Studying the electrical, mechanical, and biological properties of BCZT–HA composites

Abstract

The piezoelectric properties of natural bone and their influence on bone growth have inspired researchers to study a range of bio-piezoelectric composite materials. By exploring these materials, researchers aim to understand better, how piezoelectricity can be controlled to promote bone growth and tissue regeneration. In this work, the prominent piezoelectric material, $(\text{Ba},\text{Zr}){\text{TiO}}_3-\text{x}(\text{Ba},\text{Ca}){\text{TiO}}_3$, abbreviated as BCZT, was selected as a possible bone growth enhancer in hydroxyapatite (HA) scaffolds. Initially, BCZT and hydroxyapatite (HA) powders were synthesized using the sol–gel method. Subsequently, various composite samples of BCZT–xHA were prepared using the conventional solid-state method. After sintering the samples at 1300°C, the phase structure, microstructure, density, and electrical properties were characterized. The samples’ compressive strength was determined by analyzing the outcomes of basic compression tests. The biological behavior of the samples in terms of in vitro simulated body fluid immersion and MTT tests were evaluated. Our results revealed that among the BCZT–xHA samples, the BCZT-20HA sample had the best composition, considering its electrical, mechanical, and biological properties. A ${\text{d}}_{33}$ value of 10 pC/N, dielectric permittivity of 110, and the ${\text{g}}_{33}$ equal to 10.27 mV m/N resulted in the output voltage of 1.03 V. The results of the MTT assay test confirmed the noncytotoxic nature of the samples with the highest optical density in the BCZT-20HA sample.

1 |. INTRODUCTION

A lot of work has been done in recent years to stimulate the nanostructure biomaterials to repair damaged bone tissue. Since piezoelectricity is a mechanism that fortifies the natural bone, it seems that the piezoelectric biomaterials are good candidates to replace conventional implants. The electrical characteristics of bone consist of dielectric, ferroelectric, and piezoelectric properties, which result in improvement and reconstruction of impaired bone. Therefore, producing biomaterials with electrical and mechanical properties similar to those of natural bone is very challenging.^1–3

The most widespread bone grafting materials include calcium phosphate, hydroxyapatite (HA), bioactive glasses, zirconia, barium titanate (BT), and so forth.^4–7 Among these materials, bioactive HA has been widely developed in medical applications.^8–11 HA can be compared with the mineral part of bone and tooth and shows suitable bonding with natural bone. However, because of inferior mechanical and piezoelectric properties, its application is limited.¹² Lead-free piezoelectric materials such as BT and potassium-sodium niobate (KNN) have been used in many in vitro biological applications as implants and coating.¹³ A new solid solution based on BT with outstanding dielectric and piezoelectric properties was introduced in 2009. In recent years, the solid solution of $(\text{Ba},\text{Zr}){\text{TiO}}_3-\text{x}(\text{Ba},\text{Ca}){\text{TiO}}_3(\text{BZT}-\text{xBCT})$ has been the focus of many publications.^14–16 In 2015, Acosta et al.¹⁷ studied the chemical and biological properties (topography, wettability, chemical stability, and cytotoxicity) of some ferroelectric materials, including BZT-xBCT. According to their results, ferroelectric materials are good candidates for biological scaffolds, but their interactions with the physiological environments are complex and require further investigation for practical applications. In another study, in 2019, Poon et al.¹⁸ investigated the cell proliferation and viability of BCZT samples and compared the results with polystyrene control group. These researchers declared that BCZT has considerable potential for cell stimulation and bone graft applications. Manohor et al. studied the biological behavior of HA–xBCZT composites. They synthesized the nano powders of HA and BCZT and fabricated this composite with different amounts of BCZT from 0 to 50 wt %. They also confirmed the bioactivity and bone generation of these composites due to the presence of the piezoelectric component BCZT.¹⁹

Tang et al. provided a situation similar to reality to study the effects of piezoelectricity on bone growth in HA/BT composite. They designed a device that applied dynamic loads as large as human motion, and the sample was cultured simultaneously. They concluded that the presence of BT promotes bone repair, and the composite sample with 90 wt % BT and 10 wt % HA showed the best biocompatibility.²⁰

In another study, Zhang et al. declared that piezoelectric effect plays a significant role in bone growth, bone repair, and healing of damaged body organs. They fabricated aligned porous HA/BT composite samples with the ice templating method, studied the microstructure, piezoelectric, and mechanical properties, and evaluated the biocompatibility of the samples. Accordingly, highly porous HA/BT composite with lamellar microstructure had similar piezoelectricity to bulk samples with no cytotoxicity effect in any solid loading of the samples.²¹

In this study, BCZT and porous HA nano powders were separately synthesized via the sol–gel route. BCZT–xHA composites were fabricated by conventional pressing and sintering method using different amounts of HA from 0 to 50 wt %. For pure and composite samples, the electrical, mechanical, and biological properties were investigated in detail.

