PVTF Nanoparticles Coatings with Tunable Microdomain Potential for Enhanced Osteogenic Differentiation

Abstract

Poly(vinylidene fluoride-trifluoroethylene) (PVTF) nanoparticles coatings with electrically heterogeneous microdomains were engineered to mimic the natural electromechanical microenvironment of bone tissue, offering a novel strategy to enhance osteogenesis. Through a biphasic solvent phase separation method, PVTF nanoparticles (NPs) were synthesized and spin-coated onto substrates, followed by melt-recrystallization to achieve high β-phase crystallinity. The substrates were then subjected to corona poling, a process involving high-voltage corona discharge to electrically polarize and align the molecular dipoles. Structural and electrical characterization revealed tunable microdomain surface potentials and piezoelectric coefficients, correlating with enhanced hydrophilicity. Notably, microdomain potential—produced by controlled polarization—was shown to directly regulate cellular responses. In vitro studies demonstrated that a corona-poled PVTF NP coating significantly improved bone marrow mesenchymal stem cell (BMSC) proliferation and early osteogenic differentiation. This work establishes a surface electropatterning approach and highlights the critical role of electrical heterogeneity in bone regeneration, offering a novel strategy for bioactive biomaterial design.

1. Introduction

The surface properties of biomaterials critically regulate cellular behavior [1,2], establishing surface engineering as a key strategy to guide cellular responses. Micro/nanostructures [3,4], surface potential [5,6], and wettability [7] are key physical properties that govern cellular behavior, such as adhesion, proliferation, and differentiation. Mimicking the endogenous physical microenvironment of the native extracellular matrix (ECM) [8] is a promising strategy for designing biomaterials. For example, electroactive microenvironment mimicking [9,10] enhanced bone regeneration.

Surface potential [11], a direct modulator of electrostatic interactions, has emerged as a key driver of cellular activity, especially in bone regeneration. Electroactive materials [12,13], such as piezoelectric polymers [14,15] and conductive composites, are extensively studied for generating endogenous electrical signals that replicate bone tissue’s electromechanical microenvironment. Poly(vinylidene fluoride-trifluoroethylene) (PVTF), as a typical ferroelectric polymer, shows relatively intense piezoelectricity, thermostability, processing simplicity, and excellent biocompatibility. Poled PVTF films promote osteogenic differentiation in bone mesenchymal stem cells (BMSCs) via surface potential modulation [16], which can also promote skin wound healing through controllable surface potential [17]. These studies underscore the importance of surface potential in osteogenesis, yet most focus on homogeneous charge distributions [18].

Despite these progresses, the role of microscale and nanoscale heterogeneous surface potentials—which are closer to the natural heterogeneity of the ECM [19]—remains poorly understood. As a highly complex network, heterogeneity is a fundamental characteristic of the ECM, with extensive evidence linking matrix material physical heterogeneity to cellular behavior and stem cell fate, while the heterogeneity mainly focuses on the regulation of microtopographies [20,21]. When electrical potentials approach micro- and nanoscale dimensions [22,23], subcellular-scale electrical stimulation can directly influence specific cellular components such as membrane receptors, ion channels, and cytoskeletal structures, thereby facilitating precise modulation of cellular behavior [24,25,26]. Understanding the impact of heterogeneous electrical charge distribution on cellular behavior is equally essential to align artificial substrates with native tissue microenvironments.

To address this challenge, we engineered a ferroelectric PVTF nanoparticle (NP) coating with tunable microdomain potentials. PVTF NPs were synthesized via a biphasic solvent strategy, spin-coated onto substrates, and melt-recrystallized to enhance β-phase crystallinity, followed by corona poling for molecular dipole alignment. By correlating the microscale surface electrical potentials with wettability and cellular responses, we elucidate how microdomain surface potentials enhance BMSC proliferation and osteogenic differentiation. Our findings establish a framework for designing electroactive biomaterials that mimic natural bone’s dynamic electrical cues, advancing bone regeneration strategies.

