Correlation Between the Morphological Characteristics by Atomic Force Microscopy and the Biological Properties of Bioactive Zirconia/Polyethylene Glycol (ZrO₂/PEG) Hybrids

Abstract

Zirconia-based hybrid blends at various molecular or nanometer scales have attracted significant interest from a technological perspective. In particular, several inorganic-organic hybrids are being applied in the biomedical field. In this context, inorganic ZrO₂ and hybrids composed of ZrO₂, and polyethylene glycol (PEG) have been synthesized through the sol–gel process and characterized from both morphological and spectroscopic viewpoints to explore their potential as hybrid biomaterials. Atomic Force Microscopy (AFM) has enabled a quantitative assessment of the surface roughness of bioactive sol–gel-based materials. The findings indicated an increase in material porosity in relation to the amount of PEG present in the systems, underscoring the important role of PEG in influencing the morphological characteristics of zirconia-based blends. AFM images display the typical globular structure of PEG spread across the surface of all systems. All hybrid systems seem to be uniform, and no phase separation is evident, thereby validating that the produced materials are hybrid nanostructured ones. The simultaneous presence of both inorganic and organic phases was verified using Fourier-transform infrared spectroscopy (FT-IR). FT-IR deconvolution in 850–550 cm⁻¹ region showed that PEG progressively perturbs the Zr–O–Zr network, increasing disorder and establishing more flexible inorganic domains at high PEG content. Increasing polymer amount enhanced cell viability against NIH-3T3 cell line, while antibacterial activity decreased, with pure ZrO₂ showing the strongest inhibition against Escherichia coli (E. coli).

1. Introduction

The biological performance of the ZrO₂/PEG hybrids is strongly governed by the amount of polyethylene glycol (PEG) incorporated into the inorganic matrix, as it affects the surface morphology, hydrophilicity, and potential for forming calcium phosphate layers. Higher PEG content reduces surface cracks and enhances biocompatibility, with optimized ratios leading to improved cell proliferation and mechanical strength for biomedical applications [1,2,3]. Regarding structure and composition, ZrO₂/PEG hybrids are typically prepared via the sol–gel method, where hydrogen bonds form between Zr-OH groups and the PEG polymer chains, often resulting in amorphous, nanostructured, and bioactive materials. The process involves the hydrolysis and condensation of zirconium alkoxides in the presence of PEG, with hydrogen bonding occurring between Zr-OH and ether oxygen atoms in the polymer. These hybrid materials, suitable for biomedical applications, often exhibit enhanced flexibility, strength, and controllable drug release profiles [1,4,5]. Cell viability assays using NIH-3T3 fibroblasts show a clear enhancement in biocompatibility with increasing PEG content [1]. PEG-rich hybrids materials are highly regarded in tissue engineering and biomedical applications due to their unique surface and bulk properties, which create a more hydrophilic, flexible, and less rigid surface environment, which supports cell adhesion and proliferation [6]. The disruption of the Zr–O–Zr network by PEG reduces the density of exposed inorganic sites that can otherwise induce stress or cytotoxic responses, resulting in improved fibroblast viability at higher polymer loadings [7].

In contrast, antibacterial tests against E. coli reveal an opposite trend. Pure ZrO₂ exhibits the strongest antibacterial activity, likely due to its rigid, highly connected inorganic surface and the presence of active zirconia sites capable of interacting with bacterial membranes [8,9,10,11,12]. As PEG content increases, the antibacterial effect progressively diminishes. The polymer forms a more hydrated and compliant surface layer that partially shields the inorganic domains, reducing direct contact between bacteria and the active ZrO₂ surface. Additionally, the increased surface roughness of PEG-rich regions and the reduced availability of reactive inorganic sites contribute to the lower antibacterial efficiency. More specifically, a surface that is too rough (high PEG roughness) and lacks exposed reactive inorganic components creates a “non-active” environment that is more conducive to bacterial colonization [13,14,15].

Overall, PEG incorporation enhances mammalian cell compatibility while attenuating antibacterial activity, highlighting a tunable balance between cytocompatibility and antimicrobial performance in ZrO₂/PEG hybrid materials [16].

Microscopy methods like atomic force microscopy (AFM) are becoming essential tools for thoroughly examining morphology and structural characteristics at micro and sub-micrometric levels, to assess the temporal relationship between the physicochemical properties of biomaterials and their biological responses. AFM is not just a basic tool for analyzing surface topography; it can greatly enhance the understanding of surface and interface characteristics, thereby aiding in the optimization of biomaterials’ performance, processes, and their physical and chemical properties at the micro and nanoscale [17].

In this study, atomic force microscopy (AFM) provides a powerful means to link nanoscale surface morphology with the underlying structural modifications induced by varying PEG content in ZrO₂/PEG bioactive hybrids.