2 |. MATERIALS AND METHODS

2.1 |. Synthesis of porous HA nano powder

Raw materials of calcium nitrate $(\text{Ca}{\left({\text{NO}}_3\right)}_{2\cdot }4{\text{H}}_2\text{O}$, Merck, 99%), ammonium dihydrogen phosphate(${\text{NH}}_4{\text{HPO}}_4$, 99%), gelatin from porcine skin (Sigma-Aldrich, gel strength 300, type A), and ammonia solution (Merck, 25%) were used to synthesize HA powder.

First, stoichiometric amounts of calcium nitrate and dihydrogen phosphate were separately dissolved in deionized water to get 500 mL solutions. Gelatin powder was completely dissolved in acetic acid 2 vol/vol % at 40°C with a magnetic stirrer and was subsequently added to calcium nitrate solution. To control the pH of the solution, ammonia was added to each of the solutions to reach a pH of 11 and was stirred for another 5 min. The solution containing ammonia, calcium nitrate, and gelatin was slowly added to the ammonium dihydrogen phosphate solution. The initial sol of HA was stirred for 48 h at room temperature to reach a white precipitate. The precipitate was then washed with distilled water until a pH of 7 was reached. The produced gel was then heated at 80°C for 24 h. The dried gel was crushed into powder using a mortar. The calcination was performed at 600°C for 1 h to synthesize the porous HA nano powder.

2.2 |. Synthesis of BCZT powder

In order to synthesize the $\left({\text{Ba}}_{0.85}{\text{Ca}}_{0.15}\right)\left({\text{Zr}}_{0.1}{\text{Ti}}_{0.9}\right){\text{O}}_3$ powder with sol–gel method, raw materials of barium acetate $\left(\text{Ba}\left({\text{C}}_4{\text{H}}_6{\text{O}}_4\right)\right)$, Merck, 99%), calcium acetate $(\text{Ca}\left({\text{C}}_4{\text{H}}_6{\text{O}}_4\right)$, Merck, 99%), tetra-n-butyl orthotitanate ($\text{Ti}({\text{C}}_{16}{\text{H}}_{36}{\text{O}}_4$, merck 98%.), zirconium propoxide $(\text{Zr}\left({\text{C}}_{12}{\text{H}}_{28}{\text{O}}_4\right)$, Merck, 98%), acetylacetone (${\text{C}}_5{\text{H}}_8{\text{O}}_2$, Merck, 99%), and ammonia solution to control the pH value were used. The stoichiometric amounts of barium acetate and calcium acetate were separately solved in deionized water at room temperature to meet the specific volume. Then the solutions were mixed together, and appropriate amounts of acetic acid were added to the acetate solutions (molar ratio of 1:15) and mixed for 1 h at room temperature (solution 1). In order to prepare the second solution, stoichiometric amounts of n-butyl orthotitanate were added to ethanol with molar ratio of 1:12 and stirred at 60°C for 30 min. Afterward, the required amount of zirconium isopropoxide was added to the solution and stirred for another 30 min at 60°C. Then, acetyl acetone was added to solution 2 (the molar ratio of acetyl acetone to $\text{Zr}/\text{Ti}$, 4:3) and stirred at 60°C for 1 h. Eventually, solation 1 was added to solation 2. The addition of ammonia solution yielded a pH of 7. The resulting solution was heated at 110 C to form the BCZT gel. A yellow gel was produced and was heated at 90°C for 12 h to reach a brown gel, which was then crushed in a mortar and pestle. The calcination process was performed at 900°C for 4 h, based on the previous works.²² BCZT–xHA nanocomposite (0–50 wt %HA) was prepared in a planetary mill (Amin Asia Fanavar, Iran). These samples were abbreviated as BCZT, BCZT-10HA, BCZT-20HA, BCZT-30HA, BCZT-50HA, and HA. Milling was performed in ethanol using the zirconia container and balls with a ball-to-powder ratio of 1:20 at 220 rpm for 4 h. The required amounts of BCZT and HA were weighed and after milling, the slurry was dried at 90°C for 24 h. The powders were shaped into disks with a diameter of 10 mm and a thickness of 2 mm using a simple steed die and the final pressure of 200 MPa was applied isostatistically with an isostatic press device (CIP303/Iran). The samples were sintered at 1250–1500°C for 4 h. The Archimedes method was used to measure the bulk density of the samples. To evaluate the phase structure of the synthesized HA and BCZT powders, and the composite samples, x-ray diffraction analysis was used (XRD-Rigaku Ultima IV, Japan, Cukα radiation and steps and exposure times of 0.05 and 1 s over 2-$\theta$ range of 10–80°). Fourier transform infrared spectroscopy (FTIR) (Shimadzu 8300 model) was used to investigate the molecular structure. The morphology of the powders and the fracture surface of the samples were studied with field emission scanning electron microscopy (FESEM) (TESCAN VEGAN/XMU), and the Energy-dispersive x-ray spectroscopy (EDAX) and the corresponding maps were used to evaluate the chemical composition of the samples. The specific surface area of the powders was measured using the BET device (BELSORP MINI II, Japan). The compressive strengths of the samples (diameter of 8 mm and a thickness of 10 mm) were calculated using the data of stress–strain curves obtained from a DMG Universal testing machine (Model 7166, United Kingdom) with a speed of 0.5 mm/min (based on ASTM C1428 (28)).

