries, K.; Heinemann, F.; Gimmick, M.; Ziegler, A.; Rosenauer, A.; Fritz, M. Cryst. Growth Des. 2011, 11, 729-734. 93. Checa, A. G.; Rodríguez-Navarro, A. B.; Esteban-Delgado, J. Biomaterials 2005, 26, 6404-6414. 94. Checa, A. G.; Cartwright, J. H. E.; Willinger, M.-G. J. Struct. Biol. 2011, 176, 330-339. 95. Nakahara, H.; Bevelander, G. Tex. Rep. Biol. Med. 1969, 101-109. 96. Weiner, S.; Traub, W. FEBS Letters 1980, 111, 311-316. 97. Fritz, M.; Morse, D. E. Curr. Opin. Cell. Biol. 1998, 3, 55-62. 98. Addadi, L.; Joester, D.; Nudelman, F.; Weiner, S. A European J. 2006, 12, 981-987. 99. Nudelman, F.; Gotliv, B. A.; Addadi, L.; Weiner, S. J. Struct. Biol. 2006, 153, 176-187. 100. Cartwright, J. H. E.; Checa, A. J. Royal Interface 2007, 4, 491-504. 101.Kim, Y. Y., Ganesan, K., Yang, P. C., Kulak, A. N., Borukhin, S., Pechook, S., Ribeiro, L., Kroger, R., Eichhorn, S. J., Armes, S. P., Pokroy, B., Meldrum, F. C. Nat. Mater. 2011, 10, 890896. 102.Estroff, L. A., Cohen, I. Nat. Mater. 2011, 10, 810-811. 24

ACS Paragon Plus Environment

Page 25 of 48

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

103.Kim, Y. Y., Schenk, A. S., Walsh, D., Kulak, A. N., Cespedes, O., Meldrum, F. C. Nanoscale 2014, 6, 852-859. 104.Liu, Y.; Yuan, W.; Shi, Y.; Chen, X.; Wang, Y; Chen, H.; Li, H. Angw. Cem. Int. Ed. 2014, 53, 4127-4131.

25

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 26 of 48

Figure Captions Figure 1. Cryo-FE-SEM images of high-pressure frozen pristine hydrogels (a, b, c) and FE-SEM micrographs of the hydrogel fabric obtained after decalcification of the gel - calcite aggregate (d, e, f). (a, d) Silica, (b, e) Agarose, (c, f) Gelatin. Silica hydrogel is an assemblage of interconnected nanometric silica spheres (a). Agarose hydrogel exhibits a network of fibers (b) with large pore sizes. Gelatin hydrogel fabric (c) is slightly denser in comparison to that of agarose gel. After complete decalcification no major changes are observed in the gel fabric for silica (d), while for agarose and gelatin we see slight accumulations of hydrogel fibers (e) around voids (e, f) where the CaCO3 crystals were located. See also supplementary images S1, S2, S3, S4. Figure 2. FE-SEM images of microtome cut, polished, etched and critical point dried surfaces showing the crystal that grew in Mg-free silica hydrogel: a rhombohedral shaped calcite single crystal (a). (b, c) The interdigitation of silica gel and calcite on two different scale levels. The assembly of nanometer sized silica spheres occluded by mineral (calcite) (c). See also Figure S3. Figure 3. FE-SEM images of surfaces of a microtome cut, polished, etched and critical point dried agarose gel/calcite aggregate showing the interlinkage of subunits in the aggregate (a). Large intertwined subunits form the aggregate and they are separated from each other by agarose membranes (yellow arrows in b). Within the subunits of the gel/calcite aggregate, agarose hydrogel is developed as an evenly distributed occluded network (yellow stars in b). Enlargement showing the internal structure of the calcite/agarose composite, where calcite crystals are intimately encased by the agarose network (c). Figure 4. FE-SEM images of a microtome cut, polished, etched and critical point dried surfaces of a Mg-free gelatin - calcite aggregate showing its morphology (a) and internal structure (b, c). Large quantities of hydrogel are incorporated into the aggregate, either as thick gelatin gel membranes separating one subunit from another (white stars in b), as a highly pronounced network occluded between the mineral (subunit III in b, see also supplementary Figure S5a) or connecting patches of mineral (subunits I and II in b, see also supplementary Figure S5b). Gelatin incorporation into the aggregate is not uniform. In some subunits we observe more hydrogel (subunit III) while in others we see more mineral (subunits I, II), see also supplementary Figure S5. A high magnification SEM image of a thick agarose hydrogel membrane (white star) is shown in (c).

