In Vitro Behavior of Boron‐Doped Baghdadite/Poly(vinylidene fluoride) Membrane Scaffolds Produced via Non‐Solvent Induced Phase Separation

Abstract

This study explores the potential of boron‐doped baghdadite (BAG) powders incorporated into poly(vinylidene fluoride) (PVDF)‐based membrane scaffolds for bone tissue engineering applications. The aim is to enhance the scaffolds’ microstructure, surface wettability, thermal behavior, mechanical properties, and biological performance. Composite scaffolds are fabricated by integrating the powders into the PVDF matrix, yielding scaffolds with enhanced material characteristics and functionality. The incorporation of the powders significantly enhances the hydrophilicity of the scaffolds, as evidenced by a notable reduction in contact angle measurements. Mechanical analyses demonstrate that the addition of boron‐doped BAG powders reduces the tensile strength and elongation at the break of PVDF scaffolds, attribute to increased pore size, reduced crystallinity, and structural heterogeneity, though the values remain within the range of human cancellous bone. Furthermore, in vitro bioactivity studies reveal the superior apatite‐forming ability of the composite scaffolds, indicating their enhanced potential for biomineralization. The results of the cellular adhesion assays indicate an enhanced affinity and proliferation of cells on the membrane scaffolds, which is indicative of improved biocompatibility. In conclusion, the developed PVDF‐based membrane scaffolds, reinforce with BAG powders, show promise as effective alternatives to traditional bone graft materials, offering scalable and versatile solutions for regenerative medicine.

Boron‐doped baghdadite (BAG)‐loaded polyvinylidene fluoride (PVDF) membrane scaffolds produced by non‐solvent induced phase separation (NIPS) show higher hydrophilicity, bioactivity and cancellous‐bone‐like mechanics. The scalable BAG/PVDF composites offer a versatile alternative to traditional bone grafts for regenerative medicine.

1. Introduction

Bone tissue injuries are the main causes of skeletal system abnormalities that can result in chronic pang, fractures, and disability.^{[ 1 ]} Despite the remarkable capacity of bones to self‐renew, a dynamic tissue, this property may prove inadequate in cases of complicated critical‐size defects that necessitate external support. Bone tissue engineering offers a promising approach in the treatment of injured tissues by combining cells with biomaterials that act as scaffolds to facilitate tissue growth.^{[ 2 ]} The usage of the scaffolds constitutes a pivotal element of this methodology, providing structural support to the affected region through the establishment of a novel extracellular matrix (ECM), thereby establishing microenvironments conducive to the proliferation of bone cells.^{[ 3} , ^{4 ]} The challenge of the bone tissue engineering approach is to design this new ECM, which will replace the cellular microenvironment in the damaged tissue, with properties similar to native bone with regards to material, structure, and function.^{[ 3} , ^{5 ]} Considering the porous and hydrophilic structure, high mechanical strength, piezoelectric properties, and biocompatibility of native bone tissue, scaffolds produced using materials that meet these properties have a higher chance of application.^{[ 2} , ³ , ⁶ , ^{7 ]}

Over the last decade, a wide variety of biomaterials with specific properties have shown enormous potential as scaffold materials in preclinical and clinical studies.^{[ 8 ]} Several polymers of natural and synthetic sources are widely used as substitutes in bone tissue engineering applications.^{[ 3 ]} Natural polymers exhibit good biocompatibility, osteoconductivity, and low immunogenicity; however, they have a difficult‐to‐control degradation rate and low mechanical stability, limiting their use. A plethora of advantages have been demonstrated for synthetic polymers in comparison to their natural counterparts. These include a controlled degradation rate, the potential to deliver soluble molecules, tunable mechanical properties, prolonged shelf life, and low cost.^{[ 7 ]} Additionally, the issue of finding new approaches for innovative active scaffolds as alternatives to conventional polymers is being studied by various research groups around the world.^{[ 5 ]} In particular, the use of innovative piezoelectric synthetic polymers for this purpose has advanced bone tissue treatments to the next step.^{[ 9 ]} Piezoelectric polymers are electroactive materials that allow direct transmission of electrical impulses in the absence of an external power source.^{[ 5} , ^{10 ]} This electrical output can support a favorable regenerative microenvironment for soft tissue applications such as skin, cardiac, and nervous tissue.^{[ 5 ]} Thus, incorporating these properties into the scaffolds allows producing active bioscaffolds. Among the piezoelectric polymers, poly(vinylidene fluoride) PVDF is a semi‐crystalline polymer that has attracted attention since the 1970s.^{[ 8} , ^{11 ]} PVDF has five crystalline phases, namely α, β, γ, ε, and δ, related with different chain conformations.^{[ 12} , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ^{17 ]} Among them β‐phase has the highest piezo‐pyroelectric properties coming from the highest dipolar moment of 8 × 10⁻³⁰ cm unit⁻¹ cell.^{[ 17} , ^{18 ]} The high β‐phase content and the presence of the γ‐phase strongly suggest that the fabricated composite scaffold possesses the potential to generate piezoelectric responses under mechanical stress. This piezoelectricity is critical for applications in bone tissue engineering, as electrical stimulation has been shown to enhance osteogenic differentiation and bone regeneration.

Nevertheless, weak bone‐bonding ability, slow degradation kinetics, and lack of osteoinductivity limit the broader applications of polymeric materials. As mentioned, incorporating bioceramics within the polymer matrix is an effective approach that could address the weak points and enhance the properties of polymers.^{[ 4} , ¹⁶ , ¹⁹ , ²⁰ , ^{21 ]} Bioactive ceramics, which can be classified as calcium phosphates such as hydroxyapatite, β‐tricalcium phosphate, and calcium silicates such as akermanite, diopside, baghdadite have been widely investigated for their applications in bone tissue engineering.^{[ 22} , ²³ , ²⁴ , ²⁵ , ²⁶ , ^{27 ]} Although hydroxyapatite (HA) and beta‐tricalcium phosphate (β‐TCP) are potential candidates, a major drawback in these materials is their lack of sufficient mechanical strength rendering them unsuitable for bone as a bone analog in load‐bearing applications.^{[ 28 ]} As demonstrated in previous studies, this investigation finds that sintered HA is not completely biodegradable post‐implantation, while the degradation kinetics of β‐TCP have been shown to be slow.^{[ 28 ]} In order to overcome these limitations, calcium silicate‐based bioceramics are being investigated as an alternative to calcium phosphates.^{[ 4} , ^{19 ]} Silanol groups with calcium phosphate ions in calcium silicates promote the regeneration of injured tissues.^{[ 29 ]} Incorporating the third component into the structure of the calcium silicates, their chemical stability, bioactivity, and mechanical properties can be improved.^{[ 30 ]} Zirconium is one of the components included in the calcium silicates due to its biocompatible structure and suitable mechanical strength, which has the potential to form a network that ionically binds calcium ions.^{[ 30} , ^{31 ]} BAG (Ca₃ZrSi₂O₉), has received great attention due to the presence Ca, Si, and Zr ions which play crucial roles on bone regeneration.^{[ 32 ]} Releasing of these ions followed by degradation of ceramics, results in higher biological performance such as excellent bioactivity, great resorbability, differentiation, and proliferation of osteoblast cells in vitro.^{[ 31} , ³³ , ^{34 ]} On the other hand, the use of appropriate materials and fabrication methods can contribute to bone tissue regeneration by increasing the stability and functionality of scaffolds.

There are various techniques such as electrospinning, solvent casting, freeze drying, polymeric sponge, and phase separation to fabricate scaffolds with ideal properties, to improve them, and to achieve outstanding biological responses as well.^{[ 10} , ²⁹ , ^{35 ]} Among these, non‐solvent‐induced phase separation (NIPS) has been most extensively implemented, allowing the production of membrane structures with desired properties.^{[ 36 ]}

Previously, we investigated the production of boron‐doped BCP/PVDF‐based membrane scaffolds by non‐solvent induced phase separation technique and the effect of coagulation bath temperature on the produced scaffolds.^{[ 18} , ^{37 ]} In this research, for the first time, boron‐doped BAG powders have been incorporated into the PVDF matrix to explore their influence on the physical, thermal, and mechanical properties, as well as to maintain essential properties such as in vitro biodegradation and cellular responses required for successful bone regeneration. In addition, in vitro, biomineralization studies were performed to investigate the ability of the scaffold to induce HAp deposition while resorbing.

