Calcium Metal to Synthesize Amorphous or Cryptocrystalline Calcium Phosphates

A. Cuneyt TAS, Ph.D.

Piscataway, New Jersey 08854, USA

E_mail: c_tas@hotmail.com www.cuneyttas.com

---------------------------------------------------------------------------------------------------------------------

Article: Mater. Sci. Eng. C, 32(5), 1097-1106 (2012)

US Patent: 9,108,860 B2 Patent issued on: August 18, 2015

---------------------------------------------------------------------------------------------------------------------

Metallic calcium was never used before as the only calcium source in synthesizing bioceramics in solutions mimicking the human blood plasma (in terms of their ion concentrations). Amorphous calcium phosphate (ACP) powders were synthesized at room temperature, in synthetic mineralization solutions which contained Na⁺, K⁺, Mg²⁺, Cl^-, HCO₃^- and HPO₄^2- ions at concentrations similar to those found in the human blood plasma, by using calcium (Ca) metal as the only calcium source. The experimental conditions leading to the formation of PCA (cryptocrystalline or poorly-crystallized apatite) or CaCO₃ powders were also determined when using metallic Ca in aqueous synthesis in the mineralization solutions. The formation of calcium phosphate (CaP) in synthesis solutions was immediately initiated by the addition of calcium metal granules (or shots), at appropriate amounts, into the solutions while the solutions were being continuously stirred at room temperature (22±1°C). The synthesis reactions were reaching completion in less than 30 minutes with the final solution pH values ranging from 9 to 12, without a necessity for any external pH adjustment in the form of any strong base (such as NH₄OH, LiOH, NaOH or KOH) additions. ACP or PCA powders are useful for dentin and enamel re-mineralization applications or orthopedic (bone) defect-filling applications.

1. Introduction

The systematic synthesis and characterization of poorly-crystallized (i.e., cryptocrystalline) apatite (PCA) powders in deionized (i.e., free of Na⁺, K⁺, Mg²⁺, Cl^- and HCO₃^- ions of blood plasma) water solutions containing dissolved calcium nitrate tetrahdyrate (Ca(NO₃)₂×4H₂O) and diammonium hydrogen phosphate ((NH₄)₂HPO₄) were initiated in the early 50’s by Hayek and co-workers [1-3]. The work of Hayek et al. [1-3] taught us to raise the pH values of such cryptocrystalline apatite synthesis solutions to around 10.5-11 by the addition of ammonium hydroxide (NH₄OH). The method originally developed by Hayek et al. [1-3] to produce cryptocrystalline nanoparticles of calcium phosphate (CaP) was then adopted and popularized by Jarcho et al. [4]. Currently, the above-mentioned Hayek method of synthesizing cryptocrystalline apatitic CaP powders is one of the most preferred.

Posner and co-workers [5-7] were among the first, in the mid 50’s, to realize that the mineral of natural hard tissues consisted of non-stoichiometric pseudoapatites. Posner et al. [5] envisaged in as early as 1954 that the carbonate (CO₃)- and foreign cation-free pseudoapatites can be represented by the formula of Ca_10-xH_2x(PO₄)₆(OH)₂, where the value of x would range from zero for stoichiometric hydroxyapatite (HA) to two for an apatite with a Ca/P molar ratio of 1.333. One shall here note the similarity between that pseudoapatite of Ca/P ratio of 1.333 mentioned by Posner [5] and the compound known as octacalcium phosphate (OCP, Ca₈H₂(PO₄)₆×5H₂O or more appropriately denoted as Ca₈(HPO₄)₂(PO₄)₄×5H₂O). Posner et al. [8] later showed experimentally that the mineral of the rat bones did exhibit a very strong age-dependent (from 1 to 40 days of age) variation in their Ca/P molar ratios, i.e., the younger the rat the lower the Ca/P ratio of its bones, the very young rats having a Ca/P ratio over the range of 1.15 to 1.34, in agreement with the much earlier work of Burns and Henderson [9].

Posner and co-workers [10] have been the first to describe how to prepare synthetic amorphous calcium phosphate (ACP) powders by using CaCl₂- and (NH₄)₂HPO₄-containing distilled water solutions (i.e., not containing biologically essential ions such as Na⁺, Mg²⁺, K⁺ or HCO₃^-) whose pH values were raised to around 11 by NH₄OH additions.

Betts and Posner [11, 12] postulated that ACP actually consisted of roughly spherical clusters (also called as Posner clusters) close to 1 nm in diameter, with a Ca/P molar ratio of 1.5 and the formula of Ca₉(PO₄)₆, which were free of water. Synthetic ACP, according to Posner et al. [11, 12], consisted of roughly spherical Ca₉(PO₄)₆ clusters, which formed in water and were then aggregated randomly to produce the larger spherical particles of ACP with the inter-cluster space being filled with water. A similar aggregation process was however described previously, in 1957, by Glimcher et al. [13] in relation to mineralization in the collagen-hydroxyapatite system.

ACP, when in contact with an aqueous solution, is known to exhibit the unique ability to first nucleate OCP-like nanosize crystallites on the surfaces of its particles, which would then rapidly mature into apatitic calcium phosphate [10, 12]. This property of ACP powders was successfully exploited to prepare injectable orthopedic cements [14, 15]. Posner and his co-workers were also the first to study the interaction of casein micelles of bovine milk with ACP powders [16], and this apparently led to the development of ACP-casein phosphopeptide (CPP) [17] complexes for dental remineralization applications.

Since the early studies of Hayek [1-3] and Posner [5, 10-13], to the best of our knowledge, the ACP and PCA-related literature [18-35] did not contain any novel approaches to the synthesis of ACP powders, i.e., meaning the calcium source employed in the synthesis processes was always selected from the Ca-chloride, Ca-nitrate and Ca-acetate salt group, and the pH values of the synthesis solutions were raised to the basic range (pH ~11) by the addition of strong bases such as NH₄OH, NaOH or KOH.

The current study was originated by the following questions:

(i) Could it ever be possible to synthesize CaP powders (either ACP or PCA) in aqueous solutions totally free of nitrate (NO₃), acetate (CH₃COO) or ammonium (NH₄) ions which are not shown to be present in biological bone or tooth formation processes?

(ii) Could it be possible to synthesize ACP or PCA powders by using aqueous solutions having the pH values from 9 to 12 (which was underlined by the early works of Hayek and Posner as a necessity) without even using the smallest aliquot of a strong base such as NH₄OH, NaOH, KOH or LiOH?

(iii) Could it be possible to simulate the concentrations of inorganic ions present in human blood plasma in the synthesis solutions while strictly maintaining the conditions set by the above two questions?

In CaP synthesis, nitrate or acetate ions would be introduced into the synthesis solutions by the use of calcium nitrate tetrahydrate or calcium acetate monohydrate as the calcium source. We considered that if we were using Ca metal as the only calcium source, then that would totally eliminate any nitrate or acetate ions. Layrolle et al. [27] used Ca metal shots only to produce calcium diethoxide by reacting them with ethanol, and they did not see the Ca metal shots as a possible starting material to be quite useful in ACP or PCA synthesis in aqueous solutions. It is common knowledge [36] that Ca metal is produced by electrolysis of a molten bath of calcium chloride salt, and the produced Ca metal granules react with distilled water to raise its pH under a slow evolution of H₂ gas (i.e., in situ deprotonation). We considered that the use of Ca metal as the calcium source in ACP or PCA synthesis might have eliminated the need for using of any strong bases in raising the solution pH to the levels given in the works of Hayek [1-3] and Posner [5, 10-13].

This study, to the best of our knowledge, is the first one to use Ca metal (but not salts such as Ca-chloride, Ca-nitrate or Ca-acetate) in synthesizing calcium phosphates. This study also compared the use of Ca metal (as the Ca source) to using the above-mentioned calcium salts. This study would also be the first one to synthesize ACP (or PCA) powders in synthetic mineralization solutions developed hereby to mimick the inorganic ion concentrations of human blood plasma. Human body does not use deionized or distilled water in synthesizing the mineralized portion of bone and teeth.