26

ACS Paragon Plus Environment

Page 27 of 48

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 5. FE-SEM images of microtome cut, polished, etched and critical point dried surfaces showing the morphology of crystals grown in Mg-bearing silica hydrogel (a), the gel fabric (c) and the mineral – gel interconnection (b, c). In contrast to the rhombohedral shaped crystals that form in Mg-free silica gel the aggregates that grow in Mg-bearing silica gel are often dumbbell-shaped. See also supplementary Figure S6 which directly compares morphologies of crystals and aggregates grown in Mg-free and Mg-bearing hydrogels. The addition of Mg to the growth medium does not cause a substantial difference in the assembly of silica nanospheres. Figure 6. FE-SEM images of microtome cut, polished, etched and critical point dried surfaces of a calcite aggregate that grew in Mg-bearing agarose hydrogel. The spherical morphology of the aggregate is shown in (a) as well as the interdigitation between the gel network and the mineral (b, c). See also supplementary Figure S6. A striking feature of the aggregate are the columnar shaped calcite units (white star in b) that form when Mg is present in the growth medium. Figure 7. FE-SEM images of surfaces of microtome cut, polished, etched and critical point dried calcite aggregates grown in Mg-bearing gelatin hydrogel. Note the morphology of the aggregate (a), the interdigitation between the gel and the mineral (b, c) and the fabric of the gel network (c). See also supplementary Figure S6. Figure 8. Calcite aggregates grown in Mg-free gelatin, agarose and silica hydrogels. (a, b) Hopper calcite crystal aggregate bounded by strongly terraced {104} faces and rough morphology in the blocky surface without a well-defined orientation. The jagged character of the terrace edges is depicted in the close up of image (c). (d) Rhombohedron-shaped calcite crystal bounded by flat {104} and curved vicinal surfaces; (e) calcite radial aggregate consisting of rhombohedron-like crystals. The close-up (f) shows the high porosity of the flat {104} surfaces. (g, h) Aggregates grown in gelatin hydrogel showing rough surfaces caused by slightly misoriented, submicron sized sub-blocks with (originally hydrogel-filled) channels between them (i). Figure 9. Calcite crystal aggregates grown in gelatin, agarose and silica hydrogels containing an initial Mg concentration of 0.1M. (a, c) Dumbbell-shaped aggregates grown in silica hydrogel. The characteristics of the flat porous surfaces of these aggregates are depicted in (b) and (d). The Mg content detected by EDX analyses on the surface of the aggregates is 1.5 (a) and 3.8 (c) mol % MgCO3, respectively. (e) Sheaf-like aggregate obtained from agarose hydrogel. The aggregate has 27

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 28 of 48

two major subunits that consist of platy crystals in a sub-parallel arrangement (f). EDX analyses on the surface of this aggregate yield Mg contents between 2 and 3 mol % MgCO3. (g, h) The rounded aggregate that was obtained in agarose hydrogel consists of numerous platy crystals in a radial arrangement. EDX analyses on the surface of the aggregates shown in (e) and (g) yielded 2.5 and 5 mol % MgCO3 values, respectively. (i, k) Spherical aggregates obtained in gelatin hydrogel characterized by nanometric sub-blocks (j, l). The Mg content detected by EDX on the surface of these aggregates is 3.6 (i) and 17.9 mol % MgCO3 (k), respectively. Figure 10. EBSD crystallographic orientation for two adjacent calcite crystals (crystal 1 and crystal 2) grown in silica gel in a Mg-free environment: (a) Color-coded EBSD orientation map, (b) Corresponding pole figures showing crystallographic orientation data for crystal 1 and crystal 2. Corresponding band contrast images are shown in Figure S7a. Overall misorientation of the crystals is given in a color-coded map (c) and in a misorientation versus distance diagram (d) in comparison to the overall misorientation of a calcite single crystal precipitated from solution (e). In local misorientation versus distance diagrams (f, g) taken over the profiles A-B and C-D (shown in a) the high co-orientation of calcite c- and a*-axes is well visible. The profiles are almost 700 and 300 micrometers, respectively. Steps in the local misorientation curves (red dashed lines in f, g) with a jump in misorientation of about 0.5 degrees prove an internal structuring of the almost single crystals. Thus, these two crystals are single graded mesocrystals. Figure 11. EBSD crystallographic orientation results for a calcite-agarose hydrogel composite aggregate obtained in a Mg-free growth medium: (a) Color-coded EBSD orientation map and (b) corresponding pole figures of the aggregate. The band contrast image is shown in Figure S7c. This aggregate is composed of two large (no. 1 and 6) and 4 small (no. 2 to 5) subunits. As the pole figures in (e) and (f) show the two large subunits are single crystals. Encircled in red in e and f is the location where the poles of 61211 (subunit 6) and 3077 (subunit 1) c-axes and a-axes, respectivel fall on the same spot in the pole figure. Overall misorientation for calcite in subunits 6 and 4 is shown in (c) and (d); it is about 1.5 degrees. Local misorientation for profiles A-B (subunit 1) and C-D (subunit 6) is shown in (g) and (h). For both subunits local misorientation is less than 1 degree, we do not observe any steps in the local misorientation patterns. Thus, agarose gel induces the formation of radial aggregates composed of very few, similarly oriented, single crystalline subunits.