2. Materials and Methods

2.1. Preparation of PVTF NP Coatings

PVTF NPs were synthesized using a bi-solvent phase separation strategy [27,28], which involved a self-organizing process at the interface of two solvents with differing polarities. PVTF powder (P(VDF-TrFE) 70:30 mol%, Piezotech, Pierre-Bénite, France) was dissolved in N, N-Dimethylformamide (DMF, AR, >99.5%, Sinopharm Chemical Reagent, Shanghai, China) to prepare a 15 mg/mL precursor solution. In a 50 mL centrifuge tube, 5 mL of DMF was first added as a transition layer, followed by the slow introduction of 15 mL of the PVTF-DMF solution along the tube’s inner wall. Finally, 15 mL of deionized water was slowly injected from the bottom using a syringe. After 2 h of resting to allow stable stratification, the lower layer of the solution was collected and subjected to high-speed centrifugation at 8000 rpm for 3 min to obtain the precipitates. The obtained precipitates were then ultrasonicated for 30 min to ensure uniform dispersion in 15 mL of anhydrous ethanol (AR, >99%, Sinopharm Chemical Reagent, Shanghai, China) and stored at 4 °C.

As shown in Figure 1, to form the PVTF NP coating, a 25 μL ethanol dispersion of PVTF NPs was spin-coated onto a 1 cm × 1 cm Si substrate at 8000 rpm for 40 s. The substrate was then transferred to a hot plate, heated to 160 °C for 2 min, and allowed to cool to room temperature. To control the surface potential, the PVTF NP coating underwent corona poling under different conditions. The corona poling setup employed a high-voltage DC source connected to a needle electrode positioned 10 mm above the sample. The sample stage was grounded and equipped with a temperature-controlled heating system. The poling conditions of the PVTF NP coating were categorized as unpoled (PVTF-UN, 0 kV, 30 min, 160 °C), low-poled (PVTF-LP, 10 kV, 30 min, 160 °C), and high-poled (PVTF-HP, 12 kV, 30 min, 160 °C).

To investigate the influence of microstructure on cellular stimulation, a flat PVTF film was fabricated as a control for in vitro cell behavior evaluation. PVTF powder was dissolved in DMF at a concentration of 16.7 mg/mL. 100 μL of the solution was spin-coated onto a Si substrate at 4000 rpm for 30 s. The coated substrate was heated in ambient air using a muffle furnace. The temperature was ramped to 160 °C at 6 °C/min and held for 1 h. The resulting PVTF film was rinsed with deionized water, dried, and stored for subsequent experiments.

2.2. Characterization

The surface topography of PVTF NPs and coating was observed using scanning electron microscopy (SEM, ZEISS GeminiSEM 300, Oberkochen, Germany). Fourier transform infrared spectroscopy (FTIR, Vertex 70, Bruker, Billerica, MA, USA) and X-ray diffraction (XRD, Cu Kα, X-Pert Powder, PANalytical B.V., Almelo, The Netherlands) were employed to analyze the composition and phase. FTIR can determine the content of the polar β phase [29], using the equation:

$$\mathrm{F}\left(\mathsf{\beta }\right)={\displaystyle \frac{A_{\mathsf{\beta }}}{{({\mathrm{K}}_{\mathsf{\alpha }}/{\mathrm{K}}_{\mathsf{\beta }})A}_{\mathsf{\alpha }}+A_{\mathsf{\beta }}}}$$

(1)

A_β and A_α respectively represent the absorbance values of the β-phase at 842 cm⁻¹ and the α-phase at 763 cm⁻¹. The absorption coefficients, K_α and K_β, correspond to the wavelengths at 842 cm⁻¹ and 763 cm⁻¹, with K_{842 cm}⁻¹ = 7.7 × 10⁴ cm²/mol and K_{763 cm}⁻¹ = 6.1 × 10⁴ cm²/mol.

Crystallinity quantification was conducted following previous publications (XRD peak deconvolution method) [30,31]. The observed diffraction curves could be resolved into two peaks, crystalline and amorphous, by XRD peak deconvolution analysis. The degree of crystallinity, X_c, was evaluated using the equation:

$${\mathit{X}}_{\mathrm{c}}=\frac{A_{\mathrm{c}\mathrm{r}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{e}}}{A_{\mathrm{c}\mathrm{r}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{e}}+A_{\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{h}\mathrm{o}\mathrm{u}\mathrm{s}}}$$

(2)

where A represents the peak area.

