Synergistic antioxidant and anti-inflammatory scaffold based on ascorbic acid-enriched hydroxyapatite-PEG-ZnO composite from natural bone waste Authors: Ananthika Vijayan, Aswin Gopakumar, Sobha V. Nair, Balakrishnan Shankar, Gurunathan Saravana Kumar, Sreedha Sambhudevan Published: 2026 ABSTRACT Hydroxyapatite (HAp) derived from bovine bone waste was used to develop a bioactive composite scaffold incorporating polyethylene glycol (PEG), ZnO nanoparticles, and ascorbic acid (AA). This study aimed to fabricate a PEG-based HAp composite enriched with ZnO and AA and to evaluate the influence of these additives on the structural characteristics and biological performance of the scaffold. The resulting HAp–PEG–ZnO–AA composite was designed to provide a sustained release of Zn²⁺ ions while leveraging the antioxidant functionality of AA to enhance biological activity. X-ray diffraction (XRD) analysis indicated that PEG incorporation altered the crystallinity and grain size of HAp, while high-resolution transmission electron microscopy (HRTEM) and X-ray photoelectron spectroscopy (XPS) confirmed the successful integration of ZnO nanoparticles and AA within the HAp matrix. The composite exhibited enhanced antioxidant and anti-inflammatory activity compared to the HAp–PEG–ZnO scaffold without AA, whereas antibacterial performance was primarily attributed to the ZnO component. In vitro cytotoxicity assays using RAW 264.7 macrophage cells demonstrated that the composite was biocompatible at lower concentrations (cell viability > 90% at 6.25 µg mL⁻¹), while higher concentrations resulted in a marked reduction in viability accompanied by observable morphological changes. Overall, these findings suggest that AA incorporation synergistically improves the antioxidant and anti-inflammatory properties of the HAp–PEG–ZnO scaffold without compromising its antimicrobial activity or its biocompatibility at low dosages.Graphical Abstract Introduction Converting biological waste into high-value materials is an important strategy for sustainable biomaterial production [1]. One such material is hydroxyapatite (HAp), a calcium phosphate that can be extracted from animal bones and teeth, which serve as abundant natural sources [2]. Natural HAp differs from stoichiometric synthetic HAp by containing trace elements (e.g., Na, K, Mg, Ba, Zn, F) within its crystal lattice [3]. These substitutions make naturally derived HAp closely resemble the inorganic composition of real bone tissue. Consequently, bone waste (especially bovine and porcine bones) has been widely used to produce HAp for medical applications [4–6]. Extracted HAp from such sources is valued for its high crystallinity and biocompatibility and finds use in dental implants and bone grafts [7] HAp is well-known for its excellent osteoconductivity, bioactivity, and biocompatibility, allowing it to bond with host bone without eliciting toxicity or a severe immune response [8]. However, pure HAp is intrinsically brittle and has low tensile strength, which limits its use in load-bearing roles [9]. To overcome these mechanical limitations, HAp is often combined with polymers to form composite scaffolds that better mimic the natural bone matrix [10]. Polyethelene glycol (PEG), a hydrophilic and flexible polymer, is frequently incorporated with HAp to improve the composite’s toughness and processability [11]. PEG’s biocompatibility and chain flexibility enhance the scaffold’s structural integrity and support cell attachment and proliferation. By increasing the polymer content, the composite can achieve greater porosity and water uptake, benefiting nutrient transport and cell infiltration [12]. Thus, HAp-PEG hybrids exhibit better mechanical resilience and cytocompatibility than HAp alone, making them attractive in bone tissue engineering [13] The inclusion of functional NPs further expands the capabilities of HAp-based biomaterials. Zinc is a necessary trace element for bone production, and ZnO gives the composite its antibacterial properties. ZnO NPs are especially intriguing. ZnO is a bio-safe ceramic (classified by the U.S. FDA as “Generally Recognized as Safe”, GRAS) that is already used in food and biomedical applications. It should be noted that GRAS status pertains to specific uses and concentrations; ZnO is generally safe at moderate doses but can be toxic at high levels, so controlled loading is important [14]. Importantly, ZnO NPs can stimulate osteogenic cell growth and angiogenesis while conferring strong antibacterial properties [15]. ZnO’s antibacterial properties are mostly due to several different processes. For example, it makes reactive oxygen species (ROS) that damage parts of microbial cells, it disrupts bacterial membranes through electrostatic interactions, and it releases Zn2+ ions that interfere with microbial metabolic processes [16]. Zinc oxide is known to act against a wide range of microbes, and this activity becomes more prominent when ZnO is present in the nanoscale form because the surface area increases the number of sites available for microbial interaction. When ZnO is incorporated into an HAp–PEG system, earlier studies have reported better antibacterial performance along with enhanced protein adsorption, suggesting that the components work together rather than individually [17]. Another practical advantage is that ZnO is recognized as GRAS and, when used in controlled amounts, is considered safe for biomedical use. For implant materials, antimicrobial properties alone are not sufficient; features that support healing, such as antioxidant and anti-inflammatory behaviour, are also important [18]. Ascorbic acid (AA, vitamin C) is a familiar antioxidant in biology and is essential for collagen production and for managing oxidative stress [19]. Adequate levels of AA can help accelerate the repair of bone tissue, improve collagen matrix formation, and contribute to reducing inflammation [20]. In composite biomaterials, the addition of AA may therefore create a more supportive environment for cells by supplying local antioxidant protection and promoting extracellular matrix formation. AA is also known to assist osteogenic differentiation [21]. One limitation is that AA is highly unstable in its free form and easily undergoes oxidation [22]. This instability can be partially mitigated when AA is combined with polymers such as PEG. PEG can interact with AA through hydrogen bonding, creating a stabilising microenvironment that slows, but does not fully prevent aerobic oxidation of ascorbate. The degree of stabilization depends on the processing conditions, including temperature and exposure time, and should be verified experimentally for each formulation. HAp also contains surface hydroxyl groups that can bind AA and help retain it within the scaffold. Through these combined interactions, a HAp–PEG–ZnO–AA system can potentially release both Zn²⁺ ions and ascorbate in a controlled manner. This kind of material could offer antibacterial, antioxidant, and anti-inflammatory benefits all at once. This renders it highly advantageous for scenarios requiring simultaneous infection control and tissue restoration. Figure 1. Conceptual schematic illustrating the evolution from conventional HAp–PEG composites to the multifunctional HAp–PEG–ZnO–ascorbic acid (AA) scaffold developed in this work. While state-of-the-art HAp–PEG systems provide biocompatibility and osteoconductivity, they remain largely passive, lacking intrinsic antioxidant, anti-inflammatory, and antibacterial functionality. The incorporation of ZnO and AA addresses these gaps by creating a synergistic bioactive microenvironment suitable for bone and dental applications. This study presents the synthesis and characterisation of a new HAp-PEG-ZnO-AA composite scaffold produced from natural bovine bone debris. The objective is to examine the impact of including ZnO NPs and AA on the structural characteristics and biological efficacy of the HAp-PEG composite. Used a variety of methods (FTIR, XRD, XPS, and TEM) to verify correct component integration and analyse changes in crystallinity and chemical structure. Analyse the bioactivity of the composite using in vitro antibacterial tests against Gram-positive Staphylococcus aureus and Gram-negative Escherichia coli; anti-inflammatory tests using protein denaturation inhibition; antioxidant tests using DPPH free radical scavenging; and cytocompatibility tests using the MTT cell viability assay. By contrasting the composite with AA (S2) and without AA (S1), clarified the function of AA in augmenting the scaffold’s efficacy. The primary objective is to create a durable, multifunctional bone scaffold material that mitigates post-implantation issues such as infection, inflammation, and oxidative stress, hence enhancing the potential for bone tissue regeneration and healing.Fig. 1Schematic illustration highlighting the limitations of conventional HAp–PEG composites [13, 46] and the design strategy adopted in this work. While state-of-the-art HAp–PEG systems offer biocompatibility and osteoconductivity, they remain largely passive and lack intrinsic antioxidant, anti-inflammatory, and antibacterial functionality. The incorporation of ZnO and ascorbic acid (AA) into the HAp–PEG matrix results in a multifunctional bioactive scaffold capable of modulating oxidative stress, inflammation, and bacterial activity, while maintaining osteoconductivity Schematic illustration highlighting the limitations of conventional HAp–PEG composites [13, 46] and the design strategy adopted in this work. While state-of-the-art HAp–PEG systems offer biocompatibility and osteoconductivity, they remain largely passive and lack intrinsic antioxidant, anti-inflammatory, and antibacterial functionality. The incorporation of ZnO and ascorbic acid (AA) into the HAp–PEG matrix results in a multifunctional bioactive scaffold capable of modulating oxidative stress, inflammation, and bacterial activity, while maintaining osteoconductivity Experimental section Materials Bovine bone fragments were collected from a local butchery (8.8932° N, 76.6141° E, Kollam, India). All reagents were analytical grade and used as received without further purification. The chemicals used included sodium hydroxide (NaOH, 98%, Sigma-Aldrich), phosphoric acid (H3PO4, 99.999%, Sigma-Aldrich), ammonium hydroxide (NH4OH, 99.999%, Sigma-Aldrich), polyethylene glycol (PEG-6000, Carbowax 6000, SRL India), ZnO powder (99.9%, Techinstro, India), and L-ascorbic acid (AA, ≥99%, Nice Chem, India). Deionized water was used throughout for washing and solution preparation. Preparation of HAp NPs from bovine bone Hydroxyapatite was isolated from bovine bone by a multi-step heat and chemical treatment method. Bone fragments were initially boiled in water for approximately 24 h to eliminate soft tissues and organic material. Meticulously removed the residual tissue and thoroughly sanitized the bones. The bones were subsequently desiccated in an oven at 100 °C for 48 h. Throughout this period, the rendered lipids extracted from the bones were periodically drained to enhance the purity of the bone mineral. Immersed fragments of desiccated bone in a 0.1 M NaOH solution for 24 h to decompose any residual organic material. Subsequently, cleaned them with deionized water to eliminate the NaOH and any released contaminants. The initially treated bones were placed