In this work, atomic force microscopy (AFM) revealed that the surface morphology of the ZrO₂/PEG hybrids is strongly influenced by the amount of PEG incorporated into the inorganic matrix. Pure zirconia exhibited a relatively compact and uniform topography, consistent with a highly connected Zr–O–Zr network. As PEG content increased, the surface became progressively rougher, displaying more pronounced nanoscale features and height variations. This evolution in roughness reflects the disruption of the inorganic network caused by PEG, which promotes the formation of heterogeneous organic–inorganic domains and induces uneven shrinkage during gelation and drying. The resulting topography is characterized by a combination of polymer-rich depressions and zirconia-rich protrusions, producing a more irregular and textured surface, directly captured by AFM height maps and roughness parameters. These morphological changes align with the FT-IR evidence of increased structural disorder and contribute to the enhanced hydrophilicity and improved cell compatibility observed at higher PEG loadings. By correlating these features with spectroscopic evidence of increased structural disorder, AFM highlights how PEG content governs both the physical architecture and the biological interface of the hybrids. This relationship is crucial for understanding how PEG-driven morphological softening enhances cell compatibility while simultaneously reducing antibacterial activity. The innovation of this work lies in the use of atomic force microscopy (AFM) to analyze in detail how PEG content affects the morphology and biological properties of ZrO₂/PEG hybrids. This approach allows for the correlation of nanoscale structural changes with biological outcomes, thus optimizing the design of biomaterials for biomedical applications.

Surface roughness plays a central role in determining how biomaterials interact with their biological environment, influencing protein adsorption, cell adhesion, proliferation, differentiation, and even bacterial attachment [18]. Surface roughness is one of those deceptively simple parameters that ends up shaping almost every biological interaction a material has [19]. AFM is uniquely powerful here because it captures nanoscale features that other techniques simply average out. Because of this, the importance of AFM-based roughness evaluation in biomaterials [20] should be highlighted.

AFM roughness analysis allows researchers to detect subtle topographical variations that arise from changes in composition, processing conditions, or organic–inorganic interactions within hybrid materials. These nanoscale features often govern the initial biological response: moderately rough, hydrophilic surfaces tend to promote mammalian cell adhesion and viability, while sharper or more irregular textures can enhance bacterial colonization or trigger inflammatory reactions [21]. By quantifying parameters such as Ra, Rq, and peak-to-valley distances, AFM makes it possible to correlate structural modifications—such as polymer incorporation, network disruption, or phase separation—with functional biological outcomes [22].

In the context of hybrid systems, AFM becomes especially valuable [23]. In this regard, for the ZrO₂/PEG hybrid materials analyzed in this study, AFM reveals how PEG content alters the inorganic network, modifies surface texture, and ultimately tunes the balance between cytocompatibility and antibacterial performance. This nanoscale insight is essential for designing biomaterials with targeted biological behavior, ensuring that structural engineering at the molecular level translates into predictable and desirable interactions at the cellular level [24].

In this work, the correlation between hydrophilicity and surface roughness of ZrO₂/PEG hybrids is positive; increasing surface roughness, due to the addition of PEG, improves the hydrophilicity of the material. This increased hydrophilicity promotes cell adhesion and migration, thus contributing to better biocompatibility. In short, a rougher and more hydrophilic surface facilitates more effective cellular interactions.

The goal of this work is to articulate structure-property relationships for a study of biomaterials, namely zirconia-based hybrid blends, by trying to link the appearance of the material on a molecular scale to its biological or functional behavior. A clear structure–property link emerges when you look at the AFM roughness data together with the FT-IR evidence of network disruption. The correlation can be explained through how PEG interferes with the inorganic condensation process and how that translates into nanoscale topography. As for structure-property relationships, deconvolution of the 850–550 cm⁻¹ FT-IR region shows that increasing PEG content progressively perturbs the Zr–O–Zr network. This manifests as: greater band broadening (indicative of disorder), reduced connectivity of the inorganic framework, formation of more flexible and less cross-linked zirconia domains. The structural interpretation may be that the PEG chains interfere with Zr–O–Zr bridging, loosening the inorganic network and increasing local mobility, namely the PEG chains migrate and reorganize during solvent removal. This dynamic rearrangement increases surface undulations. Regarding the overall structure–property relationship, it can be stated that increasing PEG disrupts the Zr–O–Zr network, producing a more flexible, hydrophilic, and biologically friendly material. However, this same structural softening reduces the antibacterial potency associated with the rigid, highly connected zirconia framework. The strength of this work is the clear correlation between the morphological changes at the nanoscale, observed by AFM, and the biological properties of ZrO₂/PEG hybrids. Increased surface roughness and hydrophilicity, due to the incorporation of PEG, are directly linked to improved cellular compatibility, while disruption of the Zr–O–Zr network negatively impacts antibacterial activity. This structure-property relationship provides an important basis for designing biomaterials with desired biological behaviors. To our knowledge, there is no article in the literature that establishes such a detailed correlation between morphological changes observed by AFM and specific biological properties of ZrO₂/PEG hybrids.

