Interfacial adhesion and surface bioactivity of anodized titanium modified with SiON and SiONP surface coatings

Abstract

Titanium (Ti) surface modification via coating technologies (plasma spraying, electron-beam deposition) has been used to enhance bone-implant bonding by increasing the rate of hydroxyapatite (HA) formation, a property known as bioactivity. Regardless the enhancement in the surface activity, the high fabrication-temperature (> 600 °C) reduces coating-implant adhesion due to thermal expansion mismatch and reduces bioactivity due to increased crystallinity in the coating. Thus, amorphous surface coatings with strong Ti-substrate adhesion that can be fabricated at relatively low temperatures are crucially needed for enhanced osseointegration. Therefore, this study aimed to enhance the Ti surface bioactivity via strongly adherent bioactive thin film coatings deposited by low temperature (< 400 °C) plasma enhanced chemical vapor deposition technique on nanopore anodized-Ti (A-Ti) surface. Two groups of coating (silicon oxynitride (SiON) and silicon oxynitrophosphide (SiONP)) were deposited on anodized Ti and tested for interfacial adhesion and surface bioactivity. TEM and XPS were used to investigate the interfacial composition while interfacial adhesion was tested using nano-indentation tests which indicated a strong interfacial adhesion between the coatings and the A-Ti substrate. Surface bioactivity of the modified Ti was tested by comprehensive surface characterization (i.e., chemical composition, surface energy, morphology, and mechanical properties) and apatite formation on each surface. SiONP coating significantly enhanced the Ti surface bioactivity that presented the highest surface coverage of carbonated hydroxyapatite (HCA, ~ 40%) with a Ca/P ratio (~ 1.65) close to the stoichiometric hydroxyapatite (~ 1.67) found in bone biomineral. The HCA structure and morphology were confirmed by HR-TEM/SAED, XRD, FT-IR, and HR-SEM/EDX. MSCs in-vitro studies indicated preferable cells adhesion and proliferation on the modified surfaces without any cytotoxic effects. This study concluded that the improved surface bioactivity of Ti-SiON and Ti-SiONP coatings suggests their potential use as strongly adherent bioactive surface coatings for Ti implants.

1. Introduction

Commercially pure Ti (CP-Ti) was first introduced as an implant in 1965. Since then Ti and its alloys have been adopted broadly for orthopaedic, dental, and craniofacial applications [1,2] Ti has unique properties such as biocompatibility, favorable response of tissues on its surface, absence of allergic reaction, high corrosion resistance, and specific mechanical properties that make it a suitable metal for use as a bone implant [3–5]. But Ti implants under normal manufacturing steps lead to an oxidized, contaminated, stressed, plastically deformed, and non-uniformed surface native oxide layer which introduce many limitations and require further surface modification. One of the most common limitations of Ti-implants is aseptic loosening which is known as the failure of the bond between the implants and the bone due to poor osteointegration [6–8]. Also, metallic debris from Ti alloys induce aseptic loosening that results in severe osteolysis [8]. As a result, 75% of implant failures occur due to aseptic loosening and impaired implant fixation. Thus, various surface modification techniques have been applied to improve its biological, chemical, and mechanical properties to improve bone-implant bonding. These techniques include mechanical treatment (e.g., sand blasting), surface coatings (e.g., thermal spraying, sol-gel, plasma deposition, and ion beam-assisted deposition), chemical treatment (e.g., acid etching), and electrochemical treatment (e.g., anodic oxidation) [9–11]. Anodic oxidation (anodization) of the A-Ti is one of the most successful surface treatments over the last few decades [12]. It allows the formation of an anodic titanium oxide (ATO) layer with interconnected nanopore structure on the surface. This controlled growth layer is much thicker than the natively formed oxide layer on the Ti surface that can enhance corrosion resistance and biocompatibility [13]. The choose of surface modification technique depend on the understanding that the implant surface plays an extremely important role in the onset biological reactions between the artificial material surface and the biological environment. Improving implant bioactivity significantly improved the osseointegration with enhanced bone-implant contact and mineralization leading to rapid bone regeneration [14]. Thus, there is a crucial need for optimal surface modification that results in bioactive surface with rapid and uniform surface apatite nucleation for good bone formability on the implant’s surface.