in a furnace at 900 °C for 6 h. This phase, conducted at elevated temperatures, incinerated the remaining organic material and initiated the transformation of bone mineral into crystalline HAp. A high-velocity ball mill (PM 100, Retsch, Germany) converted the calcined material (brittle white particles) into a fine powder. The bone-derived powder was further mixed with aqueous phosphoric acid in a stoichiometric calcium-to-phosphorus ratio of roughly 1.67 to guarantee the composition aligns with that of pure HAp. The meticulous incorporation of H3PO4 restored any lost or absent phosphate (PO43–) and converted any CaO or other calcium phases into HAp. The slurry was neutralized with dilute NH4OH to achieve a pH of approximately 7 and allowed to stand at room temperature overnight to complete the crystallization of HAp. Purified the solid, rinsed it with deionized water, and then dehydrated it at 100 °C. The synthesized HAp NPs were subjected to further calcination at 900 °C for four h to enhance their purity and crystallinity. The outcome was a high-quality white hydroxyapatite powder derived from bovine bone. Preparation of PEG-HAp-ZnO-AA composite The PEG–HAp–ZnO–AA composite scaffold (S2) was synthesized by a sequential mixing and drying technique. PEG-6000 (0.45 g) and HAp powder (1 g) were combined with a little quantity of deionized water and agitated at 60 °C for 3 h (400 rpm) to achieve a homogenous PEG coating on HAp. ZnO powder (80 mg), pre-dispersed in a minimal volume of water for 15 min, was thereafter added and agitated for an additional h to attain homogeneous dispersion. L-ascorbic acid (AA, 8 mg; 0.8 wt% relative to HAp) was ultimately integrated and agitated for 1 h at 60 °C. The slurry was dried at 100 °C for 24 h, and the resultant solid was meticulously crushed to produce S2 powder. A control composite devoid of AA (S1) was created following the same technique, excluding the final AA addition. S2 comprises approximately 65 wt% HAp, 29 wt% PEG, 5 wt% ZnO, and 0.5 wt% AA, whereas S1 has roughly 65 wt% HAp, 30 wt% PEG, and 5 wt% ZnO. According to these fractions, a 1 milligram S2 sample has approximately 0.05 mg ZnO and 0.005 mg AA, while a 100 μg mL⁻¹ dispersion contains roughly 5 μg mL⁻¹ ZnO and 0.5 μg mL⁻¹ AA. Notwithstanding the minimal AA content, S2 demonstrates considerable antioxidant and anti-inflammatory properties, underscoring the efficacy of AA integration. Porous scaffolds were prepared using a freeze-drying approach. The composite powder was first suspended in deionized water to obtain a slurry containing 5–10 wt% solids and then cast into suitable moulds. The samples were frozen at −80 °C for 12 h and subsequently subjected to lyophilization for 24–48 h to remove ice crystals and generate a highly porous, sponge-like structure. When required, the scaffolds were gently heat-treated at 50 °C to improve structural stability through PEG network consolidation. Physical characterization The bovine-derived HAp was characterised using Fourier-transform infrared spectra (FTIR) (A PerkinElmer Spectrum Two), zeta potential was calculated by using Zetasizer Nano ZS, transmission electron microscope, TEM (JEOL JEM-F200), Inductively Coupled Plasma Mass Spectrometry (ICP-MS) by Agilent 7900 ICP-MS. The following methods were used to examine the synthesised HAp composite samples’ structural, chemical, and morphological properties: A PerkinElmer Spectrum Two FTIR spectrometer was used to record FTIR in the 4000–400 cm−1 range. After preparing each sample as a KBr pellet, 16 scans with a resolution of 4 cm−1 were averaged. Functional groups found in HAp, PEG, and the composites were identified using the spectra. The chemical composition of the materials was verified by analysing significant vibrational bands associated with phosphate, hydroxyl, and organic groups. XPS was conducted using a PHI VersaProbe II instrument (ULVAC-PHI, USA) with a monochromatic Al Kα X-ray source (1486.6 eV) to examine the elemental composition and chemical states of the surface. High-resolution spectra for C 1 s, O 1 s, Zn 2p, Ca 2p, and P 2p were acquired at a pass energy of approximately 47 eV, whereas survey scans (0-1100 eV binding energy) were conducted at a pass energy of around 187 eV. The binding energy scale was calibrated using the C 1 s peak of adventitious carbon at 284.8 eV. Peak fitting and quantitative analysis were conducted using PHI MultiPak software. The morphology and nanostructure were observed using a JEOL JEM-F200 TEM operated at 200 kV. For TEM imaging, samples (HAp and composites) were ultrasonically dispersed in ethanol, and a drop of the suspension was deposited onto a C-coated Cu grid. Bright-field TEM and high-resolution TEM images were acquired to examine particle size and dispersion. Selected-area electron diffraction (SAED) patterns were obtained to assess the crystallinity and phase of the nanocrystals in the composites. An energy-dispersive X-ray spectroscopy (EDX) detector attached to the TEM was used for elemental analysis to confirm the presence and distribution of Ca, P, Zn, O, and C in the samples. Powder XRD patterns were collected on a Malvern PANalytical Empyrean 3 diffractometer using Cu Kα radiation (λ = 1.5406 Å) with a Ni filter. Data were recorded over a 2θ range of 10°-90° (step size ~0.02°) at room temperature. Diffraction peaks were indexed to identify crystalline phases. The Debye-Scherrer equation was used to determine the average crystallite size of HAp from peak broadening. The XRD analysis provides information on phase identity (HAp crystallinity and any secondary phases) of the bone-derived powder and any changes in peak intensity or width upon forming the composites. In vitro antibacterial assay The antibacterial efficacy of the samples was evaluated against two representative pathogens: Staphylococcus aureus (S. aureus, Gram-positive) and Escherichia coli (E. coli, Gram-negative). An agar well diffusion method was employed on nutrient agar plates. Each test sample (S1, S2, and a reference ZnO powder control) was sterilized by heating at 110 °C for 1 h prior to testing. (This brief heat treatment ensures the powders are sterile; however, it may cause some oxidation of AA in S2. The potential impact of this step on results is considered in the Discussion.) Bacterial inocula were prepared in saline and adjusted to 0.5 McFarland turbidity ( ~ 1-2 × 108 CFU mL−1). Using a sterile cotton swab, the surface of agar plates was uniformly swabbed with each bacterial suspension to create a lawn of bacteria. Wells ( ~ 6 mm diameter) were punched into the agar and filled with 50 μL of a sterile aqueous suspension of each sample (the suspension contained 1 mg of material per well). A control well (filled with sterile water, no sample) was also included. The plates were incubated at 37 °C for 24 h. Antibacterial activity was assessed by measuring the diameter of the zone of inhibition (clear area around the well with no bacterial growth) in mm. Each test was performed in triplicate (n = 3 independent plates), and the average inhibition zone diameters (mean ± standard deviation) were calculated for each sample against each bacterium. Anti-inflammatory activity (Protein denaturation inhibition) The ability of the samples to inhibit protein denaturation (as a model of anti-inflammatory efficacy) was tested using bovine serum albumin (BSA) as the target protein. Reaction mixtures containing 0.4 mL of 3% w/v BSA solution and 0.5 mL of sample solution at varying concentrations (6.25, 12.5, 25, 50, and 100 μg mL−1 in phosphate-buffered saline, PBS) were prepared in microcentrifuge tubes. As a positive control, the anti-inflammatory drug diclofenac was tested at the same concentrations in parallel. The mixtures were incubated at 37 °C for 20 min, then heated at 80 °C for 10 min to induce protein denaturation. After cooling to room temperature, 2.5 mL of PBS (pH 6.3) was added to each tube. The turbidity of the solutions (caused by precipitated, denatured protein) was measured by absorbance at 660 nm. The percentage inhibition of protein denaturation for each sample was calculated using Eq. (1):1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${Inhibition}\, \% =\frac{{A}_{{blank}}-{A}_{{sample}}}{{A}_{{blank}}}\times 100$$\end{document}Inhibition%=Ablank−AsampleAblank×100 In this formula, Ablank is the absorbance of the heated BSA solution without any test sample (indicating maximal protein denaturation), and Asample is the absorbance in the presence of the test sample (with potentially reduced denaturation due to protective effects). A higher percentage inhibition corresponds to better stabilization of the protein (i.e., stronger anti-inflammatory activity). From the plot of % inhibition % vs. concentration, the IC50 (the concentration at which 50% inhibition of denaturation is observed) was determined for each sample and compared to that of diclofenac. DPPH radical scavenging assay (antioxidant activity) The antioxidant capacity of the composites was quantified using the DPPH free radical scavenging assay. DPPH (2,2-diphenyl-1-picrylhydrazyl) is a stable free radical with a deep purple colour (absorption maximum at 517 nm) which becomes yellow when reduced by an antioxidant. A 0.1 mM DPPH solution in methanol was prepared fresh as the radical source. In a series of test tubes, 1.48 mL of the DPPH solution was mixed with 20 μL of sample solution at various concentrations (12.5, 25, 50, 100, and 200 μg mL−1 in DMSO). AA solutions (12.5–200 μg mL−1) were used as positive controls (standard antioxidant), and a blank control contained DPPH solution with 20 μL of pure DMSO. The mixtures were incubated in the dark at room temperature for 20 min to allow the reaction to proceed. After incubation, the absorbance was recorded at 517 nm using a Shimadzu UV-1900i UV-Vis spectrophotometer. The percentage of DPPH radicals scavenged by each sample was calculated using the same equation (Eq. 1) described above (substituting absorbance at 517 nm of a DPPH-only control for Ablank, and absorbance in the presence of the sample for Asample). A higher percentage of DPPH inhibition indicates stronger antioxidant activity. The results were plotted as % scavenging vs. concentration for S1, S2, and pure AA to compare their antioxidant performances. Cytotoxicity assay (MTT assay) The in vitro cytotoxic effects of the composites were assessed using a murine macrophage cell line (RAW 264.7) and the MTT assay (a colorimetric cell viability test based on the reduction of MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] by metabolically active cells to a purple formazan product). RAW 264.7 cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum and penicillin/streptomycin antibiotics. For the assay, cells were seeded into 96-well plates at a density of 5 × 103 cells per well and allowed to attach for 24 h (37 °C, 5% CO₂). Stock dispersions of S1 and S2 were prepared by dispersing 1 mg of each sample in 1 mL of complete cell culture medium, followed by sterile filtration to remove any microbes. These stock suspensions (1 mg mL−1) were then serially diluted with culture medium to achieve final treatment concentrations of 6.25, 12.5, 25, 50, and 100 μg mL−1. After the cells had attached for 24 h, the culture medium in each well was replaced with medium containing the sample at the desired concentration (each