For the quantification of ferroelectricity and piezoelectric coefficient at the nanoscale level, piezoresponse force microscopy (PFM, Bruker Dimension Icon, Bruker, Billerica, MA, USA) was employed in contact mode. Conducting tips (cobalt-chromium (Co/Cr) coated, 150 kHz of frequency, and 5 N/m of spring constant) were used. Surface potential was measured using scanning Kelvin probe microscopy (SKPM, NTEGRA Spectra, NT-MDT, Moscow, Russia). The static water contact angles (WCA) of samples under different corona poling conditions were measured using a contact angle meter (OCA20, Dataphysics, Filderstadt, Germany).

2.3. In Vitro Cell Behavior Evaluation

2.3.1. Cell Extraction and Cell Culture

Bone marrow mesenchymal stem cells (BMSCs) were extracted from Sprague-Dawley (SD) rats aged three weeks (Hangzhou Medical College). The cells were cultured in MEM Alpha medium (GENOM, Jiaxing, China) containing 10% fetal bovine serum (Cellmax, Beijing, China), 1% antibiotic solution (1 × 10⁴ units/mL penicillin, 1 × 10⁴ μg/mL streptomycin), 1% sodium pyruvate, and 1% MEM nonessential amino acids (all from Gibco, Thermo Fisher Scientific, Waltham, MA, USA). Following treatment with 0.25% trypsin and 1 mM EDTA (Gibco, Thermo Fisher Scientific, Waltham, MA, USA), the cells were plated onto samples in 24-well plates and incubated at 37 °C with 5% CO₂.

2.3.2. Cell Viability Assay

The viability of cells was evaluated using the CCK-8 assay (Dojindo Laboratories, Kumamoto, Japan) of different materials. The cells were grown on samples at a density of 5 × 10⁴ cells/well for 1 and 3 days. The samples were moved to fresh 24-well plates and treated with a working solution (50 μL CCK-8 in 500 μL medium per sample) for 3 h in a 5% CO₂, 37 °C incubator. Subsequently, viability was quantified by measuring the optical density (OD) at 450 nm with a microplate reader (Infinite F50, Tecan, Männedorf, Switzerland).

2.3.3. ALP Activity Assay

The osteogenic differentiation of BMSCs was assessed by quantifying alkaline phosphatase (ALP) activity. BMSCs were seeded on samples at a density of 5 × 10⁴ cells/well and exposed to osteogenic inducers (10 mmol/L ascorbic acid, 1 mmol/L dexamethasone, and 1 mmol/L β-glycerophosphate sodium) starting on day 3. After 7 and 14 days of incubation, the cell lysate was harvested. ALP activity and total protein levels were individually quantified using LabAssay ALP (WakoPure Chemical Industries, Osaka, Japan) and BCA protein assay kits (Thermo Fisher Scientific, Waltham, MA, USA), following the reagent instructions. ALP activity was quantified and normalized against total protein levels to evaluate osteogenic differentiation.

2.4. Statistical Analysis

Statistical analysis was shown as mean ± standard deviation (SD). Statistical significance was determined by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test using SPSS software (IBM SPSS Statistics 27.0, New York, NY, USA). The significance levels were denoted as * p < 0.05, ** p < 0.01, *** p < 0.001, and ns represents no significant difference.

3. Results and Discussion

3.1. Topography and Composition

Figure 2 displays the surface topography of PVTF NPs and the coating. PVTF NPs self-assemble at the DMF/water interface through a bi-solvent phase separation strategy. The strong hydrophobic interaction between PVTF’s fluorine groups and the water phase drives polymer chain aggregation into NPs [27,32]. Dissolved PVTF chains migrate toward the liquid–liquid boundary to initiate NP assembly, followed by gravitational sedimentation into the water phase. The synthesized PVTF NPs exhibit spherical morphologies with smooth surfaces (Figure 2a–c), displaying Gaussian-distributed dimensions centered at ~346 nm.