To measure the electrical properties, the samples were electroded using silver paste followed by heating at 800°C to ensure ohmic contact. Dielectric properties were measured using a network analyzer (HP 8714C) in the frequency range of 1 Hz–1 MHz. The ferroelectric behavior of the unpoled samples was studied using the hysteresis loops obtained from a sawyer-tower-based device (Ferroelectric tester, (Kashan, Iran)). The poling process was performed at 40–60°C with the poling field of 3 kV/mm for 40–60 min. After aging for 24 h, the piezoelectric coefficient was measured with a ${\text{d}}_{33}$-meter (YE2730 ${\text{d}}_{33}$ meter).

In order to investigate the biological behavior of BCZT–xHA composites, an in vitro biodegradability test was carried out by immersing the samples in the simulated body fluid (SBF) solution. The solution was prepared based on the instruction of Kokubo and Takadama, which is found elsewhere.²³ Accordingly, the samples were immersed in this solution under the control of the surrounding bath; a pH value of 7.5 and temperature of 37°C for 28 days. After that the surface was dried carefully and the sedimented particles were investigated by SEM imaging, followed by EDAX analysis and mapping.

The biocompatibility of BCZT–xHA samples was evaluated by cell culture using MC3T3 cells. The viability of the cells was assessed by live/dead staining by Acridine orange (Merck) and propidium iodide (Sigma-Aldrich). Cells cultured on the membrane’s surface for 48 h were followed by removal of the medium and PBS washing three times. Loaded-cell samples were incubated in a mixture of 5 $\text{μM}$ Acridine orange and 5 $\text{μM}$ propidium iodide in 10 mL of culture medium, to simultaneously determine the existence of live (green) and dead (red) cells. After the incubation period (30 min, at room temperature), cell-membrane scaffolds were observed under a fluorescence microscope (Olympus CKX53).

The toxicity was measured using a 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazolium bromide (MTT; Sigma, St. Louis, MO, USA) colorimetric assay. Cells were seeded in a 96-well plate (NUNC, Rochester, NY, USA) (30,000 cells/well), grown for 24 h, and then treated for 24 h with culture media containing various concentrations. After 24 h, the medium was removed and was washed with PBS. Afterwards, it was replaced with an MTT-containing medium (20 μL MTT in 180 μL FBS-free medium). The formazan was dissolved by the addition of DMSO to each well (100 μL). The mixture was shaken for about 15 min. The microplate reader was used to determine the optical density at 570 nm.

3 |. RESULTS AND DISCUSSION

Figure 1 shows the XRD patterns of BCZT and HA powders synthesized at 900 and 600°C, respectively. As shown in Figure 1A, the single-phase perovskite BCZT with a tetragonal structure is formed, which is consistent with the ${\text{BaTiO}}_3$ standard card of 96-151-3253.

FIGURE 1.

X-ray diffraction patterns of sol–gel synthesized (A) BCZT and (B) hydroxyapatite (HA) powders, synthesized at 900 and 600°C, respectively, and (C) the Fourier transform infrared spectroscopy spectrum of HA powder after calcination at 600°C.

In the $\left({\text{Ba}}_{1-\text{x}}{\text{Ca}}_{\text{x}}\right)\left({\text{Zr}}_{\text{y}}{\text{Ti}}_{1-\text{y}}\right){\text{O}}_3$ composition, $x$ and $y$ values control the crystal structure, which severely affects the piezoelectric properties, as is clearly mentioned in the literature. The presence of a morphotropic phase boundary between the tetragonal and rhombohedral phases and later on, the existence of a tricritical point between the cubic, tetragonal, and rhombohedral phases were considered to be the reason for superior piezoelectric properties of ${\text{Ba}}_{0.85}{\text{Ca}}_{0.15}{\text{Zr}}_{0.1}{\text{Ti}}_{0.9}{\text{O}}_3$ composition.^16,24 Generally, in the XRD pattern of the piezoelectric ceramics the peak splitting at $2\theta =45°$ confirms the formation of a polar tetragonal structure, and overlapping of the peaks implies that a combination of tetragonal and rhombohedral phases exists. Additionally, like other piezoelectric materials, as the crystal structure of BCZT approaches the nonpolar cubic phase, it would be accompanied by inferior piezoelectric properties.²⁵

In the XRD pattern of HA powder (Figure 1B), the only candidate was hexagonal HA with reference number 96-900-3549 and no secondary phases were detected. HA crystallizes in two different structures: hexagonal and monoclinic, for which the latter exists at high temperatures (>850°C). The formation of hexagonal or monoclinic structures of HA can be controlled in wet chemical methods.