2. Materials and Methods

2.1 Materials

Sodium chloride (NaCl; Catalog No: 1.06404, Merck KGaA, Darmstadt, Germany), potassium chloride (KCl; Cat. No: 1.210517, Merck), magnesium chloride hexahydrate (MgCl_2∙6H₂O; Cat. No: 459331, Carlo Erba Reagenti, Milano, Italy), sodium bicarbonate (NaHCO₃; Cat. No: 1.06329, Merck), and disodium hydrogen phosphate anhydrous (Na₂HPO₄; Cat. No: 1.06586, Merck) were used in solution preparation.

Calcium metal (Ca; spherical granular, 2-4 mm in diameter; Cat. No: 1.02053, Merck), calcium chloride dihydrate (CaCl₂∙2H₂O; Cat. No: 1.02382, Merck), calcium nitrate tetrahydrate (Ca(NO₃)₂×4H₂O; Cat. No: 1.02121, Merck), calcium acetate monohydrate (Ca(CH₃CO₂)₂·H₂O; Cat. No: 402850, Sigma-Aldrich), and calcium hydroxide (Ca(OH)₂; Cat. No: 1.02110, Merck) were separately tested as the sources of calcium.

In some experiments diammonium hydrogen phosphate ((NH₄)₂HPO₄; Cat. No: 1.01207, Merck) was tested as the source of phosphorus instead of Na₂HPO₄. Finally, ammonium hydrogen carbonate (NH₄HCO₃, Cat. No: 1.01131, Merck) was also tested to replace NaHCO₃ in some experiments.

2.2 Solution preparation and synthesis

The solutions were prepared in 500 mL-capacity PyrexÔ glass bottles (Fisher Scientific, Cat. No: FB800500). The bottles were first cleaned by washing with 5 vol% HCl, followed by rinsing with an ample amount of doubly-distilled water, and overnight drying at 90°C. Five hundred mL of doubly-distilled water was first placed into the bottles at room temperature (RT, 22±1°C). A Teflon^®-coated (25 mm long, 5 mm in diameter) rod-shaped magnetic stirrer was then placed into the experiment bottle. All of the synthesis experiments were performed on a magnetic stir-plate and the stirring rate for all experiments was kept constant at 750 rpm. The carefully weighed chemicals were added, one by one, to the bottle, under constant stirring of the solution inside. The next chemical was not added prior to the complete dissolution of the previous one. Table 1 shows the procedure of preparing the synthesis / mineralization solutions (MS) in 500 mL distilled water (not boiled prior to use to remove any possible HCO₃^-). The chemicals were added to water in the order given. Table 1 offered three choices of solution preparation to the reader; the first one would lead to preparing a solution with 10 mM HPO₄^2-, whereas the third would result in a solution with 1 mM HPO₄^2-. All solutions shown in Table 1 were fully transparent at the time of preparation, and thus they were ready for the addition of the pre-weighed amount of Ca metal (or calcium chloride, calcium acetate monohydrate or calcium nitrate tetrahydrate in a limited number of experiments).

Table 1 Preparation of mineralization solutions (MS) 500 mL H₂O basis

Chemical g mM cation mM anion

KCl 0.1865 5 K⁺ 5 Cl^-

MgCl₂×6H₂O 0.1525 1.5 Mg²⁺ 3 Cl^-

NaCl 2.7760 95 Na⁺ 95 Cl^-

NaHCO₃ 1.1341 27 Na⁺ 27 HCO₃^-

Choices:

(1) Na₂HPO₄ 0.7098 20 Na⁺ 10 HPO₄^2-

(2) Na₂HPO₄ 0.3549 10 Na⁺ 5 HPO₄^2-

(3) Na₂HPO₄ 0.0710 2 Na⁺ 1 HPO₄^2-

________________________________________________________________

To further clarify the solution preparation technique described in Table 1; one first adds KCl to 500 mL of water, dissolves it, then performs the respective additions of MgCl₂×6H₂O, NaCl and NaHCO₃. At that moment, the solutions contain 5 mM K⁺, 1.5 mM Mg²⁺, 103 mM Cl^-, and 27 mM HCO₃^-. These ion concentrations are identical with those of the blood plasma. If one were then adding 0.7098 g of Na₂HPO₄, the solution would have a total Na⁺ ion concentration equal to 142 mM. This concentration of Na⁺ is exactly that of the blood plasma. The solution thus obtained according to the choice-1 of Table 1 was able to match the Na⁺, K⁺, Mg²⁺, HCO₃^-, Cl^- concentrations of the blood plasma, but will possess 10 times the HPO₄^2- concentration of plasma. However, the solution of choice-3 (of Table 1) will have the identical HPO₄^2- concentration with that of blood plasma.

If one were using CaCl₂×2H₂O as the calcium source (instead of Ca metal), it would not be possible to maintain the proper Cl^- ion concentration in the solution, i.e., it would have been in excess of 103 mM. Blood plasma contains exactly 103 mM Cl^-. If one were using Ca(NO₃)₂×4H₂O as the calcium source, then the synthesis medium would have contained nitrate ions, which are not present in the blood plasma. The same applies to the use of Ca-acetate, as well.

Powder synthesis began instantly by the addition of prescribed amount of calcium metal granules into the mineralization solutions stirred at 750 rpm. Reactions were continued for 25 minutes at RT (22±1°C). pH values were recorded (pH meter, Model: S40, Mettler-Toledo, w/combined pH-temperature electrode), at every 30 seconds, starting from the moment of adding Ca metal into the solutions. At the end of 25 minutes of stirring, the formed solids were immediately and quickly filtered out of their mother liquors by using a Whatman No. 2 filter paper via a Buechner funnel apparatus, backed up with a mechanical vacuum pump. The solid residues were washed with 750 mL of distilled water and then dried on watch-glasses at RT for 48 hours in an air-ventilated drying cabin. In the duplicate experiments, samples were synthesized once more as described above, but then left in the solutions overnight (i.e., at least 17 h), in the bottles, at RT. The pH values of the solutions were measured once again after that long period of RT ageing and exactly the same values were found with those measured after only 25 minutes of reaction.

2.3 Sample characterization

Prior to powder X-ray diffraction (XRD) and Fourier-transform infrared spectroscopy (FTIR) analyses, the dried samples were ground, manually, in an agate mortar by using an agate pestle. XRD runs were performed (Advance D8, Bruker, Karlsruhe, Germany) in the step scan mode, with the step size of 0.02° and preset time of 5 seconds. The powder diffractometer was equipped with a Cu tube and operated at 40 kV and 40 mA. XRD samples were prepared by gently packing the powders into the sample holder cavity of around 1 mm-deep. FTIR samples were mixed with KBr powders at the ratio of 1 mg sample-to-250 mg KBr in an agate mortar. FTIR pellets of 13 mm diameter were pressed at 10 tons. FTIR data were collected (Spectrum One, PerkinElmer, Waltham, MA) by using 256 scans. Scanning electron microscopy (Vega-3, Tescan, A.S., Brno, Czech Republic) samples were not ground and the small sample chunks were sputter-coated with a thin gold layer before imaging.

3. Results and Discussion

Until this study, the researchers working in CaP-based biomaterial synthesis have chosen either one of the following as their calcium source; calcium chloride (anhydrous or dihydrate), calcium nitrate (tetrahydrate), calcium acetate (monohydrate), calcium carbonate or calcium hydroxide. The former three of these have significant solubility even in cold water, but the latter two had much lower solubilities in comparison to the former. The major drawback associated with the use of calcium nitrate or calcium acetate was that the synthesized phases would be poisoned by the residual nitrate or acetate ions, as it was experimentally proven by Ivanova et al. [37].

One of the novelties of this study is that it used metallic calcium as the calcium source. Metallic calcium presented clear advantages: (i) it did not bring into the synthesis solutions any foreign or spectator anions, such as nitrates, chlorides or acetates, and (ii) it caused in situ deprotonation of the aqueous synthesis media resulting in a smooth and rapid pH increase (as shown below), totally eliminating the need for base (NaOH, KOH, LiOH, NH₄OH, etc.) additions to maintain the synthesis pH above neutral. These two points further define the novelty of this study. The necessity of maintaining the solution pH much above the neutral during the synthesis of either ACP (amorphous calcium phosphate) or PCA (poorly-crystallized, cryptocrystalline apatitic calcium phosphate) was well-proven throughout the previous work of Hayek [1-3], Posner [5-7] and Rey [14, 15]. Rey et al. [15], for instance, prepared their ACP powders by mixing an aqueous solution of Ca-nitrate with another solution containing disodium hydrogen phosphate, 1 M NaOH and 0.3 M NaHCO₃. The second solution [15] had a very high Na⁺ concentration (i.e., 1.3 M). In stark contrast, the Na⁺ concentration of the solutions of the current study was kept constant at only 142 mM, which is the sodium concentration of human blood plasma. Reaching pH values in the vicinity of 9 to 12 by using such low concentrations of Na⁺ is another advantage of the use of Ca metal.

