Article Cite This: Cryst. Growth Des. 2019, 19, 4162−4171
pubs.acs.org/crystal
Effect of the Ionic Radius of Alkali Metal Ions on Octacalcium Phosphate Formation via Different Substitution Modes Yuki Sugiura,*,† Yasuko Saito,‡ Takashi Endo,‡ and Yoji Makita† †
Downloaded via NOTTINGHAM TRENT UNIV on August 14, 2019 at 02:39:53 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
Health Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), 2217-14, Hayashi-cho, Takamatsu, Kagawa, Japan 761-0395 ‡ Research Institute for Sustainable Chemistry, National Institute of Advanced Industrial Science and Technology (AIST), 3-11-32, Kagamiyama, Higashi-Hiroshima, Hiroshima, Japan 739-0046 S Supporting Information *
ABSTRACT: Octacalcium phosphate (OCP), a layered calcium phosphate compound, is an attractive material for new medical combination products, which are hybrid materials consisting of both drugs and biomaterials that have shown excellent therapy scores. OCP is fabricated primarily from soluble calcium salt via an aqueous-mediated hydrolysis process. The coexisting cations are likely to be incorporated into the OCP unit lattice during this process and affect the development of the OCP crystal structure and its thermodynamic stability. However, the key parameters of the coexisting cations, such as the ionic radii and charge number, that affect OCP formation are still unclear. In this study, we focused on the ionic radius of each monovalent ion of alkali metal ions (Li+, Na+, K+, Rb+, and Cs+) for OCP incorporation and formation in a CaHPO4·2H2O (DCPD)-(NH4)2HPO4 system with alkali metal salts. There was little incorporation of Li into the OCP unit lattice at low concentrations, whereas Li was incorporated into the OCP unit lattice as a conjugate form of NH4 and K, which have ionic radii that are larger than that of Ca, resulting in a HPO4−OH layered structure in the OCP unit lattice. The Na concentration increased hyperbolically and was incorporated into the OCP unit lattice, resulting in a HPO4−OH layered structure in the OCP unit lattice. By contrast, Li, K, and Rb resulted in the HPO4−OH layer in the OCP unit lattice at low concentrations and attenuated OCP formation at high concentrations. Cs only exhibited an attenuation effect on OCP formation. Considering the differences between the Ca ionic radius (1.00 Å) and those of alkali metal ions, the alkali metal ionic radii that were smaller than ∼1.5 times the Ca ionic radius could be incorporated into the OCP unit lattice, whereas those greater than ∼1.5 times the Ca ionic radius exhibited an inhibiting effect on OCP formation. In addition, Na particularly affects the OCP formation and its crystallinity unlike other alkali metal ions because of its similar ionic radius to Ca.
■
INTRODUCTION
OCP primarily consists of two-layer structures. First, it has an apatite layer, which has the same structure as hydroxyapatite [HAp: Ca10(PO4)6(OH)2], that is sandwiched between HPO4−OH layers along the a-axis.13 There are six PO4 molecules (P1−P6) in the OCP unit lattice. P1−P4 are in the apatite layer, and P5 and P6 are in the HPO4−OH layers.13 All PO4 molecules in the OCP unit lattice are conjugated to particular Ca ions in the OCP unit lattice.18 For example, P5 and its conjugated Ca ion are strongly affected by incorporating cations, likely via substitution. This phenomenon also strongly affects the OCP crystal structure, especially its layered structural development and thermal stability. Boanini et al.19 indicated that divalent cations such as Sr2+ and Mg2+ are likely to be incorporated into the OCP unit lattice and induce collapsed OCP crystal structures during heat treatment. We introduced Na+ and NH4+ in our previous studies, which must be used as pH- or ionic-strength-adjusting agents to