28

ACS Paragon Plus Environment

Page 29 of 48

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 12. EBSD crystallographic orientation results for a calcite-gelatin hydrogel radial aggregate obtained in a Mg-free growth medium: (a) Color-coded EBSD orientation map of the entire aggregate, (b) Corresponding pole figures for the entire aggregate. The band contrast image of the aggregate is given in supplementary Figure S7e. This aggregate is composed of four large subunits (a), with the subunits being further structured on different scale levels. Orientation patterns of individual subunits are shown in pole figures (c), (d) and (e). Subunit 2 is a partial spherulite characterized by a uniform rotation of the spherulite subentities. (f). Misorientation between the entities of the spherulite is between 5 and 10 degrees (g, h). As the pole figure in (d) shows and as indicated by the slight scatter in the misorientation graph (red arrows in g) the subunits of the spherulite are mesocrystals. Subunit 1 is an assemblage of a few large interlinked mesocrystals (pole figure in c, arrows highlighting the scatter in the diagram in k). The pole figures in (e) show that subunit 3 is also a mesocrystal, however, with a very minor tilt between the subblocks of this subunit (very even color of subunit 3 shown in a). Figure 13. EBSD crystallographic orientation results for a silica hydrogel/calcite composite aggregate grown in a Mg-bearing growth medium: (a) Color-coded EBSD orientation map, (b) Corresponding pole figure (band contrast image of the entire aggregate is shown in supplementary Figure S7b). This aggregate consists of two large and one small subunits (c, e), these are irregularly oriented relative to each other. Calcite orientation patterns of subunits 1 and 2 are shown in pole figures d and f. Subunits 1 and 2 are partial spherulites. As the misorientation versus distance diagrams in Figures (g) and (h) clearly show, subunits 1 and 2 are mosaic crystals (see e.g. the subentities a, b, c, d in Figures a and h with a regular tilt of 5-6 degrees between the subentities of the partial spherulite). These subentities are further structured, calcite is in mesocrystalline state (red arrows in g and h pointing to the scatter in the misorientation curve). Figure 14. EBSD crystallographic orientation results for a calcite-agarose (a to h) and calcite-gelatin (i, j, k) hydrogel aggregates grown in a Mg-bearing growth medium. Color-coded EBSD orientation maps and corresponding pole figures are given in Figures a, c, i and b, d, k. (h) shows the band contrast image to the EBSD map shown in Figure (c), (g) shows the surface of the aggregate that was scanned with EBSD; the corresponding EBSD map is given in (c). The aggregates (a, c) and (i) have a radial spherulitic microstructure, are mosaic crystals (see the large-angle tilt between the spherulites in e, f) but do not show a mesocrystalline substructure (see the absence of scatter within a spherulite 29