3.1 Synthesizing CaCO₃ by using metallic Ca

The starting point of this study was to find an answer to the following simple question: what happens if one adds 25 mM (i.e., 10 times the calcium concentration of blood plasma) of Ca metal granules (i) into water, (ii) into saline (NaCl-, KCl- and/or MgCl₂×6H₂O-containing) water, or (iii) into carbonated (HCO₃^—containing, but no chlorides) water, and then stir the granules at RT in these solutions for only a finite time, such as 25 minutes? The experiments detailed in Table 2 summarized the design of this study.

Calcium granules stirred in doubly-distilled water for 25 minutes (with a rise in solution pH to around 12) were not dissolved (Experiment-1 of Table 2), they rather seemed to be rapidly covered with a white layer consisting of a biphasic mixture of Ca(OH)₂ and CaCO₃, as determined by their XRD data given in Figure 1a. XRD data, only in this case, were collected from the as-recovered granules, without attempting to crush them. One can further speculate here that the incident x-rays would not be able to pass through the hydroxide-carbonate layer formed on the granules to reach their still metallic cores.

25 mM of calcium granules stirred for 25 minutes in an aqueous solution containing only 5 mM K⁺, 1.5 mM Mg²⁺ and 27 mM HCO₃^- did not totally dissolve. The Cl^- ion concentration of this solution was equal to 8 mM (Experiment-2 of Table 2), but the K⁺ and Mg²⁺ concentrations were equal to that of blood plasma. Ca granules did not dissolve in distilled water (Exp-1), and they also did not totally dissolve in a solution containing 8 mM Cl^- and 27 mM HCO₃^- (Exp-2). In these two experiments, the rapid formation of a biphasic layer of Ca(OH)₂ (major phase) and CaCO₃ (minor phase) on the surfaces of the Ca metal granule was observed.

In Exp-3 (Table 2), 25 mM calcium granules were stirred in distilled water containing only 27 mM Na⁺ and 27 mM HCO₃^- (no Cl^-). Granules did not dissolve. Very small amounts of solution precipitates formed in experiments 2 and 3 proved, by XRD and FTIR, to be single-phase CaCO₃.

Cl^- concentration was increased to 95 mM in Exp-4. 25 mM of calcium granules stirred in an aqueous solution (Exp-4) containing 122 (=95+27) mM Na⁺, 95 mM Cl^-, and 27 mM HCO₃^- were dissolved completely and produced quite a significant amount of CaCO₃ precipitate in the solution in 25 minutes. We have thus experimentally determined that there seemed to be a close relationship between the complete dissolution of the Ca metal granules and Cl^- concentration of the solution into which they were placed. Ca metal granules added into aqueous solutions caused the evolution of H₂ gas (i.e., in situ deprotonation [36]), but that gas evolution slowed down by the formation of a hydroxide layer on the granule surfaces at low Cl^- concentrations. Moreover, since the granule size used in this study was 2 to 4 mm, that gas evolution was not so fierce.

We speculate that in solutions containing increased amounts of Cl^-, H₂ gas evolving at the granule surfaces was creating a microenvironment rich in HCl which could help to prevent the formation of the Ca(OH)₂ layer, and with an increase in Cl^- concentration from 0 (Exp-1) to 8 (Exp-2), then to 95 mM (Exp-4), the granules were dissolving in increasing amounts.

Experiment-5 was similar to experiment-4 but the MS solution (see Table 1) of Exp-5 also contained K⁺ (5 mM), Mg²⁺ (1.5 mM), HCO₃^- (27 mM) and Cl^- (103 mM) ions at exactly the human blood plasma levels. Half a gram of starting Ca granules was completely dissolved and produced CaCO₃ precipitates (1.228 g) at a high process yield (98.15% of theoretical). The XRD data of the samples of experiments 4 and 5 (not shown) indicated CaCO₃ of relatively high crystallinity, individual XRD datum being indistinguishable from one another. However, the FTIR data of CaCO₃ produced in MS solution (Exp. 5) was showing the O-H stretching vibration at around 3700 cm^-1, as indicated in Figure 1b. Based on observing the IR band at 1083 cm^-1, presence of very small amounts of vaterite may be suspected, although XRD data did not show this phase. The photomicrographs of the starting Ca granules and the calcite precipitated in Exp. 5 were given in Figures 1c and 1d, respectively. The calcite crystals formed by adding 25 mM calcium granules into the MS solution (Exp. 5) had a mean particle size of around 5 mm, exhibiting a high degree of agglomeration, displayed nanosize steps and kinks reminding a diffusion-controlled crystal growth kinetics on their surfaces, and by this way, they differed from the clean and smooth-surfaced rhombic morphology of calcite synthesized in distilled water as reported by Matijevic et al. [38]. The first five experiments (of Table-2) also helped to explain why an aqueous solution with a Cl^- concentration close to 100 mM was needed for use with the Ca metal granules/shots. Human blood plasma contains 103 mM Cl^-, therefore, the findings of the first five experiments were also indicating us the way to develop a solution mimicking the ion concentrations of human blood plasma. The MS solutions of this study are not SBF (synthetic body fluid) solutions since they did not contain any Tris or Hepes, which are not present at all in the human metabolism.

Figure 2 depicted the pH-time curves of the CaCO₃ synthesis experiments by using Ca granules. Ca metal granules were completely dissolved in experiments 4 and 5 at exactly the 11^th minute. However, this specific time of dissolution would surely depend on the stirring speed (750 rpm) employed, as well as the volume and geometrical shape of glass bottles in which the reactions were performed throughout this study.

All of the above solutions and numbers may seem somewhat complicated at the first sight but they actually point to a very simple fact, which could most probably be explained by the below equations.

Ca(s) + H₂O(l) ® Ca²⁺(aq) + 2OH^- (aq) + H₂ (g) (1)

Ca(s) + 2H₂O(l) ® Ca(OH)₂(s) + H₂ (g) (2a)

Ca²⁺(aq) + HCO₃^- (aq) ® CaCO₃(s) + H⁺ (aq) (2b)

Ca(OH)₂(s) + H⁺ (aq) + Cl^- (aq) ® Ca²⁺ (aq) + HCl + 2OH^- (aq) (3).

Equation-1 explains the evolution of H₂ gas and the observed rise in pH upon adding the calcium granules into the solutions. Equations (2a) and (2b) explain why the Ca granules did not dissolve in doubly-distilled water, and why the XRD data of Figure 1 showed Ca(OH)₂. Calcium hydroxide, Ca(OH)₂, is extremely prone to conversion at its surface to calcite (CaCO₃), and even many “pure,” commercial Ca(OH)₂ powders have measurable amounts of CaCO₃ in them, which can be readily confirmed by a simple FTIR run to be performed on those so-called pure and brand-new Ca(OH)₂ samples. Equation 3 explains why the Ca-metal granules readily dissolved in blood plasma-like, mineralization solutions (MS), containing significant amounts (103 mM) of Cl^- ions, in such a short time by causing such a rapid rise in pH. There shall be a strong similarity between the behavior of magnesium metal [39-43] and calcium metal in this respect.

3.2 Synthesis of CaP in HCO₃^--free solutions by using Ca metal

Ca metal shots/granules were not expected to fully react in water only containing HPO₄^2- ions. In other words, in the absence of Cl^- ions, the granules would be easily covered with Ca-hydroxide and/or Ca-carbonate and would stop reacting. This expectation was tested in experiment 6 (Table 2). 25 mM of calcium granules stirred in water only having 10 mM Na₂HPO₄ did not dissolve completely, but the pH of the solution was able to rise above 12 and the small amount of precipitates formed were found, by XRD (Figure 3a), to be comprised of biphasic mixtures of cryptocrystalline apatitic CaP (PCA) and calcite.

