Sol-Gel Synthesis of Hydroxyapatite: Applications, Methods, and a Case Study

1. Introduction

Calcium orthophosphates are chemical compounds that hold significant interest across various scientific disciplines, such as geology, chemistry, biology, and medicine, due to their natural abundance and presence in living organisms [1, 2, 3, 4].

The subject of calcium orthophosphates has been under investigation for nearly 250 years [4]. At the end of the eighteenth century, the German geologist Abraham Werner coined the term “apatite” to describe a range of phosphate minerals that are now the major global source of phosphorus [5]. In the same year, Joseph-Louis Proust and Martin Klaproth proposed that calcium apatite (at the time, all calcium phosphates were referred to as apatites [1]) was the primary inorganic component of bones [5, 6]. The initial attempts to determine the chemical composition of calcium apatites and other calcium orthophosphates were made by Jacob Berzelius in the mid-nineteenth century [6, 7, 8]. Approximately 80 years later, H. Hausen proposed the existence of different calcium phosphate phases [1, 2]. These and other milestones in the historical development of the subject, as reported by Dorozhkin [6] (in his thorough investigation) and others [1, 2, 5], are depicted in Figure 1.

All calcium orthophosphates are composed of three key elements: calcium, with a + 2 oxidation state; phosphorus, with a + 5 oxidation state; and oxygen, with a − 2 reduction state, which form part of orthophosphate anions. Additionally, many calcium orthophosphates include hydrogen as part of acidic orthophosphate anions (e.g., HPO₄²⁻ or H₂PO₄⁻), hydroxide ions (such as in Ca₁₀(PO₄)₆(OH)₂), and/or water molecules (e.g., CaHPO₄·2H₂O). The chemical composition of these compounds can vary, producing a wide range of calcium phosphates distinguished by the type of phosphate anion involved—ortho-(PO₄³⁻), meta-(PO₃⁻), pyro-(P₂O₇⁴⁻), and poly-(PO₃)_nⁿ⁻. When multi-charged anions like orthophosphates and pyrophosphates are involved, calcium phosphates are also categorized by the number of hydrogen ions associated with the anion. Examples include mono-[Ca(H₂PO₄)₂], di-(CaHPO₄), tri-[Ca₃(PO₄)₂], and tetra-(Ca₂P₂O₇) calcium phosphates, with the prefixes “mono,” “di,” “tri,” and “tetra” indicating the number of hydrogen ions replaced by calcium [6, 7, 8].

The atomic structure of calcium orthophosphates is centered around a network of orthophosphate (PO₄³⁻) groups, which provides structural stability. Most calcium orthophosphates have low solubility in water but are readily soluble in acids while remaining insoluble in alkaline solutions. Chemically pure calcium orthophosphates are typically white crystals with moderate hardness. In contrast, natural calcium orthophosphate minerals are often colored due to the presence of impurities, particularly ions of iron, manganese, and rare earth elements [7, 8].

The names, chemical formulas, and Ca/P molar ratios of the known calcium orthophosphates are depicted in Table 1.

Among the various calcium orthophosphates, hydroxyapatite (HAp) has garnered significant attention due to its structural and chemical characteristics, as well as its stability within the pH range of 4–14 at room temperature [1]. These properties make it a compound of considerable interest in catalysis, the fertilizer industry, pharmaceutical products, protein chromatography applications, water treatment processes, and the preparation of biocompatible materials. Most notably, it is the main inorganic component in calcified hard tissues (for example, bone and teeth) of vertebrates [1, 2].

HAp has been extensively studied, surpassing other apatites such as fluorapatite and oxyapatite. This is evidenced by the analysis of papers published in the SCOPUS database over the past 10 years, as shown in Figure 2.

The number of published articles was retrieved from the SCOPUS database using the fields ‘Article title, Abstract, Keywords’ for the period from 2014 to 2024, with the keywords identified in the respective plots.

To emphasize the growing importance of hydroxyapatite research, Figure 3 presents a comprehensive analysis of publication and citation trends (2000–2024), sourced from app.dimensions.ai, using the keyword “hydroxyapatite.” Publications saw a steady upward trajectory, reflecting increased research activity and output. Citations also grew sharply, particularly in the later years, surpassing 1 million annually, which suggests the field has gained considerable academic and industrial impact. The parallel growth of citations with publications highlights the increasing recognition and influence of hydroxyapatite research during this period. Therefore, it is imperative to ensure that researchers have access to the latest information and advancements.

HAp has the general formula Ca₅(PO₄)₃(OH) but is usually written as Ca₁₀(PO₄)₆(OH)₂ to emphasize that the crystal unit cell comprises two molecules [10].