Deposition of bioactive materials on the surface of Ti is of great interest due to the potential to improve implant bioactivity and bone-implant bonding. In this regard, calcium phosphate or HA based coatings have proven an elevated potential to enhance the Ti surface activity and bone regeneration capability. However, these coatings suffer from poor interfacial adhesion with the metal substrate due to the lack of miscible interface between the ionic Ti/TiO2 and the covalent structure of calcium phosphate layer [15–18]. Most recent is the plasma enhanced chemical vapor deposition (PECVD) of amorphous silica-based coatings (Si-O-N-P system) that has been developed by Varanasi and colleagues as potential bioactive coatings that adheres well to the implant surface, releases Si⁺⁴ to enhance osteogenesis, and forms surface HCA for collagen mineral attachment [15,19–21]. Our previous studies on these PECVD coatings indicated that Si-O-N-P system chemistry can be described according to a random mixing model (RMM) or random bonding model (RBM) depending on the oxygen and nitrogen content in each coating [21, 22]. In the RBM, oxygen and nitrogen incorporate in the silica tetrahedral network to form nonstoichiometric Si_wO_xN_y. Thus, oxygen-rich films were found to exhibit random bonding of nonstoichiometric Si_wO_xN_y when (x+y=4) and Si_wO_xN_yP_z when (x+y+z=4) for SiON and SiONP systems, respectively [22]. On the other hand, nitrogen-rich films exhibited random mixing of [Si−Si]_w−[Si−O]_x−[Si−N]_y for Si_wO_xN_y and [Si−Si]_w−[Si−O]_x−[Si−N]_y−[O-P]_z for Si_wO_xN_yP_z system [15,19,23]. The RMM coatings allow hydrogen bonding on the surface such as Si-H, N-H, and Si-O-H which is favorable for enhanced bioactivity. PECVD coatings have remarkable advantages compared to other coating methods [20]. This method requires a relatively low temperature (~250–400°C) that prevents mismatch between the coating and the substrate materials. PECVD coating can efficiently form a stable amorphous layer on an underlying metal surface. These new biomaterial coatings enhance osteogenesis via antioxidant and osteogenic marker expression [19]. The unique chemical structure of these amorphous PECVD SiON or SiONP is supposed to provide a hydrophilic surface with high surface free energy and readily available functional groups (e.g., Si-H, N-H, and Si-OH) once immersed/implanted in a physiological environment [20,24]. The presence of such a surface with its unique chemical structure could initiate the formation of apatite-like bone structure leading to rapid bone healing by supporting both osteoconductivity and osseointegration. However, we have not fully understood the nature of the coating-implant interface and affect these coatings have on HA formation verses bare Ti implants to gain improved understanding on how these coatings can potentially improve implant bioactivity and interfacial adhesion.

Thus, the aim of this study is to characterize the interfacial adhesion and surface activity of anodized titanium modified with SiON and SiONP PECVD surface coatings. SiON and SiONP coated A-Ti will be comprehensively analyzed and compared to clinically relevant anodized Ti fixation devices. Interfacial composition and adhesion between the coatings and Ti substrate as well as the modified surface bioactivity via rapid and uniform HA nucleation will be tested. The goal of this study is to show that SiON and SiONP coatings enhance the surface bioactivity verses bare implant surface while providing strong adhesion to the underlying Ti/TiO₂ surface. The deposited thin films in this study are RMM type and will be abbreviated as SiON and SiONP for simplicity.