concentration in triplicate wells). Untreated control wells received fresh medium without any sample. The cells were incubated with the samples for 24 h. After 24 h exposure, 30 μL of MTT solution (5 mg mL−1 in PBS) was added to each well, and incubation was continued for 4 h. Viable cells reduce MTT to insoluble purple formazan crystals, which remain in the wells. Subsequently, the supernatant was gently removed, and 100 μL of DMSO was added to each well to dissolve the formazan crystals. The absorbance of the resulting solution in each well was measured at 540 nm using a microplate reader. Cell viability (as a percentage of the untreated control) was calculated according to Eq. (2):2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${Viability}\, \% =\frac{{A}_{{sample}}}{{A}_{{Control}}}\times 100$$\end{document}Viability%=AsampleAControl×100 Here, Asample is the 540 nm absorbance of cells treated with the composite sample (after 24 h), and Acontrol is the absorbance of cells in the control wells (no sample, which represents 100% viability). Thus, if a sample has no cytotoxic effect, its treated absorbance will equal the control (100% viability), whereas lower percentages indicate a reduction in viable cell number due to the treatment. In addition to the quantitative readings, cell morphology in each well was observed under an inverted phase-contrast microscope to qualitatively assess any changes (such as cell rounding, shrinkage, or detachment) caused by the treatments. Results Structural and chemical analysis of bovine-derived HAp The zeta potential profile of the hydroxyapatite particles displays a narrow distribution with a slightly negative surface charge, as shown in Fig. 2(a). This behaviour can be attributed to the dominance of phosphate groups on the particle surface when dispersed in aqueous media. The negative surface charge (–11 eV) indicates good colloidal stability and is known to support favourable interactions with proteins and cells. The observed zeta potential values are consistent with those reported for biogenic hydroxyapatite and are suitable for biomedical applications, particularly in bone-related environments [23].Fig. 2a Zeta potential profile showing a narrow distribution with a slightly negative surface charge, b FTIR confirming the presence of characteristic functional groups associated with apatite formation c XRD peaks revealing the diffraction peaks corresponding to HAp particles (d1, d2) showing a nano-sized agglomerated cluster of HAp a Zeta potential profile showing a narrow distribution with a slightly negative surface charge, b FTIR confirming the presence of characteristic functional groups associated with apatite formation c XRD peaks revealing the diffraction peaks corresponding to HAp particles (d1, d2) showing a nano-sized agglomerated cluster of HAp FTIR, as shown in Fig. 2(b) of the bovine-derived hydroxyapatite, confirms the presence of characteristic functional groups associated with apatite formation. Distinct absorption bands observed in the region of approximately 560–605 cm⁻¹ and 960–1100 cm⁻¹ are attributed to the bending and stretching vibrations of phosphate (PO₄³⁻) groups, respectively [24]. The appearance of bands near 3570 cm⁻¹ and around 630 cm⁻¹ corresponds to the stretching and librational vibrations of hydroxyl (OH⁻) ions, indicating that the apatite crystal lattice is well preserved. Additionally, weak carbonate-related bands suggest partial carbonate incorporation, a feature commonly observed in naturally derived hydroxyapatite and considered beneficial for mimicking bone mineral composition. X-ray diffraction analysis reveals diffraction peaks that correspond well with the standard hexagonal phase of hydroxyapatite (Fig. 2(c)). The major reflections, including those indexed to the (002), (211), (300), (202), and (310) planes, confirm the successful formation of crystalline HAp without detectable secondary calcium phosphate phases [25]. The broadened nature of the diffraction peaks indicates the presence of nanoscale crystallites. Such nano-crystallinity closely resembles the mineral phase of natural bone and is advantageous for enhanced bioactivity and osteoconductive behaviour. Transmission electron microscopy images show that the hydroxyapatite consists of nanosized particles forming clustered aggregates, as indicated in Fig. 2d1, d2). The particles exhibit irregular morphologies and tend to agglomerate, which is typical of hydroxyapatite derived from biological sources [26]. The nanoscale dimensions and aggregated structure contribute to an increased surface area, which can promote biological interactions and surface functionalization. Inductively coupled plasma mass spectrometry (ICP–MS) was employed, as shown in Table 1, to determine the elemental composition of the bovine-derived HAp [27]. Calcium and phosphorus were identified as the major constituents, confirming the apatite-based composition of the material. Based on the measured elemental concentrations, the calculated Ca/P ratio was approximately 1.67, which closely matches the theoretical value for stoichiometric HAp and confirms the formation of a chemically balanced apatite phase [28]. In addition to Ca and P, a small amount of magnesium was detected, reflecting the natural trace-element composition of bone-derived apatite. Trace levels of potassium, zinc, and barium were also observed in parts-per-million concentrations, all of which were within acceptable limits. Subsequently, the presence of these minor and trace elements further supports the biomimetic nature of the synthesized hydroxyapatite and its suitability for biomedical applications.Table 1Inductively coupled plasma mass spectrometry24Mg31 P40Ca39 K66Zn138BaHAp0.6237.4562.5439.5137.930.57Unit%%%ppmppmppmDetection limit in ppm1.001.001.000.100.010.01 Table 1Inductively coupled plasma mass spectrometry24Mg31 P40Ca39 K66Zn138BaHAp0.6237.4562.5439.5137.930.57Unit%%%ppmppmppmDetection limit in ppm1.001.001.000.100.010.01 Inductively coupled plasma mass spectrometry Structural and chemical analysis of PEG-HAp-ZnO and PEG-HAp-ZnO-AA composite FTIR analysis The FTIR spectra of the PEG-HAp-ZnO composite without AA (S1) and with AA (S2) are shown in Fig. 3. Both spectra exhibit the characteristic bands of HAp, along with distinct features from PEG, and S2 exhibits the characteristic spectra for AA. A broad absorption in the 3200-3600 cm−1 region (centred around ~3420 cm−1) corresponds to O–H stretching vibrations [29]. This band is notably broadened in the composites, indicating extensive hydrogen bonding within the HAp-PEG-AA matrix. The intensity of this broad O–H band is slightly higher in S2, which can be attributed to the contributions from the –OH groups of AA in addition to the surface –OH groups of HAp and PEG. A distinct peak at 2882 cm−1 is observed in both S1 and S2, assigned to the aliphatic C–H stretching of the PEG chains (–CH2 vibrations). This peak confirms the presence of PEG in the composite, and the same peak can also be seen in S2, ensuring that the addition of AA has not significantly shifted or attenuated the PEG C–H vibrational frequency.Fig. 3FTIR spectra of the composite without AA (S1) and with AA (S2), showing the broad O–H band ( ~ 3400 cm–1), C–H stretch (2882 cm–1), phosphate vibrations of HAp ( ~ 1040, 960, 600, 560 cm–1), and other characteristic peaks FTIR spectra of the composite without AA (S1) and with AA (S2), showing the broad O–H band ( ~ 3400 cm–1), C–H stretch (2882 cm–1), phosphate vibrations of HAp ( ~ 1040, 960, 600, 560 cm–1), and other characteristic peaks In the fingerprint region, strong phosphate (PO43–) -related bands from HAp dominate both spectra. Broad bands in the ~1030-1090 cm−1 range (with a peak around 1045 cm−1) are attributed to P–O stretching modes (ν3) of the PO43– groups in HAp. Both S1 and S2 show a pronounced band near 960 cm−1 (the ν1 symmetric P–O stretch of PO43–) and appear as a shoulder around ~939 cm−1, probably broadened by overlapping contributions from other vibrations. The characteristic PO43– bending modes (ν4) appear as a doublet at ~560 and 600 cm−1, confirming the presence of the apatite phase. Additionally, a small peak at ~630 cm−1 (visible as a shoulder) can be assigned to the liberational mode of structural OH⁻ in HAp. The low-frequency region ( ~ 430-450 cm−1) shows bands that can be attributed to lattice vibrations involving PO43– and possibly Zn–O stretching from ZnO. The FTIR evidence thus confirms that the bone-derived powder retains the HAp phase in both the composites. A peak at ~1615 cm−1 is more clearly observed in S2, which can be assigned to the C=O stretching of carboxylate or the ring vibration of AA (or possibly H–O–H bending of absorbed water). The presence of this band in S2 suggests contributions from AA (ascorbate has characteristic bands in this region). Additionally, S2 exhibits slightly enhanced intensity in the region around 1050-1150 cm−1, which is due to C–O stretching vibrations from AA overlapping with the phosphate bands. These differences between S1 and S2, although subtle, indicate that AA has been successfully incorporated into the composite and contributes its functional groups to the FTIR spectrum. No obvious new bands or shifts aside from those attributable to AA were observed, implying that AA is physically adsorbed or hydrogen-bonded in the composite rather than forming a new crystalline phase. The lack of any distinct new peaks also suggests that the interactions among HAp, PEG, and AA do not produce secondary chemical compounds detectable by FTIR, but rather involve intermolecular bonding (hydrogen bonding, coordination of AA to HAp surface Ca2+, etc.). XPS analysis XPS was utilized to probe the surface elemental composition and chemical states in S1 and S2. The survey spectra (Fig. 4a) for both samples confirmed the presence of Ca, P, O, C, and Zn as expected. The absence of foreign elements and contaminants ensures the purity of the composites and the effectiveness of the synthesis method in removing impurities. The survey scan also showed that the surfaces of both composites are rich in carbon and oxygen (due to PEG and AA), with Ca and P signals from HAp and a Zn signal from ZnO. Notably, the Ca and P peaks in the survey spectra were relatively lower in intensity compared to C and O, which is reasonable given that the surface of the composite particles is coated with organics (PEG and AA) and ZnO NPs. The atomic concentration comparison of each element in S1 and S2 from the XPS has been compiled in Fig. 4b. The Ca 2p and P 2p high-resolution spectra (not shown in the figure for brevity) were consistent with stoichiometric hydroxyapatite: Ca 2p3/2 at ~347.1 eV and P 2p at ~133.3 eV, with Ca/P peak area ratios close to the expected values for HAp.Fig. 4XPS spectra of composites: a Survey spectra for S1 and S2 showing the presence of Ca, P, O, C, and Zn; b atomic concentration comparison between S1 and S2; High-resolution spectra of c C 1 s in S1, d C 1 s in S2, e O 1 s in S1, f O 1 s in S2, g Zn 2p in S1, and h Zn 2p in S2 XPS spectra of composites: a Survey spectra for S1 and S2 showing the presence of Ca, P, O, C, and Zn; b atomic concentration comparison between S1 and S2; High-resolution spectra of c C 1 s in S1, d C 1 s in S2, e O 1 s in S1, f O 1 s in S2, g Zn 2p in S1, and h Zn 2p in S2 Figure 4c to Fig. 4h