Carbonate-Hydroxyapatite Cement: The Effect of Composition on Solubility In Vitro and Resorption In Vivo

Abstract

The rate of resorption of calcium phosphate self-hardening materials for bone regeneration can be changed by changing the phase composition. The Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system is important for the synthesis of self-curing bioactive materials with variable resorption rates by changing the ratios of the initial components. Cement compositions in twelve figurative points of a four-component composition diagram at a fixed content in the α-Ca₃(PO₄)₂ system were studied with XRD, FTIR, SEM, calorimetric, and volumetric methods to obtain an idea of the effect of composition on solubility in vitro and resorption in vivo. It was found that the presence of the highly resorbable phase of dicalcium phosphate dihydrate in cement and the substitution of phosphate ions with the carbonate ions of hydroxyapatite increased solubility in vitro and resorption in vivo. The obtained results confirm the possibility of changing the solubility of a final product in the Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system by changing the ratio of the initial components.

1. Introduction

The restoration of bone volume during defect correction is a complex surgical problem. In order to solve it, it is possible to use calcium-phosphate osteoplastic materials, the composition of which is similar to the composition of an inorganic component of bone tissue.

Bone tissue is a carbonate hydroxyapatite with a content of 6-9 wt.% carbonate ions in its structure. Obtaining matrices based on carbonate hydroxyapatite (CO₃-HA) using ceramic technology is a complex technological problem due to thermal decomposition during sintering [1,2]. Possible ways to obtain CO₃-HA are low-temperature synthesis methods achieved in aqueous solutions with CO₂ with the method of dissolution–precipitation, using pressed calcium hydroxide or α-modification of tricalcium phosphate (α-TCP) blocks as a precursor [3,4,5], or with cement technology [6,7,8,9].

Hydroxyapatite cements are relatively insoluble at neutral pH, but their solubility increases at more acidic pH. Therefore, they can be resorbed by osteoclasts during bone remodeling [10]. Calcium phosphate cements based on dicalcium phosphate dihydrate are more soluble than hydroxyapatite cements and, consequently, are resorbed faster in vivo [11,12]. Since hydroxyapatite-based cements are resorbed relatively slow, their wider use in clinical applications is prevented. Thus, increasing the degradation of hydroxyapatite cements is of paramount importance for increasing their clinical use [13].

One of the approaches for increasing the rate of resorption of cement and the formation of new bone involves their chemical modification, in particular, of ion substitutions. CO₃-HA resorption increases in direct proportion to the content of carbonate ions in a hydroxyapatite structure and it is more noticeable than during decrease in hydroxyapatite crystallinity [14,15,16,17].

The rate of resorption of a material affects the concentrations of calcium and phosphorus ions in the area surrounding the material, affecting the growth and differentiation of cells involved in bone remodeling. Extracellular stimulation by Ca²⁺ ions of osteoblasts induces higher cell proliferation depending on the concentration of Ca²⁺ [18,19,20,21,22,23]. Calcium supplementation is effective for reducing the risk of postoperative hypocalcemia [24,25]. It has a positive effect on the healing of bone fractures [26]. Low-crystalline CO₃-HA increases the osteoblastic differentiation of bone marrow cells compared to hydroxyapatite due to higher expressions of genes, including the mRNA of type-I collagen, alkaline phosphatase, osteopontin, osteocalcin [27], and osteoclastogenesis [15]. Low-crystalline CO₃-HA shows high adhesion of fibroblasts. This contributes to the restoration of connective tissue [28].

In vivo experiments have found that synthetic CO₃-HA is safe and effective. Its application leads to faster bone formation than the use of deproteinized bone tissue (biological hydroxyapatite) [17,29,30].

Another technique to increase the resorption of a bone substitute is to obtain metastable phases in the final product. Materials based on dicalcium phosphate dihydrate (DCPD) were resorbed faster in vivo, unlike materials based on apatite [31]. According to the results of experimental studies, materials based on DCPD have good biocompatibility [32,33,34].

Various researchers have reported the effectiveness of two-phase bone substitutes [35,36,37]. In these cases, various characteristics, such as the solubility and osteoconductivity of various calcium phosphates, caused improvement in the activity of osteoblasts and the formation of new bone. The issues of the formation of cements of combined composition have not been previously considered in the scientific literature.

In our opinion, in the case of a two-phase material based on CO₃-HA and DCPD, it is possible to increase the dissolution rate due to the rapid release of Ca²⁺ and PO₄³⁻ ions during the decomposition of DCPD. In the Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system, it is possible to obtain multiphase compositions with different rates of resorption.

