Journal of Crystal Growth 319 (2011) 64–69

Contents lists available at ScienceDirect

Journal of Crystal Growth journal homepage: www.elsevier.com/locate/jcrysgro

Orientation and morphology of calcite nucleated under floating monolayers: A magnesium-ion-enhanced nucleation study Benjamin Stripe, Ahmet Uysal, Pulak Dutta n Dept. of Physics & Astronomy, Northwestern University, Evanston, IL 60208, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 30 November 2010 Received in revised form 31 January 2011 Accepted 31 January 2011 Communicated by S. Veesler Available online 18 February 2011

We have studied the biomimetic growth of calcium carbonate crystals under floating monolayer templates, in the presence of Mg ions, using grazing incidence X-ray diffraction and SEM imaging. Crystals grown under sulfate monolayers nucleate from the (0 0 1) plane with and without Mg ions, while undergoing substantial changes in morphology. Crystals grown under alcohol monolayers nucleate from the (1 0 4) plane in the presence of Mg. X-ray data do not detect orientation in crystals grown under acid monolayers, but at higher Mg concentrations the resulting morphologies are indicative of template-nucleated growth. These results suggest that Mg provides living organisms a way to enhance the orientation and control the morphology of acid-templated crystals. & 2011 Elsevier B.V. All rights reserved.

Keywords: A1. Biocrystallization A1. X-ray diffraction B1. Calcium compounds

1. Introduction Many biominerals are primarily calcium carbonate or phosphate, but the presence of small amounts of inorganic additives play important roles. Magnesium is the most common of these additives and is thought to be the most important. It is known that the Mg/Ca ratio in seawater has fluctuated greatly with time [1], and that this ratio strongly affects the structure of abiotic CaCO3 precipitates. Biogenic CaCO3 from some organisms, but not all, followed the same trend as abiotic CaCO3. In the laboratory, there have been many studies of the effects of Mg on bulk CaCO3 precipitation [2,3]. The addition of magnesium to the subphase has already been shown to have strong effects including delayed nucleation [4], promotion of aragonite growth [2,5] and amorphous CaCO3, and the expression of higher energy faces to the water [6]. Recently, it has been recognized that amorphous calcium carbonate (ACC) plays an important intermediate role in biomineralization, and that Mg ions stabilize ACC and slow its conversion to crystalline forms [6,7]. The role of dissolved organic molecules on bulk precipitation has also been studied [6,8,9]. For a detailed recent review, see Ref. [9]. Studies investigating the role of magnesium in the templated growth of oriented CaCO3 have used several different approaches. A large number of studies made use of solid supports such as gold/silver coated alkanethiols [10,11], silica supported polymerized 1-,12-pentacosadiyonic acid [12], or even inorganic (SnO2 coated glass) surfaces [13]. Solid-substrate-supported

n

Corresponding author. Tel.: +1 8474915465; fax: + 1 8474919982. E-mail address: [email protected] (P. Dutta).

0022-0248/$ - see front matter & 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jcrysgro.2011.01.102

templates have been very successful in nucleating CaCO3 in many different orientations, depending upon variables such as the chain length, functional group, and underlying substrate. Studies using templates without hard supports, namely floating Langmuir monolayers, have had far less success in demonstrating welloriented growth. These studies have relied mainly on ex situ imaging of irregular crystals to determine orientation [9,14–17]. Recent studies have focused on the use of grazing incidence X-ray diffraction from floating Langmuir monolayers as a way to monitor both the template structure and crystal growth in situ [18–22]. No studies have shown X-ray diffraction evidence of oriented growth under carboxylic and alcohol Langmuir monolayers, or an explanation for the lack of such orientation [9,19,20,22]. Previous studies examining heneicosanoic acid on subphases containing barium fluoride, lead carbonate, and other salts have found supercellular lattice matches and strained interfaces [23,24]. Applying the same techniques to common biominerals could be extremely informative; however, not much data is available, as pointed out recently in Ref. [14]. In situ grazing incidence X-ray diffraction on floating monolayer systems has turned up oriented growth at sulfate monolayers [20,21] but not at acid or alcohol monolayers [18,21]. This is again in direct contrast to research done on solid-supported self assembled monolayers [10,11]. The situation is made more complicated by experimental difficulties. X-ray diffraction averages over all the crystals in the ‘‘footprint’’ illuminated by the beam, and thus X-ray evidence is ideal to establish average orientation (or lack thereof) as opposed to exceptional individual crystals. However, the crystals form two-dimensional powders on the template surface, and therefore