2. Materials and methods

2.1. Study design

This study was performed in three consecutive sections to investigate our proposed hypothesis. The anodized-Ti surface was modified by depositing 300 nm thick bioactive amorphous SiON and SiONP thin film coatings. Then, surface topography, chemical structure, interfacial composition and adhesion, wettability and surface energy, and mechanical properties of the SiON- and SiONP-Ti modified surfaces were compared to an A-Ti surface. Second, cell-free in-vitro studies (i.e., only cell culture medium without cells) were performed to test the surface activity and hydroxyapatite formation on SiON- and SiONP-Ti modified surfaces compared to A-Ti. This step was performed by immersion of all samples in alpha modified essential medium (α-MEM) for 12 hours and 7 days. After each time-point, samples were removed and dried for comprehensive analysis using high resolution-scanning electron microscopy (HR-SEM) coupled with energy dispersive X-ray (EDX) analysis, optical profilometer scanning, FT-IR analysis, and transmission electron microscopy. In the third section, an in-vitro study using mesenchymal stem cells (MSCs) was carried out to investigate the cytotoxicity of SiON- and SiONP-Ti modified surfaces and their effect on cell attachment after 24 and 48 hours compared to A-Ti. An MTS cell proliferation assay was performed for quantitative analysis of the surface cytotoxicity, and a live/dead cytotoxicity assay was used for qualitative and semi-quantitative analysis of MSC adhesion on the tested surfaces. Many of the methods below are brief summaries of detailed methods given in our prior work [21,25–27]. We provide added detail below relevant to this manuscript.

2.2. SiON and SiONP thin films deposition

A-Ti plates (grade II, ASTM F67) were provided by KLS Martin Group (Mühlheim/Donau, Germany). According to the manufacturer, the surface of the provided A-Ti plates was processed by vibration grinding and glass bead blasting, and finally finished by anodic oxidation. In this study, A-Ti plates were used as substrate materials and controls. SiON and SiONP were deposited as thin film coatings on A-Ti surfaces following our prior published protocol [20,28]. A low temperature PECVD technique was used to deposit a 300 nm thin film of either SiON or SiONP coatings on the A-Ti substrate. A-Ti plates’ surfaces were cleaned using oxygen plasma cleaning through March PX-500 Plasma Asher (March Plasma Systems Inc., California, USA). The cleaned A-Ti samples were loaded into a TRION ORION II PECVD System (TRION Technology, Florida, USA) to deposit 300 nm of SiON or SiONP thin film on the A-Ti as substrates. The deposition was performed in four sequential steps: first, the chamber was purged with argon gas at a flow rate of 250 sccm for 30 seconds to ensure adequate chamber cleaning and removal of any possible dust on the sample’s surface. Second, a conditioning step was performed in which all required gases being used for deposition were run for 30 seconds to stabilize the gas flow rate and prepare the chamber for specific thin film chemistry. Third, a deposition step was performed in which all required gases were run at a specific flow rate (Table 1) with ICP power of 75 W and RIE power of 30 W. Finally, a cleaning step was carried out using argon of 250 sccm for 180 seconds to prepare the chamber for the next run. Before the process, the lower electrode was uniformly heated to 400°C. The chamber pressure was held at 900 mTorr and the excitation frequency was 13.56 MHz during all processing steps. Silane (SiH₄) and phosphine (PH₃) gases were used as sources of silicon and phosphorus, respectively, and both were diluted in argon (Ar) (15%SiH₄/2% PH₃/85%Ar). Nitrous oxide (N₂O) was used as the source for oxygen and ammonia gas was used as the source for nitrogen. The exact flow rates of each gas in sccm, the deposition rate of SiON and SiONP, and the refractive index are reported in Table 1.

Table 1.

Gas Flow Rates, deposition rate, and refractive index for Silicon Oxynitride (Si-O-N) and Silicon Oxynitrophosphide (Si-O-N-P) Layers Deposited by PECVD.