In summary, in this research work, we analyze how polyethylene glycol (PEG) content affects the morphological characteristics and biological properties of zirconia/polyethylene glycol (ZrO₂/PEG) hybrids. The aim is to correlate nanoscale structural changes with biological outcomes, thus optimizing the design of biomaterials for biomedical applications. Specifically, the aim is to understand how surface roughness and biocompatibility can be improved to promote positive interactions with the biological environment. This study was undertaken to address the growing need to develop biomaterials that can improve interaction with biological tissues, particularly in the field of biomedical applications. Zirconia (ZrO₂) is known for its bioactivity, but its integration with polyethylene glycol (PEG) could further optimize its properties, such as biocompatibility and the ability to promote cell adhesion. Understanding how variations in PEG content influence the morphology and biological properties of hybrids is crucial for designing more effective materials for use in regenerative medicine and tissue engineering. Furthermore, a thorough presentation of the data and methodologies used supports the validity of the conclusions and offers a useful reference for future research in the field of biomaterials.

It is worth noting that this study addresses the gap in understanding the interactions between the morphological characteristics and biological properties of zirconia/polyethylene glycol (ZrO₂/PEG) hybrids. Although previous research on zirconia-based materials exists, a thorough analysis of how variations in PEG composition influence the structure and biocompatibility of these hybrids is lacking. This study aims to address this gap by providing quantitative and qualitative data linking nanoscale morphology with biological performance, thus contributing to improved biomaterial design.

2. Materials and Methods

2.1. Materials

Zirconium (IV) propoxide solution (Zr(OCH₂CH₂CH₃)₄, 70 wt% in 1-propanol, Sigma-Aldrich Merck Life Science S.r.l., Milan, Italy) was employed as the precursor for the sol–gel synthesis, while ethanol (EtOH, CH₃CH₂OH, ≥99.8% purity, Sigma-Aldrich Merck Life Science S.r.l., Milan, Italy) served as the co-solvent. Acetylacetone (AcAc, CH₃COCH₂COCH₃, ReagentPlus, ≥99%) was used as a chelating agent to control the hydrolysis rate. Polyethylene glycol 400 (PEG 400, average M_v = 400 Da) was purchased to Sigma-Aldrich Merck Life Science S.r.l., Milan, Italy was incorporated as the organic phase in the hybrid materials. Potassium bromide (KBr, for spectroscopy IR grade, ACROS ORGANICS) was used for FT-IR measurements. Bacterial strains and their agar-based growth media were obtained from Liofilchem, Roseto degli Abruzzi, Italy.

2.2. Methods

2.2.1. The Sol–Gel Synthesis

ZrO₂/PEG hybrid materials (containing 0, 6, 12, 24, 50, 60, and 70 wt% of PEG) were prepared through a sol–gel route. Zirconium (IV) propoxide was firstly dissolved into ethanol, water, and AcAc solution (SolA) in a molar ratio of Zr(OC₃H₇)₄:H₂O:AcAc = 1:1:4.5. In this work, the ZrO₂/PEG hybrid material containing 0 wt% of PEG was named as ZrO₂. PEG, pre-dissolved in ethanol (SolB), was incorporated into the reaction mixture under vigorous stirring to obtain a homogenous solution (SolA + SolB) at the desired PEG content (0–70 wt% with respect to the total hybrid mass). Further information on the PEG amount used for the sol–gel synthesis is reported in Table S1.

Each PEG composition was prepared as an independent synthesis batch. For each composition, a freshly prepared zirconium precursor solution (Sol A) was obtained above. After gelation, the resulting wet gels were air-dried at 45 °C for 48 h to remove residual solvent. A schematic representation of the synthesis procedure is reported in Figure 1.

2.2.2. Atomic Force Microscopy (AFM)

Atomic force microscopy (AFM) analysis was conducted using a BrukerNanoScopeV multimodeAFM (Digital Instruments, Santa Barbara, CA, USA) device to measure the surface roughness characteristics and examine the nanoscale surface structure of sol–gel derived materials. AFM is utilized to analyze both the surface structure and the morphological arrangement of polymer materials [25]. This work discusses three types of AFM images: height, amplitude error, and phase, which together give important details about a sample’s surface. The height image, or topography image, is the most crucial and commonly used. It shows the vertical changes in the AFM tip as it scans across the surface, revealing surface roughness and features like grains, pits or protrusions. The amplitude error image shows the difference between the desired and actual oscillation of the cantilever, highlighting edges and fine details less visible in the height image. The phase image indicates the phase lag between the driver and actual oscillations, providing information on material properties like stiffness and adhesion. It helps to differentiate parts in composite materials based on their responses to the tip’s oscillation. When examining a polymer blend, these images together offer a complete view of the sample’s structure, mechanical behavior, and material composition.