B. Stripe et al. / Journal of Crystal Growth 319 (2011) 64–69

a sufficient number of crystals must be present within the X-ray footprint. If only a low density of crystals is nucleated, or the footprint is too small, the Bragg peaks seen may not be from a statistical average but rather from individual crystals that coincidentally have the correct alignment (leading to time-dependent and/or irreproducible data). Despite the inherent difficulties, nucleation under floating monolayers offers unique access to the growth interface. In this study we have sought to use the previously known effects of magnesium – slow calcite nucleation, extended life of the amorphous precursor phase, and altered morphology of the crystals – to enhance the oriented crystal growth of CaCO3 at Langmuir monolayer interfaces and thus allow the process to be studied in more detail.

2. Experimental All materials were purchased from Sigma-Aldrich and used without further purification. Heneicosanoic acid (C20H41COOH) and heneicosanol (C21H43OH) were both prepared in 10 ml chloroform to make 3 mM solutions. The arachidic sulfate sodium salt monolayer (C20H410S03Na) was prepared following the method of Hendrikx [25]. The subphases were prepared by adding 2 g CaCO3 and 0 to 16 g MgCl2  6H2O in 1 l of H2O (18.2 MO cm) and bubbling CO2 for 1 h, also known as Kitano’s method [5]. The solutions were then suction filtered with 0.2 mm filters and re-bubbled for half an hour. The final subphase had pH  5.8 and concentration of just greater than 7 mM Ca2 + as reported previously [19]. All three types of monolayers were compressed to 10 mN/m. X-ray data was collected from Mg/Ca¼2.5 samples at synchrotron beam line X-14A at the National Synchrotron Light Source, after 12 h of in situ growth time before taking data to avoid excess burning of the monolayers or amorphous layers formed in the early growth stages. The beam was incident in a grazing incidence X-ray diffraction (GIXD) geometry at a wavelength of 1.3376 A˚ on a point detector with a resolution of kxy ¼01 A˚  1 and kz ¼05 A˚  1 defined by crossed soller slits. The incident angle was approximately 0.002 rad, just below the critical angle of water. GIXD contour plots in k space were constructed by taking scans with steps sizes of kxy ¼0.005 and kz ¼0.1. Peak intensities were monitored for changes over the course of the k space contour scan, to ensure the sample had not been damaged due to radiation exposure. Additional information on the setup can be found in the article by Barton et al. [26].

65

It is also important to note the effective beam footprint of this setup. The incident mirror reflects the X-ray beam (measuring 1.5 cm horizontally by 0.7 mm vertically) downwards at a vertical angle of 0.002 rad on the water surface, creating a beam footprint of 1.5 cm  15 cm. The point detector acceptance is limited by the 2 cm total width of the soller slits. Therefore the effective footprint for averaging the 2 dimensional crystalline powder is at least 3 cm2 at 901 from the incident beam. This provides orders of magnitude better averaging than other setups [19,22]. A Leica SP2 confocal microscope was used to aid in the determination of crystal morphologies. The microscope was run in reflectivity mode and scans were taken with a 0.2 mm  0.2 mm  0.36 mm voxel resolution. The data were used to recreate a 3D model of the surface from which surface normal directions were calculated. All samples used for scanning electron microscopy (SEM) imaging were prepared on silicon substrates by surface tension transfer, and then dried under a stream of nitrogen for 10 min. Samples were then sputter coated with 10 nm of Au/Pd. All images were captured from samples left to grow for 48–72 h before transfer. Samples with higher concentrations of magnesium often did not completely dry under the nitrogen stream. The slow drying time of high Mg samples is most likely due to the formation of amorphous CaCO3 or hydrated MgCl2. Our studies were concerned with the crystals formed and not with the amorphous phase.

