A study on the effect of bioactive glass and hydroxyapatite-loaded Xanthan dialdehyde-based composite coatings for potential orthopedic applications

The most important challenge faced in designing orthopedic devices is to control the leaching of ions from the substrate material, and to prevent biofilm formation. Accordingly, the surgical grade stainless steel (316L SS) was electrophoretically deposited with functional composition of biopolymers and bioceramics. The composite coating consisted of: Bioglass (BG), hydroxyapatite (HA), and lawsone, that were loaded into a polymeric matrix of Xanthan Dialdehyde/Chondroitin Sulfate (XDA/CS). The parameters and final composition for electrophoretic deposition were optimized through trial-and-error approach. The composite coating exhibited significant adhesion strength of “4B” (ASTM D3359) with the substrate, suitable wettability of contact angle 48°, and an optimum average surface roughness of 0.32 µm. Thus, promoting proliferation and attachment of bone-forming cells, transcription factors, and proteins. Fourier transformed infrared spectroscopic analysis revealed a strong polymeric network formation between XDA and CS. scanning electron microscopy and energy dispersive X-ray spectroscopy analysis displayed a homogenous surface with invariable dispersion of HA and BG particles. The adhesion, hydrant behavior, and topography of said coatings was optimal to design orthopedic implant devices. The said coatings exhibited a clear inhibition zone of 21.65 mm and 21.04 mm with no bacterial growth against Staphylococcus aureus (S. Aureus) and Escherichia coli (E. Coli) respectively, confirming the antibacterial potential. Furthermore, the crystals related to calcium (Ca) and HA were seen after 28 days of submersion in simulated body fluid. The corrosion current density, of the above-mentioned coating was minimal as compared to the bare 316L SS substrate. The results infer that XDA/CS/BG/HA/lawsone based composite coating can be a candidate to design coatings for orthopedic implant devices.

Musculoskeletal disorders affect ~ 1.7 billion people globally.The Agency for Healthcare Research and Quality reported 750,000 knee replacements in 2017 in the US 1,2 .Hip and knee replacements are among the most performed surgeries across the globe, with the knee replacement market projected to reach USD 17.45 billion by 2030 2,3 .
Magnesium (Mg), titanium (Ti), cobalt-chromium (Co-Cr), and stainless steel (SS) can be used to fabricate orthopedic implants, including fixators (internal/external), bone plates, prosthesis, etc. Surgical grade SS (316L SS) has properties that make it an attractive option for orthopedic implants.It is bioinert, biocompatible, and has high corrosion resistance, which makes it not produce a strong immunological response or any toxicity in the body.Its mechanical properties (durability and high tensile strength), sterilizability and low cost make it viable for orthopedic implants [4][5][6] .SS implants are coated with biopolymers and bio-ceramics to promote

Sample preparation
The 316L SS strips were cut into the rectangular (length-3 cm, width-2 cm, thickness-0.1 cm) and disc shaped (diameter-10 mm, thickness-1 mm) samples using meta cut (METKON302).The disc shape samples were used for antibacterial studies.Afterward, the samples were cleaned and, abrasive silicon carbide papers (having different grit sizes: 120, 240, 360, 400, 800) were used to polish the said samples (120, 240, 360, 400, 800).The bi-distilled water was drawn over the sample`s surface to wash out the impurities/debris.The polished surface of the sample was etched out using 70% H 2 SO 4 and 30% H 3 PO 4 for 25 s followed by rinsing in bi-distilled water [40][41][42] .

Synthesis of Xanthan dialdehyde (XDA)
XDA was synthesized by the oxidation of Xanthan Gum (XG) with SMP in a dark controlled environment at 60 °C and stirred for 4-6 h 43 .Guo et al. 43 and Ngwabebhoh et al. 10 outlined the mechanism of XDA formation by using 9.6 W/V% (0.096 g/mL) SMP solution.9.6 W/V% SMP solution concentration for approximately 45% aldehyde content for XDA was reported to have the highest tensile properties.Using ratio stoichiometry, 0.5 W/V% XG required 0.096 g/mL SMP for 45% XDA content (a similar approach was employed in 43 ).A portion of 0.125 g of Xn was liquified in distilled water (25 mL) at 60 °C (0.5 W/V% XG).The SMP solution was made using 0.96 g (9.6 W/V%) of SMP in 10 mL of distilled water.Later, XG was oxidized by SMP at the concentration of 0.096 g/mL.The solution was stirred for 4-6 h in a dark environment, following 41 .The reaction was then quenched using 240µL EG, following 10 .