Topographic height images, along with amplitude error and phase images, were obtained at room temperature and analyzed using Bruker software Nanoscope Analysis 1.80 (BuildR1.126200, Digital Instruments, Santa Barbara, CA, USA). The AFM measurements were done in tapping mode, where a pointed tip on a cantilever oscillates and occasionally contacts the sample surface. Variations in amplitude or phase during scanning create topography. A laser beam reflects off the cantilever to measure deflection. The tip’s height is adjusted for consistent force, with lateral resolution typically a few nm and vertical resolution a few Å. In this study, the tip is defined by a radius of 5−10 nm, a standard spring constant of 20−100 N/m, and resonance frequencies ranging from 200 to 400 kHz. Due to the intermittent and non-continuous contact, this working method minimizes damage to the sample’s tip or surface and is especially ideal for examining soft materials like organic or biological substances. For each sample examined, multiple AFM images were captured at various locations to assess the variation in the roughness parameters and confirm their reproducibility across different scanned regions of the samples. The scanning frequency was 0.500 Hz for each scan line, containing 512 pixels per line. To assess the roughness of the surface, various roughness parameters are typically calculated and utilized. In this study, to determine the quantitative roughness, two significant height parameters, specifically the roughness average (Ra) and the root mean square roughness (Rq), have been taken into account. Specifically, Ra indicates the arithmetic average of the absolute values of the surface profile’s height, while Rq is comparable to the roughness average (Ra); however, Rq responds more acutely to peaks and valleys than Ra, as a result of the amplitude being squared in its calculation. These amplitude parameters, which define the surface by the vertical variations in the roughness map from the average surface, are widely utilized in the literature [26,27,28]. For each of the samples, the Ra and Rq parameters were assessed, based on Equations (1) and (2):

$$Ra={\displaystyle \frac{1}{l_r}}{\int }_0^{l_r}|z(x)|\ast dx$$

(1)

$$Rq=\sqrt{{\displaystyle \frac{1}{l_r}}{\int }_0^{l_r}z{(x)}^{2}dx}$$

(2)

where l_r represents the line’s length, z denotes the height, and x indicates the position.

2.2.3. Fourier-Transform Infrared Spectroscopy (FT-IR)

FT-IR spectroscopy was employed to explore the interactions occurring in the ZrO₂/PEG hybrids. The analyses were performed using a Prestige 21 spectrometer (Shimadzu, Milan, Italy). Spectral acquisition covered the 400–4000 cm⁻¹ range, with a resolution of 2 cm⁻¹ and the accumulation of 64 scans. For each measurement, pellets were prepared by mixing 198 mg of KBr to 2 mg of finely ground sample. The collected spectra were processed with IR Solution and Origin software, and the principal absorption band in the 850–550 cm⁻¹ region was deconvoluted through the Multiple Peak Fit function available in Origin 8. The FT-IR band deconvolution was performed using Gaussian peak functions. The gaussian fitted parameters have now been reported in the Supplementary Materials (Tables S1–S6) for the deconvoluted samples. The tables include peak position (cm⁻¹), full width at half maximum (FWHM), peak area (a.u.), relative area (%) that was calculated as:

$$\mathrm{Area}(\%)={\displaystyle \frac{\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{a}(\mathrm{a}.\mathrm{u}.)\mathrm{o}\mathrm{f}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{h}\mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k}}{\mathrm{s}\mathrm{u}\mathrm{m}\mathrm{o}\mathrm{f}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{s}}}\times 100$$

The FWHM values were calculated from the Gaussian width parameter obtained from the fitting procedure according to the relation:

FWHM = 2.35482⋅ω

where ω represents Gaussian width parameter of the fitted peak.

To investigate the structural evolution of the hybrids while maintaining clarity of presentation, FT-IR band deconvolution was performed on representative hybrid compositions, allowing the evaluation of the main spectral trends as a function of PEG content.

2.2.4. Biocompatibility Characterization of ZrO₂/PEG Hybrids

Cytotoxicity was assessed using the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay (indirect test) on NIH-3T3 murine fibroblasts. Extracts were obtained from disks of ZrO₂ and ZrO₂/PEG hybrid sol–gel materials containing 6, 12, 24, 50, 60, and 70 wt% PEG, after immersion for 24 h in 3.5 mL of complete culture medium at 37 °C under continuous stirring. Cells (1.0 × 10⁴/well) were seeded in 96-well plates and exposed to 100 μL of each extract [29]. After 24, 48, and 72 h, the MTT solution (0.5 mg/mL) was added and incubated for 1 h at 37 °C. The resulting formazan was dissolved in DMSO, and absorbance was measured at 570 nm using a Tecan SpectraFluor reader. Cell viability was expressed as the percentage of metabolic activity compared with the untreated control. Tests were performed in twelve replicates (n = 12) for three samples per extract.