3. Results 3.1. Sulfate monolayers SEM images of transferred CaCO3 samples grown under the arachidic sulfate monolayer with varying amounts of Mg in the subphase are shown in Fig. 1. The first image, corresponding to no Mg in the subphase, shows (0 0 1) nucleated trigonal pyramids as previously reported [20,21]. The image shows crystals resting on one of their {10.4} faces due to the surface tension transfer on the silicon substrate for imaging. However, this offers a clear view of the triangular (0 0 1) nucleation planes. The nucleation planes are not completely flat, but display a small peak in the center of the face. Our in situ GID X-ray diffraction shows that the crystals are nucleating with the (0 0 1) plane parallel to the surface. This data set is not shown here since similar data have been published previously [20,21].

Fig. 1. SEM images of crystals grown under arachidic sulfate with varying ratios of magnesium in the subphase. The inset scale bars are both 15 mm.

66

B. Stripe et al. / Journal of Crystal Growth 319 (2011) 64–69

The second image, corresponding to Mg/Ca¼2.5, shows crystals with near-hexagonal (0 0 1) faces. The wire diagram in Fig. 1 indicates the faces for one of the Mg/Ca¼2.5 crystals. The (0 0 1) face is once again peaked in the center. It is possible these may be {1 0.12} faces as previously reported for growth of sulfateterminated alkanethiols on gold-coated silicon [10]; however, not all crystals clearly display the central peak. Fig. 2 shows in situ GID X-ray diffraction data for Mg/Ca¼2.5, showing that CaCO3 is indeed still nucleating on the (0 0 1) plane. The {11.0} Debye ring scan, a scan as a function of F ¼tan  1(kz/kxy), shows that the crystals are oriented to within 51 FWHM, which is a factor of two better orientation that reported without Mg [20]. The orientation of the crystals can be verified in Table 1, which lists the calculated locations of the visible peaks assuming a [0 0 1] orientation normal to the surface. Hexagonal (0 0 1) nucleating faces have been reported before and have been attributed to increased lattice mismatch, interface strain or

impurities such that the trigonal (0 0 1) growth face does not grow as fast as the {10.4} faces [21,27]. It should be noted that while it has been shown that arachidic sulfate shares a lattice match with the (0 0 1) plane, a lattice match is not necessarily a requirement for oriented growth [9,22]. In this case the hexagonal face is caused by the intersections of the (0 1 1) and (1 0 0) crystallographic planes. The orientation of the surface normals of the (0 1 1) and (1 0 0) planes were confirmed by confocal microscopy to be within 31 of the crystallographically expected direction relative to the (0 0 1) face and also within 31 of orientation relative to each other. It is clear that the (0 1 1) and (1 0 0) planes are not completely flat; in fact, they show small features with a general planar alignment. At the opposite end from the (0 0 1) face, the crystals terminate either in small {10.4} planes or in larger crystals, in small irregular pitted or incomplete faces. The expression of these faces has been shown before in bulk growth and also in growth at SnO2 surfaces [6,13]. The third image in Fig. 1 corresponds to a ratio of Mg/Ca¼5. Once again the crystal clearly displays the hexagonal (0 0 1) nucleation plane still with a small peak at its center. The crystals still display weak (0 1 1) and (1 0 0) faces near the (0 0 1) growth face; however, as the crystal extends into it terminates with a rounded end. The remaining images in Fig. 1 show Mg/Ca¼7.5 and 10. Both images clearly display the (0 0 1) hexagonal biomimetic nucleation face. At Mg/Ca¼7.5 the rest of the crystal is completely rounded and no longer displays any visible {01.1} or {10.0} planes. The surface features of the crystal are still small and discrete, generally aligned angling away from the c-axis in a uniform manner. The crystals look like rotationally symmetric ‘‘pineapples’’. At Mg/Ca¼10 the crystals are now nearly hemispherical, no longer displaying any regular surface features seen in the previous crystals except for the nucleating face.