Experiments 7 and 8 were performed to study the effect of Ca/P molar ratio, i.e., 1.667 and 2.50, in reacting Ca granules with the MS solutions free of HCO₃^- ions. Both of these experiments produced cryptocrystalline apatitic calcium phosphate (PCA) samples in solutions with final pH values greater than 12 (Figures 3a and 3b), without any calcite. It was important to notice the characteristic stretching vibration of the O-H group at 3571 cm^-1 in the IR data (Fig. 3b) of the sample of Exp-8. Carbonates detected in the samples of Figures 3a and 3b were due to the small amounts of dissolved HCO₃^- present in the distilled water (not previously boiled) used. Calcium granules reacted completely by the end of the 11^th minute as shown in Figure 3c.

The MS solutions of experiments 7 and 8 had 115 mM Na⁺, 103 mM Cl^-, 5 mM K⁺, 1.5 mM Mg²⁺ and 10 mM HPO₄^2-, and in both experiments one would be able to freely change the Ca content without disturbing the concentration of any other ion in the solution; i.e., another advantage of using Ca metal in CaP synthesis. This would not be possible if one were using, for instance, CaCl₂×2H₂O as the calcium source.

Experiments 7 and 8, therefore, showed a simple way of producing cryptocrystalline (some call it poorly-crystalline or poorly-crystallized or nanocrystalline) apatitic CaP powders at RT, in a very short 25 minutes, without employing any external pH control technique (such as drop-wise addition of a strong base such as NH₄OH, NaOH, KOH,or LiOH) at an in situ solution pH of 12. Exp-8 had the nominal, solution Ca/P molar ratio of 2.5, which was equal to that of blood plasma. Bacteria cannot grow at a solution pH of 12, but they definitely can if the synthesis solutions were at neutral pH (6.8 to 7.6). This is another advantage of using Ca metal in PCA synthesis.

Experiments 9 and 10 (Table 2) were replacing the Na₂HPO₄ used in experiments 7 and 8 with (NH₄)₂HPO₄, while keeping all the other synthesis parameters unchanged. Although the presence of NH₄ ions in a synthesis system claiming to mimic the ions and ion concentrations in blood plasma would not be acceptable, experiment 9 produced amorphous calcium phosphate (ACP) at the Ca/P molar ratio of 1.667 and the final pH value of 11.3.

It was quite easy to distinguish between the ACP and PCA phases by using their FTIR data, as exemplified by the IR traces of experiments 9 and 7 in Figure 3b, respectively. In the IR data of ACP samples the phosphate bands over the range of 660 to 490 cm^-1 do not show that splitting, which was otherwise observed in PCA samples. When the Ca/P molar ratio was increased to 2.5 in experiment 10, the produced powders were not ACP but PCA. The solution pH in this experiment was 12. Upon repeating the experiment 9, but ageing the formed precipitates in the mother solution for 5 days at RT (solution pH dropping to 10.7 from 11.3, in 5 days), followed by filtering and drying, the obtained powders were consisted of PCA, not ACP, as shown in Figure 3a. This was quite an expected result since ACP was not a stable phase (even in its mother liquor over a period of 5 days) and it acted as a precursor to PCA, as previously shown by Cazalbou et al. [44].

3.3 Synthesis of CaP in HCO₃^--free solutions by using CaCl₂×2H₂O instead of Ca metal

Upon replacing the Ca metal with CaCl₂×2H₂O, the pH values of synthesis solutions drastically suffered from this change. Experiment 11 in comparison to experiment 6 showed that drastic drop in solution pH from 12.3 to 5.9. At such a low pH (5.9), it was inevitable to form DCPD (dicalcium phosphate dihydrate; brushite; CaHPO₄×2H₂O). The comparison of the XRD and FTIR data of experiments 6 (with Ca metal in water) and 11 (with Ca-chloride in water) was given in Figure 4.

Experiments 12 through 15 tested the formation of calcium phosphates in water and HCO₃^--free MS solutions by using Ca-chloride and diammonium hydrogen phosphate as the starting chemicals. In these experiments solution pH values remained between 5.7 and 6.5, and the obtained precipitates contained DCPD as the major phase.

If one used Ca-chloride dihydrate instead of Ca metal, as the calcium source, to synthesize CaP in HCO₃^- -free plasma-like solutions, mildly acidic DCPD would be the major phase obtained. The readers shall compare the synthesis conditions and the results of experiments 8, 10 and 13 with one another.

3.4 Synthesis of ACP in MS solutions by using Ca metal

Experiments 16 through 18 tested the synthesis conditions closest to the ionic concentrations of the human blood plasma, by using the metallic Ca granules. In experiment 16; calcium, phosphate (HPO₄^2-), bicarbonate (HCO₃^-), potassium, chloride, and magnesium ion concentrations were made identical with that of blood plasma, but in that experiment the sodium concentration was equal to 124 mM. In experiment 18, on the other hand; bicarbonate (27 mM), sodium (142 mM), magnesium (1.5 mM), potassium (5 mM) and chloride (103 mM) ion concentrations were identical with that of blood plasma. In other words, experiments 16 through 18 tested the MS solutions given in Table 1 under three different choices. The combined XRD and FTIR data of the resultant ACP samples were given in Figure 5a. The second inset of Figure 5a confirmed the absence of the octacalcium phosphate (OCP, Ca₈(HPO₄)₂(PO₄)₄×5H₂O) phase in the samples of experiments 16 to 18. At such high solution pH values it would be very difficult, if not impossible at all, to observe acidic OCP. The sample of experiment 18 showed the presence of a small amount of calcite (CaCO₃) phase in its XRD data. However, when we duplicated experiments 16 through 18, and left the precipitate-containing solutions overnight without stirring, followed by filtering and drying, the resultant XRD data of especially experiment 18 did not show that second phase of calcite. All three samples (16 through 18) depicted the characteristic XRD pattern of ACP. The Ca metal granules in experiments 16 through 18 all dissolved/disappeared at around the 11^th minute. When experiment 18 is performed (i.e., experiment 19) in doubly-distilled water (containing 10 mM HPO₄^2-, 27 mM HCO₃^-, and 47 mM Na⁺), instead of the MS solution, Ca metal granules did not dissolve and no precipitates were obtained. This again proved the role of Cl^- ions, as explained by equations (1) through (3) above.

Figure 5b showed the pH-time curves for experiments 16 through 19. The curves for experiments 16 through 18 in this figure, as well as the previous pH-time curves (Fig. 3c), exhibited a nonlinear increase of pH in a time dependent manner and they were approximated (TableCurve, v1.10, Jandel Scientific, 1993) by the logistic dose-response function (y = a + [b / (1 + (x/c)^d)]), for which the experimental parameters were given below in Table 3.

The SEM photomicrographs of samples obtained from experiments 16 and 18, were given in Figures 5c and 5d, respectively. It should be noted that these are filtered and dried samples, they were not even lyophilized upon separation from their mother liquors. Regular drying causes agglomeration of individual particles or moieties.

Table 3 Results of logistic dose-response curve fitting on the pH-time curves

__________________________________________________________________________

Parameters Exp 16 Exp 17 Exp 18 Exp 8 Exp 5

a 8.2495 8.3653 8.4338 12.6960 8.4163

b 0.9421 2.0039 3.5657 -3.7066 3.9064

c 0.7841 0.7827 1.3204 0.0222 1.4376

d -1.9020 -1.3032 -1.7912 0.4083 -1.8735

r² 0.9994 0.9985 0.9936 0.9794 0.9922

Fit Std Error 0.0043 0.0139 0.0608 0.0739 0.0760

_________________________________________________________________________

Link to a characteristic plot of solution pH versus time → http://www.cuneyttas.com/image001.jpg

Nevertheless, it was apparent from Figures 5c and 5d that the average particle diameter in these x-ray amorphous, carbonated and mesoporous CaP powders was pretty much less than 70 nm. This is the particle size directly observed by the SEM, not the crystallite size. Crystallite sizes cannot be determined by using the Scherrer equation while using the XRD data of x-ray amorphous samples (Fig. 5a).