shows the high-resolution spectra for C, O and Zn, respectively. Both S1 and S2 show a prominent peak in the C 1 s region (Fig. 4c, d) at about 284.8 eV, which is associated with C–C/C–H bonds (sp²/sp³ carbon, primarily from the PEG backbone and adventitious hydrocarbons). S1 and S2 also show a secondary C 1 s component at higher binding energy ( ~ 286.5 eV), attributable to C–O bonds (such as in C–O–C of PEG, and C–OH of AA). The presence of AA is compatible with the slightly more prominent C–O peak in S2 (vitamin C includes several hydroxyl groups, contributing extra C–O bonds). There is a little peak at about 288–289 eV that can be attributed to O–C=O (surface carbonate species or carbonyl carbons in AA). The slight increase of this peak in S2 suggests contributions from AA’s lactone carbonyl groups. As in Fig. 4e, f, the O 1 s high-resolution spectra of S1 and S2 also reflect differences due to AA. Both samples have a primary O 1 s peak near ~531 eV, which can be deconvoluted into multiple components. In S1, the O 1 s envelope includes contributions from O2− in phosphate (HAp), O in ZnO, and oxygen in PEG (ether oxygen) and surface hydroxyls. In S2, a shoulder or secondary component in the O 1 s spectrum at ~532-533 eV is more pronounced, which is attributed to oxygen in hydroxyl and carbonyl groups (O–H and C–O) introduced by AA. The higher intensity of this high-binding-energy O 1 s component in S2 confirms that the surface of S2 contains more oxygenated functional groups (from AA) than S1. The Zn 2p spectra for S1 and S2 show (Fig. 4g, h) the Zn 2p3/2 and Zn 2p1/2 doublet at ~1022 eV and ~1045 eV, respectively, with a spin-orbit splitting of ~23 eV, matching well with ZnO (Zn2+ in the wurtzite ZnO lattice). Importantly, the Zn 2p peak positions and shapes are nearly identical for S1 and S2, indicating that Zn remains in the ZnO state in both composites, and the presence of AA did not change the oxidation state of Zn or cause any additional Zn compounds to form. The XPS results confirm the FTIR findings that S2 possesses a greater abundance of oxygen-rich functional groups than S1, attributable to the presence of AA. The XPS study of surface atomic concentration indicated that both composites display significant organic content on the surface, in accordance with the PEG coating, while the Zn atomic percentage was low yet noticeable. The measured Zn surface atomic concentration was on the order of a few atomic percent (roughly 5-6 at% in both S1 and S2), consistent with the intended doping level ( ~ 5 wt% ZnO corresponds to a few at% Zn). Also noted that XPS is surface-sensitive and semi-quantitative. Nonetheless, the fact that Zn was observed and in the expected range confirms that ZnO was successfully integrated. The presence of Zn, Ca, P, C, and O on the surface demonstrates that PEG and AA did not displace ZnO, and the composite-ZnO NPs remain present alongside the HAp and organics. These XPS observations, together with the FTIR data, demonstrate that the composite’s surface is a hybrid of inorganic (HAp, ZnO) and organic (PEG, AA) components, which is ideal for a multifunctional biomaterial. XRD analysis Powder XRD was used to examine the crystalline structure of the composites S1 (HAp-PEG-ZnO) and S2 (HAp-PEG-ZnO-AA). Both samples display the characteristic reflections of hexagonal HAp, with prominent peaks at approximately 25.9° [(002)], 31.9° [(211)/(100)], 32.9° [(300)], 34.2° [(002)], 39.5° [(310)], 46.9° [(102)], 49.6° [(213)], and 53.4° [(004)] (as shown in Fig. 5 and see Table S1 in SI for more details). The corresponding recalculated d-spacings via Bragg’s equation (e.g., 3.425 Å at 26°, 2.805 Å at 31.9°, and 1.720 Å at 53.4°) match well with standard HAp, [25] confirming that the bone-derived HAp in both composites retains its expected apatite structure after phosphoric acid treatment, calcination, and composite processing. No secondary calcium phosphate phases such as β-TCP or CaO were detected, indicating that the HAp present in both S1 and S2 is phase-pure.Fig. 5XRD patterns of composite samples: S1 (HAp–PEG–ZnO) and S2 (HAp–PEG–ZnO–AA). No distinct ZnO peaks are visible, suggesting ZnO NPs are very small and well-dispersed within the composite XRD patterns of composite samples: S1 (HAp–PEG–ZnO) and S2 (HAp–PEG–ZnO–AA). No distinct ZnO peaks are visible, suggesting ZnO NPs are very small and well-dispersed within the composite A wide, diffuse peak between 15° and 30° 2θ was observed due to amorphous organic constituents such as PEG in S1 and PEG and AA in S2. These amorphous organic elements form amorphous background characteristics and do not generate any crystalline reflections. An in-depth analysis of peak widths indicates disparity between the two composites. Contrary to conventional assumptions, S2 consistently displays wider HAp reflections than S1. Due to the inverse relationship between peak broadening and crystallite coherence length, Scherrer analysis was conducted utilizing the FWHM values (refer Table S1 of SI). The (002) peak broadens from 0.173° in S1 to 0.240° in S2, whereas the (211)/ (100) region at 31.9° expands from 0.205° to 0.235°, with analogous broadening noted at elevated angles, such as 49.6° (0.205° to 0.259°). Crystaline dimensions are recalibrated using Scherrer analysis, indicating that S1 possesses bigger hydroxyapatite coherent domains (often 40-48 nm), while S2 crystallites are smaller, typically measuring 34–36 nm. The peak positions and corresponding d-spacings remain invariant across the two samples, signifying that AA does not alter the HAp lattice or produce lattice strain, but rather influences primarily domain growth. The disparities noted between S1 and S2 are most effectively elucidated by the influence of AA on HAp crystallite development. AA is anticipated to interact with Ca2+-rich HAp surfaces during composite production via moderate chelation or surface adsorption, functioning as a capping or growth-modulating agent. This interaction may inhibit coalescence and constrain crystallite formation, resulting in smaller coherent domains in S2. Concurrently, ZnO remained invisible in both XRD patterns, despite its presence in the composites. The reflections of wurtzite ZnO at around 31.8°, 34.4°, and 36.3° substantially coincide with the principal peaks of hydroxyapatite (HAp) [30]. The modest ZnO loading ( ~ 5 wt%), nanoscale dispersion, and widening effects from PEG/AA encapsulation inhibit the emergence of discrete ZnO peaks. This aligns with XPS and EDX results confirming Zn presence (see Figs. 4 and S1), despite the lack of visible ZnO peaks in XRD. Overall, the XRD data confirm that both composites contain phase-pure HAp, with AA incorporation in S2 subtly reducing crystallite coherence without altering the fundamental lattice structure. HRTEM and SAED analysis HRTEM provided direct visualization of the composite nanostructures (Fig. 6). In bright-field TEM images (Fig. 6a, b), S1 shows HAp NPs as irregularly shaped crystallites (tens of nanometers in size) embedded in or coated by an amorphous matrix (PEG). The overall morphology of S1 appears somewhat aggregated: darker regions correspond to HAp nanocrystals, whereas lighter, cloudier regions indicate amorphous PEG surrounding the inorganic particles.Fig. 6TEM images and SAED patterns of composites: a, b Bright-field TEM of S1 showing HAp nanocrystals embedded in an amorphous matrix (PEG); c SAED of S1 displaying diffuse rings characteristic of randomly oriented nanocrystals and amorphous content; d, e TEM images of S2, reveal somewhat larger and defined crystalline features within the matrix; f SAED of S2 with sharper rings and faint spots, indicating improved crystallinity and partial orientation of crystals in S2. PEG encapsulation of ZnO and HAp results in no distinct ZnO particles visible in either sample, highlighting the uniform dispersion achieved TEM images and SAED patterns of composites: a, b Bright-field TEM of S1 showing HAp nanocrystals embedded in an amorphous matrix (PEG); c SAED of S1 displaying diffuse rings characteristic of randomly oriented nanocrystals and amorphous content; d, e TEM images of S2, reveal somewhat larger and defined crystalline features within the matrix; f SAED of S2 with sharper rings and faint spots, indicating improved crystallinity and partial orientation of crystals in S2. PEG encapsulation of ZnO and HAp results in no distinct ZnO particles visible in either sample, highlighting the uniform dispersion achieved In contrast, TEM images of S2 (Fig. 6d, e) reveal slightly larger and more defined crystalline particles within the matrix. The HAp crystals in S2 appear more discernible and, in some cases, larger than those in S1, which aligns with the XRD observation of larger HAp crystallites in S2. S2’s micrographs show better contrast between the inorganic particles and the organic phase, possibly because AA presence results in a somewhat less extensive amorphous coating or different dispersion. The AA might act as a mild dispersant or capping agent, preventing excessive agglomeration of HAp during composite formation, which could explain why the particles in S2 seem better separated or outlined compared to S1. HRTEM images of S2 show lattice fringes corresponding to crystalline HAp in some of the larger particles, confirming their crystalline nature. In S1, by comparison, lattice fringes are less clear or absent in many particles, suggesting that those particles are either very small HAp crystallites or consist of multiple misoriented nanocrystals (polycrystalline aggregate) and amorphous material. This again reflects S2’s improved crystallinity. Distinct lattice fringes or particles attributable to ZnO were not evident in the HRTEM images of either S1 or S2. Given the low ZnO content and the likelihood that ZnO exists as nanometer-sized grains intimately mixed with PEG and possibly attached to HAp. The ZnO NPs in the composite may be too small (e.g., 5-10 nm or less) and with insufficient contrast to stand out against the HAp and PEG background under HRTEM. Another possibility is that some Zn2+ ions from ZnO could be adsorbing on HAp surfaces or partially substituting into HAp lattice defect sites, which would further obscure distinct ZnO phase identification (though XPS/XRD suggest Zn is mostly ZnO, not substituted in HAp). For S1 (Fig. 6c), the SAED pattern shows diffuse rings, which are characteristics of randomly oriented nanocrystalline and amorphous materials. The diffuse nature of the rings in S1 suggests that the sample contains very fine HAp nanocrystals and a substantial amorphous fraction (PEG, and possibly poorly crystalline ZnO), such that long-range order is limited. In S2 (Fig. 6f), the SAED pattern exhibits sharper rings and even faint spots on some rings. The presence of sharper rings (and occasional spots) indicates that S2 has larger or more oriented crystalline domains than S1. The faint spots superimposed on the rings suggest that a subset of HAp crystals in S2 has grown large enough or aligned enough to produce spot-like diffraction, implying partial orientation or simply larger grain size. This observation is consistent with S2’s greater HAp crystallinity from XRD. In summary, SAED confirms that S1 is highly nanocrystalline/amorphous, whereas S2, while still polycrystalline, has improved crystallinity with some domains achieving greater