2. Results and Discussion

2.1. Characterization of the Powders

The surface morphology of the sol–gel‐derived powders is depicted in Figure 1A. As demonstrated in Figure 1A, BAG powders exhibit a substantial number of irregularly shaped agglomerates that are dispersed in a homogeneous manner. These agglomerates are formed as a result of the sintering of semi‐spherical crystallites in BAG powder, as reported in previous research.^{[ 38} , ^{39 ]} The particle size of these microparticle agglomerates was determined by a particle size analyzer based on the laser diffraction principle and is given in Figure 1B. As depicted in Figure 1B, the mean particle size of BAG powders containing 0, 0.5, 1, and 2 wt.% boron was changed between 27 and 31 µm, which was consistent with previous research.^{[ 38 ]} It is worth noting that there is no obvious change in the particle size and morphology of the grains with varying boron content in the powder. The pycnometer densities of the powders shown in Figure 1B. The addition of 2 wt.% boron decreased the density of the BAG powders from 3.48 to 3.33 g cm⁻³. The observed decrease in density of BAG powders can be attributed to several factors. First, the lower atomic mass of boron (10.81 g mol⁻¹) relative to elements such as calcium, zirconium, or silicon results in a reduced overall material mass. Second, the incorporation of boron into the material results in the formation of BO₃ or BO₄ units, leading to the disruption of the more compact SiO₄ or ZrO₆ networks. Third, defects or microvoids may be introduced during the synthesis process by boron doping, which in turn reduces density. For BAG/0.5B, the density decrease is less pronounced due to the lower boron content (0.5 wt.%), resulting in fewer structural changes. These findings are consistent with those of previous studies on boron‐doped ceramics.^{[ 40} , ⁴¹ , ^{42 ]}

Figure 1.

Characterization of the powders: A) SEM images of the powders; B) Mean particle size and density of the powders; C) XRD patterns of the powders.

The formation mechanism of BAG is based on the occurrence of a covalent bond between TEOS (Si[C₂H₅O]₄) and four C₂H₅ molecules, as well as an ion exchange between Ca‐nitrate and Zr‐nitrate molecules. The C₂H₅ molecule in the TEOS structure dissolves and the TEOS accepts six electrons from Zr⁴⁺ and Ca²⁺. Then, each SiO₄ molecule takes the oxygen from the oxygen bridge formed between Si molecules and exchanges four electrons with Zr⁴⁺ and two electrons with Ca²⁺. Thus, the remaining four electrons create attraction between Zr⁴⁺ and Si₂O₇ molecules, resulting in the baghdadite.^{[ 29 ]} XRD analysis was performed to confirm the formation of the baghdadite phase. Figure 1C displays XRD patterns regarding the baghdadite powder synthesized by sol–gel method after sintering at 1200 °C. The XRD patterns clearly verify the baghdadite of the powders consisting of 0, 0.5, 1, and 2 wt.% boron. Main diffraction peaks at 2θ = 31.38° and 29.85° are distinguishable, which were indexed to the (032) and (202) planes, corresponding to the highest intensity peaks of the baghdadite (PDF card no: 00‐047‐1854). The observed peaks of the powder has unit cell parameters of a = 7.36 Å, b = 10.17 Å, c = 10.45 Å, β = 90.87°, volume = 782.8 Å³ with space group P21/c that describes the symmetry of the crystal. The results demonstrated that major peaks complied with the diffraction pattern of the BAG crystal, which is the primary component of the samples. In spite of the presence of intense peaks of the BAG phase, the presence of the larnite phase as a secondary phase was also evident. This may be because the Zr precursor in TEOS hydrolysis could not be completely dissolved.^{[ 38 ]} The crystallite size of synthesized BAG powders was 14.13, 18.7, 17.91, and 25.63 nm for 0, 0.5, 1, and 2 wt.% boron‐incorporated BAG powders based on the Scherrer equation, respectively. The size of the crystallite was close to the previously reported crystallite size of the BAG powder.^{[ 38 ]}

2.2. Characterization of the Membrane Scaffolds

2.2.1. Morphological and Surface Characterization

The surface and cross‐sectional SEM images of the membrane scaffolds are presented in Figure 2 .

Figure 2.

SEM images of the surface and cross‐section of membrane scaffolds and surface images after incubation periods of 3 and 14 days.

As seen in the SEM images, the scaffolds demonstrated an asymmetric morphology characterized by a dense surface layer and porous cross‐sectional structure.^{[ 18} , ^{37 ]} The morphology of the membranes strongly depends on the diffusion rate of the solvent and non‐solvent.^{[ 36 ]} In the coagulation bath systems, where DMF is used as solvent and water is used as non‐solvent which has a high affinity for the solvent, liquid–liquid demixing process occurs upon the polymeric film is immersed into the bath. This process takes place rapidly on the surface of the membrane as the non‐solvent penetrates into the polymer solution, resulting in the formation of a dense surface layer. On the other hand, it is known that the solvent found with water in the coagulation bath improves delayed demixing conditions by reducing the activity of the water in the bath.^{[ 40 ]} Thus, in the inner part of the membrane, the coagulation process occurs more slowly and honeycomb‐like micropores are formed.^{[ 36} , ^{43 ]} Moreover, the surface morphology of the scaffold as revealed that the incorporation of the powder into the polymer matrix did not significantly change the porous structure of the surface layer. Noticeably, all of the scaffolds consisted of regular cross‐sectional pore structures within specific limits.

Regular pores and appropriate pore size play an important role in nutrient and oxygen diffusion and waste removal, as well as strongly influence the adhesion, cell–cell interaction, and cell migration of the bone cells.^{[ 44 ]} Essentially, ≈0.2—1 µm pore size is reported to be suitable for cell adhesion.^{[ 44} , ^{45 ]} The fabricated membrane scaffolds were characterized by interconnected cross‐sectional pores that were measured as 0.67 ± 0.2, 0.77 ± 0.33, 1.17 ± 0.28, 1.24 ± 0.32, and 1.2 ± 0.35 µm for PVDF, P/BAG, P/BAG/0.5B, P/BAG/1B and P/BAG/2B samples, respectively. The P/BAG/1B sample demonstrated a cross‐sectional morphology with the highest porosity.

The scaffolds with suitable porosity must also meet the requirement of appropriate surface properties for effective attachment and interaction with cellular populations in bone tissue engineering applications.^{[ 15 ]} An ideal scaffold should have a hydrophilic surface with a droplet contact angle of less than 90°, which supports better cell adhesion and spreading compared to hydrophobic surfaces.^{[ 4 ]} To investigate the surface hydrophilicity of the PVDF‐based scaffolds, contact angle measurements using a PBS medium were performed. Figure 3 illustrates the surface hydrophilicity by quantifying the mean values of the contact angle on the sample surfaces. According to the results obtained, the contact angle of 102.83 ± 8.8° corresponds to the pure PVDF sample, which exhibited the highest contact angle among the tested scaffolds. This finding suggests that the pure PVDF surface exhibits a high degree of hydrophobicity, as values above 90° are commonly associated with such surfaces. The contact angle of 86.72 ± 9.27° corresponds to the P/BAG/2B sample (PVDF scaffold with 2 wt.% boron‐doped BAG powder). This value, being below 90°, indicates a more hydrophilic surface compared to the pure PVDF sample. The surface wettability of the samples increased with the powder addition to the polymer matrix. The contact angle values were measured as 101.53 ± 4.75°, 88.11 ± 3.44°, 79.62 ± 5.07°, and 86.72 ± 9.27° for the P/BAG, P/BAG/0.5B, P/BAG/1B, and P/BAG/2B samples, respectively. These results revealed positive effects of the powder addition to the polymer matrix on the surface properties, in line with previous studies.^{[ 10} , ^{46 ]} This may be related with the change in the surface morphology of the scaffolds by incorporating powders into the polymer matrix. The presence of powders changes the solid–liquid interface energy, leading to higher surface roughness. When the surface energy of the solid is higher than the surface tension of the liquid, the droplet spreads on the surface and the contact angle value decreases, resulting in increased hydrophilicity. Consequently, rising surface wettability provides a more active surface for cells, resulting in higher biocompatibility.^{[ 47 ]}

Figure 3.