Link to a typical XRD plot of ACP powders → http://www.cuneyttas.com/image002.jpg

The concentration of Ca metal added into the MS solutions (starting from 2.5 mM in experiment 16 and going up to 25 mM in experiment 18) was found to be quite influential on the final pH values attained in syntheses. When the Ca concentration was kept equal to that of the blood plasma (i.e., 2.5 mM in exp. 16), the pH of the solution has risen only to 9.2 and stabilized at that value. By increasing it to 12.5 mM (i.e., 5 times that of plasma in exp. 17), the pH rose to 10.3, and the pH increased to 12 when the Ca concentration in the MS solution was increased to ten times that of the blood plasma (exp. 18).

Link to a characteristic FTIR plot of the ACP powders obtained by using Ca metal → http://www.cuneyttas.com/image003.jpg

The conditions of Exp-16 was of pivotal significance for this study, since the Ca²⁺, HPO₄^2-, HCO₃^-, Mg²⁺, K⁺, Cl^- concentrations of this experiment were identical with those of human blood, and moreover, no foreign ions such as nitrate, ammonium and acetate were introduced to the synthesis process. As shown by the data of Fig. 5b, maintaining a literally constant pH in CaP synthesis, without employing any pH control (such as adding bases or acids to keep the pH constant), was never shown before to be possible. These define the novelty and practicality of the approach of using Ca metal as the sole calcium source in CaP synthesis.

Link to the SEM photomicrograph of nanosize ACP spheres obtained by using Ca metal → http://www.cuneyttas.com/image004.jpg

3.5 Synthesis of ACP in MS solutions by using Ca metal, ammonium phosphates and ammonium carbonate

The influence of the use of (NH₄)₂HPO₄ and NH₄HCO₃ salts, instead of Na₂HPO₄ and NaHCO₃ was also tested in synthesizing CaP powders by using Ca metal granules. Such a direct comparison was necessary. Experiments 20 through 24 (of Table 2) all produced ACP powders in MS solutions. The use of Na-phosphate or Na-bicarbonate (as shown in experiments 20 and 21) kept the solution pH at above 10, but when both of Na₂HPO₄ and NaHCO₃ were replaced by (NH₄)₂HPO₄ and NH₄HCO₃ the solution pH values dropped to about 9.3 to 9.5 (experiments 22 through 24). Of course, the solutions used in these experiments could not mimic the physiological solutions, since they contained significant amounts of ammonium ions which are not found in blood plasma. The XRD and FTIR data of experiments 22 through 24 were shown in Figure 6. However, the direct comparison of Exp-18 and Exp-24 would yield that it would be possible to produce carbonated ACP powders, by using Ca metal, at pH values of 12 and 9.5, respectively, without using any external pH adjustment controls.

3.6 Synthesis of PCA in MS solutions at pH 7 without using Ca metal

Experiments 25 through 30 of Table 2 studied the synthesis of CaP in MS solutions, without using Ca metal. These experiments were planned to show what difference the use of Ca metal would really cause in comparison to the more commonly preferred calcium ion sources, such as CaCl₂×2H₂O, calcium acetate monohydrate (Ca(CH₃CO₂)₂·H₂O), Ca(NO₃)₂×4H₂O, and Ca(OH)₂. Figure 7 showed the XRD traces of samples obtained in experiments 25 through 29, all indicating PCA. The inset in Figure 7, on the other hand, exhibited the IR traces of the samples of experiments 25 through 27. The IR traces of experiments 26, 28 and 29 were very similar to one another, and they all exhibited much less carbonate ion presence (according to the qualitative IR data) in comparison to, for instance, the sample of experiment 27.

Link to the characteristic XRD plot of PCA powders synthesized by using Ca metal → http://www.cuneyttas.com/image005.jpg

MS solutions were working perfectly well, at the stated ion concentrations, in providing a reaction pH of exactly 7.0 for Ca-chloride, Ca-acetate, or Ca-nitrate; without a need for any external pH adjustments by acids or bases of any kind. Ca metal granules made it possible to synthesize ACP or PCA powders at pH values higher than 7.0, without needing any base additions for pH control, in MS solutions.

Link to the typical FTIR plot of PCA powders → http://www.cuneyttas.com/image006.jpg

To synthesize PCA by using Ca metal granules, we found that one needed to eliminate HCO₃^- from the MS solutions. Using CaCl₂×2H₂O in doubly-distilled water or HCO₃^--free MS solutions containing phosphate ions, without any pH adjustments, would never allow the synthesis of PCA, since the pH of the solutions was lower than neutral (i.e., 7) and would thus only be suitable for the crystallization of brushite (CaHPO₄×2H₂O) phase, as also shown in this study.

Links to characteristic SEM photomicrographs of PCA powders synthesized by using Ca metal → http://www.cuneyttas.com/image007.jpg

http://www.cuneyttas.com/image008.jpg

http://www.cuneyttas.com/image009.jpg

3.7 Ca metal granules or Ca(OH)₂ in MS solutions?

XRD and FTIR analysis of the sample obtained in experiment 30 (Table 2), which opted for 25 mM Ca(OH)₂ to be added into the typical MS solution of this study, tried to provide an answer to the question of this section. Figure 8a compared the XRD traces of all the samples of this study which comprised of a biphasic mixture of ACP and CaCO₃ after 25 minutes of stirring at RT in the MS solutions. The main comparison should actually be made between the sample 18 (25 mM Ca) and sample 30 (25 mM Ca(OH)₂) in the chart of Figure 8a, since ammonium ions were present in the solutions of sample 20 and 21. Solution-wise, samples 20 and 21 do not compare well with those of samples 18 and 30. When Ca metal in experiment 18 was replaced by Ca(OH)₂ in experiment 30, while keeping all the other synthesis parameters constant, the amount of the secondary phase of CaCO₃ significantly increased (Fig. 8a). The FTIR data of the same experiments were given in Figure 8b. Figure 8b provided the evidence that the sample of experiment 30 was also poisoned with unreacted Ca(OH)₂, i.e., presence of the Ca(OH)₂-specific IR band recorded at around 3650 cm^-1. Moreover, the sample of experiment 30 showed the characteristic IR bands of the calcite phase at 2513, 1798, 875 and 712 cm^-1. In the duplicate experiments same results were obtained meaning that Ca-hydroxide was not able to completely react to form ACP in the MS solutions by fully consuming itself.

3.8 Significance of synthesizing CaP in mineralization solutions free of Tris or Hepes

Human blood, which provides the necessary nutrients to the trabecular/cancellous bones and the dentine of teeth, does not contain Tris (or Hepes), nitrate, acetate and/or ammonium ions. Therefore, it would be difficult to classify the synthesis (or coating) processes using Tris-HCl (or Hepes-NaOH) buffered solutions and especially the synthesis methods using one or more of the starting chemicals of Ca-nitrate tetrahydrate, Ca-acetate monohydrate, ammonium hydroxide, diammonium hydrogen phosphate or ammonium dihydrogen phosphate as properly mimicking the physiological processes [45-49].

Ammonium-, nitrate- and acetate-free synthesis recipes (especially those of experiments 7, 8, 16, 17 and 18) given in Table 2 of this study provided easy-to-reproduce and quite simple procedures to synthesize PCA (cryptocrystalline apatitic CaP) and ACP (x-ray amorphous CaP) powders at RT in glass media bottles, without requiring special reactor designs and pH adjustment/control measures. It would be naïve to assume that the PCA or ACP synthesized in such blood plasma-like solutions would be free of ionic substitutions of Na⁺, K⁺, Mg²⁺, CO₃^2- and Cl^- ions at the crystallographic Ca, PO₄ and OH sites of hydroxyapatite structure. In a follow up study, we will publish the results of ICP-AES (inductively-coupled plasma atomic emission spectroscopy) analyses on such samples in comparison to PCA or ACP synthesized in synthesis media free of K⁺, Mg²⁺ and Cl^- ions.

The ionic strength of the synthesis solutions (after the addition of Ca metal granules) of experiments 7, 8, 16, 17 and 18 of this study was adjusted to be 167.83, 184.5, 139.5, 171.5 and 211.5 mM, respectively. If one were to prepare an aqueous solution comprising 2.5 mM Ca²⁺, 1 mM HPO₄^2-, 142 mM Na⁺, 5 mM K⁺, 1.5 mM Mg²⁺, 27 mM HCO₃^- and 103 mM Cl^- (i.e., the exact ion concentrations of human blood plasma) then the ionic strength of that solution would have been 148.5 mM. The ionic strengths higher than 148.5 mM were intentionally chosen in this study to facilitate the synthesis of larger amounts of PCA or ACP powders.