3.2. Alcohol monolayers

Fig. 2. In situ GID X-ray contour scan of calcite crystals growing under arachidic sulfate with Mg/Ca ¼2.5 and accompanying {11.0} Debye ring scan where F ¼tan  1(kz/kxy).

Table 1 List of the calculated positions of peaks visible within our accessible GID range, assuming a [0 0 1] or [1 0 4] orientation. Crystal orientation (vertical lattice vector)

Vert./horiz. Diffraction peak components [0 1 2] [1 0 4] [1 1 0] [1 1  3] [2 0 2]

[0 0 1]

[1 0 4]

kz (A˚  1)

0.74

0

0.74

kxy (A˚  1)

1.42

2.52

2.91

kz (A˚  1)

0.50

0.54

0

0.77

0.50

kxy (A˚  1)

1.56

2.00

2.52

2.65

2.96

Fig. 3 shows SEM images of samples grown under heneicosanol from subphases containing various Mg/Ca ratios. The first image corresponding to Mg/Ca¼0 displays discrete {10.4} rhombohedral calcite. Previous studies have shown growth from the (1 0 4) plane on alcohol terminated alkanethiols on gold coated silicon, under Langmuir monolayers and on SnO2 surfaces [10,13,15]. However, growth under alcohol Langmuir monolayers has shown fairly poor orientation and there is no in situ X-ray data showing oriented growth [15,19]. The second SEM image in Fig. 3 corresponds to a ratio of Mg/Ca¼2.5. Similar to the crystals grown under the sulfate monolayer, the crystals clearly express the (1 0 0) and (0 1 1) faces to the water as illustrated by the wireframe. However, the growth face is now tilted  451 relative to the crystals grown under the sulfate monolayer and displays a large skew hexagonal (1 0 4) face to the monolayer. In situ GID X-ray data in Fig. 4 confirm that the crystals nucleating [1 0 4] are oriented, and the inset {10.4} Debye ring scan, where F ¼tan  1(kz/kxy), shows that the crystals are oriented within 51 FWHM. The orientation of the crystals can be verified in Table 1, which lists the calculated locations of the visible peaks assuming a [1 0 4] orientation normal to the surface. The third SEM image corresponding to Mg/Ca¼5 shows similarities to the crystals grown under the sulfate monolayers at the same concentration. The crystals once again show evidence of the (0 1 1) and (1 0 0) planes near the (1 0 4) nucleation face. Also, as the crystals grow into the water, they become rounded and terminate in a continuous curve. The surface is still rough, showing features generally aligned along the c-axis. At a concentration of Mg/Ca¼7.5 the crystals appear to begin nucleating on the (1 0 4) face, but as growth occurs the rounded

B. Stripe et al. / Journal of Crystal Growth 319 (2011) 64–69

67

Fig. 3. SEM images of crystals grown under heneicosanol with varying ratios of Mg in the subphase. The inset scale bar is 10 mm.

crystallographic planes. By Mg/Ca¼10 it is hard to say anything about the monolayer governed growth plane. The crystal morphology changes and X-ray intensity is too low to obtain an orientation.