PBS contact angle measurements of the membrane scaffolds.

2.2.2. Phase Behavior, Crystallinity and Mechanical Properties

The crystalline phase structures of the composite scaffolds were characterized by XRD spectrums, which are shown in Figure 4A. The XRD spectrum of the neat PVDF sample shows characteristic peaks ≈2θ = 20°, which are assigned to the (1 1 0) reflections of the α‐crystal of PVDF. After the incorporation of the powders into the PVDF matrix, the emergence of the reflection at ≈2θ = 20.7° instead of 2θ = 20° is observed, which generally represents the formation of β‐crystal of PVDF. Contrarily, other characteristic peaks belonging to the α phase were not found in the composite structure. This observation combined with the presence of the characteristic peaks of β‐phase inferred that the crystalline structures in the scaffolds are dominant in the β‐crystal of PVDF.^{[ 48 ]} On the other hand, the XRD peaks of the BAG phase discussed in Section 3.1 also showed its presence in the membrane scaffolds.

Figure 4.

Phase behavior and crystallinity of the membrane scaffolds: A) XRD patterns of the scaffolds; B) FTIR spectra of the scaffolds; C,D) DSC thermograms of the scaffolds.

FTIR, which is important for determining the chemical bonding or molecular structure of the materials, provided insights into the functional groups of PVDF, BAG, and boron‐containing scaffolds.^{[ 49 ]} The FTIR spectrums of the scaffolds were presented in Figure 4B and the assignments of spectra were listed in Table 1 . Several characteristic peaks of stretching and bending vibrations related to PVDF, BAG, and boron are exhibited in the spectra of the samples. The primary peaks for PVDF‐based samples were detected in 1400 cm⁻¹ and a shoulder at 795 cm⁻¹, assigned to the vibrations of CH₂ groups.^{[ 50 ]} The discernible peaks at 1068, 1147, 1179, 1166, and 1232 cm⁻¹ indicated the stretching vibrations of CF₂ groups. Besides, the absorption band at 766 cm⁻¹ was related to the bending and skeletal bending vibration mode of CF₂ groups, while the peak identified in the wavenumber of 840 cm⁻¹ was assigned to the mixed mode of CH₂ rocking and CF₂ asymmetric stretching vibrations.

Table 1.

Characteristic FTIR peaks of the membrane scaffolds.

Wavenumber [cm−1]
P	P/BAG	P/BAG/0.5B P/BAG/1B P/BAG/2B	Structural units	Ref.
2930	2930	2923	CH2 asymetric and symmetric vibrations	[52]
2856	2856	2856	CH2 asymetric and symmetric vibrations	[52]
1404	1404	1405	CH2 vibrations	[50, 52]
–	–	1232	CF2 streching vibrations	[37, 50, 52]
1174	1173	1166	CF2 streching vibrations	[50, 52]
1074	1073	1073	CF2 streching vibrations	[ 50 , 52 ]
–	1043	1043	Si─O─Zr	[31]
–	1007	1007	SiO4	[31]
–	992	992	SiO4	[31]
–	971	971	SiO4	[31]
–	948	948	SiO4	[31]
–	917	917	SiO4	[31]
–	878	863	SiO4	[31]
840	839	840	CH2 rocking and CF2 asymmetric stretching vibrations.	[50, 52]
–	–	783	Symmetric bending vibrations of BO3 3−	[52]
–	635	635	SiO4	[31]

The FTIR spectrum of powder‐incorporated composite samples was slightly differentiated from the neat PVDF sample. With the addition of powder, BAG, and boron‐doped BAG interacted with the PVDF matrix, causing a decrease in all peak intensities. In the spectrum of the composite samples, the robust peaks were identified in the wavenumber range of 1007, 992, 971, 948, 917, 863, and 635 cm⁻¹ which can be assigned to the interaction of SiO₄ tetrahedron with polymer matrix.^{[ 31 ]} In addition, a new peak located at 1043 cm⁻¹ was observed serving as an indicative marker for the presence of Si─O─Zr groups within the silica network.^{[ 31} , ^{51 ]} These peaks are associated explicitly with silicate, providing valuable insights into the formation mechanism of the silica network structure. Furthermore, the Ca─O absorption bands were found at 515 and 460 cm⁻¹.^{[ 31 ]} On the other hand, the peak located at 783 cm⁻¹ was caused by the symmetric bending vibration mode of the BO₃ ³⁻ group. The appearance of BAG and boron peaks in the composite membrane structures indicates that the synthesized boron‐doped BAG powders were successfully incorporated into the membrane scaffold structure. Significantly, the absence of these peaks in the spectra of the neat PVDF sample can also serve as evidence of the successful production of composite scaffolds.

In other respects, the bands approximately at 763 and 870 cm⁻¹ are related to the α‐phase, while the bands at ≈473, 511, and 838 cm⁻¹ are related to the β‐phase of the PVDF‐based composite scaffolds has been attributed.^{[ 35} , ^{45 ]} Moreover, the band at ≈1232 cm⁻¹ is referred to as the γ‐phase of PVDF.^{[ 37 ]} To better visualize the changes resulting from the powders in the PVDF network, the spectra specifically containing the changes are placed in the Figure 4B.

In order to further analyze the efficiency of the β‐phase of PVDF formation with different amounts of the powders, FTIR analysis was also performed to determine the percentage of β‐phase in crystalline PVDF matrix. According to the Equation (2), it can be determined that the F _β values of the PVDF, P/BAG, P/BAG/0.5B, P/BAG/1B, and P/BAG/2B samples are equal to 96.55%, 97.29%, 97.84%, 95.73%, and 97.19%, respectively. The results showed that the β phase content of all samples was significantly high. Consequently, due to the dominant presence of the β phase, the characteristic bands associated with the α‐phase were not distinctly observed in the FTIR spectra. Moreover, in this study, higher values ​​were obtained than the highest β phase content obtained in boron‐doped BCP/PVDF‐based scaffolds produced at 20 °C in our previous study.^{[ 37 ]} This indicates that BAG doping provides a better effect on the beta crystallization of PVDF‐based scaffolds compared to BCP doping can serve as evidence.

Crystallinity, a significant parameter influencing the properties of the composites, is obtained from the DSC data. The heating and cooling curves of the composite scaffolds are shown in Figure 4C,D, respectively. The melting temperatures of the β‐phase (Tm _β), crystallization temperatures (Tc) the total crystallinity and of the crystals (X_C ) are detailed in Table 2 . Based on the DSC data, the melting temperature of the PVDF‐based samples was ≈169–170 °C with heating, while the crystallization peak was detected at ≈143–146 °C with cooling of the samples. While the melting temperatures of the samples were close to each other, there was a slight increase in the crystallization temperatures with the addition of boron‐doped BAG powders. This increase is related to the fact that the addition of dopants to the PVDF matrix causes heterogeneous nucleation and so enhances the crystallization temperatures of the samples. Additionally, the PVDF sample was found to have the highest total crystallinity. On the other hand, the DSC results are consistent with XRD and FTIR results. In these results, α‐phase peaks were not observed, only β‐phase peaks were clearly seen. Furthermore, the shoulder that starts to form at ≈173 °C regarding the P/BAG/2B sample is characteristic of γ phase‐PVDF which has also piezoelectric properties.^{[ 44 ]} In our previous study,^{[ 45 ]} we developed boron‐doped BCP/PVDF‐based membrane scaffolds at 20 °C coagulation bath temperature by non‐solvent induced phase separation and observed the presence of α‐phase peaks in the produced scaffolds. In the current research, we worked at the same temperature and confirmed that β‐phase was dominant in the composite structure.