Considering the piezoelectric properties, although it is recently reported that the piezoelectric activity is more pronounced in the monoclinic HA phase, the hexagonal phase of HA is feasibly formed in the conventional low temperature chemical methods.^26,27 The FTIR spectrum of HA powder after calcination at 600°C is demonstrated in Figure 1C. The characteristic peaks of ${\text{OH}}^{-}$, and ${{\text{PO}}_4}^{-}$ at the corresponding wavenumbers between 500 and 4000 cm⁻¹ are clearly observed. Furthermore, the presence of $\text{O}-\text{H}$ and $\text{P}-\text{O}$ bands and the formation of a typical HA powder, is in good agreement with the FTIR spectra of sol–gel synthesized HA in other publications.²⁸

The SEM micrographs of synthesized powders are illustrated in Figure 2. According to Figure 2A, BCZT particles were in the range of 200–500 nm, whereas HA powder containing small agglomerates with nanometer particles, that is, average diameters less than 50 nm. The Brunauer-Emmett-Teller (BET) surface area of BCZT and HA powders were 4.3 and 118.2 m²/g, respectively. Considering the spherical morphology for both powders, the average particle size was calculated. Average particle size of BCZT with bimodal particle size distribution was 250 nm for larger particles and 20 nm for smaller particles and the determined particle size of HA powder was 20 nm. These values were in good agreement with microscopic results.

FIGURE 2.

SEM micrographs of (A) BCZT and (B) hydroxyapatite powders after sol–gel synthesis at 900 and 600°C, respectively.

It is known that HA suffers from weak mechanical properties and different solutions are proposed to improve this weakness. Among which, the fabrication of nano-bioceramics has been regarded as an effective approach.²⁶ Additionally, the mesoporous HA nanoparticles can improve the biological behavior and act as nanocarriers in drug delivery applications. As a result, a mesoporous nano HA powder would be a good candidate for biological applications.²⁹

The XRD patterns of BCZT-HA composite samples, which were sintered at 1300°C for 4 h are demonstrated in Figure 3A–F. As shown, the pure BCZT ceramic (Figure 3A) with a perovskite lattice has an orthorhombic structure, which is matched with barrio-perovskite ${\text{BaTiO}}_3$ with reference code 96-901-4493. Some traces of ${\text{CaTiO}}_3$ and calzirtite $\left({\text{Zr}}_{40}{\text{Ti}}_{16}{\text{Ca}}_{16}\right)\text{(128)}$ with JCPDS numbers 96-900-6176 and 96-901-5701, respectively is also found in this sample. On the other side of the XRD graph, the pattern of a pure HA sample is demonstrated in Figure 3E. This pattern is well fitted with hydroxylapatite, OHAP, (${\text{Ca}}_{10}{\text{P}}_6{\text{O}}_{26}{\text{H}}_2$, JCPDS: 96-901-1092), and calcium phosphate (${\text{Ca}}_{24}{\text{P}}_{16}{\text{O}}_{64}$, JCPDS: 96-210-5286) phases with hexagonal and monoclinic structures, respectively. Although characteristic peaks of HA at $2\theta =26$ and 32° are clearly seen, at this sintering temperature the decomposition reaction occurred, and calcium phosphate substituted a part of HA.

FIGURE 3.

The x-ray diffraction patterns of BCZT–xHA composite samples sintered at 1300°C, (A) pure hydroxyapatite, (B) $x=50$, (C) $x=30$, (D) $x=20$, (E) $x=10$, and (F) pure BCZT.

As it is clearly revealed in the XRD pattern of HA, the peaks in the range of $2\theta =10-50°$ are completely matched with calcium phosphate, and the rest is fitted with HA. It means that a large amount of HA is decomposed into calcium phosphate, which is rational at a high sintering temperature of 1300°C. Liao et al. studied the phase decomposition of HA from room temperature to 1500°C, and concluded that in the temperature range of 1000–1360°C, hydroxyl ions are released and HA is changed to OHAP and at higher temperatures it will decompose into tricalcium phosphate (TCP), which confirms the results of this study.³⁰ Raynaud et al. investigated the effects of sintering temperature and time in HA compositions with different Ca/P ratios. According to their results, for Ca/P equal to 1.667, the decomposition occurred at 1300°C; after 1 h, a significant amount of HA was substituted with tricalcium phosphate and tetracalcium phosphate monoxide, and after 2 h, the HA was decomposed entirely.³¹ The XRD patterns of composite samples are also shown in Figure 3B–E. Phase identification with X’Pert High Score software implied that apart from the two original phases (BCZT and HA), some intermediate compounds are also formed in these samples. The different phases are marked with individual symbols on the XRD patterns. According to the XRD patterns of Figure 3B–E, the peak at $2\theta =39°$ is intensified by increasing BCZT, as in the BCZT-10HA sample, this sharp peak overcame the highest characteristic peak of BCZT at $2\theta =32°$. In the BCZT-10HA sample with the highest BCZT content, this sharp peak is marked with calzirtite (${\text{Zr}}_{40}{\text{Ti}}_{16}{\text{Ca}}_{16}{\text{O}}_{128}$, JCPDS: 96-901-5701) and dibarium phosphate (${\text{Ba}}_6{\text{P}}_6{\text{O}}_{27}$, JCPDS: 96-200-3763) phases. It should be noted that in composite samples, a shift to higher angles is observed in BT peaks, which can be due to the Ca²⁺ substitution from the HA phase in the perovskite structure of BCZT as well as the ${\text{CaTiO}}_3$ formation. The formation of calcium titanate and barium phosphate secondary phases is reported in previous works of Baxter et al.,⁵ and Vouilloz et al.³²