Chemically pure HAp crystallizes in the monoclinic space group P2₁/b, as suggested in 1967 and later confirmed by single-crystal structure analysis [11, 12, 13], with cell parameters a = 9.4214 [8] Å, b = 2a, c = 6.8814 [7] Å, and α = β = 90°, γ = 120° [14]. The most common and initially known form is hexagonal, described in the space group P6₃/m (N° 176) with lattice parameters a = b = 9.432 Å, c = 6.881 Å, and α = β = 90°, γ = 120° [13]. The phase transition from monoclinic to hexagonal in hydroxyapatite occurs at approximately 470 K [12].

The unit cell of hexagonal HAp, comprising 44 atoms [15], is depicted in Figure 4. The 10 calcium ions (4 Ca_I and 6 Ca_II) occupy two distinct sites. The four Ca_I ions are coordinated by nine oxygen atoms from phosphate tetrahedra, while the six Ca_II ions are coordinated by six oxygen atoms from five phosphate groups and one hydroxyl ion. The Ca_I sites form distinct calcium channels parallel to the c-axis, whereas the Ca_II ions create triangular bases staggered by 60° above and below the hydroxyl ions. Similar to Ca_I, the hydroxyl ions form distinct channels parallel to the c-axis. Based on their coordination with oxygen and calcium, the oxygen atoms are categorized into three types: O_I, O_II, and O_III [13, 15, 17, 18, 19].

HAp structure allows other groups, anionic and cationic substances, to replace Ca²⁺, PO₄³⁻, and OH⁻ ions, which is considered a unique attribute since these modifications enable the tailoring of HAp for specific applications. These substitutions can affect network parameters, crystallinity, crystal dimensions, surface texture, solubility, spectral properties, and thermal stability [3, 20].

HAp can be derived from both natural and synthetic sources. Natural sources highlighted in the literature include biological materials or wastes such as mammalian bones (for example, bovine, camel, and horse), marine or aquatic sources (like fish bones and fish scales), shell sources (including cockle, clam, eggshell, and seashell), plants, algae, and mineral sources such as limestone [3, 21]. However, natural hydroxyapatite (HAp) is non-stoichiometric and may be deficient in either calcium or phosphorus. Regardless of its origin, HAp often contains trace impurities such as phosphite ions (PO₃³⁻), chloride ions (Cl⁻), and fluoride ions (F⁻). Notably, PO₃³⁻ and Cl⁻ have been reported to weaken the HAp structure, whereas F⁻ is known to enhance its strength [1].

As mentioned before, HAp is the main component of bones and teeth. It is highly biocompatible and can be a very cost-effective material due to its high availability. This, combined with the evolution of biotechnology in recent years, has increased the interest in nature-inspired materials while creating new applications beyond the initial scope. These facts have made HAp one of the most attractive and important bioceramics [22].

2. Applications

HAp has long been valued for its outstanding biocompatibility and bioactivity, establishing itself as a key material in bone tissue engineering, orthopedic implants, and dental applications [23, 24].

Furthermore, it has osteoconductive, non-toxic, non-inflammatory, and non-immunogenic properties [25].

Its capacity to bond with surrounding tissues, enhance cellular attachment, and support osseointegration has positioned it as a preferred choice for biomedical implants and surface coatings. Furthermore, more recent research indicates that HAp particles may play a role in suppressing the growth of various cancer cell types [23, 24].

As knowledge about HAp expanded, its applications evolved beyond conventional biomaterials [23]. Advances in synthesis techniques and material engineering revealed unique properties of HAp that broadened its scope to various fields. By modifying its composition, morphology, and surface features, researchers achieved improvements in mechanical strength, drug delivery efficiency, and biodegradability [23, 24]. These adjustments expanded the use of HAp to include drug delivery systems, biosensors, tissue engineering scaffolds, and applications in regenerative medicine [25, 26, 27, 28].

In addition, HAp’s outstanding biomineralization capabilities have facilitated the integration of functional ions and molecules during its synthesis. This has paved the way for creating bioactive coatings and composites with targeted therapeutic properties. Functionalized HAp materials have shown significant potential in antimicrobial coatings, controlled release systems for growth factors and drugs, catalysts in various chemical processes, sensing, and energy storage, demonstrating their potential as advanced functional materials beyond traditional biomedical applications [23, 29, 30, 31, 32, 33, 34].

Recently, HAp nanoparticles and nanostructured materials have gained significant attention in research due to their distinctive physicochemical characteristics and their potential in targeted drug delivery, imaging, and theranostic applications [23, 35, 36, 37]. The capacity to control the size, shape, and surface properties of HAp at the nanoscale has opened new possibilities for innovative strategies in personalized medicine and regenerative therapies [23, 38].

Figure 5 schematically shows some of the possible applications of HAp, highlighting its role as an advanced multifunctional material.