Table 2.

Crystallinity and β phase fraction of the membrane scaffolds.

Sample name	Tm β [°C]	T C [°C]	XC [%]	F β [%]
PVDF P/BAG	170.21 170.21	143.32 145.12	58.55 56.51	96.55 97.29
P/BAG/0.5B	170	145.64	53.75	97.84
P/BAG/1B	169.22	146.40	53.01	95.73
P/BAG/2B	169.82	146.40	56.33	97.19

Mechanical properties play an important role in scaffold fabrication in the process of restructuring bone tissue, which is a structure under static and dynamic loads.^{[ 18 ]} The scaffolds should be able to resist the proper stress through their application in specific areas and represent enough flexibility during to enable movement. The mechanical performance of prepared scaffolds as a function of powder content was assessed under tension and related stress–strain curves are plotted in Figure 5 . The results showed that the tensile strength (TS) and elongation at the break (EB) of the PVDF sample were 7.26 ± 0.06 MPa and 22 ± 1.68%, while the incorporation of BAG was 6.28 ± 0.1 MPa and 15.99 ± 1.17%, respectively. The incorporation of 0.5, 1, and 2 wt.% boron decreased the tensile strength and elongation at the break of the PVDF sample to 5.1 ± 0.06 MPa and 4.66 ± 0.05%, 4.61 ± 0.84 MPa and 8.8 ± 1.56%, and 4.78 ± 0.17 MPa and 6.69 ± 1.11%, respectively.

Figure 5.

Mechanical properties of the membrane scaffolds. Each bar represents the mean ± standard deviation. * represents statistical significance, compared with the control (PVDF) group (p < 0.05).

The mechanical properties of scaffolds are first influenced by parameters such as pore size and the uniformity of the scaffold composition.^{[ 53 ]} While larger pores are advantageous for cell ingrowth and nutrient delivery, they can compromise structural integrity by introducing void volumes that weaken the scaffold.^{[ 10} , ⁵³ , ^{54 ]} In the present study, the incorporation of BAG resulted in an augmentation of both pore size and porosity, thereby engendering a less uniform structure. This heterogeneity in pore distribution creates stress concentration points, reducing tensile strength. In contrast, the pure PVDF scaffold, with smaller and more evenly distributed pores, exhibited a more homogeneous morphology and better mechanical performance.^{[ 37 ]} Second, the crystallization behavior of the scaffolds also plays a critical role in determining their mechanical properties. In the present study, the PVDF sample with the highest total crystallinity exhibited the greatest tensile strength and elongation at break. Conversely, an increase in powder content within the polymer matrix resulted in a decrease in total crystallinity, consequently reducing mechanical properties. These findings are consistent with those reported in our previous studies^{[ 45 ]} and highlight the trade‐off between bioceramic incorporation and crystallinity.

Furthermore, the present study's findings are consistent with those of other research. For instance, Correia et al. observed that the mechanical properties of PVDF‐based 3D scaffolds decreased with increasing pore size.^{[ 55 ]} In a similar vein, Abzan et al. reported that the presence of microvoids in PVDF‐based scaffolds led to a deterioration in mechanical strength.^{[ 10 ]} These studies provide support for the observation that, while larger pores and increased porosity are beneficial for biological interactions, they can negatively impact mechanical performance. Despite the reduction in mechanical properties with the addition of BAG, it is important to note that the tensile strength of all scaffolds in our study remained close to that of human cancellous bone. This finding indicates that the studied scaffolds are still suitable for bone tissue engineering applications, where a balance between mechanical strength and biological performance is essential.

While the mechanical properties of the developed membranes, such as tensile strength and elongation at break, were found to be within the lower range of human cancellous bone, it is imperative to emphasize that these membranes are not intended for load‐bearing applications. Instead, their primary functions are osteoinduction and osteoconduction, aimed at guiding bone regeneration in non‐load‐bearing or low‐load‐bearing environments. The mechanical analyses presented herein demonstrate that the membranes possess sufficient structural integrity to support their biological role, rather than to imply equivalency with the mechanical performance of cancellous bone. In summary, the reduction in mechanical properties observed with the addition of BAG can be attributed to increased pore size, reduced uniformity, and decreased crystallinity. These factors have been exhaustively documented in the extant literature and are in alignment with the experimental findings. It is hoped that this clarification will address the concerns of the reviewer and provide a comprehensive explanation for the observed discrepancy.

2.2.3. In Vitro Apatite Mineralization and Swelling Behavior

An ideal scaffold should display bioactivity that enables the formation of a biologically active bone‐like apatite layer on its surface when implanted in the body. The formation of the apatite layer as an interface between the implant surface and living bone tissue is achieved by heterogeneous nucleation of apatite from simulated body fluid (SBF), a metastable calcium phosphate solution supersaturated with apatite.^{[ 18 ]} When silicate‐containing bioceramics are incubated in SBF solution, a series of complex biological reactions proceeding in the form of ion exchange at the material–tissue interface, organic phase accumulation, cell differentiation, attachment, proliferation, and extracellular matrix formation, resulting in the formation of a mineralized apatite layer on the surface of these materials.^{[ 37} , ^{56 ]} During incubation, hydronium ions (H₃O⁺) in the SBF replace the cations on the surface of the scaffolds and react with the silica in the bioceramic material. This leads to the breaking of Si─O─Si bonds through the hydrolysis of silica and the formation of silanol (Si─OH) groups on the surface of the scaffold. Thus, the silanol groups induce apatite nucleation and enable the formation of the mineralized apatite layer on the scaffolds.^{[ 57 ]}

In order to investigate the in vitro apatite‐forming ability of the membrane scaffolds containing boron‐doped BAG/PVDF, the 10xSBF solution prepared according to the procedure described in our previous research was used.^{[ 4} , ¹⁹ , ²⁰ , ^{58 ]} The use of concentrated 10× SBF for assessing apatite formation is a well‐established method to evaluate the bioactivity of bone‐related biomaterials. This technique is widely accepted because it closely replicates the ionic composition of human blood plasma and has been validated as a reliable indicator of in vivo bioactivity. Additionally, higher concentrations of calcium (Ca) and phosphorus (P) ions, such as those found in 10× SBF, are commonly utilized to accelerate apatite deposition, making this approach an efficient preliminary test for bioactivity.^{[ 4} , ¹⁹ , ²⁰ , ^{58 ]} Figure 2 shows SEM images of the scaffolds before and after incubation in 10xSBF at 37 °C for 3 and 14 days. As seen in the SEM images, the apatite layer on the surfaces begins to form after 3 days of incubation. The accumulation of the resulting apatite nuclei increases with longer incubation periods. After 14 days of incubation, it was observed that the nucleation on the surfaces increased. These apatites had a cauliflower‐like globular morphology, which is characteristic of the in vitro apatite growth process.^{[ 35 ]} On the other hand, these images can be used as evidence that apatite crystals formed on the surface of all scaffolds compared to SEM images regarding before incubation.

Furthermore, the presence of the Ca and P elements in the apatite crystals was confirmed using EDS. The EDS spectra demonstrated that the intensity of Ca and P peaks increased with the longer immersion period, suggesting a larger amount of apatite growth on the samples. It was found that the Ca/P ratio of the apatite crystals concentrated in the regions where nucleation occurred is particularly especially in P/BAG/0.5B and P/BAG/1B samples is higher than the Ca/P value of native bone, which is 1.67.