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 30 of 48

unit (red arrow in (f), misorientation curves shown in Figures (e) and (f)). The subentities of the spherulites are single crystals. The aggregate shown in (i) is a radial spherulite, however, without clear-cut spherulitic subunits. As the misorientation diagram in (j) shows, large-angle boundaries are present between co-oriented subunits. Corresponding band contrast images are shown in Figures (h) and S7d and S7f. Figure 15. FE-SEM micrographs of microtome cut, polished, decalcified or etched and critical point dried biopolymer matrix of hard tissues of some marine invertebrates, (a to d) entirely decalcified shell portions, (e-h) etched surfaces. (a, b) The organic matrix of a completely demineralized aragonitic shell portion of the gastropod Haliotis asinina. Well traceable in (a) are the former locations of mineral tablet assemblages to columns. (c, d) Biopolymer matrix membranes surrounding the calcite fibers of the calcitic layer of the bivalve Mytilus edulis. Biopolymer membranes (b, d) always have a porous structure. (e to h): Biopolymer network infiltrating matrix membrane lined (yellow stars in (e, f) mineral subunits (e.g. tablets, fibers). Red arrows in e and f point to organic fibrils in aragonite nacre tablets of Mytilus edulis (e) and Haliotis asinine (f), respectively. (g): Network of organic fibrils between cells of the coralline red algae Clathromorphum compactum. (h): Shell portion of the gastropod Haliotis laevigata showing the delicate biopolymer network of fibrils (red arrow in h) that permeates the calcitic part of the shell. Figure 16. FE-SEM micrograph of an assemblage of calcitic and aragonitic nanoparticles surrounded by proteinaceous matrix membranes in microtome cut, polished, slightly etched and critical point dried marine shell samples. (a) Calcite nanoparticles (yellow arrow) in a biopolymer-lined brachiopod fiber, (b) aragonite nanoparticles (yellow arrow) between polymer fibrils in a membrane bordered nacre tablet. Well visible in the nacre tablet is the organic fiber network between the mineral nanoparticles.

Table 1. Compilation of micro-, meso- and nanostructures of calcite aggregates that form in Mg-free and Mg-bearing silica, agarose and gelatin hydrogels, respectively.

30

ACS Paragon Plus Environment

Page 31 of 48

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

For Table of Content Use Only

The effect of hydrogel matrices on calcite crystal growth morphology, aggregate formation, and co-orientation in biomimetic experiments and biomineralization environments Fitriana Nindiyasaria*, Andreas Zieglerb, Erika Griesshabera, Lurdes Fernández-Díazc,d*, Julia Huberb, Paul Waltherb Wolfgang W. Schmahla aDepartment

für Geo- und Umweltwissenschaften and GeoBioCenter, Ludwig-Maximilians-Universität, 80333 Munich, Germany. bCentral Facility for Electron Microscopy, University of Ulm, Ulm, Germany. c Departamento de Cristalografía y Mineralogía, Universidad Complutense de Madrid, 28040 Madrid, Spain. dInstituto de Geociencias (UCM, CSIC). C/ José Antonio Novais 2, 28040 Madrid, Spain.

Biomimetic agarose and gelatin gel/calcite composites formed in Mg-free environments consist of very few subunits separated by hydrogel membranes, whereas silica gel/calcite aggregates are single gradient mesocrystals. Spherulites which are regular in agarose, gelatin gel/calcite aggregates and partial in silica gel/calcite aggregates form in Mg-bearing environments due to the increase in large and small angle boundaries.

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 32 of 48

Figure 1. Cryo-FE-SEM images of high-pressure frozen pristine hydrogels (a, b, c) and FE-SEM micrographs of the hydrogel fabric obtained after decalcification of the gel - calcite aggregate (d, e, f). (a, d) Silica, (b, e) Agarose, (c, f) Gelatin. Silica hydrogel is an assemblage of interconnected nanometric silica spheres (a). Agarose hydrogel exhibits a network of fibers (b) with large pore sizes. Gelatin hydrogel fabric (c) is slightly denser in comparison to that of agarose gel. After complete decalcification no major changes are observed in the gel fabric for silica (d), while for agarose and gelatin we see slight accumulations of hydrogel fibers (e) around voids (e, f) where the CaCO3 crystals were located. See also supplementary images S1, S2, S3, S4.

ACS Paragon Plus Environment

Page 33 of 48

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. FE-SEM images of microtome cut, polished, etched and critical point dried surfaces showing the crystal that grew in Mg-free silica hydrogel: a rhombohedral shaped calcite single crystal (a). (b, c) The interdigitation of silica gel and calcite on two different scale levels. The assembly of nanometer sized silica spheres occluded by mineral (calcite) (c). See also Figure S3.