3. Synthesis methods

HAp, as an advanced material in biomedical applications, is synthesized using various methods to optimize its structural, physicochemical, and biological properties. The synthesis methodologies can broadly be categorized into dry methods, wet methods, high-temperature methods, and hybrid techniques. Each method offers unique advantages and limitations depending on the desired application and characteristics of the HAp product.

A brief description of the most common methods is presented below and schematically summarized in Figure 6.

4. Sol-gel synthesis

The sol-gel synthesis method is a well-known and established process, valued for its flexibility and applicability to a diverse range of structural materials, with particle sizes spanning from the nanometer to micrometer scale [76].

One of the earliest proposed methods for the synthesis of HAp was indeed the sol-gel method [77]. However, only a limited number of studies have specifically investigated the synthesis of HAp using this process [77, 78]. Despite not being one of the most commonly applied methods for HAp synthesis, its application over the last 10 years has remained stable, as shown in Figure 7.

The data on published articles was obtained from the SCOPUS database by searching the fields “Article Title, Abstract, Keywords,” from 2014 to 2024, using the terms “hydroxyapatite” and “sol-gel.”

The sol-gel method is an effective approach for synthesizing nanophasic hydroxyapatite (HAp) due to its ability to provide strict control over process parameters [79]. This method enables molecular-level mixing of calcium and phosphorus, significantly enhancing the chemical homogeneity of HAp [79, 80].

Additional key benefits of the sol-gel process include the formation and fusion of arranged crystals at low temperatures, resulting in lower sintering temperatures compared to other methods. Moreover, the resulting powder exhibits a stoichiometric structure, a large surface area, and a small cluster size [77].

It has also been reported that HAp materials synthesized via the sol-gel process are effective in improving contact and stability at artificial/natural bone interfaces in both in vitro and in vivo environments [77, 78].

However, despite these recognized advantages, there are also drawbacks to synthesizing HAp through the sol-gel method, namely the formation of a secondary phase, commonly calcium oxide (CaO), and the high cost of certain precursors, particularly alkoxide-based compounds [77, 81].

The sol-gel process essentially involves the synthesis of an inorganic network by mixing calcium and phosphorus precursors in solution, followed by hydrolysis, gelation, and calcination [80].

It has been proven that the calcination temperature—i.e., the required temperature to form the HAp structure—depends largely on the chemical nature of the precursors [82]. Regarding the solvent, Nazeer et al. [76] investigated the effect of three different solvents on the physical and chemical properties of hydroxyapatite synthesized via the sol-gel method. The authors demonstrated that the dielectric constant and polarity of the reaction solvent had a critical impact on the nature of the product obtained during synthesis. The solvent can influence the reaction rate and hydrolysis, ultimately affecting the structure of the synthesized material. According to the authors, solvents with high dielectric constants favor the complete dissociation of electrolytes, while solvents with low dielectric constants promote ion pairing.

Figure 8 schematically presents the typical sol-gel route, highlighting some of the precursors and solvents that have been applied over the years. However, it is worth noting that, according to Livage et al. [83], phosphate esters and phosphoric acid are not suitable for the sol-gel synthesis of phosphate materials, since the first cannot be easily hydrolyzed, while the second reacts too quickly, leading to precipitation rather than gelation. The same authors showed that more convenient precursors could be obtained by dissolving phosphorus pentoxide in alcohols. Nevertheless, this is not the only possibility for synthesizing phosphates from molecular precursors. Alkyl phosphites, as well as phosphoryl chloride, can also be used as molecular precursors.

Liu et al. [82] explored two different sol-gel routes for the synthesis of hydroxyapatite (HAp): one ethanol-based and the other water-based, both utilizing the same precursors—triethyl phosphite, C₆H₁₅O₃P, and calcium nitrate, CaNO₃.

According to the authors, in anhydrous ethanol, alkyl groups may form Ca(OR)_y(NO₃)_2-y, replacing some of the nitrate groups. In contrast, calcium nitrate dissolves in water to form ionic species. Both ethanol- and aqueous-based sols were found to be relatively stable at room temperature, not forming a solid gel for at least 5 days.

After aging, the hydrolyzed phosphorus sol—which the authors hypothesized to be in the form of a phosphoric ester, PHO(OEt)₂, or more generally, P(OEt)_3-x(OH)_x — interacted with the calcium sol. This interaction likely involved Ca(OEt)_3-x(NO3)_2-y in anhydrous ethanol and Ca²⁺ ions in water, forming oligomeric derivatives containing Ca–O–P bonds.

As heating continues, solvents are eliminated, accompanied by increased thermal dehydration or the polymerization/condensation of the derivative units. This process results in the formation of additional Ca-O-P bonds in the dry gels.