Preparation of 45S5 bioglass (BG)
BG of composition: P 2 O 5 6%, CaO 24.5%, Na 2 O 24.5%, SiO 2 45% (mol%) was synthesized by first dissolving diatom biosilica (prepared following 44 ) in 35% C 6 H 8 O 7 solution.The mixture was magnetically stirred for 2 h under 120 °C.Following the complete dissolution of diatom biosilica, H 3 PO 4 was added and magnetically stirred for 45 min.Thereafter, NaNO 3 , was added followed by the addition of Ca (NO 3 ) 2 •4H 2 O which again was stirred for an hour.The attained gel was treated following the procedure executed in 45 .It was then placed in a muffle furnace (Muffle furnace LMF-E21 manufactured by Labtron) at a constant temperature increment and abatement of 10 °C/min.For 2 h, gel was kept at 700 °C followed by sintering for 3 h at 950 °C.The procedure adopted was similar to Adams et al. 44 .

Final preparation of suspension
First, BG (0.125 g) was added in XDA solution (0.5 w/v%) and continuously stirred for 6-7 h at 25 °C to disperse the BG particles.Next, 0.25 g of CS was added to the 25 mL of ethanol (1 w/v%).Later, 0.125 g of HA was added and stirred for 2-3 h at 80-100 °C and sonicated for 2.5 h 46 .In addition, 0.3 g of henna was dissolved in 15 ml of ethanol (2 w/v%) 47 and the solution was filtered using filter paper.The filtered 10 mL lawsone solution was then added to 25 mL of HA/CS solution (5:2) as shown in Fig. 1, and subsequently stirred for 30 min at 60 °C.
The final suspension for EPD was prepared by using HA/CS/lawsone solution, and XDA/BG solution (an overview is presented in Fig. 1).XDA/BG and HA/CS/lawsone solutions (1:1) were mixed and magnetically stirred at room temperature.The pH of the resulting XDA/CS/BG/HA/lawsone suspension was maintained at ~ 4.5 using AA.

Electrophoretic deposition
A 316L SS was used as substrate (cathode) and counter-electrode (anode).The anode-cathode distance was 10 mm.The coating thickness and the control over surface crack were optimized by tuning the applied deposition voltage and the deposition time given during the EPD process.The voltage was increased in 5 V increments,

Adhesion test
A tape test was used to gauge the composite coating's (n = 5) adherence strength to the substrate.According to ISO 2409 and ASTM D3359-17 standards, a crosshatch adhesion test was performed.The CC2000 basic cross hatch cutter was used to create scratches on the coated substrates.The adhesive tape was applied for 90 s on the substrate and removed afterwards maintaining an angle of ~ 180° between the hand and the coated surface.The quality of the adhesive strength was then assessed by comparing the scratches to ASTM Standards [50][51][52] .The ASTM D3363-20 standard was followed to conduct the pencil scratch hardness test (n = 5), following 53 .Standard holders with grades ranging from 8B (soft) to 2H (hard) pencils were moved over the coated substrate at an angle of 45°.The test continued until the pencil could no longer starch the covered substrate.Following the ASTM B571-97 standard, the adhesiveness of the said coating was analyzed using the bend test (n = 5).Composite coated samples were bent into a 180° U shape using pilers.Composite coatings were evaluated for their adhesive strength by studying their delamination and deformation capacities.A microscope (Novex, Holland) was used to examine the distorted samples after the bending test 54 .Scratch tests were conducted on the composite coating samples with Revetest mode (using a CSM Revetest machine).An indenter with a 200 m diameter was used to apply a progressive load (0.5-4 N) to the coated surface for scratch testing.Indenter speed and scratch length were 5.2 mm/min and 5 mm, respectively.The sample was run on eight separate tracks.By measuring acoustic emission (AE) and through optical observations critical loads were identified.The minimal load at which the initial crack begins to appear was denoted as the critical load (L C1 ) while increase in load lead to the complete delamination and that supremely enhanced load was referred as critical load (L C2 ) 55 .

Surface roughness
The coated XDA/CS/BG/HA/Lawsone and uncoated 316L SS samples were characterized in terms of average surface roughness (Ra), maximum profile peak height (Rt), and average maximum profile (Rz) by employing stylus profilometer (TMR 360).Three measurements were taken for each set of samples and the average value with the standard deviation was reported.The microscopic tip stays in contact with the surface and moves horizontally.However, the tip's vertical movement was recorded and enhanced.It was mounted vertically, using www.nature.com/scientificreports/ a stylus profilometer.As a result, the stylus is positioned above the part to be measured.On the surface of both samples, the stylus moved 5 mm.