Layered chemical compositions are attractive materials for a variety of industrial applications, including catalysts, lubricants, batteries, electrical devices, insulation, and building materials.1−5 Taking advantage of layered chemical compositions, such as their ion carrying capacity, ionic substitution ability, and controlled release property based on their specific crystal structure, has recently resulted in these layered compositions receiving more attention as adsorbing and antibacterial materials.1,6−8 Silicate materials such as muscovite, montmorillonite, vermiculite, and imogolite, which are categorized as clay minerals, have been used primarily for these types of applications.9−12 A layered calcium phosphate composition, octacalcium phosphate [OCP: Ca8(HPO4)2(PO4)4·5H2O], has recently been introduced as an attractive material for these applications because it has not only these advanced properties that silicate minerals have but also excellent biocompatibility because it is the primary inorganic component in vertebrate hard tissue.13−17 © 2019 American Chemical Society
Received: May 20, 2019 Published: June 5, 2019 4162
DOI: 10.1021/acs.cgd.9b00656 Cryst. Growth Des. 2019, 19, 4162−4171
Crystal Growth & Design
Article
considerably affect the OCP unit lattice.20−22 However, the effects of monovalent cations, ionic radii, and ionic potential on OCP formation and its layered structure are still unclear. Alkali metals should be typical monovalent cations because they have similar physicochemical properties but different ionic radii, as shown in Scheme 1.23 It is possible to only evaluate
peak) in the obtained XRD patterns of the samples were calculated via quantitative analysis using PDXL2 software (Rigaku Co., Japan). The relative rate of the yield in each sample was calculated from the d(100) and d(200) of the OCP unit lattice at ∼4.7° and ∼9.4°, respectively. The proportion of each phase (Rx: x = phase or intensity of peaks) in the samples is given as R x = C × Ix1/Ix 2
Scheme 1. Ionic Radii of Each Alkali Metal Ion with a Coordination of Six23a
where C is the phase or diffraction coefficient. Here, we use C = 1.00. The lattice parameters of the OCP crystals in the obtained XRD patterns of the samples were determined by Rietveld analysis using PDXL2 software (Rigaku Co., Japan). In this calculation, we used the structure model of the OCP crystal structure from Brown et al.13 as the initial model. The 31P chemical shifts of the samples were determined using solidstate nuclear magnetic resonance spectroscopic analysis (solid-state NMR, Varian FT-NMR (400 MHz), Agilent Technologies Co., USA) with a resonance frequency of 161.8 MHz for 31P. For all of these measurements, cross-polarization magic-angle spinning (CP-MAS) 31 P NMR spectroscopy was performed with a CP-MAS rate of 10 kHz. An Agilent 4 mm T3 CP-MAS HXY solid probe and zirconia rotors were used. The sample weight was ∼0.02 g, and the contact time for the 31P CP-MAS measurements was 5 ms, with an acquisition time of 50 ms and relaxation delays of 40 s for each measurement interval. The number of repetitions per measurement was 400. The 31 P chemical shift of (NH4)H2PO4 was used as an external reference (δ = 1.0 ppm). The chemical bonding structure of the samples was determined via Fourier transform infrared spectroscopy (FT-IR, Nicolet NEXUS670, ThermoFisher Scientific Co., USA) using a triglycine sulfate detector (64 scans, 2 cm−1 resolution) with a GeSe attenuated total reflection prism. The measurements were obtained in an air atmosphere. The fine structure of the samples was observed via field emissionscanning electron microscopy (FE-SEM, JSM-6700F, JEOL Co., Japan) at an acceleration voltage of 5 kV after Os spattering. The Ca, P(PO4), and alkali metal ion concentrations of the samples were measured via inductively coupled plasma-atomic emission spectroscopy (ICP-AES, 5110VDV, Agilent Technology Co., Japan) after dissolving the samples in 2% HNO3. The thermal stability of the samples was determined via thermogravimetry-differential thermal analysis (TG-DTA, ThermoPlus, TG8110, Rigaku Co., Japan). The heating rate was 10 °C/min to 200 °C, and Al2O3 was used as the standard. The samples that were heated to 130 and 180 °C for XRD analysis were heated following the same approach used for TG-DTA.
The color of each ion corresponds to its flame color reaction.
a
the effect of the cation ionic radius for OCP formation. Therefore, in this study, we investigated the effect of each alkali metal ion (Li+, Na+, K+, Rb+, and Cs+) on OCP formation and its layered structure development using an acidic calcium phosphate hydrolysis process with alkali metal salts.