3.3. Acid monolayers

Fig. 4. In situ GID X-ray contour scan of calcite crystals growing under heneicosanol with Mg/Ca ¼ 2.5 and accompanying {10.4} Debye ring scan where F ¼tan  1(kz/kxy).

lobe opposite the nucleation face begins to push against the monolayer surface. This creates a small flat spot on the crystal in the same plane as the larger nucleation face. However, the flat spot does not appear to be continuation of the same (1 0 4) plane because its perimeter is not regular. This suggests that instead of simply growing against the monolayer as the nucleation face does, the rounded lobe pushes against the monolayer and possibly pushes the crystal away from the monolayer, creating flat planes that are parallel to the water but are not

Growth of CaCO3 in conjunction with carboxylic acids is perhaps the most studied biomimetic nucleation process [10,15,16,20,21]. However, reported results vary widely, with nucleation on the (1–10) to (0 1 2), (0 1 5), (1 0 0), (0 0 1), and (1 1 15) planes [10,15–17]. Sommerdijk and de With [9] have discussed the number of different orientations in detail. Fig. 5 shows SEM images of crystals grown under heneicosanoic acid with various concentrations of Mg. The first image shows crystals grown with no Mg in the subphase are typical of crystals grown under heneicosanoic acid and consistent with previous results [15,20,21]. Unlike the previous two types of monolayers in the present study, there are no easily defined orientations attributable to these crystals. The only notable feature is the growth of multiple {10.4} style crystalline arms. Past in situ GID X-ray data have shown unoriented, irreproducible or powder-like patterns [20,21]. The SEM image corresponding to Mg/Ca¼2.5 is different from that seen under the previous two monolayers. The previous monolayers displayed large (0 1 1) and (1 0 0) planes. These crystals appear to display no such planes. The crystals all display a similar elongated pentagonal/hexagonal face that is peaked towards the center, similar to the crystals grown under the sulfate monolayers. The crystals also have a rough face on one side of the elongated axis giving the crystal a wedge-like conformation. The in situ GID X-ray data are inconclusive or irreproducible. At a concentration of Mg/Ca¼5, the crystals flatten out and appear to be rounded on both of the elongated ends of the crystal. The raised center of the nucleation plane is still apparent. At the highest concentrations of Mg/Ca¼7.5 and 10 the crystals have flat faces with rounded perimeters and are rounded on the other sides, forming ‘‘dumbells’’. Unfortunately, because of the slow growth and lower intensities at these high concentrations, there is not enough time for in situ synchrotron studies, and thus the average crystallographic orientation is not known. However, the shapes at the higher Mg concentrations strongly suggest oriented growth. It is possible

68

B. Stripe et al. / Journal of Crystal Growth 319 (2011) 64–69

Fig. 5. SEM images of crystals grown under heneicosanoic acid with varying ratios of magnesium in the subphase. The inset scale bars are both 30 mm.

indicate that dolomite crystals prefer expressing (1 0 0) and (1 1 0) faces. They also suggest that the segregation energy between Mg and Ca results in many crystals nucleating as calcium carbonate with Mg saturated surfaces instead of dolomite. Therefore, it is not unexpected to see faces such as the (1 0 0) and (0 1 1) in the presence of magnesium where surface preference and segregation energies can lead to calcium carbonate crystals expressing dolomite-like faces while still incorporating little or no Mg in the crystal itself.

4. Discussion

Fig. 6. Positions of ex situ [1 0 4] calcite powder diffraction peaks grown in the presence of Mg in the solution, with various Mg/Ca ratios as labeled.

that there is some average orientation nearly perpendicular to the [0 0 1] axis and possibly the (1–10) plane reported by Heywood et al. [16].