The purpose of this work is to study the effect of different concentrations of monocalcium phosphate monohydrate Ca(H₂PO₄)₂·H₂O (MCPM), calcium carbonate CaCO₃, and sodium hydrogen phosphate 12-aqueous Na₂HPO₄·12H₂O in the Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system at a fixed content in the system (α-TCP) on the predominant compositions of solid solutions of carbonate-substituted nonstoichiometric hydroxyapatite Ca_{10- x- y/2}(HPO₄)_x(CO₃)_y(PO₄)_6-x-y(OH)_2-x and DCPD and their effects on the rate of dissolution and the rate of resorption of the resulting materials.

2. Materials and Methods

Materials. α-Ca₃(PO₄)₂ (α-TCP) was prepared through a solid-state reaction by firing a mixture of CaCO₃ (99,9% pure, Vekton, St. Petersburg, Russia) and CaHPO₄ in a molar ratio 2:1 at 1400 °C for 1 h. CaHPO₄ was obtained through the heat treatment of CaHPO₄·2H₂O (99% pure, Vekton, St. Petersburg, Russia) at 250 °C for 7 h.

α-Ca₃(PO₄)₂, CaCO₃, and Ca(H₂PO₄)₂∙H₂O (99% pure, Vekton, St. Petersburg, Russia) were mixed in required quantities (mass%) and homogenized.

The components of the system were mixed in the required quantities (mass%). Na₂HPO₄·12H₂O (99.9% pure, Vekton, St. Petersburg, Russia) was introduced to the liquid phase in the form of a solution.

X-ray diffraction measurements were performed with an ARL Equinox 1000 X-ray diffractometer (Thermo Fisher Scientific INEL SAS, Saint-Herblain, France). Intensity data were collected for reflection geometry using CuKα radiation (angular range: 2θ = 5°–68°; 2θ scan step of 0.03°).

In the qualitative analysis of the resultant X-ray diffraction patterns, we used WinXPOW software and the ICDD PDF-2 database.

In the quantitative analysis, we used the reference intensity ratio method (with the I/I_c intensity ratio of the strongest lines of the substance and corundum (α-Al₂O₃) in a mixture containing 50 wt.% of both components). The weight fraction was calculated using the following relation:

$${\omega }_A=\frac{I_{iA}/(I/I_c(A))\cdot I_{iA}^{rel}}{{\displaystyle \sum I_{iK}/(I/I_c(K))\cdot I_{iK}^{rel}}}$$

where I_iA is the measured intensity of the ith reflection from phase A; I^rel_jA is the relative intensity of this reflection in the database; I/I_c(A) is the corundum number for the phase A being determined; and I_jK, I^rel_jK, and I/I_c(K) are the corresponding quantities for all the components of the mixture (including A).

FTIR spectra were captured with a Perkin Elmer Spectrum One IR Fourier spectrometer in transmission mode in the range of 4000–400 cm⁻¹ with a resolution of 1 cm⁻¹ (5 mg/50 mg of KBr powder).

The volumetric study was based on measuring the carbon dioxide released as a result of the action of a hydrochloric acid solution of 1:1 (p = 1.19) on the cement using a calcimeter.

The CO₂ content, mg, was calculated by the following formula:

CO₂ = CV

where V is the volume of CO₂ released, ml, and C is the mass of 1 mL of CO₂ at the temperature and atmospheric pressure of the experiment, mg.

A calorimetric study was carried out with a TAM Air isothermal microcalorimeter (TA Instruments, New Castle, DE, USA). The shooting temperature was 25 °C.

Microstructures were examined with a Carl Zeiss LEO SUPRA 50 VP field emission scanning electron microscope (SEM). Specimens were coated with chromium, which was sputter-deposited using a Quorum Q150T ES automatic sputtering system.

Solubility was assessed by measuring the concentration of calcium ions in a physiological solution (0.9% solution of NaCl). The particle size of the crushed cement was 250–500 microns. Cement was immersed in a physiological solution at a solid/liquid mass ratio of 1 g/10 mL. Samples were kept in an IKA KS 3000 shaker incubator (IKA^®—Werke GmbH & Co. KG, Staufen, Germany) at 37 °C for 7 days. When selecting an aliquot, it was replaced with fresh solution. The concentration of calcium ions in the physiological solution was determined with a PE540054005400 HA spectrophotometer (Ekros-Analytics LLC, St. Petersburg, Russia) at a wavelength of 570 nm during the formation of a colored complex of calcium ions with an o-cresolphthalein complex in an alkaline medium.

In order to calculate the stoichiometric formulas of the compositions at figurative points, 1 g of pre-set cement was dissolved in 20 mL of 1n HCl solution, followed by neutralization with 1n NaOH solution. The concentration of calcium ions was evaluated with a PE540054005400 HA spectrophotometer (Ekros-Analytics LLC, St. Petersburg, Russia) at a wavelength of 570 nm during the formation of a colored complex of calcium ions with an o-cresolphthalein complex in an alkaline medium. The concentration of phosphorus ions was estimated with a PE540054005400 HA spectrophotometer (Ekros-Analytics LLC, St. Petersburg, Russia) at a wavelength of 340 nm during the formation of a phosphorous-molybdenum complex of phosphorus ions with ammonium molybdate in an acidic medium.