order. These SAED differences align with the hypothesis that AA addition leads to more ordered HAp formation in S2. Figure 6 also illustrates that distinct ZnO diffraction rings are not obvious in either sample’s SAED. ZnO rings (if present) would overlap with some HAp rings (for instance, ZnO has a strong reflection at ~5.2 Å d-spacing corresponding to 2θ ~ 34°, near an HAp ring). The absence of separate ZnO rings in SAED is further evidence that ZnO is either highly dispersed and overshadowed by HAp or present in too low a quantity/crystallinity to contribute observable diffraction. This result, combined with HRTEM, suggests that ZnO is present as ultra-fine particles, likely encapsulated by PEG (and possibly intimately associated with HAp surfaces), which prevents them from aggregating or diffracting strongly. Elemental analysis by EDX (performed during TEM) further confirmed the composite composition (Figs. 7 and 4b). EDX spectra of S1 and S2 each show peaks for Ca, P, O (all from HAp), Zn (from ZnO), and C (from the organic components and the carbon support grid). No extraneous elements were detected, indicating the absence of significant impurities. The Ca/P ratio derived from EDX was roughly 1.6-1.7, close to the stoichiometric value of 1.67 for HAp, which confirms that our bone-derived HAp has an appropriate composition (within experimental uncertainty of EDX). The detection of Zn in both S1 and S2, even at low intensities, is important evidence that ZnO is present in the composites as expected. Quantitatively, the Zn signal intensity corresponded to ~5-6 atomic% of the sample, which, given the instrumental uncertainties, is in reasonable agreement with the ~5 wt% (equivalent to a few at%) added ZnO. Also emphasize that EDX provides only a semi-quantitative estimate (especially in a heterogeneous sample where matrix effects can influence X-ray yields), but the result ‘consistent with expected Zn content’ lends support to our synthesis ratio.Fig. 7EDX spectra of (left) S1 and (right) S2 obtained during TEM analysis. Both spectra show the presence of Ca, P, O (from HAp), Zn (from ZnO), and C (from the organic components and grid). The elemental analysis confirms the expected composite composition and uniform distribution of Zn in both samples EDX spectra of (left) S1 and (right) S2 obtained during TEM analysis. Both spectra show the presence of Ca, P, O (from HAp), Zn (from ZnO), and C (from the organic components and grid). The elemental analysis confirms the expected composite composition and uniform distribution of Zn in both samples Furthermore, EDX mapping (see Fig. S1, SI) showed that Zn is uniformly distributed across the observed areas of both S1 and S2. In the elemental maps, signals for Zn overlap with those of Ca and P rather than clustering separately, implying that ZnO NPs are well-dispersed throughout the HAp/PEG matrix in both samples. There was no evidence of Zn-rich regions or the loss of ZnO during composite formation; the Zn was found wherever HAp was present, suggesting intimate mixing. This confirms that the incorporation of PEG and AA did not cause ZnO to leach out or segregate; the NPs remain embedded and are found alongside HAp crystals in the composite. This uniform presence of Zn in different sampled regions supports the notion that each composite particle contains the full complement of components (HAp, PEG, ZnO, ±AA) rather than being a mixture of separate phase particles. In vitro antibacterial activity The antibacterial performance of the composites was evaluated by agar diffusion against E. coli and S. aureus, and the results are summarized in Fig. 8. Figure 8a, b shows photographs of the inhibition zones (clear rings around the wells) produced by ZnO, S1, and S2 for E. coli and S. aureus, respectively. All tested materials produced a distinct, clear zone of inhibited bacterial growth around the well, indicating antimicrobial activity in each case. The control wells (no sample) showed dense bacterial growth up to the well edge (no inhibition), confirming that the observed clear zones were due to the sample’s effect. The antibacterial properties of the samples were examined through the agar well diffusion technique, which provides a qualitative indication of bacterial inhibition.Fig. 8Antibacterial activity results: a Photograph of inhibition zones against E. coli for wells containing ZnO, S1, and S2; b inhibition zones against S. aureus for the same samples; c bar graph of the average inhibition zone diameters (mm) for each sample and organism. ZnO exhibits the largest zones (strongest antibacterial effect), while S1 and S2 show smaller but clearly present inhibition zones of similar size Antibacterial activity results: a Photograph of inhibition zones against E. coli for wells containing ZnO, S1, and S2; b inhibition zones against S. aureus for the same samples; c bar graph of the average inhibition zone diameters (mm) for each sample and organism. ZnO exhibits the largest zones (strongest antibacterial effect), while S1 and S2 show smaller but clearly present inhibition zones of similar size The neat ZnO powder, included as a positive control due to its well-established antibacterial properties, produced the largest inhibition zones for both bacterial strains. ZnO exhibited inhibition zones of approximately 17 mm against E. coli and about 19 mm against S. aureus. In contrast, the composites S1 and S2 generated noticeably smaller zones. For E. coli, S1 produced an average inhibition zone of about 9 mm, while S2 showed roughly 7 mm. Against S. aureus, both S1 and S2 produced comparable zones of around 8 mm. These results indicated that S1 and S2 exhibit similar antibacterial performance, with no statistically significant difference between them (p > 0.05). However, ZnO showed significantly greater antibacterial activity than either composite for both organisms (p < 0.01). Figure 8c summarizes these findings using mean ± SD. Anti-inflammatory activity (Protein denaturation inhibition) To assess anti-inflammatory potential, the ability of S1 and S2 to inhibit heat-induced albumin denaturation in vitro, compared with a standard anti-inflammatory drug (diclofenac), has been measured. Figure 9 shows the percentage inhibition of protein denaturation as a function of sample concentration (6.25–100 μg mL–1) for S1, S2, and diclofenac.Fig. 9Protein denaturation inhibition (anti-inflammatory assay) results for S1, S2, and Diclofenac (standard), across concentrations 6.25-100 µg mL–1. Both composites show a dose-dependent increase in inhibition of albumin denaturation. S2 consistently exhibits higher inhibition percentages than S1 at each concentration, approaching the performance of Diclofenac at certain doses (e.g., at 25 µg mL–1). The improved efficacy of S2 highlights the contribution of AA to anti-inflammatory potential. Error bars represent standard deviation (n = 3) Protein denaturation inhibition (anti-inflammatory assay) results for S1, S2, and Diclofenac (standard), across concentrations 6.25-100 µg mL–1. Both composites show a dose-dependent increase in inhibition of albumin denaturation. S2 consistently exhibits higher inhibition percentages than S1 at each concentration, approaching the performance of Diclofenac at certain doses (e.g., at 25 µg mL–1). The improved efficacy of S2 highlights the contribution of AA to anti-inflammatory potential. Error bars represent standard deviation (n = 3) The corresponding UV–Vis absorption spectra from the percentages are provided in Fig. S2, SI. Both composites exhibited a dose-dependent inhibition of protein denaturation. At the lowest concentration tested (6.25 μg mL–1), neither composite showed significant inhibition ( < 20%). However, as the concentration increased, the inhibition percentage increased for both S1 and S2. At the highest concentration (100 μg mL–1), S2 achieved a protein denaturation inhibition of roughly ~60%, whereas S1 reached ~50% inhibition. In the same assay, diclofenac (the positive control) showed ~85% inhibition at 100 μg mL–1. Thus, both S1 and S2 impart anti-inflammatory effects; S2 is more effective than S1, and neither is as potent as the pharmaceutical agent diclofenac at equivalent mass doses. Notably, at intermediate concentrations around 25 μg mL–1, S2’s performance was higher than that of diclofenac. S2 achieved nearly ~40% inhibition at 25 μg mL–1, which was slightly higher than the inhibition by diclofenac at the same concentration. In contrast, S1 at 25 μg mL–1 only caused ~35% inhibition. This suggests that at lower-to-moderate concentrations, S2 can be as effective as diclofenac in this protein stabilization model, highlighting a strong anti-inflammatory capability for a biomaterial. The fact that S2 outperforms S1 at these doses clearly highlights the role of AA in boosting anti-inflammatory activity. Across the entire tested range, S2 consistently showed higher inhibition percentages than S1 at each concentration (the two curves in Fig. 8 show S2 above S1 at all points). Error bars (SD, n = 3) indicate the variation was modest, and the differences between S2 and S1, especially at 12.5 μg mL–1 and above, were statistically significant (p < 0.05 by ANOVA/Tukey). Also determined IC50 values from these data: S2 had an IC50 of ~70 μg mL–1, whereas S1’s IC50 was ~95 μg mL–1 (diclofenac’s IC50 ~ 50 μg mL–1). A lower IC50 indicates higher potency; thus, S2 is markedly more potent than S1 in preventing protein denaturation (IC50 reduced by ~25 μg mL–1). Antioxidant activity (DPPH radical scavenging) The DPPH assay results (Fig. 10, with spectra in Fig. S3, SI) quantify the free radical scavenging capacity of S1 and S2, with pure AA tested as a reference standard. In this assay, a higher % DPPH scavenged corresponds to stronger antioxidant power. The tested composite concentrations up to 200 μg mL–1, and for comparison, equivalent concentrations of pure AA (12.5–200 μg mL–1) were also tested.Fig. 10DPPH radical scavenging activity of S1 and S2 at various concentrations, compared with the AA standard. S2 shows a higher % scavenging than S1 across all concentrations tested. Even though both composites require a high concentration (200µgmL-1) to exceed 40% scavenging, S2 reaches over 45%, whereas S1 remains around 40%. Pure AA (25 µg mL-1) by comparison achieves ~60% scavenging, indicating the current low loading of AA in S2. The enhancement in S2’s antioxidant capacity over S1 validates the inclusion of AA for oxidative stability DPPH radical scavenging activity of S1 and S2 at various concentrations, compared with the AA standard. S2 shows a higher % scavenging than S1 across all concentrations tested. Even though both composites require a high concentration (200µgmL-1) to exceed 40% scavenging, S2 reaches over 45%, whereas S1 remains around 40%. Pure AA (25 µg mL-1) by comparison achieves ~60% scavenging, indicating the current low loading of AA in S2. The enhancement in S2’s antioxidant capacity over S1 validates the inclusion of AA for oxidative stability Both S1 and S2 showed very low scavenging activity at the lowest concentrations (12.5 and 25 μg mL–1), typically <20% DPPH inhibition. As the concentration increased, radical scavenging improved for both samples. However, S2 demonstrated significantly higher antioxidant activity than S1 at all corresponding concentrations. For instance, at 100 μg mL–1, S1 