The influence of synthesis pH on the CaP formation seemed to be not receiving the required attention in the previous literature. To the best of our knowledge, there are very few studies to mention the basicity of apatitic CaP forming in solutions with pH values around 11. The current study obtained pH values from 9 to 12 without adding any base. Liu et al. [50] used the Ca-nitrate/(NH₄)₂HPO₄ route and studied the ACP and apatitic CaP precipitation at pH 10 to 11, whereas the high pH values in that study were apparently obtained by NH₄OH additions. Liu et al. [50] study was not designed to measure the basicity of the CaP formed. The lack of previous studies on the basicity of apatitic CaP may even force the field researchers to think that apatite (which is basically a hydroxyl-containing phosphate in its formula and structure) is not a compound with a significantly basic surface, which is not true. However, the work of Tsuchida et al. [51] deliberately and quantitatively studied the surface basicity of apatitic Ca/P, by again using the Ca-nitrate//(NH₄)₂HPO₄ route of synthesis (with ammonia additions during synthesis) and found that (i) the solution pH had the greatest influence on the Ca/P ratio of apatitic CaP produced and (ii) the basic site density in apatite depended only on the Ca/P ratio of the sample. Therefore, the current study using Ca metal provided a very simple method of synthesizing CaP at the high pH values (from 10 to 12) studied separately by Liu et al. [50] and Tsuchida et al. [51].

Most of our samples produced at pH values 9 to 12 were of poor crystallinity. Nelson [52] investigated the reason for the poor crystallinity of Na⁺- and CO₃^2--containing apatitic CaP samples (usually prepared by high temperature (90°<T<250°C) processes) by using TEM and found out that the reason was not a decrease in overall particle size but the fact that each particle consisted of agglomerates of small crystalline domains each having a different orientation. The domain sizes appeared to decrease with an increase in the carbonate content and could become as small as 8 nm. Nelson [52] also found out that the simultaneous incorporation of Na⁺ and CO₃^2- ions into the apatitic CaPs resulted in an increased rate of dissolution for the solid. The formula for the Na- and CO₃-doped CaP could be as complex as Ca_10-xNax[(PO₄)_6-x(CO₃)_4x/3][(OH)_2-2x/3], where 0 £ x £ 3 [53]. A similar formula for K- and CO₃-doped CaP shall be expected. Nonstoichiometric CaP phases would usually have slightly higher aqueous solubility with respect to their perfectly stoichiometric counterparts [54].

Although the size of the Mg²⁺ ion (0.066 nm) is quite smaller than that of Ca²⁺ (0.101 nm), magnesium ions can substitute for Ca in a number of CaP phases, including whitlockite (Ca₃(PO₄)₂) [55]. The incorporation of Mg into amorphous CaP has been relatively well studied. Termine et al. [56] found that the elapsed time between the precipitation of ACP and its solution-mediated transformation into cryptocrystalline apatitic CaP (PCA) may be increased considerably with the addition of small amounts of Mg²⁺ ions. The current study was not focused on the hydrothermal transformation of ACP into PCA or vice versa, however our synthesis solutions (MS) contained Mg²⁺.

The author’s lab has been the first to synthesize cryptocrystalline apatitic CaP powders in Tris-buffered SBF (synthetic body fluid) solutions (by using Ca-nitrate) at 37°C and to show (via ICP analyses) that Mg and Na were indeed incorporated into the obtained powders [46]. Such biomimetic apatite powders were also shown to possess unprecedented high stability against thermal decomposition [46].

For readers who may ask the question of why one would need a solution pH as high as 9.2 (as in Exp-16) to synthesize CaP mimicking the physiological processes, it is a well-known fact that alkaline phosphatase (ALP) enzyme is secreted in bones by the osteoblast cells while depositing nanosize apatitic CaP crystals, and the optimum pH of ALP secretion is between 9.5 and 10.5 [57-59]. Synthesis procedures described for Experiments 16 and 17 in Table 2 were both able to produce the ACP phase at or around this biomimetic pH value of ALP secretion.

It was also shown in this study, in contrast to some previous reports, that the use of synthetic polymers, which do not have any place in the human metabolism, was not necessary at all to synthesize ACP in aqueous media [60].

Magnesium metal was already tested [39-43] as a starting material for biomedical scaffolds, the current study may initiate the use of Ca metal for the same purpose.

4. Conclusions

Metallic calcium was used for the first time in synthesizing CaCO₃, poorly-crystalline (cryptocrystalline) apatite (PCA) or x-ray amorphous calcium phosphate (ACP) powders.

Calcium phosphate synthesis with metallic Ca was tested both in doubly-distilled water and in water containing ions found in human blood.

The use of metallic Ca eliminated the need for external pH control in calcium phosphate synthesis solutions in the form of adding strong bases such as NaOH, KOH, LiOH or NH₄OH.

The use of metallic Ca made it possible to synthesize PCA or ACP powders in solutions completely free of foreign ions such as ammonium, nitrate or acetate, which are not encountered in human blood.

Notes

Certain commercial equipments, instruments, or chemicals are only identified in this paper to foster understanding. Such identification does not imply recommendation or endorsement by the author, nor does it imply that the equipment or materials identified are necessarily the best available for the purpose.

References

[1] E. Hayek, F. Mullner, and K. Koller, “Zur Kenntnis des Hydroxylapatits,” Monatsh. Chem. 82 (1951) 958-969.

[2] E. Hayek, J. Lechleitner, and W. Bohler, “Hydrothermal Synthese von Hydroxylapatit,” Angew. Chem. Int. Edit. 67 (1955) 326-326.

[3] E. Hayek and H. Newesely, “Pentacalcium Hydroxyorthophosphate,” Inorganic Syntheses, Volume VII, pp. 63-65. McGraw-Hill, Inc., 1963.

[4] M. Jarcho, C.H. Bolen, M.B. Thomas, J. Bobick, J.F. Kay, and R.H. Doremus, “Hydroxylapatite Synthesis and Characterization in Dense Polycrystalline Form,” J. Mater. Sci. 11 (1976) 2027-2035.

[5] A.S. Posner, C. Fabry, and M.J. Dallemagne, “Defect Apatite Series in Synthetic and Natural Calcium Phosphates: The Concept of Pseudoapatites,” Biochim. Biophys. Acta 15 (1954) 304-305.

[6] A.S. Posner, J.M. Stutman, E.R. Lippincott, “Hydrogen-bonding in Calcium-deficient Hydroxyapatites,” Nature 188 (1960) 486-487.

[7] M.I. Kay, R.A. Young, and A.S. Posner, “Crystal Structure of Hydroxyapatite,” Nature 204 (1964) 1050-1050.

[8] A.S. Posner, R.A. Harper, S.A. Muller, and J. Menczel “Age Changes in the Crystal Chemistry of Bone Apatite,” Ann. NY. Acad. Sci. 131 (1965) 737-742.

[9] C.M. Burns and N. Henderson, “Influence of Age on the Mineral Constituents of Bones from Pups and Kittens,” Biochem. J. 30 (1936) 1207-1213.

[10] N.C. Blumenthal, J.M. Holmes, and A.S. Posner, “Effect of Preparation Conditions on the Properties and Transformation of Amorphous Calcium Phosphate,” Mater. Res. Bull. 7 (1972) 1181-1190.

[11] F. Betts and A.S. Posner, “An X-ray Radial Distribution Study of Amorphous Calcium Phosphate,” Mater. Res. Bull. 9 (1974) 353-360.

[12] A.S. Posner and F. Betts, “Synthetic Amorphous Calcium Phosphate and Its Relation to Bone Mineral Structure,” Acc. Chem. Res. 8 (1975) 273-281.

[13] M.J. Glimcher, A.J. Hodge, and F.O. Schmitt, “Macromolecular Aggregation States in Relation to Mineralization: The Collagen-Hydroxyapatite System as Studied in vitro,” P. Natl. Acad. Sci. USA 43 (1957) 860-867.

[14] A. Tofighi, S. Mounic, P. Chakravarthy, C. Rey, and D. Lee, “Setting Reactions Involved in Injectable Cements based on Amorphous Calcium Phosphate,” Key Eng. Mat. 192-1 (2000) 769.