3.4. Magnesium incorporation Ex situ powder X-ray data from calcite crystals collected at the water surface, Fig. 6, shows lattice shifts as a function of Mg concentration in the subphase. Our results are consistent with those of Nishino et al. [13], who characterized the bulk structural changes in surface-nucleated crystals at high Mg concentrations. Their results show crystals composed of roughly 20 nm domains with slightly misaligned c-axes leading to the observed rounding. We also show a change in the lattice spacing from calcite towards a dolomite spacing consistent with Vegard’s law [29]. The lattice shift corresponds to about 6% Mg incorporation when grown from a 40 mM Mg subphase. There is also a notable change in X-ray intensity as the magnesium concentration increases. The expression of the (1 0 0) and (0 1 1) faces with increased Mg concentration (Figs. 1 and 3) is consistent with the results of experiments studying the recrystallization of high-Mg biogenic calcite along with the accompanying biological macromolecules [6]. Moreover, simulations of the nucleation of dolomite (CaMg(CO3)2) and high magnesium calcite by Titiloye et al. [30]

We have shown that crystals of CaCO3 grown under arachidic sulfate maintain their [0 0 1] orientation over a large range of Mg concentrations. While the diffracted X-ray intensity drops and the crystals incorporate smaller and smaller crystalline domains with increasing Mg [13,18] the omnipresent hexagonal nucleating plane and morphology of the crystal is of clear evidence of the continued templated/oriented growth. On a side note, in all three experiments the Mg/Ca ratios do not match very well with commonly reported ratios found previously in the literature [2,11,13,21]. However, kinetic differences in preparation methods could explain these differences, and further investigation is required. In the case of heneicosanol, GID X-ray evidence in the absence of Mg is inconclusive. However, at Mg/Ca¼2.5 the GID X-ray data clearly shows [1 0 4] oriented growth. This enhancement of the (1 0 4) nucleation can be described by a simple energy savings argument based on the calculations of Duffy et al. [28]. In the absence of Mg the (1 0 4) crystal–monolayer interfacial energy is only slightly lower than the (1 0 4) crystal–water interfacial energy, and so unoriented nucleation has a small energy disadvantage and a large entropy advantage. With the addition of Mg, the crystal–Mg solution interfaces prefer the higher energy {10.0}, {01.1} planes as evidenced by the planes surface areas. This implies that Mg ions increase the (1 0 4)–Mg solution interfacial energy relative to the (1 0 0)/(0 1 1)–Mg solution interfacial energy; however, it is reasonable to assume that Mg solution does not penetrate to the (1 0 4)–monolayer interface and thus do not affect this interface energy. Therefore the net energy savings for the (1 0 4)–monolayer interface is increased. The monolayer still wishes to interact with the (1 0 4) face while

B. Stripe et al. / Journal of Crystal Growth 319 (2011) 64–69

the solution wishes to interact less with the {10.4} faces, leading to more energetically favorable nucleation from the (1 0 4) plane. Much like the crystals grown under arachidic sulfate, the crystals maintain their orientation over most of the range as Mg concentration increases, as evidenced by the continued presence of the (1 0 4) growth plane and the morphology showing the 451 growth direction relative to the (0 0 1) plane of crystals grown under arachidic sulfate. As discussed above, a possible explanation for the lack of oriented growth in the absence of Mg may be that the energy advantage of pairing the neutral (1 0 4) surface with the neutral monolayer is low, while the entropy advantage for unoriented {10.4} rhombohedral growth is high. Given the high concentration of acid groups in the organic layers of natural biominerals, and given the reports of oriented growth at supported acid templates [10,11,18], the fact that no one has seen any X-ray averaged orientation in CaCO3 nucleating under floating acid monolayers [19–21] has not been obtained. Several possibilities as to why oriented X-ray data elude us exist, including favored orientations that are not stable against the monolayer and nucleation of multiple orientations [14,15,27]. However, our studies of sulfate and alcohol templates provide us with clues to interpret the acid template imagery. As the Mg concentration increases, the crystals begin to develop large flat nucleation planes and the same distinctive dumbbell-like morphologies seen in previous cases. Unfortunately, at high Mg concentrations there is not enough signal for us to obtain statistically-averaged X-ray data, but the observed morphological trends as a function of Mg concentration strongly suggest that acid monolayers do nucleate oriented calcite. The crystals grown under the acid monolayer appear to be growing oriented nearly perpendicular to the [0 0 1] axis, possibly on the (1 1 0) plane reported by Mann et al. [16]. It is of course widely known that Mg slows the nucleation rate of CaCO3. The fact that it also enhances oriented growth under Langmuir monolayers suggests that Mg can be used as a tool to allow more detailed studies of templated calcite nucleation than have been possible so far. Our studies point to a possible biological function by slowing the growth of calcite biominerals, Mg provides living organisms a way to enhance the orientation and control the morphology of acid-templated calcite crystals.