scavenged only about 25–30% of DPPH, whereas S2 scavenged roughly 35-40%. At the highest concentration of 200 μg mL–1, S2 achieved around 45–50% DPPH inhibition, compared to ~40% for S1. These values are moderate in an absolute sense (neither composite fully quenched the DPPH radical at the tested doses), but the key point is the consistent enhancement in S2 relative to S1, demonstrating that AA improved antioxidant capability. To better interpret these findings, it is important to consider the actual amount of AA present in the S2 composite. Pure AA is a strong antioxidant; for instance, 50 μg mL⁻¹ of pure AA scavenged ~60% of DPPH as in Fig. 10. In contrast, even 200 μg mL⁻¹ of S2 achieved only ~45% scavenging. This difference is expected because S2 contains only a very small fraction of AA as low as ~0.5 wt% of S2 based on the composite formulation. Therefore, at an S2 concentration of 200 μg mL⁻¹, the actual amount of AA present is only about 1 μg mL⁻¹. This corresponds to approximately 0.5% of the AA mass used in a 200 μg mL⁻¹ pure AA sample. The markedly lower AA content explains why S2 exhibits significantly lower antioxidant activity than pure AA, yet still shows appreciable scavenging efficiency. Notably, despite its very low AA loading, S2 consistently outperformed S1 at all concentrations, confirming that the incorporated AA contributes measurably to the enhanced antioxidant activity of the composite. However, it is notable that S2 still achieved nearly 45-50% DPPH scavenging at 200 μg mL⁻¹. Although this is still far lower than the 50 μg mL⁻¹ of pure AA that produced ~55% scavenging in our control experiment, the fact that 1 μg mL⁻¹ of embedded AA supports ~50% scavenging is striking. Also see that S1 (with no AA) is not completely inert in the DPPH assay, reaching ~40% inhibition at 200 μg mL–1. This residual activity could be due to PEG at high concentration exhibiting a weak radical scavenging or interference effect (some studies suggest PEG can have a very mild antioxidant behaviour, possibly by virtue of its ether oxygens quenching radicals at high concentrations) [31]. Additionally, HAp itself might adsorb some DPPH or radicals [32]. Regardless, S1’s activity is modest and only appears at high concentration, whereas S2 shows an elevated curve across all concentrations. According to the UV-Vis data and Beer-Lambert’s law, all samples showed a roughly linear decrease in absorbance as concentration increased, indicating stronger inhibition at higher doses (Fig. S3, SI). Gradual decrease in absorbance at 517 nm for all samples as concentration increases, consistent with increasing DPPH radical scavenging. Across all concentrations, S2 exhibits slightly lower absorbance values than S1, indicating modestly higher scavenging efficiency, while the standard shows the greatest reduction in absorbance. At 100 µg mL⁻¹, the absorbance values are approximately 0.50 for S1, 0.46 for S2, and 0.20 for the standard. These values correspond to minimal inhibition for S1, marginally higher inhibition for S2, and substantial inhibition for the standard. Although the composites do not approach the activity of the standard, the consistently lower absorbance of S2 relative to S1 confirms that incorporation of AA enhances antioxidant performance. Overall, the dose-dependent decline in absorbance reflects increased quenching of DPPH radicals, with S2 performing better than S1 at all concentrations tested. Statistically, the differences between S1 and S2 were significant at all concentrations tested (p < 0.05), confirming that S2 possesses greater antioxidant capability. It requires the highest tested concentration (200 μg mL⁻¹) to exceed 50% DPPH inhibition, whereas pure AA achieved comparable scavenging at only ~20–25 μg mL⁻¹. These trends suggest that higher antioxidant performance could be achieved by increasing the AA loading in the composite, provided material stability is not compromised. While a strictly linear enhancement is unlikely due to the saturating nature of radical scavenging kinetics, increasing AA content would reasonably be expected to raise the overall scavenging response. In vitro cytotoxicity (MTT assay) Cytocompatibility of the composites was evaluated by treating RAW 264.7 macrophages with various concentrations of S1 and S2 and measuring cell viability after 24 h (Fig. 11). Both composites exhibited a concentration-dependent decrease in cell viability, a common trend observed in NPs and particulate materials, where higher doses can induce stress or toxicity in cells. However, across the entire range of concentrations, S2 maintained higher cell viability than S1 (i.e., S2 was less cytotoxic than S1 at equivalent doses).Fig. 11Cell viability of RAW 264.7 macrophages after 24 h exposure to various concentrations of S1 and S2, as measured by the MTT assay. Both composites show dose-dependent cytotoxicity, but S2 (with AA) maintains higher viability at each concentration compared to S1. At 6.25 µg mL–1, both are essentially non-toxic ( > 90% viability). At 12.5 µg mL–1, S2 still shows ~90% viability while S1 is around 80%. Even at 100 µg mL–1, S2 retains ~66% viability versus ~56% for S1. These results demonstrate improved cell compatibility of the AA-loaded composite. Error bars indicate standard deviation (n = 3); the dashed line at 70% marks a common threshold for cytotoxicity evaluation Cell viability of RAW 264.7 macrophages after 24 h exposure to various concentrations of S1 and S2, as measured by the MTT assay. Both composites show dose-dependent cytotoxicity, but S2 (with AA) maintains higher viability at each concentration compared to S1. At 6.25 µg mL–1, both are essentially non-toxic ( > 90% viability). At 12.5 µg mL–1, S2 still shows ~90% viability while S1 is around 80%. Even at 100 µg mL–1, S2 retains ~66% viability versus ~56% for S1. These results demonstrate improved cell compatibility of the AA-loaded composite. Error bars indicate standard deviation (n = 3); the dashed line at 70% marks a common threshold for cytotoxicity evaluation At the lowest concentration tested (6.25 μg mL–1), both S1 and S2 were essentially non-toxic to the cells, with viabilities >90% of the control. In fact, S1 yielded ~91% viability and S2 ~ 96% at 6.25 μg mL–1, which is within the margin of error of the untreated control (100%). This indicates that at very low concentrations, neither the presence of ZnO nor AA has any acute detrimental effect on macrophages, and both materials can be considered biocompatible at these levels. As the concentration increased to 12.5 μg mL–1, differences began to emerge. S1’s viability dropped to around 80%, whereas S2’s viability was still around 90%. This suggests that S2 was more biocompatible than S1 at moderate concentrations, likely due to AA mitigating some toxic effects. A viability of ~80% for S1 at 25 μg mL–1 is borderline in cytotoxicity terms (often >70% viability is considered non-cytotoxic by ISO standards). S2, on the other hand, remained well within the non-cytotoxic range at these levels, with ~90% viability at 12.5 and still ~80% at 25 μg mL–1 (whereas S1 might have dipped to ~70% by 25 μg mL–1). Thus, AA’s presence extends the range of concentrations over which the material is cyto-compatible. At higher concentrations, the viability continued to decrease for both materials, but S2 consistently showed a higher percentage than S1. For instance, at 50 μg mL–1, S1 dropped to around 60–65% viability, while S2 was around 75%. At the highest concentration of 100 μg mL–1, S1 had about 56% viability, whereas S2 had about 66% viability. So, at 100 μg mL–1 (which is quite a high concentration for particulate exposure in vitro), both materials showed a cytotoxic effect (viability well below 70%), but S2 was markedly less harmful than S1 (absolute difference of ~10 percentage points). Considering the viability cutoff of 70% (often used to define cytocompatibility), S1 crosses that threshold between 25 and 50 μg mL–1, whereas S2 stays above 70% until somewhere between 50 and 100 μg mL–1. In other words, S2 offers a wider “therapeutic window” or safety margin in terms of dosage. For practical applications, one would aim to keep the local concentration of particles (e.g., those leaching from an implant into surrounding tissue) low. Studies suggest that the local concentrations of NPs that cells experience in vivo are often quite low (on the order of a few μg mL–1 or less). In this context, if the scaffold releases particulate or dissolves, the concentrations might remain in the range where both S1 and S2 are fine (say <25 μg mL–1). But importantly, S2 extends the biocompatible range; if, for instance, a higher local concentration transiently occurs, S2 is more likely to remain non-toxic compared to S1. Aside from the quantitative viability, the observed cell morphology under the microscope. At low concentrations (6.25–12.5 μg mL–1), cells in both S1 and S2 conditions appeared healthy and spread out, like the control. At intermediate concentrations ( ~ 25–50 μg mL–1), S1-treated cells began to show signs of stress; a few cells appeared rounded or less spread (indicative of early cytotoxic effects), whereas S2-treated cells largely maintained normal morphology at the same doses. At the highest dose (100 μg mL–1), both S1 and S2 caused noticeable cell stress: many cells were rounded and detached in S1, while S2 had some rounded cells but still a greater fraction of adherent cells than S1. These qualitative observations align with the viability measurements, reinforcing that AA inclusion helps preserve cell health to a point. The MTT assay results confirm that S2 is more cytocompatible than S1 when interacting with mammalian cells. The enhancement is attributable to AA’s antioxidant action, mitigating the oxidative and toxic effects of ZnO at higher doses. Importantly, at low concentrations (up to ~12.5 μg mL–1), both composites can be considered biocompatible (viability ~80–100%). This is promising, as in a real implant scenario, the concentration of leached particles or ions would likely be in that low range (especially considering the scaffold is a solid material and only a small fraction might leach out over time). Therefore, both S1 and S2 are suitable from a biocompatibility standpoint for low exposure levels, but S2 provides an added buffer of safety at moderately higher exposures. Given that tissue engineering scaffolds ideally should not provoke cell death, S2’s improved cytocompatibility is a valuable advantage. Discussion It is intrinsically difficult to design a bone scaffold that is simultaneously cytocompatible, antibacterial, anti-inflammatory, and antioxidant. When host cells are nearby, ZnO, the same chemical that kills bacteria by producing reactive oxygen species and releasing Zn2+ ions, is also known to cause oxidative damage and inflammatory cascade activation [14, 33]. The primary goal of adding ascorbic acid (AA) to the PEG–HAp–ZnO system was to alleviate this tension because of AA’s well-established dual function as a metal-ion chelator and a radical scavenger [34]. The four components coexist with chemical integrity, according to the structural characterisation data: XRD showed that AA functions as a crystallite-capping agent that moderates HAp domain growth without destabilizing the lattice [30], FTIR revealed hydrogen-bonding interactions consistent with