Subsequent spin-coating and thermal processing yielded PVTF NP coatings. Thermal annealing was conducted at 160 °C (ΔT = +8.5 °C relative to PVTF’s melting point of 151.5 °C). The resulting coating exhibited characteristic acicular crystalline domains (Figure 2d), indicative of controlled crystallization. Controlled cooling induced crystallization, producing lamellar structures [33]. Homogeneous substrate coverage was achieved (Figure 2e,f) featuring domains with a mean diameter of 1.195 μm. The observed size discrepancy arises from thermal-induced structural reorganization: molten NPs undergo substrate spreading and coalescence during the melting-agglomeration phase, ultimately forming larger composite domains.

The composition and crystallinity of PVTF NPs were characterized via FTIR and XRD analyses, comparing as-synthesized (pristine) and melt-recrystallized samples. PVTF exhibits four crystalline phases: α, β, γ, and δ. Among these phases, the β-phase demonstrates the highest dipole moment per unit cell (8 × 10⁻³⁰ C·m) [34]. Three characteristic β-phase absorption bands were identified in the FTIR spectra (Figure 3a): 1397 cm⁻¹ (CH₂ wagging), 842 cm⁻¹ (CH₂ rocking), and 1285 cm⁻¹ (CF₂ asymmetric stretching) [35,36,37]. While γ-phase identification via FTIR remains challenging due to spectral overlap with β-phase features, the absence of γ-phase signatures (e.g., 1234 cm⁻¹) [35] in Figure 3a confirms β-phase dominance. Therefore, the three distinct absorption bands collectively confirm the prevalence of polar β-phase. Quantitative analysis revealed β-phase fractions of 79.5% (pristine) and 82.2% (melt-recrystallized) in PVTF NPs.

Deconvolution of the pristine PVTF NPs’ XRD profile (Figure 3b) reveals a predominant β-phase diffraction at 19.8° (indexed to (110)/(200) planes) [35,38] superimposed on an amorphous halo (15–25° 2θ). Melt-recrystallization induced crystallographic refinement, evidenced by strong β-phase diffraction at 19.8° ((110)/(200)) and suppressed amorphous background scattering [28]. The degree of crystallinity, X_c, increased from 31.9% (pristine) to 47.3% (melt-recrystallized). Moreover, the reduced full-width-at-half-maximum (FWHM) from 0.90 (pristine) to 0.70 (melt-recrystallized) quantitatively confirms enhanced crystallinity through thermal processing [39].

The structural evolution induced by thermal treatment—increased β-phase fraction and crystallinity degree—confers enhanced dipole alignment capability in PVTF NPs. Such crystallographic optimization provides a mechanistic foundation for the superior piezoelectric response observed in annealed samples.

3.2. Nanoscale Piezo- and Ferro-Electricity

The polarization switching characteristics of the PVTF NP coating were analyzed by piezoresponse force microscopy (PFM) to confirm ferroelectric functionality [40]. PFM characterization of PVTF-LP was conducted under a drive amplitude of 852 mV. Figure 4 displays the surface topography and corresponding piezoresponse mappings (amplitude and phase) of PVTF-LP. Spatial heterogeneity in mechanical deformation is revealed between PVTF NP-decorated regions and bare substrate areas, while phase mapping demonstrates uniform polarization orientation across all NPs. Cross-sectional line profiles of a representative PVTF NP and adjacent bare substrate (white line) reveal coordinated variations in topography, piezoresponse amplitude, and phase orientation. A height differential of Δh ≈ 150 nm indicates nanoparticle protrusion morphology. Contrasting piezoresponse amplitudes at NP sites reflect localized deformation behavior. A uniform phase retention of 130° across the NP domain confirms successful dipole alignment.

Figure 5a presents characteristic PFM hysteresis loops demonstrating stable polarization reversal. A characteristic 180° phase shift (ΔΦ) emerges during polarization reversal, confirming stable domain switching. Applied tip bias induces polarization reversal, confirming the ferroelectric behavior in the coating.

Butterfly loop trajectories under ±10 V DC bias (Figure 5b) reveal the piezoelectric response. Progressive positive biasing (0→10 V) induces 1.2 nm expansion, with subsequent voltage reduction (10→0 V) causing contraction—a behavior mirrored in negative bias cycles. Negative biasing (0→−10 V) drives dipole reorientation, achieving 0.5 nm displacement at −5 V. Voltage cycling completes the hysteretic trajectory through dipole contraction, regenerating the initial configuration at 0 V bias.