[15] D.D. Lee, C.Rey, M. Aiolova, and A. Tofighi, “Method of Preparing a Poorly Crystalline Calcium Phosphate and Methods of Its Use,” U.S. Patent No: 7,517,539 April 14, 2009.

[16] T.C.A. McGann, R.D. Kearney, W. Buchheim, A.S. Posner, F. Betts, and N.C. Blumenthal, “Amorphous Calcium Phosphate in Casein Micelles of Bovine Milk,” Calcified Tissue Int. 35 (1983) 821.

[17] M. Bannon, R.H. Hammond, and E.C. Reynolds, “Amorphous Calcium Phosphate-Casein Phosphopeptide (ACP-CPP) as a Dentinal Hypersensitivity Treatment Agent,” J. Dent. Res. 74 (1995) 754.

[18] H. Fleisch, R.G.G. Russell, S. Bisaz, J.D. Termine, and A.S. Posner, “Influence of Pyrophosphate on Transformation of Amorphous to Crystalline Calcium Phosphate,” Calc. Tiss. Res. 2 (1968) 49.

[19] E.D. Eanes, “Thermochemical Studies on Amorphous Calcium Phosphate,” Calc. Tiss. Res. 5 (1970) 133.

[20] A.L. Boskey and A.S. Posner, “Magnesium Stabilization of Amorphous Calcium Phosphate -Kinetic Study,” J. Dent. Res. 52 (1973) 167.

[21] R.Z. LeGeros, W.P Shirra, M.A. Miravite, J.P. LeGeros, “Biological and Synthetic Amorphous Calcium Phosphates,” J. Dent. Res. 53 (1974) 117.

[22] M.J. Glimcher, L.C. Bonar, M.D. Grynpas, W.J. Landis, A.H. Roufosse, “Recent Studies of Bone Mineral – Is the Amorphous Calcium Theory Valid,” J. Cryst. Growth 53 (1981) 100.

[23] M.S. Tung and W.E. Brown, “An Intermediate State in Hydrolysis of Amorphous Calcium Phosphate,” Calcified Tissue Int. 35 (1983) 783.

[24] H.A. Lowenstam and S. Weiner, “Transformation of Amorphous Calcium Phosphate to Crystalline Dahllite in the Radular Teeth of Chitons,” Science 227 (1985) 51.

[25] L. Brecevic, V. Hlady, and H. Furedi-Milhofer, “Influence of Gelatin on the Precipitation of Amorphous Calcium Phosphate,” Colloid. Surface. 28 (1987) 301.

[26] D. Skrtic, E.D. Eanes, and J.M. Antonucci, “Dissolution Behavior of Amorphous Calcium Phosphate Methacrylate Composites,” J. Dent. Res. 73 (1994) 302.

[27] P. Layrolle, A. Ito, and T. Tateishi, “Sol-gel Synthesis of Amorphous Calcium Phosphate and Sintering into Microporous Hydroxyapatite Bioceramics,” J. Am. Ceram. Soc. 81 (1998) 1421.

[28] A. Rodrigues and A. Lebugle, “Behavior in Wet Atmosphere of an Amorphous Calcium Phosphate with an Atomic Ca/P Ratio of 1.33,” J. Solid State Chem. 148 (1999) 308.

[29] M. Kazanci, P. Fratzl, K. Klaushofer, E.P. Paschalis, “Complementary Information on In Vitro Conversion of Amorphous (precursor) Calcium Phosphate to Hydroxyapatite from Raman Microspectroscopy and Wide-angle X-ray Scattering,” Calcified Tissue Int. 79 (2006) 354.

[30] T. Tsuji, K. Onuma, A. Yamamoto, M. Iijima, and K. Shiba, “Direct Transformation for Amorphous to Crystalline Calcium Phosphate facilitated by Motif-programmed Artificial Proteins,” P. Natl. Acad. Sci. USA 105 (2008) 16866.

[31] Z.Z. Zyman, D.V. Rokhmistrov, and V.I. Glushko, “Structural and Compositional Features of Amorphous Calcium Phosphate at the Early Stage of Precipitation,” J. Mater. Sci. Mater. M. 21 (2010) 123.

[32] H.H. Pan, X.Y. Liu, R.K. Tang, and H.Y. Xu, “Mystery of the Transformation from Amorphous Calcium Phosphate to Hydroxyapatite,” Chem. Commun. 46 (2010) 7415.

[33] D. Rabadjieva, R. Gergulova, R. Titorenkova, S. Tepavitcharova, E. Dyulgerova, C. Balarew, and O. Petrov, “Biomimetic Transformations of Amorphous Calcium Phosphate: Kinetic and Thermodynamic Studies,” J. Mater. Sci. Mater. M. 21 (2010) 2501.

[34] D. Lee and P.N. Kumta, “Chemical Synthesis and Characterization of Magnesium Substituted Amorphous Calcium Phosphate (Mg-ACP), Mat. Sci. Eng. C 30 (2010) 1313.

[35] J.L. Moreau, L.M. Sun, L.C. Chow, and H.H.K. Xu, “Mechanical and Acid neutralizing Properties and Bacteria Inhibition of Amorphous Calcium Phosphate Dental Nanocomposite,” J. Biomed. Mater. Res. B 98 (2011) 80.

[36] L. Pauling, General Chemistry, Dover Publications, New York, 1988, p. 627.

[37] T.I. Ivanova, O.V. Frank-Kamenetskaya, A.B. Koltsov, and V.L. Ugolkov, “Crystal Structure of Calcium-deficient Carbonated Hydroxyapatite; Thermal Decomposition,” J. Solid State Chem. 160 (2001) 340.

[38] L. Wang, I. Sondi, and E. Matijevic, “Preparation of Uniform Needle-like Aragonite Particles by Homogeneous Precipitation.” J. Colloid Surf. Sci. 218 (1999) 545.

[39] M.P. Staiger, A.M. Pietak, J. Huadmai, and G. Dias, “Magnesium and Its Alloys as Orthopedic Biomaterials: A Review,” Biomaterials 27 (2006) 1728.

[40] Y. Xin, K. Huo, H. Tao, G. Tang, and P.K. Chu, “Influence of Aggressive Ions on the Degradation Behavior of Biomedical Magnesium Alloy in Physiological Environment,” Acta Biomater. 4 (2008) 2008.

[41] X.N. Gu, W.R. Zhou, Y.F. Zheng, Y. Liu, and Y.X. Li, “Degradation and Cytotoxicity of Lotus-type Porous Pure Magnesium as Potential Tissue Engineering Scaffold Material,” Mater. Lett. 64 (2010) 1871.

[42] M. Tomozawa, S. Hiromoto, and Y. Harada, “Microstructure of Hydroxyapatite-coated Magnesium prepared in Aqueous Solution,” Surf. Coat. Tech. 204 (2010) 3243.

[43] J.Y. Uan, S.H. Yu, M.C. Lin, L.F. Chen, and H.I. Lin, “Evolution of Hydrogen from Magnesium Alloy Scraps in Citric Acid-added Seawater without Catalyst,” Int. J. Hydrogen Energ. 34 (2009) 6137.

[44] S. Cazalbou, C. Combes, D. Eichert, C. Rey, and M.J. Glimcher, “Poorly Crystalline Apatites: Evolution and Maturation in vitro and in vivo,” J. Bone Miner. Metab. 22 (2004) 310.

[45] D. Bayraktar and A.C. Tas, “Chemical Preparation of Carbonated Calcium Hydroxyapatite Powders at 37°C in Urea-containing Synthetic Body Fluids,” J. Eur. Ceram. Soc. 19 (1999) 2573.

[46] A.C. Tas, “Synthesis of Biomimetic Ca-Hydroxyapatite Powders at 37°C in Synthetic Body Fluids,” Biomaterials 21 (2000) 1429.

[47] E. Landi, A. Tampieri, G. Celotti, R. Langenati, M. Sandri, and S. Sprio, “Nucleation of Biomimetic Apatite in Synthetic Body Fluids: Dense and Porous Scaffold Development,” Biomaterials 26 (2005) 2835.

[48] N. Nassif, F. Martineau, O. Syzgantseva, F. Gobeaux, M. Willinger, T. Coradin, S. Cassaignon, T. Azais, M.M. Giraud-Gille, “In vivo inspired conditions to synthesize biomimetic hydroxyapatite,” Chem Mater. 22 (2010) 3653.