molecular-level integration rather than phase separation [20], XPS revealed a surface richer in oxygenated functional groups in S2 than in S1, and TEM confirmed that ZnO nanoparticles are uniformly dispersed within the PEG encapsulant with no evidence of aggregation [12]. These results are significant because they demonstrate that S2 is a structurally integrated composite rather than just a physical mixture, with each component playing a specific chemical role. This is necessary for the coordinated biological behaviour that follows. Comparing results from all four assays reveals a concept that makes the biological data interpretively coherent: the ratio of AA to ZnO at the quantities pertinent to each biological scenario governs the functional impact of AA in S2, which is context-dependent rather than fixed. The small AA reservoir is stoichiometrically unable to quench the radical output when ZnO operates at high local concentrations against bacteria, resulting in a sustained ROS flux. As a result, antibacterial potency is maintained, which is consistent with published findings that antioxidant loadings below ~5 wt% produce no detectable attenuation of ZnO antibacterial activity [33]. The same AA loading crosses a biologically significant threshold and offers quantifiable protection against oxidative damage, ionic Zn2+ toxicity, and NF-κB-mediated inflammatory signaling against host macrophages, where particle concentrations are orders of magnitude lower and ZnO-derived oxidative stress is proportionately reduced [18, 34]. This is a concentration-calibrated selectivity that the composite achieves by design, based on the ~10:1 ZnO-to-AA mass ratio and the spatial co-localization of both agents on the HAp surface [20, 35]. It is not a compromise between competing functionalities. Each biological finding is examined in turn in the ensuing subsections, with special focus on the molecular mechanisms underlying this selectivity and how the findings compare with similar systems in the published literature. Retention of antibacterial efficacy with AA addition The antibacterial equivalence between S1 and S2 is not surprising when examined mechanistically. Through direct electrostatic membrane disruption, ROS-mediated oxidative damage, and Zn2+ ion interference with microbial enzyme function, ZnO exerts its bactericidal action [33]. Only the ROS pathway has the potential to be inhibited by AA. The ascorbate-to-ZnO molar ratio, however, is much below the threshold at which antioxidant quenching of ZnO-generated radicals becomes observable due to the ~0.5 wt% AA loading in S2; existing benchmarks in polyphenol–ZnO systems suggest this threshold is above ~5 wt% antioxidant content [33]. Practically speaking, there is no difference in the amount of ROS that reaches the bacterial lawn between the two composites, and the observed zone size equivalency is a predictable result rather than a drawback. More significantly, it confirms a fundamental design goal: AA can be added for host-tissue advantages without sacrificing antibacterial efficacy. Rather than decreased efficacy, the smaller zone widths of both composites in comparison to pure ZnO reflect composite architecture. Because ZnO makes up just about 5% of each formulation and its nanoparticles are contained in a PEG matrix, the polymer network limits the pace at which bactericidal substances diffuse into the agar. While non-encapsulated Zn/HAp formulations produce larger zones precisely because the lack of a polymer diffusion barrier allows for faster Zn2+ release [36], this is mechanistically consistent with published data from analogous PEG-encapsulated HAp–ZnO systems [12, 30], which report comparable zone dimensions at similar ZnO loadings. For implant applications, where a quick ZnO burst would pose a cytotoxic risk to host cells, but a prolonged, lower-flux release provides long-lasting protection without local toxicity, this controlled-release characteristic is a desired feature rather than a disadvantage. Enhanced anti-inflammatory response The enhanced protein-stabilizing capacity of S2 over S1 in the denaturation inhibition assay reflects a convergence of three molecular roles that AA performs simultaneously. It neutralizes thermally produced ROS before they can oxidize vulnerable albumin side chains by acting as a two-electron reductant (E° = +0.08 V) [34]. The trace Fe3+ and Cu2+ ions that would otherwise accelerate Fenton chemistry and increase protein cross-linking are chelated by its enediol moiety [34]. Additionally, AA directly protects albumin’s folded conformation against heat unfolding by forming hydrogen bonds with surface-exposed arginine and lysine residues [20]. Through an excluded-volume effect that thermodynamically disfavours the unfolded state, PEG strengthens all three contributions [37]. The combination of these mechanisms explains why S2 approaches the protein-protective efficacy of diclofenac at intermediate concentrations and achieves a significantly lower IC50 than S1. The assay’s protein chemistry is only one aspect of its anti-inflammatory implications. Denatured proteins operate as danger signals in the peri-implant milieu, triggering NF-κB-mediated transcription of TNF-α, IL-1β, and COX-2 by activating macrophage pattern-recognition receptors (TLR4, NLRP3) [18]. AA interferes with this cascade at two upstream points: it suppresses IKK phosphorylation, which prevents NF-κB nuclear entrance, and it scavenges the ROS that function as NF-κB second messengers [34]. In a ZnO-biomaterial setting, Abdulhameed et al. accurately illustrated this anti-inflammatory cycle by demonstrating that ascorbate addition in polycaprolactone scaffolds significantly decreased macrophage cytokine release on ZnO-containing surfaces [19]. S2’s performance in the protein denaturation assay, a validated physicochemical surrogate for this cellular process, suggests significant anti-inflammatory capacity at the implant interface—as long as released AA concentrations reach the effective range in situ, which dissolution kinetics studies must verify. A cautious reading of the diclofenac comparison is necessary. Diclofenac uses nanomolar COX inhibition in cell-based settings to provide pharmacological anti-inflammation; its behavior in this assay is more indicative of non-specific protein-stabilizing characteristics than enzymatic inhibition. As a result, the comparison does not establish S2 as a COX inhibitor, but it does demonstrate that the composite functions within a range of protein-protective concentrations that could be attained at a healing implant surface—a clinically significant benchmark that encourages further release-kinetics characterization. Radical scavenging and antioxidant activity The most interpretively telling aspect of the antioxidant data is the disproportionate radical-scavenging contribution of a small, matrix-embedded AA quantity. Based on the experimentally measured IC50 of the pure standard, the actual ascorbate in the assay amounts to just ~1 µg mL−1 at the highest composite concentration tested, a level at which free AA in solution would contribute little DPPH scavenging. However, compared to free ascorbate, this matrix-bound AA generates a statistically significant increase above S1, suggesting an approximately 10-fold amplification of functional antioxidant activity per unit mass. The hallmark of S2’s antioxidant action is its amplification, which shows that the HAp–PEG matrix functions as an active functional enhancer rather than just a passive carrier. The amplification can be explained by two structural characteristics of the composite. Throughout the experiment, a larger steady-state concentration of active radical-quenching species is maintained because PEG hydrogen-bonds to AA’s vicinal diol groups, slowing the spontaneous aerobic oxidation of ascorbate to its inactive dehydro form [22]. Concurrently, ascorbate is anchored at the inorganic surface and co-localized with ZnO, the main site of ROS formation, through coordinative adsorption of AA onto HAp via Ca2 + –hydroxyl interactions. This allows radical interception at the point of origin instead of depending on bulk diffusion [20]. According to Bien et al., surface retention transformed a modest polyphenol loading into disproportionately efficient cellular antioxidant protection in a mechanistically identical amplification for quercetin in HAp composites [35]. The relevance for formulation design is that antioxidant output is more strongly governed by matrix architecture than by simple mass loading; it is anticipated that subsequent iterations with a marginally higher AA content will result in significantly better scavenging. From the standpoint of the implant, even a little amount of surface ROS interception during the acute foreign body reaction might change the local redox equilibrium in a way that is more favourable to osteoblast survival and the production of new bone [19]. The possible partial oxidation of AA during the 110 °C/1 h dry-heat sterilizing procedure applied to powders before antibacterial testing is a drawback of the current investigation that needs to be openly acknowledged. The degree of conversion under these conditions in a solid powder matrix is not completely described here. Ascorbic acid undergoes thermally accelerated oxidation first to dehydroascorbic acid (DHAA) and then to physiologically inert breakdown products [38]. The observed DPPH scavenging increment of S2 over S1 is difficult to reconcile with fully degraded AA products alone, and the XPS and FTIR spectral features of S2 are consistent with the retention of a significant fraction of the ascorbate enediol functionality after composite processing and sterilization. Nevertheless, a conclusive measurement of the amount of intact ascorbate left after sterilization—for instance, using an enzymatic colorimetric assay or HPLC—was not carried out and is still a top concern for further evaluation. Therefore, rather than focusing only on the contribution of pure, unmodified ascorbic acid, it is more accurate to say that the antioxidant and anti-inflammatory activity of S2 as measured here represents the combined contribution of any intact AA, DHAA, and residual degradation products present after sterilization. In order to better maintain AA integrity and enable the deconvolution of the contributions of intact vs oxidized species, future research will investigate gentler sterilization techniques, such as filter sterilization of freshly formed suspensions and gamma irradiation at room temperature. In vitro cytocompatibility and safety The potential of AA to reduce each of the three known mechanisms of ZnO cytotoxicity is demonstrated by the consistently enhanced macrophage viability with S2 across the measured dose range [14]. Ascorbate, which is transported intracellularly via SVCT1/2, protects mitochondrial membranes and inhibits caspase activation by neutralizing the hydroxyl radicals and superoxide generated by internalized nanoparticles [34]. Abdulhameed et al. [19] directly confirmed the relevance of this mechanism in a ZnO-biomaterial setting. The enediol moiety of AA reduces the bioavailable ionic load that macrophages experience and dampens the lysosomal dissolution arm of toxicity by chelating Zn2+ ions produced from ZnO inside the composite matrix at the ionic level [34]. In HAp–ZnO composites, Bien et