A preclinical assessment of biocompatibility and resorption of the obtained calcium phosphate matrices was carried out in in vivo studies on Wistar rats using a model of bone perforation of a critical size (2/3 diameter) on the tibia. In all cases, the material was implanted directly into the area of the defect and tamped into the medullary canal with a small excess outside the bone. The study was conducted on five animals. In four animals, two samples were installed (in the right and left lower extremities), while in one animal, one sample was installed. Thus, three cement compositions were studied in three repetitions.

A tomographic examination was carried out in vivo after surgery, as well as posthumously after 3 months of implantation. Microcomputed tomography (micro-CT) was performed with a Bruker SkySkan 1178 scanner (BRUKER BELGIUM, Kontich, Belgium) at a voltage of 65 kV and a current of 615 μA with a 0.5 mm A1 filter. The spatial resolution was 84 microns/pixel. Sections were reconstructed using NRecon v1.6.10.4 software. 3D reconstructions were created in FEI Avizo 9.0.1.

For histological examination, samples were fixed in neutral buffered 10% formalin, decalcificated in a special acid solution (SoftyDek, Biovitrum, St. Petersburg, Russia), subjected to standard alcohol wiring, and embedded into paraffin blocks. Paraffin section of 4 μm thick were stained with hematoxylin and eosin, as well as picrosirius red, according to the standard protocols. The obtained microscopic slides were examined with a universal LEICA DM4000 B LED microscope (Leica Microsystems Wetzlar GmbH, Wetzlar, Germany) equipped with a LEICA DFC7000 T digital video camera using standard light, phase contrast, and polarized microscopy methods.

For statistical analysis, the normality of the distribution was checked using the Shapiro–Wilk test. A one-way ANOVA and a comparative Tukey test were used. The significance level was p ≤ 0.05. Statistics were processed in Origin Pro 2021. The results were presented as average value ± standard deviation. Compact letter display was used to show the statistical difference (p < 0.05) between different materials at one time point.

3. Results

3.1. Composition of Cement at Figurative Points

In order to determine the effect of the ratio of components in the α-Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system on the final cement composition, twelve figurative points on the diagram were selected in accordance with Figure 1 and

Table 1. Content of components at figurative points.

	Quantity, wt.%
Composition	1	2	3	4	5	6	7	8	9	10	11	12
CaCO3	10	8	4	5	13	10	10	6	6	8	12	8
Ca(H2PO4)2∙H2O	3	6	10	12	1	4	6	6	8	4	4	8
Na2HPO4∙12H2O	7	6	6	3	6	6	4	8	6	8	4	4

.

3.1.1. X-ray Diffraction (XRD)

The phase compositions of the cement obtained after 30 days of hardening in wet conditions for all twelve compositions of figurative points were represented by carbonate-substituted hydroxyapatite Ca₁₀(PO₄)₄(CO₃)₂(OH)₄ [01-085-7370] (CO₃-HA), calcium-deficient hydroxyapatite Ca_8,86(PO₄)₆(H₂O)₂ (CDHA) [01-082-1943], dicalcium phosphate dihydrate CaHPO₄·2H₂O [01-072-0713] (DCPD), and β-tricalcium phosphate β-Ca₃(PO₄)₂ [01-070-2065] (β-TCP) (Figure 2). Calcium-deficient and carbonate-substituted hydroxyapatites were not marked in separate phases on the diffractogram due to the coincidence of the main peaks. The calculations were performed using the characteristic peaks at interplane distances of 2.7309 and 1.9489 for CDHA and at 2.8066 and 2.7782 for CO₃-HA during the quantitative phase analysis. The identification of calcium carbonate through an X-ray phase analysis was impossible since there were no characteristic peaks (not coinciding with the peaks of the present phases) in this system. Calcium carbonate was not taken into account when calculating the quantitative phase composition.

The results of the quantitative determination of the CaHPO₄·2H₂O phase in the cement at figurative points are presented in

Table 2. The amounts of CaHPO₄·2H₂O in cement compositions at figurative points.

Phase/Composition	Quantity, wt.%
	1	2	3	4	5	6	7	8	9	10	11	12
CaHPO4·2H2O	3.0 ± 0.3	3.4 ± 0.3	6.4 ± 0.5	5.7 ± 0.5	2.0 ± 0.3	3.7 ± 0.4	3.8 ± 0.3	4.2 ± 0.4	4.6 ± 0.4	3.3 ± 0.3	4.3 ± 0.4	4.4 ± 0.5

.