Acknowledgments We thank Jianming Bai for his assistance at X-14A. This work was supported by the US Department of Energy under Grant no. DE-FG02-84ER45125. In situ grazing incidence X-ray data was collected at the National Synchrotron Light Source beam line X-14A. The National Synchrotron Light Source is supported by USDOE Contract #DE-AC02-98CH10886. The SEM images were recorded at the Electron Probe Instrumentation Center, Northwestern University References [1] S.M. Stanley, Effects of global seawater chemistry on biomineralization: past, present, and future, Chemical Reviews 108 (2008) 4483–4498. [2] G. Falini, S. Fermani, G. Tosi, E. Dinelli, Calcium carbonate morphology and structure in the presence of seawater ions and humic acids, Crystal Growth & Design 9 (2009) 2065–2072. [3] Y. Kitano, A. Tokuyama, T. Arakaki, Magnesian calcite synthesis from calcium bicarbonate solution containing magnesium and barium ions, Geochemical Journal 13 (1979) 181–185.

69

[4] E. DiMasi, S. Kwak, F.F. Amos, M.J. Olszta, D. Lush, L.B. Gower, Complementary control by additives of the kinetics of amorphous CaCO3 mineralization at an organic interface: in-situ synchrotron X-ray observations, Physical Review Letters 97 (2006) 45503. [5] Y. Kitano, A study of the polymorphic formation of calcium carbonate in thermal springs with an emphasis on the effects of temperature, Bulletin of the Chemical Society of Japan 35 (1962) 1980–1985. [6] S. Albeck, J. Aizenberg, L. Addadi, S. Weiner, Interactions of various skeletal intracrystalline components with calcite crystals, Journal of the American Chemical Society 115 (1993) 11691–11697. [7] E. Loste, R. Wilson, R. Seshadri, F. Meldrum, The role of magnesium in stabilising amorphous calcium carbonate and controlling calcite morphologies, Journal of Crystal Growth 254 (2003) 206–218. [8] D. Wang, A. Wallace, J. De Yoreo, P. Dove, Carboxylated molecules regulate magnesium content of amorphous calcium carbonates during calcification, PNAS 106 (2009) 21511–21516. [9] N. Sommerdijk, G. de With, Biomimetic CaCO3 mineralization using designer molecules and interfaces, Chemical Review 108 (2008) 4499–4550. [10] J. Aizenburg, A. Black, G. Whitesides, Oriented growth of calcite controlled by self-assembled monolayers of functionalized alkanethiols supported on gold and silver, Journal of the American Chemical Society 121 (1999) 4500–4509. [11] Y. Han, L.M. Wysocki, M.S. Thanawala, T. Siegrist, J. Aizenberg, Templatedependent morphogenesis of oriented calcite crystals in the presence of magnesium ions, Agnewandte Chemie International Edition 44 (2005) 2386–2390. [12] K. Sato, Y. Kumagai, T. Kogure, K. Watari, J. Tanaka, Polymorph and orientation controls of calcium carbonate crystals achieved by organic matrices, Journal of the Ceramic Society of Japan 114 (2006) 754–759. [13] Y. Nishino, Y. Oaki, H. Imai, Magnesium-mediated nanocrystalline mosaics of calcite, Crystal Growth & Design 9 (2009) 223–226. [14] C. Lendrum, K.M. McGrath, Toward controlled nucleation: monolayer chemistry with monolayer