Figure 4 illustrates the FESEM micrographs of thermal etched surfaces of BCZT, HA, and composite samples sintered at 1300°C. The grain size distribution curves of pure HA and pure BCZT, which were plotted from the ImageJ results, are demonstrated in the corner of each image. The uniform and dense microstructure of Figure 4A belongs to pure HA with the most frequent grains of less than 1 μm.

FIGURE 4.

Field emission scanning electron microscopy micrographs of BCZT–xHA composite samples sintered at 1300°C, (A) HA, (B) $x=50$, (C) $x=30$, (D) $x=20$, (E) $x=10$, and (F) BCZT.

On the other side of the micrographs a similar microstructure with larger grains is observed in Figure 4F, which is related to pure BCZT. The grain distribution is demonstrated in the corresponding ImageJ graphs. With increasing the BCZT content from 50 (Figure 4B) to 90 wt % (Figure 4E), the grain size of composite samples increased, as it was about 500 nm in the BCZT-50HA and reached more than 3 μm in the BCZT-10HA. The bimodal microstructure of composite samples was observed in Backscattered (BS) photographs, which are further analyzed with EDS and mapping.

The elemental analysis of all samples revealed the compositional difference as confirmed by the results of EDS spectra at different points. It was performed for all samples, and among the composite samples, the typical EDS results of BCZT-20HA sample as well as EDS mapping are demonstrated in Figure 5. For BCZT-20HA, the EDS spectra of points A to D were analyzed in detail and the amounts of the constituent elements in atomic percent are reported in Table 1. Accordingly, in points B and D, in which the atomic percent of P is zero, the calculated perovskite phases are ${\text{Ba}}_{0.088}{\text{Ca}}_{0.912}{\text{Zr}}_{0.253}{\text{Ti}}_{0.9286}{\text{O}}_{3-\text{x}}$ and ${\text{Ba}}_{0.849}{\text{Ca}}_{0.156}{\text{Zr}}_{0.253}{\text{Ti}}_{0.747}{\text{O}}_{3-\text{x}}$, based on calcium titanate and BT compositions, respectively. It seems that the dominant phase at point A is ${\text{Ca}}_{3.93}{\text{P}}_6{\text{O}}_{25.06}$, a calcium phosphate composition. Considering the EDS mapping, point C is calcium deficient, but a considerable amount of Ba and P along with small amounts of $\text{Ti}$ and $\text{Zr}$ in this region implies that barium phosphate and calzirtite are the prevailing phases. The estimated phases are in good agreement with the XRD results of Figure 2.

FIGURE 5.

The EDS spectra and the elemental maps of BCZT-20HA composite sample sintered at 1300°C.

TABLE 1.

A summary of electrical properties of BCZT–xHA samples.

Sample	Dielectric permittivity	Dielectric loss	${\text{d}}_{33}$ (pC/N)	${\text{g}}_{33}$ (mV m/N)	Output voltage (V)
HA	40	0.013	0.5	5.65	0.57
BCZT-50HA	51	0.011	5	11.1	1.11
BCZT-30HA	74	0.007	7	10.84	1.08
BCZT-20HA	110	0.006	10	10.27	1.03
BCZT-10HA	799	0.005	21	2.97	0.30
BCZT	3041	0.003	140	5.2	0.52

Abbreviation: HA, hydroxyapatite.

Figure 6A shows the variations in bulk and relative densities of pure HA, pure BCZT, and BCZT–xHA composites with the weight fraction of HA. According to the graphs, increasing the amount of HA in the BCZT–xHA composites will decrease the bulk density of the composite, which is attributed to the lower theoretical density of HA compared with BCZT. As the amount of HA increases, the possibility of the formation of secondary phases with lower density is increased. Therefore, the lower density of composite samples can be attributed to these secondary phases.³³ Accordingly, among the composite samples, BCZT-10HA has the highest relative density.

FIGURE 6.

The variations in the (A) density and (B) compressive strength of BCZT–xHA composite samples sintered at 1300°C. HA, hydroxyapatite.

Figure 6B shows the compressive strength values of BCZT–xHA composite samples sintered at 1300°C. The calculated compressive strengths of pure HA and BCZT samples are equal to 153 and 200 MPa, respectively. According to this graph, the compressive strength of composite samples shows a descending trend by increasing the HA content. Accordingly, the compressive strength of the BCZT-10HA sample is 198 MPa, which is higher than other composite samples. Although BCZT has a higher compressive strength than HA, the compressive strengths of BCZT-30HA and BCZT-50HA samples are lower than pure HA. As mentioned earlier, the presence of secondary phases at the sintering temperature of 1300°C leads to the formation of porosities in the BCZT-HA samples and reduces the compressive strength. Additionally, the density of the biological ceramics severely affects the mechanical properties, such as compressive strength, as is declared in the previous studies.³³ Considering that the compressive strength of normal bone under vertical forces is 133 MPa,³⁴ our BCZT–xHA samples have similar or even better mechanical properties.