2.2.5. Antibacterial Activity

The antimicrobial activity of the samples was evaluated through a slightly modified Kirby-Bauer method. The powders were finely ground, pressed into 200 mg discs, and sterilized under UV light for 1 h prior to being placed in direct contact with microbial suspensions (10⁵ CFU/mL). Two bacterial strains were employed: E. coli ATCC25922 (Gram-negative) and Enterococcus faecalis (E. faecalis) ATCC29212 (Gram-positive), both selected for their clinical importance as common causative agents of nosocomial infections in hospital settings. E. coli and E. faecalis were cultured on Tryptone Bile X-Gluc medium and Slanetz–Bartley agar base, respectively, and incubated in the presence and absence of the discs. After incubation, inhibition zones were recorded, with four measurements taken per Petri dish. Data were expressed as mean ± Standard Deviation (SD).

3. Results

3.1. AFM Morphological Characterization and Surface Roughness Evaluation

The surface roughness and nanoscale surface morphology of the sol–gel-synthesized zirconia-poly(ethylene glycol) (ZrO₂/PEG) hybrid systems were assessed using Atomic Force Microscopy (AFM) in relation to the varying PEG concentrations. Specifically, the AFM approach allowed for the measurement of the surface roughness of the samples and the assessment of the impact of varying PEG concentration on them. In order to illustrate the connection between the surface roughness of the examined samples and their porosity, a thorough AFM investigation was used in this work to examine the morphology of nanostructured hybrids formed by PEG and zirconia (ZrO₂) using the sol–gel method in which intermolecular interactions (hydrogen bonds) between the Zr-OH groups and PEG (terminal OH or ethereal oxygen) are clearly observed. The findings showed that the amount of PEG in the hybrid systems increased the material porosity, which increases adhesion and produces sol–gel hybrid materials with excellent bioactivity. Increased porosity causes the ZrO₂/PEG nanohybrid’ surface roughness to increase, resulting in variances and irregularities in the materials’ surface topography and morphology.

Figure 2, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8 show the morphological analysis of ZrO₂ and ZrO₂/PEG hybrids. In particular, for each sample, the three AFM-2D images, namely Height, Amplitude Error and Phase, are reported on the top in (a) while the corresponding AFM-3D images are reported on the bottom in (b).

Regarding phase distribution, looking at AFM images, while lower percentages of PEG create a more refined, disperse structure, increasing the concentration of PEG can lead to agglomerations causing significant surface irregularities and increased uneven porosity. These findings are in accordance with scanning electron microscopy (SEM) observations, showing that the morphology of the materials changes progressively with increasing PEG content. Indeed, samples containing low PEG amounts (6–12 wt%) appear compact and homogeneous, with a relatively smooth morphology, while at intermediate PEG contents (24–50 wt%) the microstructure becomes more heterogeneous and filament-like features with an average diameter of about 10 μm appear, and at higher PEG concentrations, the materials exhibit a markedly irregular surface morphology [30,31].

To obtain repeatable and comparable surface roughness measurements and information on the actual substructure and morphological characteristics of the ZrO₂ matrix alone and ZrO₂/PEG hybrids characterized by different amounts of PEG (6, 12, 24, 50, 60 and 70 wt% with respect to the zirconia content), the same scan size (2 µm), which represents one of the critical experimental conditions to be maintained constant, was used for the different samples analyzed. The quantitative parameters Ra and Rq are defined mathematically, but their value changes if the size of the dataset changes. In conclusion, using the same scan size therefore ensures that the same scale of phenomena is observed, the resolution is comparable, the roughness parameters are comparable, and the comparison is scientifically valid. From a comparison of the different samples, we can easily deduce that AFM images of height, amplitude error, and phase are crucial for studying ZrO₂–PEG hybrid materials because they allow us to see the actual topography (height image), identify the structural details of the nanoparticles (amplitude error image), and distinguish the organic and inorganic components thanks to mechanical contrast (phase image). It is precisely the combination of the three channels that allows us to understand how ZrO₂ and PEG are distributed, interact, and organized on the surface. The 3D view of AFM images allows for a spatial understanding of the surface and a more accurate assessment of roughness and roughness distribution. 3D AFM images provide information about slope and inclination that are difficult to capture in 2D, where the slope is “compressed” into color.