As a result of the two routes investigated, the ethanol-based synthesis yielded a thermally stable calcium hydroxyapatite phase, whereas the water-based synthesis produced a calcium-deficient apatite.

5. Short review of hydroxyapatite synthesized by the sol-gel method

HAp is typically used in powdered form, and its efficacy is influenced by various properties, such as phase purity, crystallinity, precursor sources, solvent or surfactant materials, particle size, morphology, and surface area [23, 38].

A wide variety of routes have been adopted for the synthesis of HAp using the sol-gel method. Efforts have been made to control geometry, crystallinity, size, stoichiometry, and degree of particle agglomeration (for different applications) by employing new routes or modifying pre-existing synthesis methods [23].

This section provides a review, also summarized in Table 2, though not exhaustive, of some of these routes, focusing on the combination of different precursors and solvents, as well as on the phase purity and morphology of the obtained material.

Balamurugan et al. [79] used triethyl phosphite, dissolved in a mixture of ethanol and distilled water, and a stoichiometric amount of calcium nitrate tetrahydrate, according to the Ca/P ratio of 1.67, which was dissolved in ethanol and added dropwise to the first solution. This procedure was performed under vigorous stirring, which was continued for an additional 10 minutes after the mixing. As a result of this process, a clear solution was obtained and aged at room temperature for 16 hours. The mixture was then dried at 60°C until a viscous liquid was obtained, and further drying of the viscous liquid at 60°C resulted in a white gel. The gel was ground with a mortar and pestle into a fine powder and subjected to different calcination treatments, ranging from 300 to 900°C, for 2 hours.

According to the authors, a pure monophasic crystalline HAp phase was obtained with the calcination performed at 900°C, exhibiting a nano-structured, ellipse-like morphology.

In the sol-gel process presented by Bogdanoviciene et al. [84], ammonium hydrogen phosphate and calcium acetate monohydrate were selected as P and Ca precursors, respectively, with a Ca/P molar ratio of 1.67. Calcium acetate was first dissolved in ethanol at 65°C, and to this solution, ammonium hydrogen phosphate, dissolved in distilled water, was added. The resulting mixture was stirred for 1 hour at the same temperature. In a subsequent step, ethylene diamine tetra-acetic acid or tartaric acid, as complexing agents, were added to the solution. After concentrating the solution by slow evaporation at 65°C under stirring, a procedure that took 15 hours, the sol turned into a transparent gel, which was dried at 100°C. The obtained material was ground and annealed for 5 hours at 1000°C in air, with a heating rate of 10°C/min.

According to the X-ray Diffraction (XRD) analysis, regardless of the complexing agent used, no impurities were detected. Regarding the morphology, Scanning Electron Microscopy (SEM) micrographs of the HAp powders derived from the ethylene diamine tetra-acetic acid route suggest that the sample is composed of agglomerates with a broad size distribution, ranging from 500 to 8 μm. The HAp powders derived from the tartaric acid route present a different morphology, characterized by well-shaped ultrafine elongated particles (microrods and/or microsticks) with lengths of about 2–10 μm and widths of 1–4 μm.

For the synthesis of HAp nanorods, Padmanabhan et al. [85] used calcium nitrate tetrahydrate and potassium dihydrogen phosphate as calcium and phosphorus precursors, respectively, both dissolved in deionized water, with concentrations adjusted to ensure a Ca/P ratio of 1.67. This synthesis was conducted at pH = 9, using ammonia solution. The resulting mixture was stirred for 1 hour and aged at room temperature for 48 hours. After repeated procedures of filtration and washing with double-distilled water to remove NH4⁺ and NO₃⁻ ions, the obtained precipitate was dried at 60°C for 24 hours in a dry oven. The dried powder was subsequently heated over a temperature range of 300–700°C for 30 minutes in an electric furnace in air, with a heating rate of 10°C/min.

XRD analysis showed that the sample heat-treated at 700°C exhibited a high degree of purity, with no secondary phases detected. Regarding the morphology, the authors reported the formation of nanorods with diameters of 70–90 nm and lengths of 400–500 nm, which has been attributed to the “oriented attachment” mechanism, where HAp crystallites adhere to high-energy planes and template growth in a specific direction.

A simple but time-consuming route using trimethyl phosphite and calcium nitrate tetrahydrate as precursors was presented by Chen et al. [86]. These were first dissolved in absolute ethanol, with deionized water added to the trimethyl phosphite solution under vigorous stirring. The solutions were mixed with a Ca/P molar ratio of 1.4, 1.5, and 1.67 (S1.4, S1.5, and S1.67, respectively). The prepared mixtures were continuously stirred for about 10 minutes and then kept at room temperature for approximately 16 hours. Afterwards, they were heated in a water bath at 60°C for 6 days, resulting in a white dried gel. The dried gel was then calcined in a furnace at 600°C for 3 hours.