3.1.2. Fourier-Transform Infrared Spectroscopy (FTIR)

Figure 3 shows the FTIR spectra of the cement compositions at figurative points.

The most intense and high-frequency band of 1435 cm⁻¹ of the calcite spectrum was not a characteristic of the FTIR spectra of the cements. It merged with the doublet at 1409–1418 cm⁻¹ and 1449–1474 cm⁻¹, belonging to mode ν₃ of the antisymmetric stretching characteristic of CO₃²⁻ ions. These bands are characteristic of carbonate-substituted hydroxyapatites. The band at 875 cm⁻¹ was a characteristic of calcite but was hidden by a band of HPO₄²⁻ ions. Thus, the content of CO₃²⁻ ions could be inferred only by the band at 712 cm⁻¹. It referred to the asymmetric (ν₄) deformation oscillation of CO₃^2– and was absent in the spectra of the cements.

The doublet characteristic of carbonate ions contained peaks characteristic of A (substitution of OH⁻ ions) and B types (substitution of PO₄³⁻ ions) of carbonate hydroxyapatite. In it, the bands at 1418 and 1470 cm⁻¹ were characteristic only for the B type of carbonate hydroxyapatite, and the bands at 1450 and 1544 cm⁻¹ were only for the A type. In the compositions of figurative points 1, 5, 6, and 11, peaks at 1450 and 1470 cm⁻¹ were clearly distinguished, whereas in other compositions, the peaks merged into one wide peak.

Peaks at 600 and 559–560 cm⁻¹ were characteristic of orthophosphoric tetrahedral ions in a hydroxyapatite-type structure. They were wide, merging into a doublet. Peaks at 578 and 631 cm⁻¹ correspond to well-crystallized hydroxyapatite and were absent in the spectra.

Fluctuations in the HPO₄²⁻ anion in the range of 1400–1750 cm⁻¹ were characteristic. In connection with the superposition of δ (POH) oscillations on the fluctuations of ν₃ (CO₃), we considered the fluctuations in the range of 1600–1750 cm⁻¹ as decisive. Peaks at 1642 cm⁻¹ were clearly defined on the spectra, which determined the ions of HPO₄²⁻ calcium-deficient hydroxyapatite and dicalcium phosphate dihydrate.

A semiquantitative analysis based on comparing the concentrations of the same ions in different compositions was carried out using normalized FTIR spectra by finding the integral in the range of wave numbers from 1350 to 1525 cm⁻¹. The dependence of the amount of carbonate ions on the initial compositions of cements was determined with the following formula:

$$K_{{CO}_3^{2-}}={\int }_{1350}^{1525}\frac{dI}{d\lambda }$$

The intensity integrals from 1350 to 1525 cm⁻¹ are presented in

Table 3. Integral values of peak intensity in the range of wavenumbers from 1350 to 1525 cm⁻¹.

Composition	Area, Conv. Unit2	Composition	Area, Conv. Unit2
1	3745	7	3282
2	2388	8	1575
3	1139	9	2589
4	1000	10	2439
5	5158	11	4871
6	3505	12	1953

.

3.1.3. Volumetric Study

The amount of carbonate ions was determined using the volumetric technic with the determination of the volume of released CO₂ when the product was exposed to a solution of hydrochloric acid. The concentrations of CO₃²⁻ ions in the final compositions at figurative points are presented in

Table 4. Concentrations of CO₃²⁻ ions in the final compositions at figurative points.

Composition	CO32−, mg/1 g Cement	Composition	CO32−, mg/1 g Cement
1	56.37	7	50.37
2	41.00	8	23.06
3	20.33	9	30.22
4	14.17	10	29.62
5	69.18	11	65.48
6	41.00	12	30.22

.

3.1.4. Calorimetric Study

The formation of different amounts of CaHPO₄·2H₂O at different figurative points was confirmed by data on the kinetics of heat release (

Table 5. The kinetics of heat release of cements.