Contact angle
Distilled water was used to measure surface wettability.Distilled water droplets of 5 µL were used to quantify the contact angle of the bare 316L SS, XDA coated 316L SS, and XDA/CS/BG/HA/lawsone coated 316L SS substrate.A micropipette was used to drop the droplet on bare 316L SS, XDA coated 316L SS, and XDA/CS/BG/ HA/lawsone coated 316L SS.Images were taken after every 5 s for 30 s and analyzed through ImageJ software using drop analysis libraries (LB-ADSA) 56,57 .Tests were repeated five times, and averaged values were reported along with standard deviations.

Energy dispersive spectroscopy (EDX) analysis
The EDX (OXFORD) installed with FESEM (MIRA, TESCAN) was employed to access the elemental composition of XDA/CS/BG/HA/lawsone coatings (acceleration voltage was kept at 25 kV).www.nature.com/scientificreports/FTIR The functional groups in XDA/CS/BG/HA/lawsone coating were evaluated using OMNIC paradigm software of an attenuated total reflection Fourier transformation infrared (ATR-FTIR) spectroscope (Thermofisher Nicolet Summit Pro).The transmittance spectrum was obtained in the region of 4000-400 cm −1 with 4 cm −1 resolution after omitting noise.

Corrosion behavior
The Gamry instrument (Potentiostat/Galvanostat/ZRA, reference 600) evaluated the corrosion resistance of the composite coatings and bare 316L SS.Gamry device contains three electrodes: working electrodes (coated/ uncoated substrates), counter electrode (graphite), and reference electrode (Ag/AgCl).SBF was used as an electrolyte at 37 °C.At 2.5 mV/sec scanning rate, potentiodynamic polarization scans were carried out at ± 500 mV potential.Tafel plot corrosion potential was calculated as (E corr ) and corrosion current density was calculated as (I corr ).During an EIS test, an open circuit potential was used with 10 mV perturbation.Frequencies were kept between 100 kHz and 1 MHz.In each case, with measurements in triplicate 58,59 .

In-vitro bioactivity studies
The biocompatibility (generally ability to bind or attach bone cells in case of orthopedic implants) of XDA/ CS/BG/HA/lawsone coatings was accessed using SBF synthesized by using the procedure outlined by Kokubo et al. 60 .The composite coatings (1.5 cm × 1.5 cm) were submerged in SBF filled falcon tubes and placed in an shaking incubator (30 rpm) for 1, 7, 14, 21 and 28 days at 37 °C.To simulate physiological conditions, the SBF was replaced every 24 h.After the stated periodic time interval of a week, coated strips were carried out of falcon tubes, rinsed with distilled water and processed for further characterization.SEM and FTIR analyses were conducted on the said samples both before and after immersion in SBF.

Antibacterial studies
The basic qualitative antibacterial efficacy of XDA/CS/BG/HA coated 316L SS (Control) and the XDA/CS/BG/ HA/lawsone composite coated sample was investigated using the Kirby-Bauer test.Henna was loaded in two different concentrations 0.01 mg/mL and 0.02 mg/mL as of literature and subsequent coatings were tested 61 .Nutrient Agar (Oxoid-UK) was used to prepare agar medium, which was then autoclaved at 121 °C for 20 min, subsequently; 20 mL of prepared media was transferred to the sterile petri dishes.Composite coated and XDA/ CS/BG/HA 316L SS were exposed to UV light for an hour.20 µL of S. aureus and E. coli owing optical density (OD 600 ) of 0.015 ± 0.002 was rolled out over solidified agar medium using a sterilized glass slider.Coatings were then placed over the uniformly distributed bacteria in such a way that coated side-faced bacteria, followed by round the clock incubation at 37 °C.Afterward, Pictures were taken, by using ImageJ software the inhibition zone was measured.

Morphological analysis
The surface topography of the XDA/CS/BG/HA/lawsone composite coating deposited on 316L SS was evaluated using FESEM.SEM micrographs of composite coating at low and high magnification revealed a plain, homogeneous, and all-alike coating with no cracks on the surface, as shown in Fig. 3A, B. Through the decomposition of water into hydrogen and oxygen (O) (electrolysis), that usually occurs during EPD, a limited number of micropores created.The coating exhibited uniform distribution of BG and HA particles (FTIR and EDX verified the existence of HA and BG) in the polymeric matrix due to invariable particle size in final suspension 39 .The FESEM images show the agglomeration of BG and HA particles (Fig. 3A).The outcome of the current study is like the one revealed by 62,63 .HA particles are induced and embedded in the polymeric matrix depicting quasi spherical morphologies.The pores present in the polymeric matrix are filled by HA particles and their agglomerates, ultimately intensifying the mechanical stability of the coating.Similar morphologies for HA-based composite coating were proclaimed by Ahmed and Ur Rehman et al. 64 and Louvier-Hernández et al. 65 .The cross-sectional micrographs of XDA/CS/BG/HA/lawsone composite coating authenticates the uniform coating with thickness of ~ 3 μm, as shown in Fig. 3B.Similar coating thicknesses were reported in previous studies 66,67 .