The XRD analysis showed that sample S1.4 was composed of beta-tricalcium phosphate, sample S1.5 was a mixture of beta-tricalcium phosphate and HAp, with HAp being the main phase, while sample S1.67 was pure HAp.

The SEM micrographs revealed a similar morphology among the synthesized powders, showing essentially a mixture of particles of different sizes, with each particle being an assembly of numerous distinguishable grains sized approximately between 90 and 150 nm.

Figueroa et al. [87] proposed a considerably faster approach, using tetrahydrated calcium nitrate and phosphorus pentoxide as precursors, combined in a stoichiometric ratio of Ca/P = 1.67. The starting solutions were prepared separately, using ethanol as the solvent, and the first solution was added dropwise to the second solution.

The final mixture was kept under stirring for an hour at room temperature and then dried at 56°C for 12 hours. Finally, the obtained powder was ground and heat-treated at 600, 900, and 1200°C for 2 hours.

XRD analysis confirmed the formation of a crystalline HAp phase; however, a small trace of calcium phosphate was also identified.

Transmission electron microscopy (TEM) analysis revealed particles with relatively homogeneous shapes, averaging 40 nm in size for the heat-treatment temperature of 600°C, and 500 nm for 900 and 1200°C, indicating that the heat-treatment temperature may control particle size.

To explore the influence of aging time on the synthesis of HAp, Bakan et al. [88] used calcium nitrate tetrahydrate and ammonium dihydrogen phosphate as starting Ca and P precursors, respectively, maintaining a Ca/P ratio of 1.67. The first precursor was dissolved in ethanol and the second in water, with ammonia (NH₃, Merck) used to adjust the pH of the solution. The resulting mixture was kept under vigorous stirring at room temperature until a white precipitate was obtained, which was subsequently aged for 12, 24, and 48 hours, also at room temperature.

The gel obtained after the aging process was first filtered and washed repeatedly with double-distilled water to remove NH₄⁺ and NO₃⁻ ions, and then washed with pure acetone. The obtained material was dried at 80°C for 12 hours in an oven, and the dried powder was calcined at 750°C for 4 hours in an electric furnace, with a heating rate of 10°C/min in air.

The XRD studies revealed that the samples synthesized with 12-hour and 24-hour aging times contained a small amount of beta-tricalcium phosphate along with HAp. After 48 hours of aging, the secondary phase gradually disappeared, and pure HAp was obtained.

From the SEM and TEM micrographs, the HAp particles were identified as hexagonal, nano-needle-like structures with a mean particle diameter of 12 nm and a mean particle length of 65 nm.

To study the influence of the solvent nature and calcination time, Türk et al. [89] synthesized HAp using calcium hydroxide and phosphoric acid, considering a Ca/P atomic ratio of 1.67. Two separate solutions of calcium hydroxide were prepared: one with double-distilled water and the other with ethanol. For the preparation of the phosphorous solutions, the same procedure was followed. The prepared calcium solutions (water- and ethanol-based) were added dropwise into the phosphate solutions (water- and ethanol-based) under stirring, with the pH adjusted to 11 using ammonium hydroxide. The solutions were then aged for 24 hours at room temperature and exposed to thermal treatment at 105°C in an air oven until white-dried gels were obtained. The dried powders were ground in a mortar and calcined for 1, 2, and 4 hours at 950°C, with a heating rate of 5°C/min.

XRD characterization revealed that, despite the calcination time, the samples obtained by the ethanol-based synthesis presented calcium oxide and calcium hydroxide as impurities, while with the water-based synthesis, the only secondary phase was calcium oxide, with the 1-hour calcination yielding the best result.

The SEM analysis showed that the powders obtained from both solvents were non-uniform, and the arrangement of cluster-like particles formed nano-sized spherical aggregates for all calcination times. However, it can be said that, to some degree, homogeneous particles with a narrow size range and lower degree of aggregation occurred in the water-based samples.

Farazin et al. [90] synthesized HAp to develop a chitosan/hydroxyapatite composite scaffold for use as a bone replacement tissue. The authors used diammonium hydrogen phosphate and calcium nitrate tetrahydrate as precursors, with a Ca/P molar ratio of 1.67. Diammonium hydrogen phosphate was dissolved in deionized water, and calcium nitrate tetrahydrate was dissolved in a mixture of ethanol and deionized water in a 1:12 ratio. Ammonia solution was used to maintain the pH between 10 and 12. The first solution was added to the second one, with this step being carried out in a low-temperature water bath (38–42°C) for 4 hours under magnetic stirring. The resulting material was filtered and washed several times with deionized water.