Figure 7A shows the variations in the dielectric permittivity $(ϵ)$ and dielectric loss $(\text{tan}\delta )$ for pure and composite samples at measuring frequency of 1 kHz. As it is shown, by increasing the HA content, the dielectric loss encountered an ascending trend and dielectric permittivity decreased. The dielectric permittivity depends on characteristics such as the nature of dielectric material, microstructure, distribution of crystalline phase, and the density of the sample.³⁴ As a result, the combination of a low dielectric constant HA with BCZT causes a sudden decrease in the dielectric constant of the composite samples compared with pure BCZT, which is similar to the reported data for BT-HA samples.³⁵ Accordingly, the BCZT-10HA composite sample has the highest amount of dielectric permittivity among the composite samples. As seen in Table 1, the lowest dielectric permittivity of 40 corresponds to pure HA, whereas other composite samples have higher dielectric values. Jiao et al. synthesized BT/HA composite via the hydrothermal method and reported the highest dielectric permittivity of 27.4 for composite powders prepared in the solvent of ethanol.³⁶ In another study, Bowen et al. measured the dielectric properties of HA-xBT (0–100 vol %) samples sintered at 1300°C. Their results revealed that the dielectric permittivity of HA-xBT samples at the frequency of 100 Hz was in the range of 30–800.³⁷ There are some factors influencing the piezoelectric constant, of these, dielectric permittivity can be mentioned. According to the following equation, a high dielectric constant $\left({\epsilon }_{33}\right)$ would result in a large piezoelectric coefficient (${\text{d}}_{33}$).³⁸

$${\text{d}}_{33}=2P_{\text{s}}{\epsilon }_{33}{Q}_{33.}$$ (1)

FIGURE 7.

The variations in (A) dielectric properties and (B) piezoelectric coefficient ${\text{d}}_{33}$ with hydroxyapatite (HA) content in BCZT–xHA samples sintered at 1300°C.

In this equation, ${\text{d}}_{33}$ is the piezoelectric constant, $P_{\text{s}}$ is the spontaneous polarization, ${\epsilon }_{33}$ is the relative permittivity, and $Q_{33}$ is the electrostrictive coefficient.

After polarizing the samples, their piezoelectric coefficient ${\text{d}}_{33}$ was measured, and the corresponding diagram is shown in Figure 7B. The piezoelectric coefficient of pure BCZT and pure HA sintered at 1300°C was 140 and 0.5 pC/N. The corresponding values for composite samples are 21, 10, 7, and 5 pC/N, respectively. Manohor et al. reported HA-BCZT composites with different BCZT values from 0 to 100 wt %. In this paper, the piezoelectric coefficient of HA was not mentioned, but the ${\text{d}}_{33}$ value of the BCZT-50HA composite sample (sintered at 1350°C and polarized under 10 kV/mm electric field at 95°C for 4 h) was 7 ± 1pC.¹⁹ Bowen et al. reported a ${\text{d}}_{33}$ value of 6 pC/N for BT-HA samples containing higher than 90 vol % BT, which is lower than the corresponding values of our BCZT-10 wt % HA sample.³⁷ Considering the dielectric permittivity and piezoelectric ${\text{d}}_{33}$, the voltage coefficient of piezoelectric materials is calculated by the following formula:

$${\text{g}}_{33}=\frac{{\text{d}}_{33}}{\epsilon }\left({\text{Vm}}^{-1}{\text{Pa}}^{-1}\right).$$ (2)

As a result, increasing the amount of HA in BCZT–xHA samples first increased the ${\text{g}}_{33}$ value of samples, so as it was maximized in the BCZT-50HA sample. In addition, having the lowest values of ${\text{d}}_{33}$ and $ϵ$, pure HA showed the lowest value of the piezoelectric voltage coefficient. Applying a moderate pressure of 1 MPa to these composite samples with a specified thickness of 1 μm, the maximum output voltage of more than 1 V is generated in BCZT–xHA samples with 0.2 < × < 0.5, which is twice the values of previous reports in similar condition.

Figure 8 shows the room temperature ferroelectric behavior of pure BCZT, HA, and composite samples after sintering at 1300°C. As, there is a relationship between piezo-, pyro-, and ferroelectricity, the importance of ferroelectricity in biological applications cannot be ignored. A ferroelectric material having piezoelectric, pyroelectric, and dielectric properties would be a promising candidate in biological applications.³⁹ As a result, it would be interesting to evaluate the ferroelectric nature of the so-called piezo-bio composite materials.

FIGURE 8.

Ferroelectric hysteresis loops of BCZT–xHA samples sintered at 1300°C, measured by applying the electric field of 3000 V/mm.