FTIR analysis was conducted to further assessment the apatite formation on the surface of the scaffolds. The related FTIR spectra of selected P/BAG/1B samples before and after incubation in 10 × SBF for 3 and 14 days were presented in Figure  6B. Compared with the spectra before incubation, the characteristic peaks belonging to PVDF, BAG, and B, although their intensities decreased, were still observed after incubation. In addition, additional bands representing the symmetric and asymmetric stretching modes of phosphate (PO₄ ⁻³) groups of CaP providing valuable insights into the structural composition of the apatites. These bands at 560 and 600 cm⁻¹ were associated explicitly with apatite attributed to P─O bending vibrations, while a broad band at ≈870 cm⁻¹ and between ≈1375 and 1555 cm⁻¹ belonged to (PO₄ ⁻³).^{[ 45 ]} In addition, the previously visible presence of OH vibrations adsorbed by H₂O in the apatite layer was emerged as a band at ≈1650 cm⁻¹ and a broad band between 3100 and 3600 cm⁻¹.^{[ 37 ]} The formation and growth of apatite on the scaffolds were demonstrated by XRD analysis. The overlapping XRD diffractograms for the P/BAG/1B sample before and after 3 and 14 days of incubation are shown in Figure 6C. The diffractograms of the apatite layer show crystalline peaks at 2θ = 32° and 46°, as we have explained in our previous study.^{[ 45 ]} Upon comparing the diffraction patterns, an increase in the intensity of the peaks was observed. This increase also enhanced as the incubation time increased, which can be attributed to the apatite growth. These findings clearly demonstrate the bioactivity of the membrane scaffolds, highlighting a significant increase in apatite formation on the surface of the P/BAG/1B sample.

Figure 6.

Biological properties of the membrane scaffolds: A) EDS spectra of the scaffolds, B) FTIR spectra of the scaffolds before and after incubation in SBF, C) XRD patterns of the scaffolds before and after incubation in SBF, and D) swelling ratio of the scaffolds.

The swelling ability, which enables the absorption of body fluids, the transfer of cell nutrients and metabolites within the scaffold, plays a major role in regulating biological interactions.^{[ 45} , ^{59 ]} Scaffolds exhibit differential swelling behaviors in response to the surrounding environment, which is critical for the fitting to the body. Herein, we assessed the swelling kinetics of the scaffolds by immersing the cut samples in PBS (37 °C, pH:7.4) for up to 28 days. According to our visual inspection, the samples could maintain their structural integrity, following the absorption of body fluid.

As seen in Figure 6D, all scaffolds swelled rapidly during the first 24 h, after which the swelling stabilized at an equilibrium level. The highest degree of swelling was observed in the P/BAG/1B sample compared to the others after 24 h, which can be attributed to its enhanced hydrophilicity and lower contact angle, as supported by the surface contact angle results in Chapter 3.2.1. Interestingly, the swelling rate of the samples with powder additives (BAG and boron‐doped BAG) stabilized later compared to the PVDF sample. While the swelling in the PVDF sample decreased after day 7, the swelling in the samples with powder additives increased until day 14 and then gradually decreased. This behavior is primarily due to the scaffolds reaching their maximum swelling capacity and undergoing structural stabilization, rather than evaporation. The differential swelling behaviors among the samples are influenced by their composition, with powder additives delaying the stabilization process.^{[ 45 ]}

In general, the incorporation of powders into the polymer matrix led to a decrease in the swelling rate until day 28. When comparing the powder‐containing samples, those with higher boron‐doped BAG content showed lower swelling rates compared to the BAG‐containing sample without boron. The different swelling behaviors and the low swelling tendency of the scaffolds with lower powder content at the end of 28 days were also complemented by the previously described surface contact angle results in Chapter 3.2.1. The low contact angle and increased hydrophilicity indicate a high affinity to for water and better water absorption. Moreover, the fact that low powder content improves particle dispersion and produces better performance highlights potential applications in scaffolds.

The present findings underscore the importance of material composition in tailoring scaffold properties for specific tissue engineering applications. The observed swelling kinetics, particularly in the P/BAG/1B sample, demonstrate that scaffolds with enhanced hydrophilicity, such as those with boron‐doped additives, are better suited for applications requiring optimal fluid exchange. Furthermore, the gradual stabilization of swelling in powder‐containing samples suggests the critical role of bioactive additives in regulating scaffold performance, with potential applications in areas such as bone regeneration and wound healing, where controlled fluid absorption and structural stability are crucial.

2.2.4. In Vitro Cytotoxicity and Cellular Adhesion Evaluation

In vitro cytotoxicity analysis of the scaffolds was conducted with HOB cells for a period of 72 h, and the results are presented in Figure 7 . The cytotoxicity results demonstrate the biocompatibility of PVDF and its composites with BAG and boron‐doped BAG (BAG/0.5B, BAG/1B, and BAG/2B), evaluated through cell viability assays over 24, 48, and 72 h. The PVDF sample, serving as the control, exhibited consistent cell viability across all concentrations (25%, 50%, 75%, and 100%), with viability generally remaining above 90%. This aligns with the literature, where PVDF is noted for its non‐toxic nature and inert properties, supporting cellular adhesion and growth under standard conditions.^{[ 60} , ^{61 ]} For P/BAG, cell viability initially decreased within the first 24 h, reaching a minimum of 66.41% at higher concentrations. However, a marked improvement was observed over the next 24 h, with viabilities exceeding 100%, consistent with BAG's bioactive nature and its ability to stimulate osteoblast activity and proliferation.^{[ 59 ]} By 72 h, cell viability stabilized at values close to or slightly above 100%, indicating the material's potential to support long‐term cell health. The P/BAG/0.5B sample showed moderate variations in cell viability, with initial decreases to 72.94% at higher concentrations during the first 24 h, followed by a significant recovery by 48 h (viabilities ranging from 100% to 116.9%). This trend persisted at 72 h, with viabilities generally maintaining levels near 100%, suggesting that low‐level boron doping enhances cellular responses, consistent with previous studies reporting boron's osteogenic potential and its role in promoting cell proliferation.

Figure 7.

In vitro cytotoxicity results of HOB cells after cultivation with scaffolds for 72 h. (The p‐value below 0.05 was considered statistically significant and is indicated on the graphs.).

For P/BAG/1B, viability values were consistently higher, with minimal reductions during the first 24 h (lowest value: 98.75%) and a slight increase by 48 h. After 72 h, viability remained stable, reflecting an optimal balance of BAG and boron doping, likely enhancing the material's bioactivity without introducing cytotoxic effects. Similar observations have been reported in boron‐doped bioactive glass systems, which enhance cellular differentiation and proliferation without adverse effects.^{[ 62 ]}

Interestingly, the P/BAG/2B sample demonstrated initial reductions in viability during the first 24 h, particularly at higher concentrations (minimum of 77.68%). While partial recovery was observed after 48 h, cell viability remained slightly lower compared to the other samples at the same time point. By 72 h, however, viability values approached 100%, indicating that while higher boron content may delay initial cell attachment or proliferation, it does not exhibit long‐term cytotoxic effects. This result is in agreement with studies suggesting that excessive boron levels may temporarily inhibit cellular functions before normalizing over time.^{[ 63 ]}

In summary, the results highlight that the incorporation of BAG and boron‐doped BAG into PVDF matrices generally enhances cell viability over time, with optimal performance observed at moderate boron concentrations (e.g., P/BAG/0.5B and P/BAG/1B). These findings emphasize the potential of boron‐doped bioactive materials for applications in bone regeneration, where controlled ion release and biocompatibility are critical. Further studies focusing on long‐term evaluations and in vivo conditions could validate these observations and optimize the composite formulations.

The cellular adhesion on the membrane scaffolds was visualized via the fluorescent labeling of nuclei in cells with DAPI, and the results were analyzed with fluorescent microscopy and scanning electron microscopy. The findings are presented in Figure  8A,B.

Figure 8.

A) Fluorescent microscopy images of membrane scaffolds following a 72‐hour cultivation period with samples. B) SEM images of samples after being cultivated with HOB cells.