The drying process comprised three steps: 14 hours at 40°C, 2 hours at 60°C, and finally 18 hours at 80°C, followed by heat treatment at 700°C for 1 hour, with a heating rate of 10°C/min.

XRD analysis showed that, besides HAp, beta-tricalcium phosphate and calcium phosphate were formed as secondary phases. On the other hand, FESEM micrographs revealed that a nanostructured powder was obtained, with an average size of 48 nm (Table 2).

6. Case study

This case study explores the synthesis and characterization of HAp. For the synthesis, the sol-gel route proposed by Figueroa et al. [87] was used. However, to enable upscaling of the process, the mass of the precursors was doubled along with the volume of the solvent. The obtained powder was thermally analyzed and heat-treated at four different temperatures. Structural characterization of the four samples was performed using X-ray diffraction, while the morphology and electrical behavior of the most promising sample were analyzed using scanning electron microscopy and impedance spectroscopy, respectively.

6.1 Materials and methods

6.1.1 Materials

For the hydroxyapatite synthesis, the starting materials were calcium nitrate tetrahydrate (Ca(NO₃)₂·4H₂O, EssentQ®, Scharlab), diphosphorus pentoxide (P₂O₅, EssentQ®, Scharlab), and ethanol (C₂H₅OH, 96% v/v, Aga) as the solvent.

6.1.2 Synthesis of HAp

As mentioned, the synthesis of HAp was performed using the sol-gel method, following a modified approach originally presented by Figueroa et al. [87]. The first solution was prepared by dissolving 8.52 g of P₂O₅ in 175.0 ml of ethanol, and the second solution was prepared by dissolving 47.20 g of Ca(NO₃)₂∙4H₂O in 584.0 ml of ethanol in order to achieve a stoichiometric Ca:P ratio of 1.67.

Afterward, the first solution was added to the second solution in a dropwise fashion. The final solution was stirred for an hour at room temperature and then dried at 80°C for 36 hours to evaporate the solvent.

To evaluate the temperatures at which crystalline phases form, the obtained powder was subjected to thermal characterization using differential scanning calorimetry and thermogravimetric analysis. Based on these results, and after grinding, the powder was heat-treated at 500, 600, and 700°C for 2 hours, with a heating rate of 5°C/min.

Figure 9 shows the flowchart of the steps involved in the synthesis of the HAp powders.

6.1.3 Characterization techniques

The thermogravimetric analysis, TGA, was performed on a Netzsch TG 209 F1 Libra in a nitrogen atmosphere, with a heating rate of 5°C/min, ranging from 30 to 1095°C. The differential scanning calorimetry, DSC, was carried out between 25 and 500°C with a Netzsch DSC 204 F1 Phoenix, also in a nitrogen atmosphere, at the same heating rate of 5°C/min.

X-ray Diffraction, XRD, analyses were carried out using a PANalytical X’Pert PRO diffractometer with Cu Kα (λ = 1.54060 Å) radiation, operating at 45 kV and 40 mA. The tests utilized parallel beam geometry, with a step size of 0.04° and an exposure time of 3 s per step.

The morphology analysis was conducted by Scanning Electron Microscopy, SEM, utilizing a TESCAN Vega 3 microscope. All measurements were taken at a voltage of 15 kV.

The impedance spectroscopy analysis was made on the sample having a disk shape with a diameter of 13.10 mm and a thickness of approximately 1.36 mm. Silver electrodes were prepared by applying a thin layer of silver conductive ink onto the opposite sides of the sample.

The measurements were performed using an Agilent 4294A precision impedance analyzer in C_p–R_p configuration (capacitance in parallel with resistance), with an applied voltage of 0.5 V, at different temperatures over the frequency range of 40 Hz to 1 MHz. During these measurements, the sample was maintained in a helium atmosphere to improve the heat transfer and eliminate moisture.

6.2 Results

6.2.1 Thermal, structural, and morphological characterization

Figure 10(a) shows the TGA curve of the HAp powder at a heating rate of 5°C/min, revealing a total weight loss of 45.94%, with the most significant mass loss occurring below 500°C.

In the first stage, between 33 and 79°C, the weight loss can be attributed to the evaporation of physically adsorbed ethanol and water, which corresponds to the first endothermic peak in the DSC curve, Figure 10(b), centered at 52°C [91, 92, 93, 94, 95].

The second stage, in the range of 79 to 360°C, exhibits weight loss due to the release of adsorbed and lattice water, as well as the partial decomposition of organic matter and crystallization of HAp. The release of the adsorbed and lattice water corresponds to the endothermic peaks in the DSC curve, centered at approximately 118 and 140°C, while the decomposition of organic matter and crystallization of HAp can be assigned exothermic peaks centered at 310 and 341°C, respectively [91, 94, 96, 97, 98].