■
(1)
EXPERIMENTAL SECTION
Materials and Solutions. All reagents were purchased from Wako Pure Inc., Japan (reagent grade). (NH4)2HPO4 and various alkali metal chlorides (LiCl, NaCl, KCl, RbCl, and CsCl) were dissolved in distilled water to form solutions. The experimental solutions consisted of 1.0 mol/L (NH4)2HPO4 and 0−1000 mmol/L of a given alkali metal chloride. A 2.39 g sample of dicalcium hydrogen phosphate dihydrate [DCPD: CaHPO4·2H2O] (14 mmol) was immersed in 20 mL of the above-mentioned solutions at 60 °C for 1 day. The initial and final pH values of the solutions were recorded at room temperature using a pH electrode (LAQUA ToupH 9615S-10D with pH meter D-72, Horiba Co., Kyoto, Japan). The treated samples were washed with distilled water several times to remove the residual solutions and then placed in a drying oven at 60 °C for several hours. Characterization. Crystallographic information on the samples was obtained via X-ray diffraction (XRD, MiniFlex600, Rigaku Co., Japan) at an acceleration voltage of 40 kV and amplitude of 15 mA. The diffraction angle was continuously scanned over the 3−70° 2θ range at a scanning rate of 2°/min for characterization and over the 2−12° range at a scanning rate of 0.8°/min for crystallographic parameter analysis. The integral intensities of each peak (Ix: x = each
■
RESULT AND DISCUSSION White precipitates, which were characterized as β-Li3PO4, were observed during the preparation of >500 mmol/L LiCl solutions (Figure S1). However, no precipitation and no
Figure 1. Initial (a) and final (b) pH values of the treated solutions. ●: Li, ◆: Na, ■: K, 4163
▼:
Rb, and
▲:
Cs. DOI: 10.1021/acs.cgd.9b00656 Cryst. Growth Des. 2019, 19, 4162−4171
Crystal Growth & Design
Article
Figure 2. Wide-range XRD patterns of the samples obtained from the following: (a) LiCl, (b) NaCl, (c) KCl, (d) RbCl, and (e) CsCl solutions. Each number corresponds to the solution concentration (mmol/L). The reference OCP XRD pattern calculated by the RIETAN-FP-VESTA3 program from ref 13 is mentioned for comparison.32,33
color changes were observed for any of the other initial alkali metal solutions. Although it was unclear what effect these precipitates would have on OCP formation, in this study, we performed a DCPD hydrolysis experiment on this precipitate formed in solution because we believed that maintaining the total molecular amount constant in the reaction system was important rather than removing the precipitate from solution. In addition, DCPD hydrolysis experiments were conducted to dissolve the precipitates as the reaction proceeded. The initial and final pH values of the reacting solution were measured because their values would likely change during the reaction (Figure 1). The initial pH values of the >500 mmol/L LiCl solutions, which yielded white precipitates, were only ∼7.20, whereas the initial pH values of the other reacting solutions were almost constant at ∼8.10. Although the initial pH values of the >500 mmol/L LiCl solutions were lower than the others, the final pH value of each reacting solution was 6.4− 6.7, indicating weak acidic conditions. The NaCl solutions yielded final pH values that were slightly lower than those of the other solutions. The treated samples were characterized using XRD analysis, as shown in Figure 2. Each sample possessed a peak at ∼4.7° that corresponded to an OCP (100) reflection. Furthermore,
the 1000 mmol/L LiCl solution possessed several extra peaks at ∼11.5°, ∼23.4°, and ∼29.3° that corresponded to DCPD. A significant peak was observed at ∼10.5° for the alkali metal-free solution and for the 1000 mmol/L RbCl and CsCl solutions that corresponded to an HAp d(100) reflection. How can the effect of alkali metal ions on the OCP crystal structure be evaluated? Our previous studies revealed the key evaluation methods: estimating the development of an OCP layered structure, evaluation of P5 conjugated to Ca, which is likely to be replaced by other cations, and estimation of the stability of the HPO4−OH layered structure via thermal analysis.21,22 Furthermore, Jiang et al.24 indicated that a significant evolution of the powder XRD patterns occurred when a monolayer was stacked into multilayers. Therefore, first, we employed the relative d(100)/d(200) peak intensities for our analysis. The small-angle XRD patterns (2−12°) of the samples were analyzed to estimate the development of the layered structures in the OCP phase, as shown in Figure 3. Furthermore, the ratios of the peak d(100)/d(200) intensity revealed that high d(100)/d(200) ratios suggested a well-developed OCP layered structure (Figure 4). There was almost no evidence of OCP layered structure development in the