Pure BCZT with a typical ferroelectric behavior has a coercive field, $E_{\text{c}}$, of 0.4 kV/mm and its remanent polarization, $P_{\text{r}}$, is 6.5 μC/cm². Among the composite samples, a suitable ferroelectric behavior is observed in the BCZT-10HA sample with saturation polarization and residual polarization of 1.6 and 0.35 μC/cm², respectively. The corresponding values of $P_{\text{s}}=0.51$ and $P_{\text{r}}=0.016$ μC/cm² were obtained for BCZT-20HA sample, which were much larger than the ferroelectric properties of the dry cortical bone ($P_{\text{r}}=0.00068$ μC/cm²).⁴⁰ By increasing the weight percentage of HA in other composites, an approximate linear behavior with very low residual polarization was observed, and they did not reach the saturation state by applying the voltage of 3 kV/mm. Mezzourh et al.⁴¹ synthesized BCZT by sol–gel method and sintered at 1420°C; they reported the saturation polarization of 9.3 μC/cm², lower than the results of this study ($P_{\text{s}}=11.5$ μC/cm²). According to the results of Buatip et al.,⁴² a BCZT sample sintered at 1300°C for 4 h has a coercive field of 0.43 kV/mm, and a remanent polarization of 1.243 μC/cm, which is lower than the corresponding values of our BCZT.

The graphs of the dielectric constant versus frequency are demonstrated in Figure 9. As it is clearly seen in Figure 9A, BCZT has a different behavior and sudden variations occurred at resonance and antiresonance frequencies, which is the characteristic of piezoelectric materials. Among the composite samples, a similar behavior with lower permittivity values is observed in BCZT-10HA samples. By increasing the amount of HA, as is illustrated in Figure 9B, the dielectric values decreased and almost analogous variations with frequency are observed in these composite samples. Although a sudden change in dielectric permittivity of BCZT and BCZT-10HA samples occurred at 400 kHz, other samples experienced slow variations before 1 kHz. These results agree with BT-HA composites of previous studies,^34,37 which declared a similar permittivity dispersion for BT-HA samples in frequencies lower than 10 kHz. It is clear that in dielectric solids increasing the frequency paralyzes the polarization mechanisms one by one, resulting in lower dielectric permittivity.⁴³ Additionally, the decrease in the dielectric constant with increasing frequency is justified by higher conductivity at higher frequencies, which is attributed to ionic conductivity in dielectric materials.⁴⁴

FIGURE 9.

The variations in dielectric permittivity of BCZT–xHA samples with (A, B) frequency and (C, D) temperature.

Figure 9C,D demonstrates the variations in dielectric permittivity with temperature measured in the range of room temperature to 150°C at 1 kHz. Pure BCZT and BCZT-10HA samples have similar trends with Curie temperatures of 111 and 89°C, and maximum permittivity of 6985 and 1604, respectively. For other composite samples, inferior maximum primitivities and lower Curie temperatures were observed. Since the Curie temperature arises from the ferroelectric nature of dielectric materials, BCZT–xHA composites with HA content higher than 20 wt % hardly demonstrate the typical Curie temperature. Dang et al. studied the temperature dependence of dielectric permittivity in BT-HA composites. They reported a more noticeable variation in composites with BT content higher than 26 vol % and attributed this behavior to the phase interactions in adjacent BT particles.³⁵

Pure BCZT, pure HA, and BCZT–xHA composite samples sintered at 1300°C for 4 h were immersed in SBF solution with a pH value of 7.5 at a temperature of 37°C for 28 days. The SEM micrographs of Figure 10 depict the morphology of surface deposits and the amount of sedimented particles is reported in Table 2.

FIGURE 10.

SEM micrographs showing the deposited particles on the surface of BCZT–xHA samples sintered at 1300°C, after 28 days immersion in simulated body fluid solution; the corresponding energy dispersive x-ray spectroscopy spectra of the dominant sediments are also demonstrated near each image.

TABLE 2.

The variations in weight of different BCZT–xHA samples after immersion in simulated body fluid solution for 28 days.

Sample	Weigh before immersion (g)	Weigh after immersion (g)	Weigh changes (%)
HA	0.102	0.104	1.96
BCZT-50HA	0.150	0.153	1.330
BCZT-30HA	0.165	0.167	1.210
BCZT-20HA	0.197	0.200	1.500
BCZT-10HA	0.224	0.226	0.890
BCZT	0.241	0.243	0.820

Abbreviation: HA, hydroxyapatite.

As is clearly seen in the SEM images, nearly uniform deposits are formed on the surface of all the samples. However, some cuboid particles are obviously observed in BCZT-50HA sample. In order to verify the chemical composition of deposits, EDAX analysis was performed in three different points of SEM images and the results are summarized in Table 3. Apart from pure HA with the largest amount of bonelike apatite sediments, among the composite samples, the surface of BCZT-20HA composite sample was covered with the highest amount of apatite particles, which is confirmed by EDAX analysis of Figure 10. Increasing the weight percentage of HA in the composite, the amount of sedimented apatite particles is expected to increase. However, the formation of cuboid NaCl crystals on the surface of the BCZT-50HA sample lowered its biocompatibility.