PFM hysteresis loops and butterfly-shaped trajectories confirm the ferroelectricity of PVTF NPs, demonstrating electric-field-controlled dipole reorientation and enhanced piezoelectric tunability. Vertical PFM mode enables quantitative evaluation of the microscopic piezoelectric coefficient (d₃₃) in PVTF NP coatings [41]. Statistical analysis of piezoelectric properties was performed through PFM measurements at over 10 test points per sample (Figure 5c), ensuring representative characterization. The ascending d₃₃ values (PVTF-UN: 49.5 ± 8.5 pm/V, PVTF-LP: 78.3 ± 12.5 pm/V, PVTF-HP: 88.0 ± 10.0 pm/V) arise from progressively optimized dipole alignment under enhanced poling fields.

3.3. Micro-Domain Potential of PVTF NP Coatings

SKPM was employed to characterize corona-poled PVTF NP coatings under varying poling fields. Figure 6 presents correlated topography and surface potential for PVTF-UN, PVTF-LP, and PVTF-HP samples.

PVTF-UN exhibits microscale domains in topography (Figure 6a), spatially aligned with SEM-observed particle assemblies. Surface potential mapping (Figure 6d) reveals commensurate microscale electrical domains, demonstrating strong structure–property correlation. Surface potentials at PVTF sites are consistently reduced compared to bare substrate regions. Progressive polarization intensification induces global surface potential reduction, as quantified in Figure 6e,f. Post-poling analyses maintain observed structure–potential correlations.

Quantitative potential analysis under three poling conditions (Figure 7a) reveals systematic polarization-dependent surface potential modulation. Surface potentials of PVTF NPs decrease progressively with poling intensity: 319 mV (PVTF-UN) → 186 mV (PVTF-LP) → −70 mV (PVTF-HP). The absolute potential difference (|ΔV|) between PVTF NPs and bare substrate regions increases progressively from 80 mV to 267 mV under enhanced poling conditions.

The measured surface potential decreases predominantly arose from β-phase dipole realignment in PVTF nanostructures, and conductive Si substrates showed negligible response to corona poling. The corona poling setup employed a high-voltage DC source connected to a needle electrode above the sample. As shown in Figure 7c, atmospheric ionization produced positive oxygen/nitrogen ions that migrated to the grounded PVTF NP coating, accumulating positive surface charges. The sample was mounted on a temperature-controlled grounded metallic stage. The downward poling field induced the alignment of CF₂ dipoles with electronegative terminals (-CF₂) oriented upward, as determined by the dipole moment direction. Progressive field intensification enhanced dipole coherence, yielding a 77% d₃₃ improvement from 49.5 to 88.0 pm/V. This dipole configuration reduced the surface potential of PVTF by 389 mV (from 391 to −70 mV), as confirmed by SKPM.

Surface potential variations at the microscale directly govern macroscopic wettability [42]—a critical determinant of cellular responses through interfacial interactions. The wettability of solid materials is described by Young’s equation, ${\gamma }_{\mathrm{s}\mathrm{v}}-{\gamma }_{\mathrm{s}\mathrm{l}}={\gamma }_{\mathrm{l}\mathrm{v}}\cos \theta$, where θ refers to the contact angle and γ represents the surface interfacial energies between phases (solid, liquid, and vapor) with corresponding phase-denoted subscripts. Interfacial charge accumulation reduces γ_sl through electrostatic repulsion effects that decrease surface expansion energy [43], causing an appreciable increase in wettability. Water contact angle (WCA) measurements (Figure 7b) revealed potential-induced hydrophilicity enhancement, decreasing from 66.7° (PVTF-UN) to 46.7° (PVTF-LP) and 44.2° (PVTF-HP). This potential-induced wettability modification creates electrochemically active interfaces that profoundly mediate cell-material crosstalk.