Time, h Composition	Integral							General
	0.25	0.5	1	3	6	12	24
Point 1	13.5 ± 0.12 g	4.3 ± 0.10	1.5 ± 0.15	1.9 ± 0.19	3.4 ± 0.10	4.1 ± 0.12	5.5 ± 0.09	40.1 ± 1.32 cde
Point 2	19.8 ± 0.05 d	6.0 ± 0.32	2.2 ± 0.12	2.1 ± 0.09	3.6 ± 0.11	3.8 ± 0.18	4.3 ± 0.11	49.2 ± 0.85 cd
Point 3	26.7 ±0.42 b	7.6 ± 0.15	2.4 ± 0.05	1.9 ± 0.12	3.5 ± 0.16	4.0 ± 0.22	4.7 ± 0.23	58.9 ± 0.79 a
Point 4	27.0 ± 0.43 b	6.5 ± 0.48	2.0 ± 0.16	2.8 ± 0.43	4.7 ± 0.25	4.2 ± 0.09	3.7 ± 0.12	58.6 ± 0.70 a
Point 5	6.0 ± 0.78 h	3.9 ± 0.11	1.6 ± 0.09	1.5 ± 0.22	2.4 ± 0.12	2.8 ± 0.24	4.1 ± 0.19	27.7 ± 1.00 g
Point 6	15.4 ± 0.08 ef	5.1 ± 0.09	1.9 ± 0.20	2.1 ± 0.08	3.4 ± 0.27	3.5 ± 0.70	4.5 ± 0.27	42.7 ± 0.63 f
Point 7	15.2 ± 0.09 fg	7.1 ± 0.20	2.2 ± 0.18	2.2 ± 0.10	3.6 ± 0.19	3.6 ± 0.17	4.2 ± 0.41	46.0 ± 1.03 e
Point 8	20.8 ± 1.17 d	5.5 ± 0.37	1.9 ± 0.08	2.0 ± 0.56	3.5 ± 0.18	3.8 ± 0.08	4.9 ± 0.29	49.4 ± 0.77 cd
Point 9	21.2 ± 0.66 c	7.1 ± 0.23	2.4 ± 0.37	2.2 ± 0.16	3.7 ± 0.09	3.9 ± 0.19	4.6 ± 0.23	53.4 ± 0.78 b
Point 10	17.3 ± 0.47 e	4.7 ± 0.08	1.7 ± 0.21	2.0 ± 0.12	3.4 ± 0.20	3.8 ± 0.35	5.0 ± 0.65	44.2 ± 1.01 f
Point 11	14.8 ± 0.08 a	9.2 ± 0.28	2.2 ± 0.18	2.0 ± 0.14	3.3 ± 0.18	3.3 ± 0.12	4.2 ± 0.15	46.9 ± 0.18 de
Point 12	22.3 ± 0.15 c	6.9 ± 0.17	1.8 ± 0.07	2.0 ± 0.12	3.3 ± 0.19	3.4 ± 0.16	4.0 ± 0.12	50.2 ± 0.50 c

Tukey test; means that do not share a letter are significantly different (p < 0.05).

).

An increase in thermal power in the case of all the compositions was registered in the first 15 min. It indicated an exothermic reaction. Due to the high rate of formation of CaHPO₄·2H₂O during the acid–base interaction, heat release in the initial periods of hardening was attributed to the formation of this phase.

The main reactions of DCPD formation were dissolution of the monohydrate of monocalcium phosphate, dissolution of α-tricalcium phosphate, dissolution of calcium carbonate, and formation of CaHPO₄·2H₂O. The dissolution of the initial components occurred mainly with the release of heat:

Ca(H₂PO₄)₂∙H₂O → Ca²⁺ + 2HPO₄²⁻ + 2H⁺ + H₂O (Q = 0 kJ/mol or 0 J/g);

α-Ca₃(PO₄)₂ + 2H+ → 3Ca²⁺ + 2HPO₄²⁻ (Q= −81.2 kJ/mol or −261.94 J/g);

CaCO₃ + 2H⁺ → Ca²⁺ + CO₂↑ + H₂O (Q= −14.83 kJ/mol or −14.83 × 1000/100 = −148.3 J/g)

According to the enthalpies of formation, it can be concluded that the formation of CaHPO₄·2H₂O occurred with the release of heat during the interactions of MCPM with TCP and MCPM with calcium carbonate:

Ca²⁺ + 2HPO₄²⁻ + H₂O + 3Ca²⁺ + 2HPO₄²⁻ + 7H₂O → 4 CaHPO₄∙2H₂O (Q= −6.4 kJ/mol or −37.21 J/g);

Ca²⁺ + 2HPO₄²⁻ + 2H⁺ + 2H₂O + Ca²⁺ + CO₂↑ + H₂O → 2 CaHPO₄∙2H₂O (Q= −175.29 kJ/mol or −509.56 J/g).

The total heat release by reaction was as follows:

α-Ca₃(PO₄)₂ + Ca(H₂PO₄)₂∙H₂O → 4 CaHPO₄∙2H₂O (Q = −299.15 J/g);

CaCO₃ + Ca(H₂PO₄)₂∙H₂O → 2 CaHPO₄∙2H₂O + CO₂↑ (Q = −657.86 J/g).

These calculations of heat release do not reflect the course of the entire set of reactions, but are appropriate for the main ones. They make the greatest contribution to the initial total heat release.