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

Page 34 of 48

Figure 3. FE-SEM images of surfaces of a microtome cut, polished, etched and critical point dried agarose gel/calcite aggregate showing the interlinkage of subunits in the aggregate (a). Large intertwined subunits form the aggregate and they are separated from each other by agarose membranes (yellow arrows in b). Within the subunits of the gel/calcite aggregate, agarose hydrogel is developed as an evenly distributed occluded network (yellow stars in b). Enlargement showing the internal structure of the calcite/agarose composite, where calcite crystals are intimately encased by the agarose network (c).

3

ACS Paragon Plus Environment

Page 35 of 48

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 4. FE-SEM images of a microtome cut, polished, etched and critical point dried surfaces of a Mg-free gelatin - calcite aggregate showing its morphology (a) and internal structure (b, c). Large quantities of hydrogel are incorporated into the aggregate, either as thick gelatin gel membranes separating one subunit from another (white stars in b), as a highly pronounced network occluded between the mineral (subunit III in b, see also supplementary Figure S5a) or connecting patches of mineral (subunits I and II in b, see also supplementary Figure S5b). Gelatin incorporation into the aggregate is not uniform. In some subunits we observe more hydrogel (subunit III) while in others we see more mineral (subunits I, II), see also supplementary Figure S5. A high magnification SEM image of a thick agarose hydrogel membrane (white star) is shown in (c). 4

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 36 of 48

Figure 5. FE-SEM images of microtome cut, polished, etched and critical point dried surfaces showing the morphology of crystals grown in Mg-bearing silica hydrogel (a), the gel fabric (c) and the mineral – gel interconnection (b, c). In contrast to the rhombohedral shaped crystals that form in Mg-free silica gel the aggregates that grow in Mg-bearing silica gel are often dumbbell-shaped. See also supplementary Figure S6 which directly compares morphologies of crystals and aggregates grown in Mg-free and Mg-bearing hydrogels. The addition of Mg to the growth medium does not cause a substantial difference in the assembly of silica nanospheres. 5

ACS Paragon Plus Environment

Page 37 of 48

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 6. FE-SEM images of microtome cut, polished, etched and critical point dried surfaces of a calcite aggregate that grew in Mg-bearing agarose hydrogel. The spherical morphology of the aggregate is shown in (a) as well as the interdigitation between the gel network and the mineral (b, c). See also supplementary Figure S6. A striking feature of the aggregate are the columnar shaped calcite units (white star in b) that form when Mg is present in the growth medium.

6

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 38 of 48

Figure 7. FE-SEM images of surfaces of microtome cut, polished, etched and critical point dried calcite aggregates grown in Mg-bearing gelatin hydrogel. Note the morphology of the aggregate (a), the interdigitation between the gel and the mineral (b, c) and the fabric of the gel network (c). See also supplementary Figure S6.

7

ACS Paragon Plus Environment

Page 39 of 48

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 8. Calcite aggregates grown in Mg-free gelatin, agarose and silica hydrogels. (a, b) Hopper calcite crystal aggregate bounded by strongly terraced {104} faces and rough morphology in the blocky surface without a well-defined orientation. The jagged character of the terrace edges is depicted in the close up of image (c). (d) Rhombohedron-shaped calcite crystal bounded by flat {104} and curved vicinal surfaces; (e) calcite radial aggregate consisting of rhombohedron-like crystals. The close-up (f) shows the high porosity of the flat {104} surfaces. (g, h) Aggregates grown in gelatin hydrogel showing rough surfaces caused by slightly misoriented, submicron sized sub-blocks with (originally hydrogel-filled) channels between them (i). 8