Tape test
Adhesion tests are conducted on composite coating to evaluate their mechanical strength.Samples that underwent tape testing were analyzed using an optical microscope (Novex, Holland) and categorized using a tape test according to ASTM D3359 (Fig. 4).First, we scratched the coated portion of the substrate with a crosshatch cutter, as shown in Fig. 4A.After 90 s, the tape was quickly removed and the image was acquired Fig. 4B.4B adhesion strength was achieved by the composite coatings (only minor delamination of 5%); this implies that at intersections, small fragments of the coating become dislodged, impacting less than 5% of the surface area, as per the ASTM standards 68 .Prior studies shows that HA/BG-based composite coatings rated as 3B (approximately 5-15% of the lattice area suffer from small flakes detaching along edges and cuts) shows good adhesion strength but in our case coating adhesion rated as 4B much good adhesion strength than already reported studies.The robust adhesion strength observed in the composite coating can be attributed to the resilient polymeric network formed by XDA and CS, as discussed in "FTIR" section, as well as the presence of HA/BG 69 .The developed composite coatings presented better adhesion than other reported coatings, that utilized biopolymers as well.M. A. U. Rehman et al. 62 conducted the tape test (ASTM D3359-97 and B571-97) on composite coating of CHI/ gelatin/Cu-doped-BG on a 316L SS substrate, the coatings were reported to delaminate at the edges.The better adhesion strength of the composite coatings is attributed to the defect-free coatings (microstructure), enhanced cross-linking capability of aldehyde groups in XDA, and etching of the substrate prior to the coatings which may have assisted in the mechanical and chemical linking of the composite coating with the substrate 43,70 .Thus, the coating was rated as "4B" according to the ASTM standard.Zhang et al. 71 observed increased adhesion of www.nature.com/scientificreports/coatings, coinciding with the uniform distribution of HA particles, of Bovine-HA deposited on Ti-12Cr substrate.Chen et al. 72 utilized EPD to develop Alginate/BG coatings on 316L SS and remarked their adhesive strength to be "3B" according to ASTM D3359-B, after conducting the tape test.

Bend test
An ASTM B57197 bend test was used to assess the bending strength of the composite coating upon bending to 180°. Figure 4C, and D show the images of an optical microscope (Novex, Holland) captured after the bend test.
The composite coating exhibited sufficient resistance to bending loads throughout substrates, as no delamination was observed even at the edges of the composite coating.In this study, composite coatings presented better adhesion strength upon bending compared to the 62 where CHI-based composite coatings delaminated at the edges.
The better adhesion strength of the composite coatings is attributed to the defect-free coatings (microstructure), enhanced cross-linking capability of aldehyde groups in XDA, and etching of the substrate prior to the coatings which may have assisted in the mechanical and chemical linking of the composite coating with the substrate 43,70 .Thus, the coating was rated as "4B" according to the ASTM standard.

Pencil hardness test
The adhesion strength of the XDA/CS/BG/HA/lawsone coating with the 316L SS was evaluated by the pencil test.
The findings showed the composite coating to be able to withstand a range of pencil hardness values without showing any signs of cracking or flaking till H.However, a slight powder delamination was observed when the coating was scratched with the 2H pencil, as shown in Fig. 4E.Said results are due to the presence of HA/BG 69 and strong polymeric network of XDA and CS (discussed in "FTIR" section).

Scratch test
XDA/CS/BG/HA/lawsone composite coating adhesion strength was evaluated with the scratch test shown in Fig. 5.As shown in Fig. 5, XDA/CS/BG/HA/lawsone composite coating experienced a 15 N L c1 value, and circular cracks appeared on its surface, despite remaining intact.According to the measured value of second critical load (L c2 ), at L c2 there is still no significant delamination of coatings showing that the XDA/CS/BG/HA/ lawsone composite coating has better adherence.After the L c2 composite coatings delaminated from the surface.A scratch test indicated that the composite coatings were suitable for designing coatings for orthopedic implant devices Fig. 5 73,74 .