al. discovered the same chelation process as a cytoprotective contribution of quercetin [35], confirming that co-incorporated antioxidants can significantly alter metal-ion bioavailability as a general protective tactic. There is mechanistic significance to the choice of RAW 264.7 macrophages as the test model. Macrophages are the main immune effectors at a newly implanted surface, and their polarization status determines the fate of the scaffold: M2 polarization promotes vascularization and osseointegration, while M1 activation causes fibrous encapsulation and implant failure [18]. Through ROS-driven NF-κB activation, ZnO at cytotoxic quantities encourages M1 polarization, which is directly detrimental to bone repair. The idea that AA’s antioxidant inhibition of NF-κB also moderates this polarization shift and preserves macrophage morphologies more favourable to tissue integration is raised by the prolonged ISO 10993 viability window shown with S2. A top goal for subsequent mechanistic research is to confirm this using M1/M2 cytokine profiling (TNF-α, IL-1β against IL-10, Arg-1) [39, 40]. Synergistic multi-functionality: the interplay of ZnO and ascorbic acid ROS are produced by ZnO as part of its antibacterial action, but they also harm host cells and increase inflammation around implants. The data collectively show a concentration-selective duality in which the same AA loading is stoichiometrically insignificant at the high ZnO concentrations needed to kill bacteria but biologically significant at the much lower concentrations that host tissue cells experience. This allays the worry that adding an antioxidant would compromise antibacterial function. In the cytocompatibility and anti-inflammatory regimes, the correspondingly decreased ZnO output falls within a range where AA significantly buffers both oxidative and ionic stress; in the antibacterial regime, ZnO overwhelms the limited AA reservoir and functions at full oxidative capacity. The ~10:1 ZnO-to-AA mass ratio and the HAp surface chemistry that co-localizes both components at the same interfacial point are the structural causes of this imbalance, which is not accidental. Within this architecture, each S2 component has a unique, non-redundant purpose. ZnO and AA are anchored in spatial proximity by the osteoconductive framework and surface chemistry provided by HAp. Through hydrogen bonding, PEG prolongs AA’s active half-life, modifies release kinetics, and helps stabilize excluded-volume proteins. ZnO exhibits long-lasting, all-around antibacterial action. Furthermore, without interfering with ZnO’s bactericidal function, AA selectively reduces the oxidative output’s harmful effects on the host. No more straightforward combination of these components could produce the multifunctional behaviour that results—simultaneous infection control, inflammatory modulation, oxidative stress reduction, and intact biocompatibility. To properly optimize the formulation and validate this multi-functionality under physiologically realistic settings, a systematic investigation of the ZnO/AA ratio space will be required, along with in vitro release kinetics characterization and macrophage polarization experiments. A comparative analysis with published multifunctional bone scaffold systems is shown in Table 2.Table 2Comparative performance of S2 (PEG–HAp–ZnO–AA) against published antioxidant-bearing bone scaffold compositesComposite SystemDPPH Antioxidant ActivityAnti-inflammatory IC50/InhibitionAntibacterial (ZOI/Method)Cell ViabilityReferencePEG-HAp-ZnO-AA(S2) [This Work]Reaches over 45% at 200 µg mL–1~70 µg/mL (BSA assay; vs. diclofenac ~50 µg/mL)7–9 mm (E. coli & S. aureus)>70% up to 50 µg/mL (RAW 264.7)Present studyHAp-Quercetin composite (QHaP)~70% relative to control (cellular DPPH surrogate, 100 µg/mL)Not reported (BSA assay not performed)Not reported153.8% membrane integrity vs. control (fibroblasts, HFF2)Bien et al., Sci. Rep. [35]Quercetin-PEG nanosuspension (Qu-PEG NS)~65–70% @ 100 µg/mL (DPPH)Significant NO inhibition in LPS-stimulated RAW 264.7 (concentration-dependent)Not reported>80% @ 100 µg/mL (RAW 264.7, MTS assay)Kang et al., Molecules [41]Polyphenol-functionalized HAp/TCP scaffold (Synergoss Red)72.8% (DPPH, composite bulk)Not reportedNot reportedNot reported (in ovo model used)Sponchioni et al., Biomater. Res. [42]Lignin–ZnO nanohybrid~84% @ 500 µg/mL (H₂O₂ scavenging)Comparable to diclofenac (BSA denaturation assay, ~100–500 µg/mL)Broad-spectrum; specific ZOI not reportedNot reportedSamal et al., ACS Omega. [43]PCL–ascorbic acid electrospun scaffold>90% radical scavenging (0.3 wt% AA in PCL fibers)Not reportedNot reported>90% viability (HUVEC cells, 6-day culture)Handley et al., J. Appl. Polym. Sci. [44]Sr/Zn co-doped nHAp–PLGA scaffoldNot reportedNot reportedSignificant S. aureus inhibition (CFU reduction)>80% (osteoblast-like cells)Hassan et al., Polymers [45]ZOI zone of inhibition, BSA bovine serum albumin denaturation inhibition assay, DPPH 2,2-diphenyl-1-picrylhydrazyl radical scavenging assay. Table 2Comparative performance of S2 (PEG–HAp–ZnO–AA) against published antioxidant-bearing bone scaffold compositesComposite SystemDPPH Antioxidant ActivityAnti-inflammatory IC50/InhibitionAntibacterial (ZOI/Method)Cell ViabilityReferencePEG-HAp-ZnO-AA(S2) [This Work]Reaches over 45% at 200 µg mL–1~70 µg/mL (BSA assay; vs. diclofenac ~50 µg/mL)7–9 mm (E. coli & S. aureus)>70% up to 50 µg/mL (RAW 264.7)Present studyHAp-Quercetin composite (QHaP)~70% relative to control (cellular DPPH surrogate, 100 µg/mL)Not reported (BSA assay not performed)Not reported153.8% membrane integrity vs. control (fibroblasts, HFF2)Bien et al., Sci. Rep. [35]Quercetin-PEG nanosuspension (Qu-PEG NS)~65–70% @ 100 µg/mL (DPPH)Significant NO inhibition in LPS-stimulated RAW 264.7 (concentration-dependent)Not reported>80% @ 100 µg/mL (RAW 264.7, MTS assay)Kang et al., Molecules [41]Polyphenol-functionalized HAp/TCP scaffold (Synergoss Red)72.8% (DPPH, composite bulk)Not reportedNot reportedNot reported (in ovo model used)Sponchioni et al., Biomater. Res. [42]Lignin–ZnO nanohybrid~84% @ 500 µg/mL (H₂O₂ scavenging)Comparable to diclofenac (BSA denaturation assay, ~100–500 µg/mL)Broad-spectrum; specific ZOI not reportedNot reportedSamal et al., ACS Omega. [43]PCL–ascorbic acid electrospun scaffold>90% radical scavenging (0.3 wt% AA in PCL fibers)Not reportedNot reported>90% viability (HUVEC cells, 6-day culture)Handley et al., J. Appl. Polym. Sci. [44]Sr/Zn co-doped nHAp–PLGA scaffoldNot reportedNot reportedSignificant S. aureus inhibition (CFU reduction)>80% (osteoblast-like cells)Hassan et al., Polymers [45]ZOI zone of inhibition, BSA bovine serum albumin denaturation inhibition assay, DPPH 2,2-diphenyl-1-picrylhydrazyl radical scavenging assay. Comparative performance of S2 (PEG–HAp–ZnO–AA) against published antioxidant-bearing bone scaffold composites ZOI zone of inhibition, BSA bovine serum albumin denaturation inhibition assay, DPPH 2,2-diphenyl-1-picrylhydrazyl radical scavenging assay. Conclusion In summary, developed a novel HAp-PEG-ZnO-AA composite (S2) that outperforms the HAp-PEG-ZnO composite without AA (S1) in multiple facets of performance, while matching it in others. S2 exhibited improved cytocompatibility (maintaining higher cell viability at given doses), superior anti-inflammatory activity (inhibiting protein denaturation) and antioxidant capability (scavenging free radicals) due to the presence of AA, and enhanced HAp crystallinity (larger crystal domains) as a result of AA’s influence on the mineralization process- all while maintaining antimicrobial efficacy comparable to S1. The synergistic combination of components in S2, HAp for osteoconductivity and structural support, ZnO for antibacterial function, and AA for antioxidant and anti-inflammatory roles, all integrated within a biocompatible PEG matrix, yielded a versatile biomaterial that addresses several critical challenges in bone tissue engineering. S2 leverages the natural bone mineral (HAp) as a scaffold, ZnO NPs as a dispersed nano-antibacterial agent, and AA as a bioactive molecule that can scavenge ROS and support collagen synthesis. The result is a composite that balances structural stability with biological activity. It can provide localized antimicrobial action to fight infection (through controlled Zn2+ release and ZnO’s contact-killing mechanisms) without the need for external antibiotics, dampen the inflammatory response (through AA’s protein-protective and radical-scavenging effects), and reduce oxidative stress in the implant vicinity (through the antioxidant function of AA). This multi-functionality is highly desirable for advanced biomedical applications such as bone grafts, coatings for orthopaedic implants, or tissue engineering scaffolds. In these contexts, an ideal material should not only serve as a template for new bone growth but also actively enhance the healing process and prevent complications. S2 composite makes progress toward that ideal by simultaneously addressing infection, inflammation, and oxidative stress, three factors that commonly hinder healing post-implantation. Moreover, by deriving HAp from bovine bone waste, the material aligns with principles of sustainability and cost-effectiveness, giving a second life to what would otherwise be agricultural waste. Future work will focus on further optimizing the composite and validating its performance in more complex models. For instance, one could experiment with higher AA loadings or alternative antioxidant additives to increase the antioxidant capacity, while monitoring any effects on mechanical properties or ZnO’s function. Long-term release studies would be valuable to determine how Zn2+ and ascorbate are released over time from S2, and whether a sustained release profile is achieved. Mechanical testing (compression, tensile properties) of the scaffold will also be needed to ensure it meets the requirements for bone-support applications. The presence of PEG likely improves toughness, and the higher HAp crystallinity in S2 might enhance stiffness. In vitro co-culture studies or osteogenic differentiation assays could assess whether S2’s environment (with Zn and AA) promotes bone cell activity. Ultimately, in vivo animal model studies will be crucial to confirm that the composite’s antimicrobial and anti-inflammatory benefits translate to better healing outcomes (e.g., faster bone regeneration with fewer complications). Also, a detailed evaluation of the 3D printability of HAp-PEG-ZnO-AA composite will be undertaken in future work. Nonetheless, even at this proof-of-concept stage, S2 presents a compelling case as a multifunctional bone scaffold material. By harnessing the synergy of its components, it offers a combination of properties that individually would otherwise require multiple separate treatments or coatings. S2 acts as a scaffold that not only supports new bone growth but also actively improves the local physiological environment by reducing infection risk and moderating the immune response. This work demonstrates the feasibility and advantages of incorporating antioxidant molecules into inorganic-biopolymer composites for biomedical use, and sets the stage for further development of smart, bioactive composite scaffolds for tissue engineering. Supplementary information Supplementary Information Supplementary Information