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 40 of 48

Figure 9. Calcite crystal aggregates grown in gelatin, agarose and silica hydrogels containing an initial Mg concentration of 0.1M. (a, c) Dumbbell-shaped aggregates grown in silica hydrogel. The characteristics of the flat porous surfaces of these aggregates are depicted in (b) and (d). The Mg content detected by EDX analyses on the surface of the aggregates is 1.5 (a) and 3.8 (c) mol % MgCO3, respectively. (e) Sheaf-like aggregate obtained from agarose hydrogel. The aggregate has two major subunits that consist of platy crystals in a sub-parallel arrangement (f). EDX analyses on the surface of this aggregate yield Mg contents between 2 and 3 mol % MgCO3. (g, h) The rounded aggregate that was obtained in agarose hydrogel consists of numerous platy crystals in a radial arrangement. EDX analyses on the surface of the aggregates shown in (e) and (g) yielded 2.5 and 5 mol % MgCO3 values, respectively. (i, k) Spherical aggregates obtained in gelatin hydrogel characterized by nanometric sub-blocks (j, l). The Mg content detected by EDX on the surface of these aggregates is 3.6 (i) and 17.9 mol % MgCO3 (k), respectively.

9

ACS Paragon Plus Environment

Page 41 of 48

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 10. EBSD crystallographic orientation for two adjacent calcite crystals (crystal 1 and crystal 2) grown in silica gel in a Mg-free environment: (a) Color-coded EBSD orientation map, (b) Corresponding pole figures showing crystallographic orientation data for crystal 1 and crystal 2. Corresponding band contrast images are shown in Figure S7a. Overall misorientation of the crystals is given in a color-coded map (c) and in a misorientation versus distance diagram (d) in comparison to the overall misorientation of a calcite single crystal precipitated from solution (e). In local misorientation versus distance diagrams (f, g) taken over the profiles A-B and C-D (shown in a) the high co-orientation of calcite c- and a*-axes is well visible. The profiles are almost 700 and 300 micrometers, respectively. Steps in the local misorientation curves (red dashed lines in f, g) with a jump in misorientation of about 0.5 degrees prove an internal structuring of the almost single crystals. Thus, these two crystals are single graded mesocrystals. 10

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 42 of 48

Figure 11. EBSD crystallographic orientation results for a calcite-agarose hydrogel composite aggregate obtained in a Mg-free growth medium: (a) Color-coded EBSD orientation map and (b) corresponding pole figures of the aggregate. The band contrast image is shown in Figure S7c. This aggregate is composed of two large (no. 1 and 6) and 4 small (no. 2 to 5) subunits. As the pole figures in (e) and (f) show the two large subunits are single crystals. Encircled in red in e and f is the location where the poles of 61211 (subunit 6) and 3077 (subunit 1) c-axes and a-axes, respectivel fall on the same spot in the pole figure. Overall misorientation for calcite in subunits 6 and 4 is shown in (c) and (d); it is about 1.5 degrees. Local misorientation for profiles A-B (subunit 1) and C-D (subunit 6) is shown in (g) and (h). For both subunits local misorientation is less than 1 degree, we do not observe any steps in the local misorientation patterns. Thus, agarose gel induces the formation of radial aggregates composed of very few, similarly oriented, single crystalline subunits. 11

ACS Paragon Plus Environment

Page 43 of 48

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 12. EBSD crystallographic orientation results for a calcite-gelatin hydrogel radial aggregate obtained in a Mg-free growth medium: (a) Color-coded EBSD orientation map of the entire aggregate, (b) Corresponding pole figures for the entire aggregate. The band contrast image of the aggregate is given in supplementary Figure S7e. This aggregate is composed of four large subunits (a), with the subunits being further structured on different scale levels. Orientation patterns of individual subunits are shown in pole figures (c), (d) and (e). Subunit 2 is a partial spherulite characterized by a uniform rotation of the spherulite subentities. (f). Misorientation between the entities of the spherulite is between 5 and 10 degrees (g, h). As the pole figure in (d) shows and as indicated by the slight scatter in the misorientation graph (red arrows in g) the subunits of the spherulite are mesocrystals. Subunit 1 is an assemblage of a few large interlinked mesocrystals (pole figure in c, arrows highlighting the scatter in the diagram in k). The pole figures in (e) show that subunit 3 is also a mesocrystal, however, with a very minor tilt between the subblocks of this subunit (very even color of subunit 3 shown in a). 12