Table 1. Roughness parameters Ra and Rq for the ZrO₂ matrix alone and ZrO₂/PEG hybrids (ZrO₂/PEG 6 wt%, ZrO₂/PEG 12 wt%, ZrO₂/PEG 24 wt%, ZrO₂/PEG 50 wt%, ZrO₂/PEG 60 wt%, ZrO₂/PEG 70 wt%).

Sample	Ra (nm)	Rq (nm)
ZrO2	105	134
ZrO2/PEG 6 wt%	133	167
ZrO2/PEG 12 wt%	175	212
ZrO2/PEG 24 wt%	279	351
ZrO2/PEG 50 wt%	480	592
ZrO2/PEG 60 wt%	562	685
ZrO2/PEG 70 wt%	645	768

shows the roughness parameters Ra and Rq for the ZrO₂ matrix alone and ZrO₂/PEG hybrids characterized by different amounts of PEG (6, 12, 24, 50, 60 and 70 wt% with respect to the zirconia content).

The surface roughness parameters Ra and Rq increase from ZrO₂ to ZrO₂/PEG hybrids and with higher PEG content due to the incorporation of PEG into the ZrO₂ matrix, which enhances porosity and creates irregularities in the surface structure. As PEG concentration rises, it disrupts the Zr–O–Zr bridging network, leading to a more complex and rougher surface morphology. This increased surface roughness is indicative of the morphological changes that occur as the organic phase (PEG) interacts with the inorganic phase (ZrO₂).

The distribution of the inorganic phase (represented by ZrO₂) and the organic phase (represented by PEG) along the three white lines is evaluated using AFM section analysis for the ZrO₂ matrix and ZrO₂/PEG hybrid systems in Figure 9. The distribution profile of the two phases, which are visible on the right side of the AFM image in the three distinct colors of green, blue, and red, is represented by each white line drawn along the sample surface section.

Although it is feasible to observe the degree of irregularity of the material surface as the weight percentage of PEG increases, the profiles show a fairly regular distribution of the interpenetrated phases for all the samples because the overall structure remains interconnected and cohesive. The PEG incorporation leads to localized surface irregularities without significantly disrupting the uniformity of the phase distribution across the samples. This indicates that while surface morphology becomes more complex, the interpenetration of the ZrO₂ and PEG phases maintains a consistent arrangement throughout the material.

3.2. FT-IR Spectroscopy

FT-IR model Prestige 21 (Shimadzu, Kyoto, Japan) was used to investigate the interactions between the inorganic and organic phases in ZrO₂/PEG hybrids. Figure 10 shows the spectra of ZrO₂ with increasing PEG content (6–70 wt%), compared to pure ZrO₂ and PEG. ZrO₂ synthesized via sol–gel at room temperature and treated at 45 °C to remove organic solvents showed, by X-ray diffractometer (XRD) Philips PW 1730 (Philips, Almelo, The Netherlands), broad diffuse halos typical of amorphous materials, indicating the absence of long-range crystalline order, as reported elsewhere in a previous work [30]. The broad band around 3400 cm⁻¹ is attributed to OH stretching vibrations from residual absorbed water and Zr–OH groups involved in hydrogen bonding. Increasing PEG content caused a shift in this band to lower wavenumbers, up to 22 cm⁻¹ in ZrO₂/PEG 70 wt%, indicating an increased interaction between PEG and ZrO₂. In 1800 and 1200 cm⁻¹ range, ZrO₂ spectra shows peaks related to the presence of AcAc in the matrix, used during sol–gel synthesis. As previously reported, residual AcAc signals appear at 1583, 1530, 1424, 1364, 1282, and 1030 cm⁻¹ [32], and their intensity progressively decreases as the PEG content increases [31].

The incorporation of PEG is strongly highlighted in the FT-IR spectra of the hybrid materials. In particular, the bands at 2915 and 2873 cm⁻¹ correspond to the asymmetric and symmetric C–H stretching vibrations, respectively. In pure PEG, these bands increase in intensity with increasing PEG content. Similarly, the intensity of the methylene C–H bending band at 1455 cm⁻¹ (in pure PEG) increases proportionally with the PEG content. The band at 1250 cm⁻¹, attributed to the alcoholic ν(C–O) stretching, undergoes a slight shift toward lower wavenumbers (down to ~1247 cm⁻¹) in hybrids with a high PEG content. This downshift is consistent with the formation of hydrogen bonds between the terminal CH₂–OH group of PEG and the Zr–OH groups of the inorganic matrix, which weaken the C–O bond and reduce its vibrational energy. The broader ether band of pure PEG, centered at ~1100 cm⁻¹ and attributed to ν(C–O–C) stretching, becomes more intense and better defined in the hybrids, also showing a shift toward lower wavenumbers. These changes indicate that PEG is effectively incorporated into the ZrO₂ matrix and interacts with the residual AcAc and hydroxyl groups, gradually dominating the spectral features as its concentration increases [33].