The SEM images revealed robust cell attachment and spreading on the membrane scaffolds, particularly those containing boron‐doped BAG. The extended lamellipodia and filopodia observed indicate active cellular interaction with the scaffold surfaces. The SEM images (Figure 8) demonstrate that cells cultured on PVDF+BAG and boron‐doped BAG/PVDF membrane scaffolds exhibit extended lamellipodia and filopodia, and a more confluent morphology compared to PVDF‐only scaffolds. These observations suggest that the enhanced cell adhesion and spreading behavior observed in the PVDF+BAG and boron‐doped BAG/PVDF membrane scaffold cultures is indicative of an improved cell–matrix interaction. The enhanced wettability (reduced contact angle) and augmented porosity exhibited by the P/BAG/1B scaffold further substantiate these observations, as both of these characteristics are conducive to cellular attachment and proliferation. Enhanced wettability, as demonstrated by a reduction in contact angle, facilitated by boron doping, contributed to improved surface hydrophilicity and cell adhesion. Consistent with Wei et al. (2009),^{[ 64 ]} hydrophilic surfaces promote protein adsorption (e.g., fibronectin and vitronectin) critical for osteoblast adhesion and proliferation. Among the tested membrane scaffolds, P/BAG/1B exhibited optimal wettability and the highest degree of cell attachment, correlating with its lowest contact angle. Although the fluorescent microscopy images in Figure 8A offer a qualitative overview of cell attachment, it is crucial to acknowledge that the assertion regarding P/BAGB1 demonstrating the most significant degree of cell attachment is substantiated by exhaustive quantitative data, encompassing DAPI staining and image analysis utilizing ImageJ. These quantitative measurements, in conjunction with statistical analysis, consistently demonstrate that P/BAGB1 promotes superior cell attachment compared to other samples.

The bioactivity of BAG, widely documented in the literature, is integral to its role in promoting osteogenic cell growth and differentiation. Studies such as Forogh et al. (2024)^{[ 63} , ^{65 ]} demonstrate that BAG enhances biodegradability, mechanical properties, and hydroxyapatite formation, thereby improving fibroblast viability and osteoblast attachment. In alignment with these findings, the SEM results of this study show that the incorporation of BAG into PVDF membrane scaffolds significantly enhanced cell adhesion and spreading compared to PVDF‐only membrane scaffolds. This indicates that BAG provided a bioactive surface conducive to HOB cell adhesion. Similarly, Najar et al. (2024)^{[ 64 ]} highlighted that BAG‐modified scaffolds foster osteogenic differentiation and calcium deposition, essential processes for bone regeneration.

The enhanced cellular morphology observed in Figure 8, characterized by well‐developed filopodia and lamellipodia, is attributed to the bioactivity of BAG particles and the osteoinductive effects of boron doping. The enhanced wettability, optimal pore structure, and increased porosity contributed to creating a favorable environment for cell adhesion, as supported by prior studies.^{[ 63} , ⁶⁴ , ^{65 ]} These observations are consistent with the existing literature, which demonstrates that the incorporation of BAG and boron‐doped BAG facilitates osteoblast attachment, spreading, and early osteogenic activity.

The observed cellular morphology in this study suggests the early stages of osteogenesis on BAG‐reinforced PVDF membrane scaffolds. The incorporation of boron into the membrane matrix serves to augment its bioactivity. Boron‐doped BAG/PVDF membrane scaffolds have been demonstrated to enhance cell viability and adhesion, consistent with the osteogenic potential of boron ions (e.g., borate (BO₃ ³⁻) or metaborate (BO²⁻)), which has been shown to stimulate osteoblast activity.^{[ 66} , ⁶⁷ , ^{68 ]} The P/BAG/1B membrane scaffold, with moderate boron content, exhibited the most favorable cell adhesion, whereas excessive boron levels (e.g., P/BAG/2B) may have induced slight cytotoxic effects or oversaturation, as noted in previous studies. The observed improvements in bioactivity due to boron doping align with findings in the literature. For example, Jodati et al. (2023)^{[ 69 ]} reported that boron‐containing scaffolds enhance porosity, accelerate degradation, and promote new bone formation, all of which are crucial for treating bone defects. Additionally, BAG scaffolds have been shown to support substantial bone formation and integration in vivo.^{[ 70 ]} These scaffolds not only create a bioactive environment for osteoblast differentiation but also mimic the natural bone microenvironment, making them highly suitable for clinical applications.

The performance of a scaffold is closely associated with its structure, shape, porosity, and pore size. The material exhibits a range of exceptional properties, such as effective energy absorption, high strength, low weight, and suitability for the design of biomimetic bone scaffolds.^{[ 71 ]} It is well‐established that porosity and pore size play a critical role in facilitating the diffusion of oxygen and nutrients, as well as influencing the adhesion and migratory behavior of osteoblast cells. Studies have reported that a pore size ranging from 1 to 2 µm is optimal for cell adhesion. Furthermore, the P/BAG/1B sample demonstrated a distinct cross‐sectional morphology.

Among the samples, P/BAG/1B exhibited the highest porosity and the largest pore size (1.24 ± 0.32 µm), which may enhance nutrient and gas exchange and promote cellular migration. As previously discussed, a droplet contact angle of less than 90° is considered beneficial for cellular adhesion and spreading. The P/BAG/1B sample was reported to have the lowest contact angle (79.62 ± 5.07°), which is expected to improve cellular adhesion. During the synthesis of the membranes, it was observed that the presence of a solvent mixed with water in the coagulation bath reduced the activity of the water, leading to the formation of honeycomb‐like micropores in the inner structure of the membrane. This honeycomb‐like structure is known to enhance osteoblast bioactivity.^{[ 71 ]} Additionally, as mentioned earlier, the incorporation of baghdadite has been shown to positively influence cell bioactivity. Pahlevanzadeh et al. (2019) reported that the addition of baghdadite (BAG) nanoparticles to bone cement improved the attachment and viability of MG63 osteoblasts.^{[ 72 ]} Lu et al. (2014) demonstrated that baghdadite scaffolds have significant potential in promoting the osteogenic differentiation of human osteoblasts through the upregulation of key osteogenic genes. The distinctive composition of these scaffolds, coupled with their favorable mechanical properties, renders them a promising candidate for applications in bone tissue engineering and regenerative medicine.^{[ 73 ]}

In a more recent study, Lu et al. (2020) investigated the effects of baghdadite ceramic scaffolds on senescent human primary osteoblast‐like cells (HOBs) and aged rats. They reported that baghdadite‐conditioned media corrected mitochondrial dysfunction in senescent HOBs and improved bone healing in critical‐size defect models in aged rats.^{[ 74 ]} Furthermore, Vaez et al. (2021) demonstrated that a chitosan‐baghdadite (1 wt.%) nanocomposite coating enhanced the viability and adhesion of MG63 cells on SS 316L implants. In light of the literature, the addition of baghdadite has been shown to improve both the structural and biological properties of scaffolds. These improvements are supported by evidence from contact angle measurements, pore size analysis, cell viability assays, and cell adhesion imaging.^{[ 75 ]}

The non‐toxicity of boron‐doped BAG/PVDF membrane scaffolds, confirmed by MTT assays, further supports their potential for bone tissue engineering. The SEM observations align with these findings, showing that these scaffolds foster an osteoinductive environment conducive to bone regeneration. It is important to note that the outcomes of the cell viability assessment (Figure 7) and the SEM‐based cell attachment observations (Figure 8) reflect different phases and dimensions of cellular behavior. The MTT assay quantifies the overall metabolic activity of cells, influenced by initial adhesion and subsequent proliferation, as well as cellular health over time. In contrast, SEM images provide a morphological evaluation of cell attachment and spreading at specific early time points, primarily after 24 h. This distinction is well‐supported in literature, where initial attachment does not necessarily correlate linearly with longer‐term proliferation due to additional factors such as cell signaling, scaffold bioactivity, and extracellular matrix deposition influencing later stages.^{[ 64} , ^{73 ]}

In summary, the incorporation of boron‐doped BAG into PVDF membrane scaffolds markedly enhances their bioactivity, as evidenced by improved cell adhesion, spreading, and viability. These findings, supported by the literature, suggest that P/BAG/1B membrane scaffolds are promising candidates for advanced bone tissue engineering applications.