3.4. Biocompatibility and Osteogenesis of Micro-Domain Potential

CCK-8 assays (Figure 8) demonstrated significantly higher viability (p < 0.001) of BMSCs on a PVTF NP coating relative to flat PVTF film. The coated samples exhibited sustained viability advantages on days 1 and 3, indicating that microdomain structures promote adhesion and proliferation behaviors. These findings suggest that the PVTF NP coating exhibits good biocompatibility. For poled samples (PVTF-LP/HP), polarization-induced surface potential modulation created distinct cellular responses: reduced initial adhesion coupled with enhanced subsequent proliferation.

The observed proliferation enhancement aligns with previous studies, where surfaces with controlled charge states and optimized hydrophilicity promote cellular proliferation [11,44]. Reduced primary adhesion likely stems from competing interfacial effects: the negative charge is unfavorable for cell adhesion, while hydrophilicity is favorable [7]. The combined effects generated a lower amount of cell adhesion.

ALP activity assays were conducted to evaluate early-stage osteogenic differentiation, with results shown in Figure 9. Initial ALP analysis (Day 7, Figure 9a) showed no significant osteogenic differentiation, while marked material-dependent variations emerged by Day 14 (Figure 9b). Comparable ALP levels between PVTF film and PVTF-UN (p > 0.05) suggest limited osteoinductive effects from microdomain structures alone. Poled samples (PVTF-LP/HP) demonstrated significantly enhanced ALP activity, indicating that negatively charged surfaces promote early osteogenic differentiation.

Comparative analysis reveals that a PVTF NP coating with microdomain structures significantly enhances cellular viability and facilitates coordinated adhesion–proliferation dynamics compared to flat PVTF film, though demonstrating limited osteogenic induction capacity. Surface potential engineering via polarization establishes dual bioeffects: marked proliferation enhancement coupled with early-stage osteogenic differentiation, as evidenced by the ALP activity assay.

Poled PVTF NP coatings generate localized negative potential microdomains on the surface, creating a unique electroactive environment that influences cellular behavior. The molecular mechanisms and signaling events associated with surface potential heterogeneity-driven osteogenesis remain unclear and are still under investigation. Studies have shown that integrins [45,46] are key mechanosensitive structures that translate mechanical cues into biochemical signals, thereby driving bone adaptation [8,47,48]. One plausible mechanism involves modulating integrin binding kinetics through electrostatic effects, where surface charges affect the conformation of adsorbed proteins. This, in turn, guides integrin clustering and activates downstream differentiation pathways [3,49,50]. Recent studies support this mechanism. Sun et al. [25] demonstrated that an electrical microenvironment enhances osteogenic differentiation of BMSCs by modulating integrin α5β1-mediated mechanosensing, while Li et al. [51] showed that surface charge stimulates enhanced neural differentiation of stem cells by activating the FAK-ERK mechanosensing pathway. Additionally, Bai et al. [26] demonstrated that heterogeneous surface potential promotes mechanosensitive differentiation through integrin overexpression. While the osteoinductive effects of poled PVTF coatings show promise, further research is required to fully elucidate the structure–function relationships governing these processes. This study lays the groundwork for developing electroactive biomaterials with tunable surface potentials for regenerative medicine applications.

4. Conclusions

PVTF NP coatings were fabricated using a bi-solvent phase separation strategy combined with spin-coating, which successfully created bioactive surfaces with tunable micro-domain potentials. The key findings include:

• Enhanced piezoelectric properties: Thermal annealing increased β-phase fraction from 79.5% to 82.2% and crystallinity from 31.9% to 47.3%.
• Effective dipole alignment: Corona poling enhances the piezoelectric coefficient d₃₃ from 49.5 pm/V to 88.0 pm/V while generating localized surface potential.
• Significant potential modulation: The poled surfaces showed a −389 mV reduction in surface potential compared to unpoled controls.

In vitro cell behavior evaluations demonstrated that the micro-domain structures significantly improved cellular adhesion and proliferation. Poled coatings, which created a micro-domain negative surface potential, were found to accelerate cell proliferation and promote early osteogenic differentiation of BMSCs.

This study establishes a novel surface electropatterning strategy that couples microdomain morphology with tunable potential for regenerative biomaterials, providing a new strategy for smart biomaterial design.