The FT-IR spectra also show that the characteristic doublet of amorphous zirconia, corresponding to Zr–OH (∼655 cm⁻¹) [34] and Zr–O–Zr (∼623 cm⁻¹) in pure ZrO₂, is affected by the formation of the hybrids. Indeed, a clear modification of this double band is observed upon PEG incorporation, indicating an alteration of the local Zr–O interaction environment induced by the hybrid network [31].

To better understand the structural evolution of the inorganic network upon PEG incorporation, the FT-IR region between 850 and 550 cm⁻¹ was deconvoluted (Figure 11). More details about the fitting parameters are reported in Supplementary Material (Tables S1–S6).

In pure zirconia, the spectrum is characterized by the typical doublet of amorphous ZrO₂, with the ν(Zr–O) mode at ~683 cm⁻¹ (yellow band) and the Zr–O–Zr bridging vibration at ~623 cm⁻¹ (violet band), together with the Zr–OH band at ~655 cm⁻¹(blue band) and the lower-frequency Zr–O–Zr deformation at ~577 cm⁻¹ (pink band) [35,36,37].

The band with a maximum at 683 cm⁻¹ in ZrO₂ shows an area increasing upon addition of PEG up to 24 wt%, indicating a higher degree of local disorder in partially condensed Zr–O domains due to PEG perturbation of the inorganic ZrO₂ network. Above 50 wt% PEG, conversely, this band decreases, in accordance with the transition to a polymer-dominated hybrid in which the relative fraction of zirconia is reduced and the inorganic clusters of Zr–O–Zr become more isolated. The 655 cm⁻¹ band shows a slight decrease up to 24 wt% PEG, likely due to the partial screening of surface Zr–OH groups involved in hydrogen bonding with PEG, as also confirmed by the observed downshift occurred [22] while increasing PEG content. At higher polymer loadings (PEG amount > 50 wt%), the intensity of the band at 655 cm⁻¹, related to surface Zr–OH groups, is not affected by PEG amount. Conversely, the 623 cm⁻¹ band intensity decreases almost linearly with PEG content, confirming the progressive disruption of the Zr–O–Zr bridging network by the PEG chains [38]. Finally, the 577 cm⁻¹ component increases at high PEG loadings, reflecting the formation of more flexible and less-condensed Zr–O–Zr environments.

Conversely, the 623 cm⁻¹ band decreases almost linearly with PEG content, confirming the progressive disruption of the Zr–O–Zr bridging network caused by the organic chains of PEG. The 577 cm⁻¹ band in pure ZrO₂ is almost negligible, but upon polymer incorporation it appears even at low PEG content (6 wt%) as a shoulder. In this region, the FTIR spectrum of pure PEG shows a contribution from wagging/rocking modes of –CH₂– groups bound to oxygen [39], suggesting that the polymer induces local perturbations in the Zr–O–Zr network. As PEG loading increases, the intensity of the 577 cm⁻¹ band grows markedly, particularly at 70 wt% PEG, reflecting the transition to a polymer-dominated hybrid and the formation of more flexible and disordered Zr–O–Zr environments.

3.3. Biocompatibility Characterization

The ZrO₂/PEG hybrids were tested as pressed disks to evaluate their cytotoxicity toward murine fibroblastic NIH-3T3 cells at increasing exposure times (24, 48, and 72 h). Pure zirconia is inherently non-cytotoxic, showing cell viability values above 70% [40]. The addition of PEG to the hybrid system further increases cell viability in proportion to the PEG content, due to its low immunogenicity and toxicity. Moreover, PEG acts as a scaffold material: its hydrophilicity enhances cell attachment and migration, contributing to the improved biocompatibility. This beneficial effect combines with the well-known bioactivity of ZrO₂, which promotes bone cell activity, thereby explaining the overall enhancement in biocompatibility [41,42].

These data are consistent with the information obtained by AFM (Table 1): the increase in surface roughness, proportional to the PEG content in the zirconia matrix, facilitates cell adhesion and proliferation. As a result, biocompatibility increases linearly, reaching up to 92% cell viability for the ZrO₂/PEG 70 wt% hybrid (

Table 2. Cell viability of ZrO₂ and ZrO₂/PEG hybrid materials against NIH-3T3 cells.

Sample	24 h CV %	48 h CV %	72 h CV %
ZrO2	78.20 ± 1.80	77.30 ± 1.50	76.10 ± 1.20
ZrO2/PEG 6 wt%	82.40 ± 1.60	80.50 ± 1.20	78.90 ± 1.00
ZrO2/PEG 12 wt%	85.10 ± 1.00	82.60 ± 0.80	80.40 ± 0.70
ZrO2/PEG 24 wt%	87.30 ± 1.30	84.20 ± 1.00	82.50 ± 0.90
ZrO2/PEG 50 wt%	90.10 ± 1.50	86.30 ± 1.20	84.70 ± 1.00
ZrO2/PEG 60 wt%	91.20 ± 1.00	87.40 ± 0.90	85.30 ± 0.80
ZrO2/PEG 70 wt%	92.00 ± 2.50	88.10 ± 2.00	86.20 ± 1.00

) [1].