Wettability
The wettability of the surface was measured by contact angle, as shown in Fig. 6.The optimal contact angle for facilitating initial protein attachment typically falls within the range of (35°-80°) 75 .The measured values of bare 316L SS, XDA coated 316L SS and composite coating were 64° ± 1°, 56 ± 1° and 48° ± 4°, respectively.The contact angle of XDA coated 316L SS is lower than the bare 316L SS, indicating increased hydrophilic character introduced by the XDA coating.This enhanced hydrophilicity is attributed to XDA`s intrinsic properties, particularly its hydroxyl and aldehyde functional groups 76,77 .Furthermore, the composite coating exhibited lowest contact angle signifies the highest-level hydrophilicity in comparison to XDA coated 316L SS and bare 316L SS Substrate, this could be due to the hydrophilic nature of HA, BG (silanol groups led to the hydrophilic nature of BG) and CS 78 .The presence of Sulfate and carboxylate functional groups in CS and hydroxyl groups in HA makes the composite coatings hydrophilic 10 .Additionally, XDA also showed a hydrophilic nature due to the introduction of aldehyde groups into the polymer chain, leading to enhanced hydrophilicity 77 .Chen et al. 72 reported the relationship of contact angle and surface roughness, surface roughness increases as contact angle decrease and the reported results are in agreement with 72,79 .Manzur et al. shows the contact angle of Poly(lactic-co-glycolic) Acid/henna/Copper-doped Mesoporous bioactive glass nanoparticles (PLGA/henna/Cu-MBGNs) is 41° on Mg substrate which agrees with reported studies 57 .Shi et al. 79 also conducted study of hydrophilicity on CHI/HA/GO coating on Ti substrate and reported a contact angle of 53.3 ± 1.3° and reported that enhanced hydrophilicity is desirable to facilitate the adhesion of cells and cell-adhesive proteins, promoting osteoconductivity.

Surface roughness test
Surface topography plays a significant role in influencing the cellular interaction with induced biomedical implants.Related cells and proteins should be able to adhere to the implant surface 49,80 .Surface topography was investigated by measuring the average surface roughness (Ra), maximum profile peak height (Rt), average maximum profile (Rz) of the XDA/CS/BG/HA/lawsone composite coatings deposited on 316L SS via EPD, as given in Table 2.The composite coatings exhibited relatively higher values of Ra, Rz, and Rt.The composite coatings exhibited the Rz values of 1.78 µm, which can allow the osteoblast cells to attach to the coated surface 81,82 owe to the favorable surface topography.Osteoblast attachment and proliferation can be influenced by other factors than just surface topography.These factors include wettability and surface chemistry, which should also be considered.This study found that composite coatings had suitable wettability, roughness, and surface chemistry for protein attachment and later cell attachment and proliferation.To achieve optimal tissue integration, the growth and differentiation of the relevant cells are important (in the current case osteoblast cells) 83,84 .www.nature.com/scientificreports/

Chemical properties
EDX analysis EDX analysis indicated the presence elements attributed to the biopolymers (XDA/CS), bioceramics (HA/BG) and lawsone.The peaks of O and Carbon (C) may denote the subsistence of CS and XDA in the composite coatings, as shown in Fig. 7. Although, it is not accurate to spot the C and O via EDX.Therefore, to verify the subsistence of XDA and CS in the composite coatings, FTIR was executed.Additionally, the occurrence of HA in the coating was verified by the identification of peak spots, associated with P and Ca.Furthermore, the area EDX analysis portrayed Mg, Si and P indicating the presence of BG in the composite coatings.Although, the Ca and P are present in HA and BG.The appearance of Si is associated with the presence of BG in composite coating.