[49] C. Mossaad, M. Starr, S. Patil, and R.E. Riman, “Thermodynamic Modeling of Hydroxyapatite Crystallization with Biomimetic Precursor Design Considerations,” Chem. Mater. 22 (2010) 36.

[50] C. Liu, Y. Huang, W. Shen, and J. Cui, “Kinetics of Hydroxyapatite Precipitation at pH 10 to 11,” Biomaterials 22 (2001) 301.

[51] T. Tsuchida, J. Kubo, T. Yoshioka, S. Sakuma, T. Takeguchi, and W. Ueda, “Influence of Preparation Factors on Ca/P Ratio and Surface Basicity of Hydroxyapatite Catalyst,” J. Jpn. Petrol. Inst. 52 (2009) 51.

[52] D.G.A. Nelson, “The Influence of Carbonate on the Atomic Structure and Reactivity of Hydroxyapatite,” J. Dent. Res. 60C (1981) 1621.

[53] D.G.A. Nelson, G.J. Wood, J.C. Barry, and J.D.B. Featherstone, “The Structure of (100) Defects in Carbonated Apatite Crystallites- A High Resolution Electron Microscope Study,” Ultramicroscopy 19 (1986) 253.

[54] P. Koutsoukos, Z. Amjad, M.B. Tomson, and G.H. Nancollas, “Crystallization of Calcium Phosphates- Constant Composition Study,” J. Am. Chem. Soc. 102 (1980) 1553.

[55] L.W. Schroeder, B. Dickens, and W.E. Brown, “Crystallographic Studies of the Role of Mg as a Stabilizing Impurity in b–Ca₃(PO₄)₂. II. Refinement of Mg-containing b–Ca₃(PO₄)₂,” J. Solid State Chem. 22 (1977) 253.

[56] J.D. Termine, R.A. Peckauskas, and A.S. Posner, “Calcium Phosphate Formation in Vitro. II. Effects of Environment on Amorphous-Crystalline Transformation,” Arch. Biochem. Biophys. 140 (1970) 318.

[57] M. Harada, N. Udagawa, K. Fukasawa, B.Y. Hiraoka, and M. Mogi, “Inorganic Pyrophosphatase Activity of Purified Bovine Pulp Alkaline Phosphatase at Physiological pH,” J. Dent. Res. 65 (1986) 125.

[58] R. Koncki, B. Rozum, and S. Glab, “pH-metric Detection of Alkaline Phosphatase Activity as a Novel Biosensing Platform,” Talanta 68 (2006) 1020.

[59] T. Yabe, “The Effect of pH on Alkaline Phosphatase Activity in Serum of the Rat and Other Species,” Arzneimittelforschung 35 (1985) 193.

[60] Y. Li, T. Wiliana, and K.C. Tam, “Synthesis of Amorphous Calcium Phosphate using various types of Cyclodextrins,” Mater. Res. Bull. 42 (2007) 820.

Table 2 Details of select experiments

Experiment	P source	Ca source	CO3 source	P (mM)	Ca (mM)	CO3 (mM)	Final pH	Phases/XRD	Medium
1	--	Ca	--	--	25	--	12.6	Ca(OH)2+CaCO3	H2O
2	--	Ca	NaHCO3	--	25	27	9.9	CaCO3	5 KCl+ 1.5 MgCl2
3	--	Ca	NaHCO3	--	25	27	9.9	CaCO3	H2O
4	--	Ca	NaHCO3	--	25	27	12.3	CaCO3	95 NaCl
5	--	Ca	NaHCO3	--	25	27	12.3	CaCO3	MS
6	Na2HPO4	Ca	--	10	25	--	12.3	PCA+CaCO3	H2O
7	Na2HPO4	Ca	--	10	16.667	--	12.2	PCA	MS w/o HCO3
8	Na2HPO4	Ca	--	10	25	--	12.4	PCA	MS w/o HCO3
9	(NH4)2HPO4	Ca	--	10	16.667	--	11.3	ACP	MS w/o HCO3
10	(NH4)2HPO4	Ca	--	10	25	--	12.0	PCA	MS w/o HCO3
11	Na2HPO4	CaCl2×2H2O	--	10	25	--	5.9	DCPD+PCA	H2O
12	(NH4)2HPO4	CaCl2×2H2O	--	10	16.667	--	6.5	DCPD	MS w/o HCO3
13	(NH4)2HPO4	CaCl2×2H2O	--	10	25	--	6.5	DCPD+PCA	MS w/o HCO3
14	(NH4)2HPO4	CaCl2×2H2O	--	10	50	--	5.7	DCPD+PCA	MS w/o HCO3
15	(NH4)2HPO4	CaCl2×2H2O	--	10	16.667	--	6.1	DCPD+PCA	H2O
16	Na2HPO4	Ca	NaHCO3	1	2.5	27	9.2	ACP	MS
17	Na2HPO4	Ca	NaHCO3	5	12.5	27	10.3	ACP	MS
18	Na2HPO4	Ca	NaHCO3	10	25	27	12.0	ACP+CaCO3	MS
19	Na2HPO4	Ca	NaHCO3	10	25	27	9.0	No ppt^s	H2O
20	(NH4)2HPO4	Ca	NaHCO3	10	25	27	10.4	ACP+CaCO3	MS
21	Na2HPO4	Ca	NH4HCO3	10	25	27	10.1	ACP+CaCO3	MS
22	(NH4)2HPO4	Ca	NH4HCO3	6.667	16.667	27	9.4	ACP	MS
23	(NH4)2HPO4	Ca	NH4HCO3	10	16.667	27	9.3	ACP	MS
24	(NH4)2HPO4	Ca	NH4HCO3	10	25	27	9.5	ACP	MS
25	Na2HPO4	CaCl2×2H2O	NaHCO3	10	25	27	7.0	PCA	MS
26	Na2HPO4	CaCl2×2H2O	NaHCO3	10	25	27	7.0	PCA	H2O
27	(NH4)2HPO4	CaCl2×2H2O	NH4HCO3	10	25	27	7.0	PCA	MS
28	Na2HPO4	Ca-acetate	NaHCO3	10	25	27	7.0	PCA	MS
29	Na2HPO4	Ca-nitrate	NaHCO3	10	25	27	7.0	PCA	MS
30	Na2HPO4	Ca(OH)2	NaHCO3	10	25	27	11.7	ACP+CaCO3	MS

Figure Captions (see all the below Figures in http://www.cuneyttas.com/Ca-metal.pdf)

Fig 1a XRD traces of as-received Ca granules (bottom) and Ca granules stirred in H₂O or MS (top, Exp. 1)

Fig 1b FTIR traces of the samples of experiments 3, 4, and 5

Fig 1c Macrophotograph of as-received Ca metal granules (shots)

Fig 1d SEM photomicrograph of the CaCO₃ samples of experiment 5

Fig 2 pH-time curves for experiments 2, 3, 4, and 5 (the moment of dissolution of Ca granules were indicated for experiments 4 and 5)

Fig 3a XRD traces of the samples of experiments 6, 7, 8, 9, and 10 (solution pH values, at the end of 25 min of stirring, were shown on the traces)

Fig 3b FTIR traces of the samples of experiments 1, 6, 7, 8, and 9

Fig 3c pH-time curves for experiments 1, 5, and 8 (the moment of dissolution of Ca granules was indicated by the arrows for experiments 5 and 8)

Fig 4 Combined XRD and FTIR traces for the samples of experiments 6 and 11 (the bottom XRD trace for PCA of experiment 6, the XRD trace for DCPD of experiment 11 shown on top)

Fig 5a Combined XRD and FTIR traces for the samples of experiments 16, 17, and 18 (CaCO₃ peaks were indicated by + in the XRD trace of experiment 18)

Fig 5b pH-time curves for experiments 16, 17, 18, and 19 (the dissolution time of Ca granules was indicated by the straight dashed line)

Fig 5c SEM photomicrograph of the sample of experiment 16

Fig 5d SEM photomicrograph of the sample of experiment 18

Fig 6 Combined XRD and FTIR traces for the samples of experiments 22, 23, and 24

Fig 7 Combined XRD and FTIR traces for the samples of experiments 25 through 29

Fig 8a XRD traces of the samples of experiments 18, 20, 21, and 30

Fig 8b FTIR traces of the samples of experiments 18, 20, 21, and 30