The total heat release by 15 min. consisted of the heat of the formation of CaHPO₄·2H₂O and the heat of the dissolution of α-TCP. A decrease in the pH of the cement dough due to an increase in the MCPM content affected the dissolution of α-TCP. In this regard, the increase in heat release by 15 min. linearly depended on the increase in the amount of MCPM in the initial composition (Figure 4).

In the initial period of hydration, when α-TCP particles came into contact with water, crystal dissolution reactions began. The liquid phase near the particles was saturated with Ca²⁺ and PO₄³⁻ ions. The liquid phase was supersaturated with ions. The final compounds were precipitated.

Heat dissipation was higher in compositions with higher concentrations of Na₂HPO₄·12H₂O with the same amount of MCPM in most cases. When using a solution of Na₂HPO₄·12H₂O as a mixing fluid, an additional source of PO₄³⁻ ions appeared. The Ca/P ratio shifted toward lower values. Probably, DCPD crystals were formed [38]. They subsequently rearranged into calcium-deficient hydroxyapatite. The formation of DCPD crystals led to an increase in heat release and a new dissolution of α-TCP was provoked, increasing the rate of formation of the hardened cement.

According to the data of integral heat release, the amount of heat released decreased by 1–3 h. Then, it rose slightly until 24 h. This was due to the slowing down of the reaction. On the surface of the α-TCP particles, new crystals of DCPD and hydroxyapatite were formed, which made it difficult to diffuse to the nonhydrated part of the α-TCP particles. The reaction rate and heat release increased when the shells were destroyed by growing crystals.

3.1.5. Stoichiometric Formulas

Based on the data of the volumetric research and the results of the spectrophotometric determination of calcium and phosphorus, the stoichiometric formulas of pre-set cement compositions at figurative points were derived (

Table 6. Stoichiometric formulas of compositions at figurative points.

Composition	Stoichiometric Formula
1	Ca7.9Na0.3(HPO4)1.4(PO4)3.5(CO3)(OH)0.6
2	Ca8Na0.3(HPO4)1.6(PO4)3.6(CO3)0.8(OH)0.4
3	Ca7.6Na0.3(HPO4)2(PO4)3.4(CO3)0.6
4	Ca8Na0.1(HPO4)1.7(PO4)4.0(CO3)0.3(OH)0.3
5	Ca8Na0.3(HPO4)1.2(PO4)3.6(CO3)1.2(OH)0.8
6	Ca8.2Na0.3(HPO4)1.2(PO4)4(CO3)0.8(OH)0.7
7	Ca8Na0.2(HPO4)1.5(PO4)3.7(CO3)0.8(OH)0.5
8	Ca8.2Na0.4(HPO4)1.4(PO4)4.1(CO3)0.5(OH)0.6
9	Ca7.9Na0.3(HPO4)1.6(PO4)3.8(CO3)0.6(OH)0.3
10	Ca8.3Na0.4(HPO4)1.2(PO4)4.2(CO3)0.5(OH)0.8
11	Ca7.9Na0.2(HPO4)1.5(PO4)3.4(CO3)1.1(OH)0.5
12	Ca8.1Na0.2(HPO4)1.5(PO4)4(CO3)0.5(OH)0.5

). The amounts of (HPO₄)²⁻, (PO₄)³⁻ and (OH)⁻ ions were determined according to the following formula:

Ca_{10–x–y/2–z/2}Na_z(HPO₄)_x(CO₃)_y(PO₄)_6–x–y(OH)_2–x

subject to the following conditions:

-

the neutrality of the molecule;

-

the amount of Ca²⁺ = 10 − (HPO₄)²⁻ − (CO₃)^2v/2 − Na⁺/2;

-

the amount of OH⁻ = 2 − (HPO₄)²⁻.

3.1.6. Scanning Electron Microscopy (SEM)

The microstructures of cement compositions at figurative points were represented mainly by irregular crystals of hydroxyapatite and a small amount of unreacted α-TCP particles (Figure 5). According to SEM, CaHPO₄·2H₂O was formed both on the surface of the pores and in the bulk of the cement (Figure 5).

3.2. Solubility of Cement Formed in the Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O System at Figurative Points

The solubility of the cement was evaluated at physiological temperature values of 37 °C in a physiological solution. The concentrations of calcium ions in the physiological solution after holding pre-set cement granules formed in the Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system at figurative points are shown in Figure 6.

The solubility of pre-set cement depended on the content of DCPD and the amount of CO₃²⁻ ions. In general, the solubility was inversely proportional to the concentration of Na₂HPO₄·12H₂O.

3.3. In Vivo Experiments

Biocompatibility and resorbability were investigated in in vivo experiments on Wistar rats in a model of tibial perforation. Granules of 250–500 microns in size of the compositions at figurative points 3, 5, and 8 were implanted into bone perforations of critical size in the tibia.