3. Conclusion

In the present study, the NIPS method was successfully employed to develop membranes for bone scaffolds based on boron‐doped BAG/PVDF composites. The incorporation of boron‐doped BAG powder significantly enhanced the physical and biological properties of the membrane scaffolds. The results revealed that the membrane scaffolds exhibited interconnected cross‐sectional pores beneath a dense surface layer, a feature critical for nutrient transport and cellular infiltration. Notably, increasing the boron‐doped BAG content enhanced surface hydrophilicity, as evidenced by the reduction in contact angle values, which promotes better cell–material interactions.

While the addition of boron did not significantly alter the thermal stability of the materials, it contributed to an increased β‐crystalline phase fraction, which is advantageous for certain mechanical characteristics such as flexibility and toughness. The mechanical properties of the scaffolds were found to fall within the range suitable for bone tissue applications, ensuring functional structural support. However, the incorporation of BAG and boron‐doped BAG led to increased pore size and porosity, resulting in a more heterogeneous structure. This heterogeneity introduced local stress concentration points, leading to a decrease in tensile strength compared to the neat PVDF scaffold, which exhibited a more uniform morphology. Therefore, although the overall mechanical profile—including flexibility and energy absorption—was improved through β‐phase enhancement, the tensile strength specifically was compromised by the morphological irregularities. Despite this, the tensile strength values of all scaffolds remained within the range of human cancellous bone, indicating their continued suitability for bone tissue engineering applications where a balance between mechanical resilience and biological performance is essential. Additionally, boron doping reduced the degree of swelling, thereby improving the scaffold's structural stability in physiological conditions. Biologically, the scaffolds demonstrated excellent in vitro bioactivity, as shown by their enhanced apatite formation in simulated body fluids. Cellular studies confirmed that boron‐doped BAG significantly supported osteoblast viability, proliferation, and adhesion, with no signs of cytotoxicity. The presence of boron likely facilitated these effects through its known role in bone metabolism and cellular activity.

In conclusion, the boron‐doped BAG/PVDF membrane scaffolds developed in this study offer a promising platform for bone tissue engineering. The integration of enhanced bioactivity and cellular compatibility in these membrane scaffolds underscores the significance of boron as a pivotal element in the design of next‐generation bone scaffolds.

4. Experimental Section

Materials and Reagents

For the synthesis BAG powder, tetraethyl orthosilicate (Si[C₂H₅O]₄; TEOS), zirconium (IV) oxynitrate hydrate (ZrO[NO₃]₂.xH₂O; Zr‐nitrate) and calcium nitrate tetrahydrate (Ca(NO₃)₂.4H₂O; Ca‐nitrate) were purchased from Alfa Aesar, USA. Nitric acid (HNO₃) utilized was obtained from Merck, Germany. Boric acid (H₃BO₃) as a boron source was obtained from Merck, Germany. To prepare the PVDF membrane scaffold, PVDF (Solef 1015) was used as the primary constituent material. N,N‐Dimethylformamide (DMF) obtained from Fisher, USA was used as the solvent. Distilled water (DI) and DMF were used as the non‐solvent in the coagulation bath. Reagents such as sodium chloride (NaCl), potassium chloride (KCl), calcium chloride dihydrate (CaCl₂⋅2H₂O), magnesium chloride hexahydrate (MgCl₂⋅6H₂O), and sodium dihydrogen phosphate monohydrate (NaH₂PO₄⋅H₂O) were obtained from Merck (Germany). Sodium bicarbonate (NaHCO₃) was also purchased from Sigma‐Aldrich (Germany) to use in in vitro biomineralization studies. For using in vitro swelling experiments, Dulbecco's phosphate buffered saline (PBS, pH 7.4) was obtained from Gibco (USA). Dulbecco's modified media (DMEM), fetal bovine serum (FBS) and Penicilin‐Streptomycin (Pen‐Strep) were obtained from PAN Biotech (Germany) and 3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5‐diphenyl tetrazolium bromide (MTT), dimethyl sulfoxide (DMSO) and 4,6,‐di amidino‐2‐2‐phenylindole (DAPI) were obtained from Merck (Germany). Human osteoblast cell line (HOB) (PromoCell, Merck) was provided by Professor Ebru Toksoy Öner.

Powder Synthesis

BAG powder was synthesized by sol–gel method as described in previous studies.^{[ 38 ]} Briefly, TEOS, distilled water, and HNO₃ (2 m) with a molar ratio of TEOS: water: HNO₃ = 1:8:0.16 were mixed for 30 min. After hydrolysis, Zr‐nitrate and Ca‐nitrate were added into the solution with a molar ratio of Zr‐nitrate: Ca‐nitrate: TEOS = 1:3:2. The reactants were stirred at room temperature for 5 h. Afterward, the solution was kept at 60 °C for 24 h to obtain wet sol and dried at 100 °C for 48 h to acquire a dry gel. Finally, the resulting gel was annealed at 1200 °C for 3 h. The synthesis of boron‐doped (0.5, 1, and 2 wt.%) BAG powder followed similar steps as described above. The only distinction was that H₃BO₃ was added to the solution as a boron source. The synthesized powder was grounded to prepare the membrane scaffolds.

Scaffold Fabrication

PVDF‐based membrane scaffolds comprising boron‐doped BAG with 0, 0.5, 1, and 2 wt.% boron were fabricated using the NIPS process, in accordance with the methodology established in the previous studies.^{[ 18 ]} Briefly, the synthesized powder was added to DMF to ensure the uniform dispersion of powder in the PVDF matrix (10% wt of PVDF). Then, PVDF was added to this solution keeping the total solid concentration constant at 18% wt, and stirred at 60 °C for 24 h. Before casting, the solutions were sonicated for 10 min in order to enhance homogeneity. Then, the solution was casted onto a glass plate using a 200 µm casting knife followed by soaking in a coagulation bath for 3 h. The bath containing DMF and water with a volume ratio of 60:40 was preset to 20 °C, which was the optimized temperature in the previous research.^{[ 37 ]} Subsequently, the membranes were carefully detached from the glass plate and rinsed in a freshwater bath to remove any remaining solvent. Then, the membranes lyophilized at −80 °C for 4 h (Toros TRS‐2/2 V), assuring the elimination of possible remaining DMF traces. The obtained membrane scaffolds were coded based on their BAG and boron‐doped BAG content and designated as PVDF, P/BAG, P/BAG/0.5B, P/BAG/1B, and P/BAG/2B.

Powder Characterization

The morphological characterization of the powders was performed using a field emission scanning electron microscopy (FE‐SEM, Carl Zeiss/Gemini 300) operating at an accelerating voltage of 15 kV and equipped with an energy dispersive spectrometer (EDS). The surface of the powders was coated with Au‐Pd for improving their conductivity prior to imaging. The mean particle size (D _mean) of the synthesized powders was measured using a Malvern Mastersizer 3000E particle size analyzer. The densities of the powders were measured using a gas pycnometer (Micromeritics, AccuPyc II 1340). The presence of crystalline phases of the powders was detected using an X‐ray diffractometer (XRD, Panalytical X'Pert Pro) using Cu‐Kα (λ = 1.5406 Å) radiation at 40 mA and 40 kV conditions. The 2θ range was set to 20°–60°, a step size of 0.02, and a rate of 2° min⁻¹. Crystalline phases were defined using the International Centre for Diffraction Data (ICDD) powder diffraction database. Using XRD data, the Scherrer Equation (1) was applied to estimate the size of the crystallite of the powder as follows:^{[ 31} , ⁷⁶ , ⁷⁷ , ^{78 ]}

$$D=\frac{k\lambda }{\beta \cos \theta }$$ (1)

where β is the half of the diffraction peak width, θ is the Bragg diffraction angle, k  is the Scherrer constant (≈0.9), λ is the wavelength of X‐ray (1.5405 Å), and D is the size of crystals.

Scaffold Characterization

The surface and cross‐sectional morphologies of the scaffolds were examined with a field emission scanning electron microscope (FE‐SEM, Carl Zeiss/Gemini 300) by applying a changing voltage of 5 and 10 kV to the samples under vacuum. The samples were sputter‐coated with a thin layer of Au‐Pd before analysis. To obtain cross‐sectional images, sputter coating was performed after fracturing of the samples in liquid nitrogen. In order to determine the pore sizes of the scaffolds, cross‐sectional SEM images were analyzed by measuring at least 100 pore sizes using the ImageJ v1.53s software.