FTIR analysis of composite coatings
The identification of the aldehyde groups in XDA can be confirmed by FTIR spectroscopy.The region of 1730-1750 cm −1 shows an evident peak which identify the aldehyde group in XDA 70 .In addition, the peak in the region of 1600-1650 cm −1 associated to the C=C stretching vibration and peak at 2978 cm −1 is associated with C-H stretching vibrations, respectively 10 .Moreover, the band of O-H stretching vibrations appeared in the range of 3200-3600 cm −1 .The peaks in the region of 1050-1100 cm −1 correspond to C-O-C stretching vibrations 10 .
In FTIR spectra of CS, the peaks in between 3000-3600 cm −1 represent -NH and -OH stretching vibrations.The broad peak appeared at 2907 cm −1 correspond to the C-H bond 13 .The sulfate region of CS can typically be found between 1000 and 1300 cm −1 as shown in Fig. 8A.This region indicated the presence of a strong peak at 1236 cm −1 , attributed to the disproportioned and uneven stretching vibrations of the sulfate groups (S-O bond)) 85 .The peak at ~ 1039 cm −1 can also be observed, which corresponded to the symmetric stretching vibration of the sulfate groups (C-O-S ring)) 85 .The amide region of CS was found between 1700 and 1500 cm −1 .The  www.nature.com/scientificreports/peak at 1606 cm −1 introduced due to the amide I vibration of the peptide backbone 86 , similarly peak at around 1560 cm −1 was introduced due to the amide II vibration of the peptide backbone 85,87 .FTIR spectra of bioactive glasses showed Si-O-Si stretching vibration, at ∼703 cm −1 and 468 cm −1 and Si-O stretching of nonbridging oxygen atom at ∼925 cm −188,89 .This peak is manifesting the SiO 2 network structure in BG, which is critical for its bioactivity.The phosphate peak (PO 4 −3 ) appeared at ∼610 cm −189 .The peak at ~ 910 cm −1 in HA, is accredited to the stretching vibration of phosphate (PO 4 −3 ) groups in HA.An extensive band in the range of 500-600 cm −1 , depicted bending vibration of phosphate (PO −3 ) groups in HA 90 .A weak peak at around 738 cm −1 , corresponded to the deformation vibration of carbonate (CO 3 −2 ) groups in HA as shown in Fig. 8B.Moreover, a solid peak for carbonate (CO 3 −2 ) appeared at 1454 cm −190,91 .The infrared spectra of lawsone with transmittance peaks showed at ∼3320 cm −1 indicating an O − H bond 92 .The peak spot at about 1755 cm −1 denotes the carbonyl group (C=O), while the aliphatic C-H stretching peaks associated with ∼2925 cm −1 .The phenolic groups in henna were evident by the climax at ∼1000 cm −193 .
The FTIR spectra of composite coating showed the genesis of Schiff base between XDA and CS 86 .XDA exhibited characteristic peaks of the aldehyde group (C=O) at around 1730-1740 cm −1 .However, the amine group (N-H) in CS appeared at around 3300-3400 cm −1 .After the crosslinking reaction, a new peak corresponding to the imine group (-C=N-) in the Schiff base appears at a typical range of 1620-1660 cm −194 This shift in the peak position from the aldehyde and amine groups to the imine group can serve as an indicator for the formation of the Schiff base between XDA and CS.The -OH band existed in the final composite coating showing the presence of henna and XDA.The overlapping peaks at 1022 cm -1 is attributed to phosphate group of HA and phenolic group of henna 90,93 .However, the peak of dispersed BG in the composite coating appeared at 703 cm −1 corresponded to Si-O-Si stretching vibrations 95 .The phosphate group peak due to HA appeared at 533 cm −186 .

Potentiodynamic polarization
The potentiodynamic polarization behavior of bare and XDA/CS/BG/HA/lawsone coated 316L SS in SBF is represented in Fig. 9.The values of E corr and i corr were determined using the Tafel method.From Fig. 9 the composite coated 316L SS shows the cathodic slope is lower as compared to the bare 316 L SS.Similarly, the anodic slope of the composite coated 316L SS is suppressed by three orders of the magnitude and have a good agreement with the cathodic slope.The composite coated 316L SS shows reduced corrosion current density (i Corr ) and higher corrosion potential (E corr ) than bare 316L SS.The polarization resistance (R p ) of the XDA/CS/BG/HA/lawsone coated 316L is increased in two ordered of the magnitude than bare 316L SS.This indicates that the composite coating provides the significant corrosion resistance to 316L SS in SBF solution 13,96,97 .

Electrochemical impedance spectroscopy (EIS)
The plot of the EIS of uncoated 316L SS compared to XDA/CS/BG/HA/lawsone coated 316L SS is shown in Fig. 10A.The Nyquist plots of the bare and coated samples present a high semi-capacitive arc.It indicates that the same process of corrosion is occuring at a lower rate for the coated substrate 47 .Figure 10B shows the impedance values are almost equal at low frequencies (úZê 0.01 Hz ), but at higher frequency, coated 316L SS has higher impedance values-dispensing information about the electrode/solution interface.Bode phase plot shows coated 316L SS exhibits an increased capacitive response over the broad frequency range with a lesser phase angle than bare 316L SS 59,98 .
Figure 10C, D shows that the electric equivalent circuits (EEC) )) indicates the electrochemical activity between uncoated and coated 316L SS.R s (Q 1 (R 1 (Q 2 (R 2 W)))) is the preferred choice for coated 316L SS.According to the reported data, numerous interpretations have been provided for the circuit components 99 .Specifically, Q 1 and R 1 represent the porous layer capacitance and resistance, while Q 2 and R 2 show the barrier layer resistance when connected (series) with a Warburg diffusion element (W).The Table 3 shows that the R 1 porous layer has a higher corrosion resistance than R 2 , thus BG/HA/lawsone contributed more to corrosion resistance than R 2 resistance.Potucek et al. 100