Partial resorption of the matrix and the formation of new bone tissue occurred after 3 months of implantation, according to computer microtomography (

Table 7. 3D model and orthogonal projections of the tibia of a laboratory animal created using CT results.

	Figurative Point
	3	5	8
3D model and orthogonal projections Initial
3D model and orthogonal projectionsAfter implantation period of 3 months

3D model and orthogonal projections are shown in one example for each composition.

).

The best healing of the defect, consisting of complete closure of the trepanation opening with the formation of a clearly X-ray-distinguishable cortical bone, was observed after three months of implantation in the case of using the cement of composition 3. The granules of the material remained radiopaque in the area of the bone marrow canal. Some decrease in their density was noted. No exostoses were observed.

An increase in the radiopacity of composition 5 on the side facing the defect indicated that the formation of cortical bridges deepened into the bone marrow canal. They were not fully connected to each other and were not yet fully formed. There was also an overgrowth of bone from the borders of the defect toward the implant.

Composition 8 showed the least pronounced degree of closure of the bone defect. There was a bright callus in the projection of the trepanation hole, with bone growth from the defect boundaries by an elongated, thin cortical bridge. The maturity of this regenerate was estimated as minimal compared to other formulations.

There was good bone regeneration with filling of the defect area with newly formed bone tissue after 3 months of the implantation of granules into the holes of the tibia in almost all cases according to the morphological examination (

Table 8. Micrographs of morphological examination of cements.

Point	Type of Microscopy
	Standard Light	Standard Light	Polarization	Standard Light	Phase Contrast
	Staining with Hematoxylin–Eosin	Staining with Picrosirius Red	Staining with Picrosirius Red	Staining with Hematoxylin–Eosin	Staining with Hematoxylin–Eosin
3
5
8

*—newly formed bone tissue, X—muscles, arrows—remains of the implants, tips of arrows—connective tissue capsule around remains of the implants.

). At the same time, inflammatory reaction was absent or minimal. The remains of the implant were detected in bone defects in most cases. The most intensive bone regeneration was characteristic of the compositions at figurative points 3 and 5 compared to composition 8. Moreover, the newly formed bone had a sufficiently mature structure with a large number of osteocytes.

The remains of the implant, in addition to bone, were found in muscles near the defect area when examining samples of different compositions. In this case, they were surrounded by a connective tissue capsule consisting of collagen fibers with fibroblasts, macrophages, and an admixture of giant, multinucleated cells.

4. Discussion

The Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system is an astringent system; when interacting with water, the product is a pre-set cement of variable composition.

Regulation of the solubility of pre-set cement was formed in the Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system, possibly due to the presence of a more resorbable phase in the composition compared to hydroxyapatite due to a decrease in energy of the hydroxyapatite crystal lattice. The latter could be achieved by replacing PO₄³⁻ ions with HPO₄²⁻ or CO₃²⁻ and Ca²⁺ with Na⁺.

The hydroxyapatite detected through X-ray had variable compositions: the numbers of substituted PO₄³⁻, Ca²⁺, and OH⁻ ions varied within the pre-set cement and formed a series of solid solutions. In addition to carbonate-substituted and calcium-deficient hydroxyapatites, more soluble phases of DCPD and β-TCP were found in the compositions of pre-set cement at physiological pH values.

The dissolution of the initial calcium phosphates, saturation with Ca²⁺ and PO₄³⁻- ions of the solution, and precipitation of reaction products from the solution occurred during the interaction of the Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system with water. As soon as the crystals of the reaction products were formed, calcium ions and phosphate ions saturated the solution again. At the same time, calcium carbonate also decomposed into Ca²⁺ and CO₃²⁻ ions in the aqueous solution. There was continuous precipitation–dissolution. In the presence of PO₄³⁻ ions, the solution became supersaturated with respect to CDHA or DCPD. DCPD could recrystallize into CDHA. The PO₄³⁻, CO₃²⁻, and Ca²⁺ ions in the solution were precipitated as CO₃-HA since, under such conditions, CO₃-HA was the most thermodynamically stable phase [39,40].

Presumably, DCPD was formed during the acid–base interaction of calcium carbonate and α-TCP with MCPM at the early stage of hydration. Acidic MCPM dissolving reduced the pH of the medium, affecting the thermodynamic stability. It increased the solubility of α-TCP and CaCO₃ and led to the formation of phases with lower Ca/P ratios. This was confirmed by the straightforward relationship between MCPM and DCPD. The formation of DCPD led to the appearance of a seed for the crystallization of hydroxyapatite and accelerated the dissolution of α-TCP. DCPD crystals were formed on the surface of the initial particles. They gradually dissolved, giving way to newly formed crystals.