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 44 of 48

Figure 13. EBSD crystallographic orientation results for a silica hydrogel/calcite composite aggregate grown in a Mg-bearing growth medium: (a) Color-coded EBSD orientation map, (b) Corresponding pole figure (band contrast image of the entire aggregate is shown in supplementary Figure S7b). This aggregate consists of two large and one small subunits (c, e), these are irregularly oriented relative to each other. Calcite orientation patterns of subunits 1 and 2 are shown in pole figures d and f. Subunits 1 and 2 are partial spherulites. As the misorientation versus distance diagrams in Figures (g) and (h) clearly show, subunits 1 and 2 are mosaic crystals (see e.g. the subentities a, b, c, d in Figures a and h with a regular tilt of 5-6 degrees between the subentities of the partial spherulite). These subentities are further structured, calcite is in mesocrystalline state (red arrows in g and h pointing to the scatter in the misorientation curve). 13

ACS Paragon Plus Environment

Page 45 of 48

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 14. EBSD crystallographic orientation results for a calcite-agarose (a to h) and calcite-gelatin (i, j, k) hydrogel aggregates grown in a Mg-bearing growth medium. Color-coded EBSD orientation maps and corresponding pole figures are given in Figures a, c, i and b, d, k. (h) shows the band contrast image to the EBSD map shown in Figure (c), (g) shows the surface of the aggregate that was scanned with EBSD; the corresponding EBSD map is given in (c). The aggregates (a, c) and (i) have a radial spherulitic microstructure, are mosaic crystals (see the large-angle tilt between the spherulites in e, f) but do not show a mesocrystalline substructure (see the absence of scatter within a spherulite unit (red arrow in (f), misorientation curves shown in Figures (e) and (f)). The subentities of the spherulites are single crystals. The aggregate shown in (i) is a radial spherulite, however, without clear-cut spherulitic subunits. As the misorientation diagram in (j) shows, large-angle boundaries are present between co-oriented subunits. Corresponding band contrast images are shown in Figures (h) and S7d and S7f. 14

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 46 of 48

Figure 15. FE-SEM micrographs of microtome cut, polished, decalcified or etched and critical point dried biopolymer matrix of hard tissues of some marine invertebrates, (a to d) entirely decalcified shell portions, (e-h) etched surfaces. (a, b) The organic matrix of a completely demineralized aragonitic shell portion of the gastropod Haliotis asinina. Well traceable in (a) are the former locations of mineral tablet assemblages to columns. (c, d) Biopolymer matrix membranes surrounding the calcite fibers of the calcitic layer of the bivalve Mytilus edulis. Biopolymer membranes (b, d) always have a porous structure. (e to h): Biopolymer network infiltrating matrix membrane lined (yellow stars in (e, f) mineral subunits (e.g. tablets, fibers). Red arrows in e and f point to organic fibrils in aragonite nacre tablets of Mytilus edulis (e) and Haliotis asinine (f), respectively. (g): Network of organic fibrils between cells of the coralline red algae Clathromorphum compactum. (h): Shell portion of the gastropod Haliotis laevigata showing the delicate biopolymer network of fibrils (red arrow in h) that permeates the calcitic part of the shell. 15

ACS Paragon Plus Environment

Page 47 of 48

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 16. FE-SEM micrograph of an assemblage of calcitic and aragonitic nanoparticles surrounded by proteinaceous matrix membranes in microtome cut, polished, slightly etched and critical point dried marine shell samples. (a) Calcite nanoparticles (yellow arrow) in a biopolymer-lined brachiopod fiber, (b) aragonite nanoparticles (yellow arrow) between polymer fibrils in a membrane bordered nacre tablet. Well visible in the nacre tablet is the organic fiber network between the mineral nanoparticles.

16

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 48 of 48

Table 1. Compilation of micro-, meso- and nanostructures of calcite aggregates that form in Mg-free and Mg-bearing silica, agarose and gelatin hydrogels, respectively.

Hydrogel

Mg-free Aggregate

Subunits

Silica

Single mesocrystal with one overall orientation with a slight orientational gradient

-

Agarose

radial aggregate comprising a few large highly co-oriented major subunits

single crystals with very little mesocrystalline spread

Gelatin

radial aggregate comprising a few not co-oriented subunits

fan-like mosaic crystal (partial spherulite) consisting of mesocrystals

Mg-bearing

Hydrogel

Silica Agarose

Aggregate

Subunits

radial aggregate with very few not co-oriented subunits

partial spherulite (mosaic crystal) consisting of mesocrystals

spherulite

mosaic crystals with single crystals as subentities

mosaic crystals with

Gelatin

spherulite

single crystals as subentities

17

ACS Paragon Plus Environment