In-vitro SBF studies
The XDA/CS/BG/HA/lawsone composite coating was immersed in SBF to track the possible changes in the morphology and the formation of the apatite-like layer on the surface of the coatings.It is widely accepted that the genesis of apatite-like layer indicates the osteoinduction and osteointegration of the designed implants 84 .The HA layer formation was confirmed via FTIR, SEM, EDX and X-ray diffraction (XRD).Dissolution of the BG particles results in the release and diffusion of Ca and P ions, the ionic exchange that then happens assists in the formation of apatite-like structure on the surface of the coatings (Fig. 11A, B) 20,21,101 .The dissolution of BG was evident by the decrease of the intensity of silica peaks in the FTIR spectra (the reduction in silica peak may also be credited to the initiation of the HA layer after being dipped in SBF).The magnitude of the silica peaks reduces with the increase in immersion time whereas the intensity of phosphate bands increases (Fig. 11C).After 28 days of immersion in SBF, the FTIR spectra faded away BG-associated Si-O-Si elongated peaks 102 .The bioactive glass peaks were formed at 460 and 703 cm −1 in pure composite coating and after immersion in SBF till 21 days, as shown in (Fig. 11C).The carbonate (CO 3 −2 ) and phosphate (PO 4 −3 ) peaks attributed to HA (1)   Agathopoulos et al. 102 also reported the appearance of HA layer associated peaks in FTIR after immersion in SBF.
The reported results are in agreement with 103,104 .The possible mechanism for the formation of HA crystals upon dipping in SBF is attributed to the release of Ca and P ions from the composite coatings, which then react with the SBF ultimately devising a concentrated solution of Ca Phosphate.Thus, resulting in the formation of HA crystals on the surface of the coating through nucleation and growth 105 .The HA layer formation on composite coating after immersion in SBF for 28 days was first validated by FTIR followed by SEM analysis.
HA layer formation on BG coatings was also confirmed using FESEM after immersion in SBF, as shown in (Fig. 11A, B).The XDA/CS/BG/HA/lawsone coatings are designed to create a bioactive layer on implant surfaces (to induce mineralization), which is crucial for successful osteointegration 106 .FESEM analysis showed the appearance of typical spherical or needle-shaped morphology.
XRD was utilized to examine XDA/CS/BG/HA/lawsone coatings immersed in SBF for 28 days to verify the existence of a HA layer, as depicted in Fig. 11D.The results of the XRD analysis unequivocally substantiated the formation of a HA layer before immersion in SBF due to the presence of HA in the developed coatings.The increase in the intensities of HA associated peaks after immersion in SBF for 28 days confirmed the formation of HA layer due to BG in coating 107 .The XRD pattern exhibited a peak corresponding to HA at angles of 25.8° and 31.5°1 08,109.
EDX analysis after 28 days of immersion in SBF showed the peaks of Ca and P while Si peak vanished (agrees with FTIR analysis).The Ca/P ratio calculated from the EDX analysis before and after immersion in SBF was 0.544 and 1.54, respectively as shown in Fig. 11E, F. Ca/P ratio of 1.54 after immersion in SBF is very close to the stoichiometric ratio of the HA 101 .The results agree with 110,111 Antibacterial studies Significant antimicrobial effects of composite coatings were seen because of lawsone and naphthoquinone release.Figure 12A shows that XDA/CS/BG/HA coated 316L SS did not form any zone of inhibition, while 0.01 mg/ mL and 0.02 mg/mL lawsone loaded composite coated samples showed the zones of 13.32 mm ± 0.5 mm and 21.65 mm ± 0.5 mm against E. coli, respectively.While Fig. 12B shows that XDA/CS/BG/HA 316L SS couldn't show any bacteriostatic or bactericidal properties while 0.01 mg/mL and 0.02 mg/mL lawsone loaded composite coated samples exhibited zones of 16.48 mm ± 0.5 mm and 21.04 mm ± 0.5 mm against S. aureus, respectively.The clear zone of inhibition portrayed that 0.02 mg/mL of lawsone was enough to create strong antibacterial effect.Lawsone (2-hydroxy-1,4-naphthoquinone) the major component of henna, generates reactive oxygen species (ROS) and shows to apoptotic cell death of bacteria 112 .The results are consistent with those reported by Rubiay et al. 113 .
Novel and functional bio-composite coating development has been a success.Apatite induction was confirmed through in-vitro SBF studies, while antibacterial potential due to the release of lawsone was established and evaluated through Kirby-Bauer test against E. coli and S. aureus.Corrosion resistance of 316L SS was enhanced by first order of magnitude, examined via Potentiodynamic polarization scan and EIS analysis.The composite coatings developed in the current study showed the strong potential to coat orthopedic implant devices.The composite coatings can provide bone regeneration and can prevent biofilm formations.The future studies will involve cell biology, biomarkers quantification, alkaline phosphate activity, drug release kinetics, quantitative microbiology studies and in-vivo studies.It is expected that the designed coatings will help in the differentiation and proliferation of the osteoblast cells which will eventually help in the growth of bone tissues.The biochemical process aided through XDA/CS/BG/HA lawsone is proposed to overcome the challenges associated with orthopedic implants, as shown in Fig. 13.