Sample	Gas flow rate (sccm)				Deposition Rate (nm/min)	Refractive index (n)
	SiH4/PH3/Ar	N2O	N2	NH3
SiON	24	3	225	50	41.0	1.82
SiONP	24	16	225	50	38.5	1.68

2.3. Surface characterization

2.3.1. Surface morphology and chemical composition

Scanning electron microscopy (SEM, S-3000N, Hitachi, Japan) coupled with energy dispersive X-ray spectroscopy (EDS) was used to study the surface microstructure, morphology, and chemical composition before and after surface modification. SEM working distance was set to 15 mm with accelerating voltage of 20 kV and various SEM images were captured at different magnifications to compare the surface morphology before and after the addition of thin film coatings. Ultra HR-SEM (Hitachi S-4800 II FE SEM; Hitachi) was used to capture the nanostructure of the surfaces before and after coating. Images were acquired at a working distance of 10 mm under 20 kV at different magnifications. For the EDX analysis, the accelerating voltage was held constant at 15 kV and areas of interest were scanned for composition analysis. FT-IR was also used to reveal the surface chemical structure and indicate the surface functional groups after surface modification with SiON and SiONP compared to uncoated A-Ti. The interfacial analysis of Ti-SiONx coatings was performed using X-ray photoelectron spectroscopy (Thermo Scientific K-Alpha XPS, MA, USA) and transmission electron microscopy (TEM, Hitachi HF-3300, Japan) at the Center for Nanophase Materials Sciences (CNMS) at Oak Ridge National Laboratory, Oak Ridge, TN (ORNL). Monochromatic Al-K alpha source with an energy of 1486.6 eV and Ar⁺ beam sputtering operated at an energy of 4.2 keV was used for XPS analysis. Samples were prepared using a Hitachi NB5000 dual-beam SEM focused ion beam (SEM-FIB) onto Cu grids prior to imaging by TEM.

2.3.2. Surface roughness

An optical profilometer (Wyko NT9100, Veeco Instruments Inc., USA) was used to capture the surface topography and measure the surface roughness parameters according to surface texture parameters in ISO 25178 standards. A large area (310 μm * 232.5 μm) of the samples’ surface was scanned via non-contact mapping in three dimensions (3D) and 9 measurements per each sample were captured for analysis. As suggested by Wennerberg and Albrektsson, the 3D surface roughness parameters are superior compared to the 2D measurements [29]. Thus, 3D surface roughness measurements were performed, and the preferred height, hybrid, and features parameters were reported. For height, the arithmetic mean height (Sa) and root mean-square height (Sq) were estimated for all samples. Sa expresses the difference height of each point compared to the arithmetical mean of the surface, while the S_q indicates the root mean square value of ordinate values within the definition area [30,31]. Measured hybrid parameters are root mean square gradient (S_dq) and developed interfacial area ratio (S_dr) which represent the root mean square of slopes at all points in the definition area and the developed surface area ratio, respectively. Finally, feature parameters included the Summit density (S_ds) which counts the number of summits per unit area of the surface, and the Mean Summit Curvature (S_sc) which is an indicator of the shape and size of the higher areas of a surface. Surface roughness parameters were used to compare the A-Ti surface before and after modification with SiON and SiONP coatings as well as after HA deposition after 12 hours and 7 days in α-MEM.

2.3.3. Wettability and surface energy

Surface wettability was assessed using contact angle measurements. Surface contact angles were measured using 250-U4 Goniometer/Tensiometer with a super speed German-made camera and DROP-image Advanced software (Ramé-hart instrument co., NJ, USA). The sessile ´ drop technique was used to measure the surface contact angle of all samples. Deionized water and diiodomethane (99% purity, Sigma-Aldrich Co., St. Louis, MO, USA) were used as probing liquids to measure the contact angle of each sample at 25°C. For each coating, three samples were tested with 3 repetitive drops of DI water and diiodomethane on each sample for a total reading of nine measurements per coated sample as well as A-Ti samples. Contact angle images were captured on each surface using a high-resolution camera within 10 seconds of droplet seeding. Using the contact angle results, the free surface energy of each sample was calculated using the Owens-Wendt-Kaelble (OWK) equation [32]. The OWK Eq. (1) relates the contact angle (ϴ) of a solid surface to its total surface free energy, broken into its polar and dispersive components [32].

$${\gamma }_{LV}(1+\cos \theta )=2\left(\sqrt{{\gamma }_L^d{\gamma }_s^d}+\sqrt{{\gamma }_L^p{\gamma }_s^P}\right)$$ (1)

Where γ_LV represents the total surface free energy of a solid surface, and S and L represent the solid and liquid phases, respectively. γ^d and γ^p refer to the dispersive and polar surface tension components. The values of surface energy and its corresponding dispersive and polar components of deionized water and diiodomethane were previously reported [32] as shown in Table 2. These values were used in our calculations.