According to Tõnsuaadu et al. [99], the lattice water is irreversibly lost between 200 and 400°C. Such temperature variations can be ascribed to differences in the samples and the experimental synthesis procedures [93, 100].

The third stage, occurring between 360 and 500°C, likely represents the loss of water due to the decomposition of chemically bonded water, as well as the decomposition of the organic matrix and carbonates. In the DSC curve, these events correspond to the broad endothermic peak centered at 425°C [91, 95, 101]. Above 500°C, the weight loss becomes minimal.

Based on this analysis, heat treatments at 500, 600, 700, and 800°C were selected to observe the progression from crystal growth and decomposition at lower temperatures to the stability and crystallization expected at higher temperatures.

The crystalline phases were identified by XRD, as shown in Figure 11(a). In addition to the patterns of the sintered powders, the reference pattern for HAp (ICDD 01-074-0565 [102]) is also displayed.

The formation of hexagonal HAp is observed even at the lowest heat treatment temperature of 500°C. However, along with HAp, three secondary phases—Ca₂P₂O₇, Ca₃(PO₄)₂, and Ca₃N₂—were identified. These phases were also present in the samples heat-treated at 600 and 700°C, with the heat treatment at 800°C resulting in the decomposition of both HAp and Ca₃N₂.

The chemical composition of the sintered samples was estimated using Rietveld refinement, performed with the software Profex [103], with the goodness of fit ranging from 1.39 to 1.81. The weight percentage of each phase is presented in Figure 11(b), showing that as temperature increases, the HAp content decreases.

These results indicate that when using the approach proposed by Figueroa et al. [87] as a starting point, the upscaling of the process cannot be governed solely by the concentration of the precursor solutions. Using the same concentrations reported by Figueroa et al., but doubling the mass of calcium nitrate tetrahydrate and diphosphorus pentoxide and the corresponding ethanol volumes, the efficiency of the process remains lower than that achieved by the mentioned authors, who reported HAp as the main component with only a small trace of calcium phosphate.

Based on the obtained results, the sample heat-treated at 500°C demonstrated the most favorable properties, making it the most suitable candidate for further characterization.

Figure 12 presents the SEM micrographs of this sample captured at two different magnifications, providing insights into its surface morphology. Signs of agglomeration are visible, along with particles of varying shapes and highly diverse sizes, ranging from the nanometric to micrometric scale.

6.2.2 Impedance spectroscopy measurements

Impedance spectroscopy is a reliable technique for quantitatively interpreting the electrical properties of materials and studying the processes occurring within them. It enables a direct correlation between the response of a real system and an idealized model circuit composed of discrete electrical components, allowing the separation of grain and grain boundary contributions to the transport properties of materials [104, 105].

Since Z*(ω) is a complex quantity, it can be represented in polar coordinates or Cartesian coordinates. This representation is given by [106]:

Z∗=Z′−jZ′′E2

where Z′ and Z″ are the real and imaginary parts of the complex impedance, respectively, and can be calculated using the equations [105]:

Z′=RP1+ωRPCP2E3

Z″=ωRP2CP1+ωRPCP2E4

where C_p and R_p are the measured capacitance and resistance, respectively, and ω is the angular frequency.

Figure 13 shows the imaginary part of the impedance, Z″, as a function of frequency at different temperatures.

Analyzing the variation of Z″ with the frequency, one can infer the appearance of peaks at a particular frequency, which became evident for temperatures above 300 K, and the convergence of the spectra at higher frequencies.

This behavior suggests the presence of a thermally activated relaxation mechanism in the system in the frequency and temperature ranges studied [107].

In addition to complex impedance, the modulus formalism can also be applied to study the dielectric behavior of a ceramic material, with several authors emphasizing the advantages of the combined use of both formalisms, specifically because the Z″ plots highlight the phenomenon of the largest resistance, whereas M″ highlights those of the smallest capacitance [108, 109].

The dielectric modulus, M*, is defined as a quantity directly proportional to the complex impedance [108, 110]:

M∗=M′+jM″=jωC0Z∗E5

where M′ and M″ are the real and imaginary parts of the complex modulus, respectively, and can be obtained from [111]:

M′=−ωC0Z″E6

M″=ωC0Z′E7

C₀ is the capacitance of the empty (vacuum) measurement cell and is given by [112]:

C0=ε0AdE8

where ε₀ is the vacuum permittivity (8.8542 × 10⁻¹² F/m), d is the sample thickness, and A is the electrode area.

In the inset of Figure 13, the imaginary part of the modulus is depicted as a function of frequency at the same temperatures. The frequency dependence of M′′ clearly exhibits a maximum for each temperature, once again above 300 K. It is observed that the position of the peak shifts toward higher frequencies as the temperature increases, confirming the relaxation mechanism previously mentioned.