The formation of DCPD with a release of carbon dioxide occurred when MCPM interacted with calcium carbonate. The number of embedded CO₃² ions in the hydroxyapatite crystal lattice decreased in this regard. Thus, the amount of DCPD increased, and the amount of CO₃²⁻ decreased when the figurative points were located closer to the stoichiometric amount of calcium carbonate and MCPM. The resulting carbonate-substituted hydroxyapatite was predominantly B-type. No calcium carbonate was found in the final composition of the cement.

Thus, the solubility of pre-set cement formed in the Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system depended both on the amount of DCPD and carbonate ions in the hydroxyapatite structure. The highest solubility values were characteristic of compositions 3 and 4. Since these compositions contained the largest amounts of DCPD, the Ca/P ratio was 1.4.

The low solubility of the compositions of figurative points 8 and 10 containing high concentrations of Na⁺ ions may be associated with the formation of Na-containing hydroxyapatites. Their solubility may be lower. Sodium ions of the system could be embedded in the structures of carbonate and calcium-deficient hydroxyapatites and could compensate for the difference in charges when replacing PO₄³⁻ ions with CO₃²⁻ or HPO₄²⁻.

The compositions of points 8 and 10 contained the least amount of HPO₄²⁻ and CO₃²⁻ ions with the highest contents of sodium ions in accordance with the stoichiometric formulas obtained empirically. Na-containing hydroxyapatites lost Na⁺ and turned into calcium-deficient hydroxyapatite Ca₉(HPO₄)(PO₄)₅OH when immersed in aqueous solutions in an open system for about 2 months, according to other authors’ research [41].

All the compositions studied in in vivo experiments showed high biocompatibility, as well as good bone regeneration with filling of the defect area with newly formed bone tissue. In vivo resorption was observed in higher compositions with high contents of DCPD and carbonate groups (points 3 and 5). When using composition 3, the formation of a clearly X-ray-distinguishable cortical bone was observed.

It should be taken into account that in vivo resorption was caused not only by chemical dissolution of the cement in extracellular fluid (passive resorption), but also by active resorption due to cell activity (osteoclasts, giant cells, and macrophages) [42]. Passive resorption (chemical dissolution) depended on the properties of the cement stone, particularly the amounts of DCPD and carbonate ions, and on properties of the environment, particularly pH. The active resorption of the calcium phosphate cements is mediated by giant cells and osteoclasts, with the absorption of fragmented cement particles by macrophages [43,44]. Osteoclasts, which are tightly attached to a mineral surface, decrease the pH locally near a material and dissolve the inorganic calcium phosphate that is located under them [45,46,47]. A local decrease in pH is possible with the dissolution of DCPD. This may slightly affect the solubility of carbonate hydroxyapatite since the pH of DCPD is about 4-5. Figure 7 shows the scheme of the resorption of calcium phosphate cement particles depending on composition.

The concentration of Ca²⁺ should be higher than the physiological norm in order to increase proliferation and differentiation [16,48]. Consequently, the increased level of Ca²⁺ as a result of cement dissolution may be one of the key factors in the formation of new bone. The rate of resorption of calcium phosphate cements affected cellular proliferation by binding the released calcium to the extracellular calcium-sensitive receptor that was associated with the G-protein [23]. In this regard, cell proliferation was higher in apatite cements with substituted PO₄³⁻ on CO₃²⁻ ions [49] compared to cements without substitutions. The content of the highly resorbable phase of DCPD was also a key factor affecting the formation of new bone tissue.

5. Conclusions

The Ca₃(PO₄)₂/CaCO₃/Ca(H₂PO₄)₂·H₂O/Na₂HPO₄·12H₂O system is important for the synthesis of self-hardening bioactive materials with variable resorption rates. It allows the adjustment of the characteristics of the final product in wide intervals through changing the ratios of the initial components.

The formation of carbonate-substituted hydroxyapatite and dicalcium phosphate dihydrate in the system increased passive resorption (solubility) due to the substitution of PO₄³⁻ with CO₃²⁻ ions and the distortion of the hydroxyapatite crystal lattice, as well as due to the formation of the crystals of a metastable compound in the composition of cement stone under physiological conditions. An increase in active resorption occurred due to the binding of the released calcium to the extracellular calcium-sensitive receptor associated with the G-protein. This increased the proliferation and phenotypic expression of bone cells and led to formation of new bone in direct contact with biomaterial. The high rate of resorption promoted high bone regeneration with filling of the defect area with newly formed bone tissue.

Calcium phosphate cements are bioactive, osteoconductive, and resorbable. They integrate into bone tissue without the formation of a connective tissue capsule. Increasing the rate of resorption of calcium phosphate cements based on hydroxyapatite is an important task for material science and may contribute to its wider use in clinical practice.