fluidity, Crystal Growth & Design 10 (2010) 4463–4470. [15] C.D. Lendrum, K.M. Mcgrath, Calcite nucleation on mixed acid/alcohol monolayers, Crystal Growth & Design 9 (2009) 4391–4400. [16] B.R. Heywood, S. Mann, Molecular construction of oriented inorganic materials—controlled nucleation of calcite and aragonite under compressed langmuir monolayers, Chemistry of Materials 6 (1994) 311–318. [17] S. Mann, D. Archibald, J. Didymus, T. Douglas, B. Heywood, F. Meldrum, N. Reeves, Crystallization at inorganic–organic interfaces—biominerals and biomimetic synthesis, Science 261 (1993) 1286–1292. [18] S. Kwak, E. DiMasi, Y. Han, J. Aizenberg, I. Kuzmenko, Orientation and Mg incorporation of calcite grown on functionalized self-assembled monolayers: a synchrotron X-ray study, Crystal Growth & Design 5 (2005) 2139–2145. [19] E. DiMasi, S. Kwak, B. Pichon, N. Sommerdijk, Structural adaptability in an organic template for CaCO3 mineralization, CrystEngComm 9 (2007) 1192–1204. [20] S. Kewalramani, K. Kim, B. Stripe, G. Evmenenko, G. Dommett, P. Dutta, Observation of an organic–inorganic lattice match during biomimetic growth of (0 0 1)-oriented calcite crystals under floating sulfate monolayers, Langmuir 24 (2008) 10579–10582. [21] K. Kim, A. Uysal, S. Kewalramani, B. Stripe, P. Dutta, Effects of chitosan on the alignment, morphology and shape of calcite crystals nucleating under Langmuir monolayers, CrystEngComm 11 (2009) 130–134. [22] E. DiMasi, M. Olszta, V. Patel, L. Gower, When is template directed mineralization really template directed?, CrystEngComm 5 (2003) 346–350. [23] J. Kmetko, C.J. Yu, G. Evmenenko, S. Kewalramani, P. Dutta, Organic-templatedirected nucleation of strontium fluoride and barium fluoride: epitaxy and strain, Physical Review B 68 (2003) 085415. [24] S. Kewalramani, G. Evmenenko, C.J. Yu, K. Kim, J. Kmetko, P. Dutta, Evidence of surface reconstruction during ‘bioinspired’ inorganic nucleation at an organic template, Surface Science 591 (2005) L286–L291. [25] Y. Hendrikx, Interactions in mixed films of docosyltrimethylammonium bromide and sodium docosyl sulfate—effect of cetyl alcohol upon the properties of these Films, Journal of Colloid Interface Science 69 (1979) 493–501. [26] S.W. Barton, B.N. Thomas, E.B. Flom, S.A. Rice, B. Lin, J.B. Peng, J.B. Ketterson, P. Dutta, X-ray diffraction study of a langmuir monolayer C21H43OH, Journal of Chemical Physics 89 (1988) 2257–2270. [27] B. Pokroy, J. Aizenberg, Calcite shape modulation through the lattice mismatch between the self-assembled monolayer template and the nucleated crystal face, CrystEngComm 9 (2007) 1219–1225. [28] D. Duffy, J. Harding, Simulation of organic monolayers as templates for the nucleation of calcite crystals, Langmuir 20 (2004) 7630–7636. [29] J. Goldsmith, D.L. Graf, H. Heard, Lattice constants of the calcium-magnesium carbonates, American Mineralogist 46 (1961) 453–457. [30] J. Titiloye, N. De Leeuw, S. Parker, Atomistic simulation of the differences between calcite and dolomite surfaces, Geochemica et Cosmochimica Acta 62 (1998) 2637–2641.