Figure 14 shows the plots of the imaginary part of impedance and modulus versus frequency at 300 K. It can be seen from the figure that Z”_max and M”_max do not occur at the same frequency, indicating non-Debye-type relaxation [113].

Figure 15 shows the complex impedance plots, Z″ as a function of Z′. In the temperature range previously explored, the Nyquist plots consist of semicircular arcs. For low temperatures, the large radius indicates the high-insulating nature of the material. With further increasing temperature, the radius of the distorted semicircular arc decreases, indicating that the conductivity increases with increasing temperature. For higher temperatures, the curves bend toward the real Z’-axis to form distorted semicircular arcs that support the non-Debye-type relaxation phenomenon.

The distorted nature of the semicircular arcs, as highlighted in the inset for 300 K, may imply the existence of multi relaxation processes in the sample, which were not revealed by the previous analyses. Also, in the inset of Figure 15, the Bode-phase diagram allows us to infer the existence of one additional relaxation process in the low-frequency region, since a step-like change is starting to appear [114]. To make a clearer understanding of the multirelaxation behavior in the sample, a simulation of complex impedance data was performed with equivalent electric circuits.

Often, a single data set can be reasonably fitted by multiple equivalent circuits, making the choice dependent on both simplicity and consistency with the known physical and chemical processes occurring in the system [115].

In this study, the impedance data was modeled employing an equivalent circuit, containing a series combination of two parallel (R//CPE) elements (as also depicted in the inset of Figure 15), having the model circuit parameters been obtained using the EIS Spectrum Analyzer fitting software [116]. The simulated response of the equivalent circuit for 300 K is also shown in the inset of Figure 15.

The impedance of a constant phase element is given by:

Z=Af−ncosnπ2−jsinnπ2E9

where A is the impedance observed for the angular velocity of 1 rad/s, f is the frequency, and n is an exponent equal to 1 for the case of a pure capacitor. The CPE was used to replace capacitance to account for non-ideal Debye behavior [117].

The semicircular arc at high frequencies arises mainly from the grain properties, while the appearance of a second semicircle at a lower frequency can be attributed to the grain boundary properties [118].

Table 3 summarizes the estimated resistance and capacitance values by fitting the experimental data using the mentioned software.

T (K)	R1 (MΩ)	R2 (MΩ)	Q1 (pS)	n1 (a.u.)	Q2 (pS)	n2 (a.u.)
300	49.61	245.63	11.13	0.966	13.78	0.977

Table 3.

Fitted results for the grain and grain boundary resistance, R₁ and R₂, and corresponding admittances, Q₁ and Q₂, of the CPE elements, at 300 K.

7. Final remarks and outlooks about HAp synthesis methods

Over the years, substantial progress has elevated HAp from a basic biocompatible material to a sophisticated functional substance with diverse applications.

The synthesis of HAp is a multifaceted field that encompasses a wide range of methods, each with its advantages and limitations. Dry methods, such as solid-state and mechanochemical synthesis, offer scalability and simplicity but often yield products with lower purity and irregular morphology. Wet methods, including chemical precipitation, hydrolysis, and sol-gel synthesis, enable precise control over particle size and composition, making them suitable for high-performance applications. High-temperature methods and hybrid techniques further expand the toolkit for HA synthesis, providing avenues for producing materials with tailored properties.

Although numerous methods have been explored for HAp synthesis, only a few are assumed viable when considering both performance and economic factors. The primary challenges associated with most techniques include the reliance on various precursors, solvents, or surfactants, along with issues such as broad particle size distribution, complex and costly procedures, significant agglomeration, and the presence of phase impurities.

On the other hand, biogenic synthesis represents a promising direction for sustainable HAp production, aligning with the global emphasis on resource conservation and waste valorization. Biogenic synthesis involves deriving HAp from natural sources such as animal bones, eggshells, and plant materials. These methods capitalize on the natural composition of biogenic materials, reducing the need for synthetic precursors. Techniques such as thermal calcination and chemical precipitation are employed to extract and process HAp from these sources. Biogenic HAp often contains trace elements that enhance its bioactivity. Integrating advanced technologies like microwave-assisted synthesis and additive manufacturing has the potential to revolutionize the production and application of HAp in biomedical engineering.

Acknowledgments

This research is sponsored by national funds through FCT—Fundação para a Ciência e Tecnologia, under projects UIDB/00285—Centre for Mechanical Engineering, Materials and Processes and LA/P/0112/2020, and by i3N (LA/P/0037/2020, UID-B/50025/2020, and UID-P/50025/2020) financed by the COMPETE 2020 Program and national funds through FCT/MEC and FEDER under the PT2020 Partnership Agreement.

Conflict of interest

The authors declare no conflict of interest.