The surface properties of the scaffolds were designated by performing contact angle measurements. A contact angle goniometer (Attention Theta Lite) was employed to quantitatively probe the static CA of the surfaces of the samples. The CA analysis performed using PBS with the sessile drop approximation at ambient temperature. Before measurement, drops of 5 µL volume were automatically pipetted onto the surfaces of the samples, which were cut into equal sizes and adhered to a glass substrate with double‐sided adhesive tape. The contact angle values of the samples were determined by examining the droplet images taken 5 s after contact. Each sample was tested on at least 3 positions and the average value was used as the reported one.

The crystalline phase structure of the scaffolds was identified at 3 and 14 days of soaking in SBF using an X‐ray diffractometer (XRD, Panalytical X'Pert Pro)with Cu‐Kα radiation. The X‐ray diffraction patterns were recorded using a Cu‐Kα radiation source (1.54060 Å) operated at 40 mA and 40 kV, with a scan range of 10° to 35° and a step size of 0.02° at a scan rate of 0.5 s step⁻¹. Crystalline phases were defined using the ICDD database.

The chemical structures and interactions of the scaffolds were analyzed using a Fourier transform infrared (FTIR) spectroscopy (Thermo Nicolet, iS50). Analyses were performed by measuring absorbance values in the frequency range of 400–4000 cm⁻¹. The relative β phfase content F _β in the PVDF phase was calculated by comparing the absorbances of vibration band peaks at 838 cm⁻¹ (CH₂ rocking) and 763 cm⁻¹ (CF₂ bending and skeletal bending) based on the Lambert–Beer law (Equation 2):^{[ 45 ]}

$$F_{\beta }=\frac{A_{\beta }^{838}}{1.26A_{\alpha }^{763}+A_{\beta }^{838}}$$ (2)

where A _α and A _β are the absorbance intensities of the samples at 763 and 838 cm⁻¹, respectively.

The thermal behavior and total crystallinity of the scaffolds were analyzed using Differential scanning calorimetry (DSC, TA Instrument/DSC25). The sample (≈7—8 mg) sealed in an aluminum pan was first heated from 0 to 210 °C and then cooled 210–0 °C at 10 °C min⁻¹ under a nitrogen atmosphere. The crystallization and melting temperatures were taken as the maximum peak temperatures of the cooling and heating curves, respectively.

The total crystallinity (X_C ) was calculated according to the following Equation (3):

$$X_C=\frac{\mathrm{\Delta }{H}_f^1}{\mathrm{\Delta }{H}_f^0}\times 100$$ (3)

where $\mathrm{\Delta }{H}_f^1$ is the melting heat and $\mathrm{\Delta }{H}_f^0$ is the melting heat of 100% crystalline PVDF, which is 104.5 J g⁻¹ for neat β‐PVDF respectively.

The mechanical properties of the scaffolds were measured using a static mechanical tester (Shimadzu AGS X) with a 1 kN load cell at a strain rate of 50 mm min⁻¹. A total of five samples were subjected to tensile strength testing, and the mean tensile strength data was reported with the standard deviation of at least five samples.

In vitro, the swelling behavior of the scaffolds was investigated by the gravimetric method. First, the samples were cut into ≈1.5 × 1.5 cm pieces and their dry weights (W _d) were measured. Then, the samples were immersed in 5 mL of phosphate‐buffered saline (PBS, pH 7.4, 37 °C) solution. The fluid‐absorbing samples were taken from the fluid at the end of the incubation periods of 1, 7, 14, 21, and 28 days, fluid on the surface was gently blotted and immediately weighted (W _w). The degree of swelling (W _s ) _t of the samples at various time intervals was calculated by the following Equation (4):

$${(W_s)}_t=\left(W_w-W_d\right)/W_d$$ (4)

where W_d is the dry weight, W_w is the wet weight of the samples. Measurements were made in triplicate and the standard deviation was calculated for each case.

In vitro, biomineralization experiments were performed by in 10xSBF solution which prepared according the procedure of the previous reports.^{[ 4} , ¹⁹ , ^{20 ]} Briefly, the samples cut to equal pieces were incubated in the 10xSBF with the following ionic concentrations (in units of mm): 1000 NaCl^‐, 5 KCl⁻, 25 CaCl₂.2H₂O, 5 MgCl₂.6H₂O, 3.62 NaH₂PO₄.H₂O at 37 °C. After some of the prepared solution was placed in a glass beaker, NaHCO₃ was added at a concentration of 10 mm to ensure that the pH was 7.4. After the incubation period of 3 and 14 days, the samples were removed from the solution and gently rinsed with distilled water. Finally, the samples were frozen at −20 °C and lyophilized at −80 °C for 4 h. The surface morphology of the samples was visualized using SEM and the Ca/P ratios of bone‐like apatite formed on the surfaces were characterized by EDS. Also, the chemical bonds of the mineralized CaP layer were characterized by FTIR, while the CaP formation on the surface was investigated by XRD.

The in vitro cytotoxicity of the scaffold was evaluated following the ISO‐10993‐5: Biological Evaluation of Medical Devices—In vitro Cytotoxicity Test.^{[ 79 ]} HOB cells were cultured in DMEM supplemented with 10% FBS and (FBS) and 1% penicillin‐streptomycin. The cells were maintained at 37 °C in a humidified atmosphere containing 5% CO₂ and were passed every two days. The MTT assay for cell viability was performed as described by Tornaci et al. (2024).^{[ 80 ]} MTT, a tetrazolium salt, was metabolized by the mitochondria of viable cells into insoluble purple formazan crystals, which accumulate intracellularly and were used as an indicator of cellular viability.

For the assay, 1 × 10⁴ cells well⁻¹ were seeded into 96‐well plates and incubated overnight. Scaffolds were washed with PBS, sterilized with UV light on both sides and extracted in DMEM at 37 °C according to the ISO‐10993‐12 extraction protocol.^{[ 81 ]} The resulting extraction media were applied to cells at different concentrations (100%, 50%, 25%, and 0%, with 0% representing cells incubated in DMEM alone as a control). Cells were incubated with the extraction media for 24, 48, and 72 h. At the end of each incubation period, the media were removed, and 110 µL of a 5 mg mL⁻¹ MTT solution prepared in PBS was added to each well. Following a 4‐hour incubation, 100 µL of DMSO was added to dissolve the formazan crystals, and the absorbance was measured at 570 nm with a reference wavelength of 630 nm.

Cell adhesion studies were conducted using fluorescence microscopy and SEM. Scaffolds were re‐sterilized, fortified with DMEM, and seeded with cells at a density of 5 × 10⁵ cells well⁻¹ in 24‐well plates. After 24, 48, and 72 h of incubation, the media were removed, and the scaffolds were washed with PBS and fixed in 2.5% glutaraldehyde for 20 min. Samples intended for fluorescence microscopy were stained with DAPI, washed with PBS, and dehydrated using increasing concentrations of ethanol (70%, 80%, 90%, and 100%). For SEM analysis, the samples were similarly dehydrated after fixation and PBS washing.

All experiments were performed in triplicate, and quantitative data were reported as mean ± standard deviation (SD). Statistical analysis for cell culture experiments was conducted using one‐way ANOVA followed by the Kruskal–Wallis multiple comparison test. The p‐value of less than 0.05 was considered statistically significant.

Ethical Statement

Since, an international database, which was open to all researchers was used, an ethics committee decision was not taken for this research.

Conflict of Interest

The authors declare no conflict of interest.

Mutlu B., Demirci F., Erginer M., and Duman Ş., “In Vitro Behavior of Boron‐Doped Baghdadite/Poly(vinylidene fluoride) Membrane Scaffolds Produced via Non‐Solvent Induced Phase Separation.” Macromol. Biosci. 25, no. 9 (2025): 25, e00619. 10.1002/mabi.202400619

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.