TABLE 3.

The summary of energy dispersive x-ray spectroscopy analysis from the deposited particles on the surface of BCZT–xHA samples sintered at 1300°C, after 28 days of immersion in simulated body fluid solution.

		Element (at %)
Points		O	Na	P	Cl	Ca
HA	A	48.81	4.40	10.64	19.06	17.08
	B	68.62	0.55	12.97	0.48	17.37
	C	31.63	15.96	6.56	35.92	9.93
BCZT-50HA	A	52.73	5.25	10.71	15.82	15.49
	B	6.27	49.08	0.08	44.46	0.11
	C	66.47	4.89	4.49	19.38	4.78
BCZT-30HA	A	69.79	1.84	7.60	12.77	8.00
	B	25.25	28.05	3.93	40.43	2.33
	C	41.10	22.54	4.37	27.87	4.12
BCZT-20HA	A	51.92	21.78	4.16	18.01	4.13
	B	51.11	1.43	10.35	21.89	15.22
	C	69.92	4.65	7.03	8.53	9.87
BCZT-10HA	A	10.65	24.01	2.77	59.99	2.58
	B	64.52	4.93	6.16	16.49	7.90
	C	64.55	6.41	7.50	10.16	11.39
BCZT	A	61.86	8.73	5.00	17.19	7.22
	B	37.65	20.23	4.66	32.89	4.57
	C	10.71	36.24	3.24	47.17	2.64

Figure 11 shows the results of the cell culture experiment, which was performed with MC3T3 cells. The viability and proliferation of MC3T3 cells on the composite samples were qualitatively determined by green/red fluorescence imaging, so that the green and red colors indicate the live and dead cells, respectively. As shown in Figure 11A–F, no obvious evidence of cell death (red fluorescence) was observed in these images.

FIGURE 11.

The microscopic images of MC3T3 cells cultured on (A) hydroxyapatite (HA), (B) HA 50-BCZT50, (C) HA30-BCZT70, (D) HA20-BCZT80, and (E) HA10-BCZT90, after 7 days of culture, and (F) the plot of OD versus composition.

The results of the cytotoxicity test (MTT) are summarized in the graph of Figure 11F. The cell density after 7 days confirms the suitable biocompatibility and non-cytotoxicity of the samples. As is declared in the literature,⁴⁵ a poled piezoelectric material can stimulate bone growth. Hence, the materials with a larger piezoelectric constant are expected to be more efficient in biological applications. In the BCZT–xHA composites of this study, there should be a balance between the piezoelectric nature of the BCZT and the biological behavior of HA. As a result, there is a composite with the best piezoelectricity-affected biological behavior. As it is seen in Figure 11, the Optical Density (OD) value of the BCZT-20HA is higher than pure HA. The better cell distribution and proliferation of the BCZT-20HA sample confirms the effective role of piezoelectricity on cell culture.

Manohar et al. studied the metabolic activity of HA-x BCZT samples in an MTT test. According to their results, the HA-10BCZT sample showed the maximum proliferation among the composites. They also attributed this improved biological behavior to the piezoelectric nature of BCZT, which persuades bone growth.¹⁹ In another study, Tavangar et al. investigated the biocompatibility of HA-xBT composites and reported better cell proliferation in HA-BT samples than in pure HA.³³ As declared in the literature,⁴⁶ the mechanical stimulation of a piezoelectric scaffold can provide electrical stimulation, which results in tissue regeneration. Additionally, the poling process itself electrically stimulates the piezoelectric scaffold, which results in better cell culture and proliferation of the biological scaffold. As a result, the piezo-bio composite of this study is a promising candidate in bone tissue regeneration applications. Furthermore, studying the effect of the piezoelectric nature of this composite on nonunion bone fracture healing would be very interesting.

4 |. CONCLUSION

In this study, BCZT–xHA composite samples were fabricated from sol–gel synthesized powders. The samples sintered at 1300°C were characterized, and the electrical, mechanical, and biological properties were investigated in detail. Among the different composite samples, the one with 20 wt % HA demonstrated the optimum behavior. The electrical properties of $ϵ=110$, ${\text{d}}_{33}=10$ pC/N, $g_{33}=10.27$ mV m/N, and the calculated output voltage of 1.03 V, as well as the compressive strength of 196 MPa made this sample a good piezoelectric candidate for biological applications. The in vitro biodegradability was investigated by immersing the samples in SBF solution for 28 days, and the proliferation and the cell viability higher than pure HA confirmed the effective role of piezoelectricity on bone growth.

In vitro studies in the body-simulating solution showed that apatite was formed on the surface of the composites, which indicated the bioactivity of the BCZT–HA samples. Furthermore, the MTT cytotoxicity analysis confirmed the biocompatibility of BCZT–xHA samples.

DATA AVAILABILITY STATEMENT

The data are available on the reasonable request.