Conclusions
The XDA/CS crosslinking in the polymeric network was confirmed by FTIR analysis, resulting in the formation of a robust coating of XDA/CS/BG/HA/lawsone (coating thickness ~ 3 µm) using EPD.The results of the pencil hardness test, tape test, scratch test, and bend test confirmed suitable adhesion strength of the composite coating with the 316L SS substrate.The bioactivity and antibacterial tests, combined with the potentiodynamic polarization assessment, demonstrated that the 316L SS coated with XDA/CS/BG/HA/lawsone exhibited promising characteristics for the development of orthopedic implants.The coated substrate showed significant corrosion resistance, as well as apatite induction and antibacterial potential, making it a highly attractive prospect for further investigations and eventually the translation to the clinic.Overall, the results suggested that the fabricated coated implant has the potential to be considered to design orthopedic implant devices (Table 4).Coating demonstrated higher surface roughness than bare substrate Ra (coating) = 0.32 ± 0.015 µm Ra (bare) = 0.32 ± 0.015 µm Cheng et al. 72

Figure 1 .
Figure 1.Schematic representation of the XDA/CS/BG/HA/lawsone suspension preparation steps conducted prior to the EPD process.

Figure 2 .
Figure 2. Schematic representation of Direct current EPD setup showing the deposition of XDA/CS/BG/HA/ lawsone composite coatings deposited on 316L SS substrate.

Figure 3 .
Figure 3. SEM images (A) shows uniformly distributed and induced BG and HA particles, respectively at 5 kX and 150 kX magnification, (B) shows cross-sectional SEM image of XDA/CS/BG/HA/lawsone.

Figure 4 .
Figure 4.The results of the adhesion strength test: (A, B) Images of the Tape test before and after Tape test, (C, D) Images showing the results of the bend test (red arrows represent the area of maximum stress), (E) Image showing the result of the pencil hardness test.

Figure 6 .
Figure 6.Contact angles comparison of bare 316L SS, XDA coated 316L SS, and XDA/CS/BG/HA/lawsone composite coating deposited on 316L SS (Digital images illustrating the presence of a water droplet on the bare surface of 316L SS, XDA coated 316L SS and XDA/CS/BG/HA/lawsone composite coating deposited on 316L SS).
indicated the solution chosen has significant impact on the resistance R values generated from the EIS data, while the capacitance (C) values can be used to accurately determine the electrochemical properties associated with the solution selected, irrespective of the solution condition.Equation 1 allows us to calculate the different elements`s equivalent capacitance values in Fig. 10C, D by combining values (Q and n) with their respective resistances.

Figure 11 .
Figure 11.(A) SEM image of composite coatings in SBF for 28 days at 25kX, (B) SEM image of composite coatings in SBF for 28 days at 10kX, (C) FTIR spectra before and after immersion in SBF for 1, 7, 21 and, 28 Days and, (D) XRD pattern for 0 Day and 28 Day, (E) EDX analysis of composite coating immersed in SBF for 28 days with Ca/P ratio of 1.54 and (F) EDX analysis of composite coating before immersion in SBF with Ca/P ratio of 0.544.

Figure 13 .
Figure 13.The proposed mechanism of biochemical action of XDA/CS/BG/HA/lawsone composite coatings to facilitate bone regeneration and to inhibit the growth of biofilm formation.

Table 1 .
EPD Parameters and the suspension composition used for the optimization of EPD process.

Table 3 .
EIS circuit elements fitted from EIS data in SBF.
formation in SBF were observed after 28 days.The slightly shifted and intensified carbonate peaks appeared at 680 and 1415 cm −1 .Moreover, broad, and more intense peaks of phosphate group appeared at 550 and 980 cm −1 .

Table 4 .
Comparative analysis of reported results with previously reported results.Adhesion tests 4B adhesion strength confirmed via tape and bend test.Adhesion suitable for orthopedic implants Cheng et al. 72 -Alginate/BG coatings, '3B' Ur Rehman et al. 62 -CHI/gelatin/Cu-doped-BG coatings, '4B' Increased adhesion due to strong polymeric network of XDA/CS and uniform particle distribution Contact angle Substrate and coating contact angles respectively; 64° ± 1° and 48° ± 4°S hi et al. 81 -CHI/HA/GO coating, 63.3° ± 1.3°L ower values of contact angle due to hydrophilic nature of constituents Surface roughness