A modified agar diffusion test was used to evaluate the antimicrobial activity of the ZrO₂/PEG hybrids against two bacterial strains: E. coli as a Gram-negative model and E. faecalis as a Gram-positive one (Figure 12). The results show that pure ZrO₂ exhibits an inhibition halo of 1.8 cm in accordance with scientific literature evidence [43] which decreases proportionally with increasing PEG content, down to 1.3 cm. This value corresponds to the diameter of the sample placed at the center of the plate, meaning that inhibition occurs only at the point of contact. This trend is expected since PEG has no intrinsic antibacterial activity [44], and therefore its higher content dilutes the antibacterial effect. For E. faecalis, pure ZrO₂ shows an inhibition halo of 1.3 cm, and this value remains approximately constant even for the ZrO₂/PEG 70 wt% hybrids. These results are significant for a greater understanding of aspects directly related to the antibacterial properties of materials for various applications [45,46,47,48].

The results we obtained help to highlight the importance of ZrO₂ as promising material also on the basis of the recent applications (nuclear technology, luminescence devices and gas sensors), studies, and achievements related to ZrO₂ [49,50,51].

4. Conclusions

The morphological characteristics of ZrO₂/PEG hybrids materials (containing 0, 6, 12, 24, 50, 60, and 70 wt% of PEG) prepared through a sol–gel route were analyzed by AFM and include surface topography and roughness. AFM images show an increase in porosity and the formation of irregularities in the morphology with increasing PEG content. Furthermore, roughness parameters, such as Ra and Rq, increase significantly with increasing PEG concentration, highlighting the changes in the surface structure of the materials. The surface roughness of ZrO₂/PEG hybrids increases with increasing PEG content, which promotes cell adhesion and proliferation, thus improving biocompatibility. However, increasing PEG concentration reduces antibacterial activity, as PEG has no intrinsic antibacterial activity and dilutes the effect of ZrO₂. Therefore, there is an inverse correlation between surface roughness and antibacterial activity, while surface roughness is positive for biocompatibility. The most salient results of the morphological characterization of ZrO₂/PEG hybrids are reported below:

• Atomic force microscopy (AFM) revealed an increase in porosity and surface roughness with increasing PEG content. As PEG concentration rises, it disrupts the Zr–O–Zr bridging network, leading to a more complex and rougher surface morphology. This increased surface roughness is indicative of the morphological changes that occur as the organic phase (PEG) interacts with the inorganic phase (ZrO₂).
• FT-IR spectroscopy confirmed the interaction between the inorganic and organic phases. In particular, FT-IR deconvolution in the 850–550 cm⁻¹ region indicates that PEG incorporation progressively perturbs the Zr–O–Zr network. As PEG content increases, the inorganic framework becomes more disordered and develops more flexible domains, suggesting reduced crosslinking within the zirconia matrix.
• Biocompatibility increased with increasing PEG content, reaching 92% cell viability for the 70 wt% ZrO₂/PEG hybrid. This improvement is attributed to the low immunogenicity and toxicity of PEG, which promotes cell adhesion and migration. In fact, biological assays reveal that higher PEG amounts enhance the viability of NIH-3T3 fibroblasts, likely due to increased hydrophilicity and reduced surface rigidity.
• The antibacterial activity of ZrO₂ decreased with increasing PEG content, highlighting the lack of intrinsic PEG activity, with pure ZrO₂ exhibiting the strongest inhibitory effect against E. coli. This trend suggests that PEG softens or shields the inorganic surface, diminishing its bactericidal efficiency.
• The PEG content significantly affects the surface roughness of ZrO₂/PEG hybrids, with an increase in roughness (Ra and Rq) proportional to the amount of PEG. For example, the roughness Ra increases from 105 nm for pure ZrO₂ to 645 nm for the 70 wt% ZrO₂/PEG hybrid. This increase in roughness facilitates cell adhesion and proliferation, improving the material’s biocompatibility.
• AFM reveals that PEG in zirconia blends functions as a structure-directing agent, enhancing surface roughness through its role as a pore-former, and facilitating the creations of organic-inorganic hybrids.
• This work stands out for its in-depth analysis, which combines morphological observations with biological performance assessments, offering new insights into the field of biomaterials. Therefore, the research significantly contributes to the understanding of the interactions between structure and functionality in these hybrid systems.