Table 2.

Surface energy and surface tension components of water and diiodomethane.

Liquid	${\gamma }_L^d$ (mJ/m2)	${\gamma }_L^p$ (mJ/m2)	γLV (mJ/m2)
Water	21.8±0.7	51.0	72.8
Diiodomethane	49.5	1.3	50.8

2.3.4. Mechanical properties

Vickers hardness and nanoindentation tests were performed to investigate any changes in the mechanical properties of the A-Ti compared to the SiON and SiONP modified surfaces. Vickers microhardness was measured using a LECO LM 300 AT Micro Hardness Tester (LECO Co., MI, USA) operated at a load of 500g for 10 seconds and 50x objective was used to visualize the indent after each indentation. After Vickers hardness testing, the samples surfaces were scanned using HR-SEM to capture the indentation dents on each surface. The two diagonals (d1 and d2) of the diamond shaped indentation were measured and the hardness (H_V) was determined using the following equation [33]:

$$H_V=0.1891\ast F/d^2$$ (2)

where F is the applied load, and d is the average of indentation diagonals. The H_V is given in kgf/mm² which can be converted into GPa by multiplying the previous equation by 0.009806. For nanoindentation, a 3600 μm² surface area was scanned using scanning probe microscopy (SPM) and the area of interest was indented with a Berkovich tip using a Hysitron Ubi-1 Nanoindenter (Hysitron, Minneapolis, MN). For each sample, 12 nanoindentations were performed on the selected area according to the following loading function: ramped up to 10 mN at a rate of 250 μN/s, held for 10 seconds, and then unloaded at 250 μN/s. The unloading slope of the load-displacement curve was used to calculate the reduced elastic modulus (E_r) and the hardness (H) based on the Oliver-Pharr method [34,35]. E_r can be defined as the elastic modulus that contains the elastic contributions of the specimen and the diamond indenter tip according to the following equation [36]:

$$\frac{1}{E_r}=\frac{1-v_s^2}{E_s}+\frac{1-v_i^2}{E_i}$$ (3)

here, E_s and E_i indicate the elastic modulus of the sample and the indenter tip, while ν_s and ν_i represent the Poisson’s ratio of the sample and the indenter tip, respectively. Then, the E_r can be calculated using the slope of the linear part of the unloading curve (S) and the projected contact area of the Berkovich tip (A_c) according to Eq. (4) [34,35]. The projected contact area of the indenter tip was determined by calibrating the tip at various indentation depths with fused quartz.

$$E_r=\frac{\sqrt{\pi }}{2}\frac{s}{\sqrt{A_C}}$$ (4)

Then, the hardness can be defined as the maximum applied load (P_Max) divided by the projected contact area of the indenter tip (A_c) as follows [34]:

$$H=P_{Max}/A_c$$ (5)

Nano-scratch tests were performed to investigate the interfacial adhesion strength of PECVD thin film coatings on the A-Ti surface. Briefly, the indenter tip was dragged horizontally on the surface while the vertical force is increased simultaneously. This coupled motion can detach the thin film coatings from the substrate if the adhesion strength of the films was exceeded. The nano-scratch tests left behind a scratch mark on the surface that could be visualized using scanning probe microscopy (SPM). A Hysitron Ubi-1 Nanoindenter (Hysitron, Minneapolis, MN) was used to perform nano-scratch tests on the PECVD coatings. Load controlled nano-scratch tests were performed on the surface using the following load function: the normal load started from 0 mN and gradually increased until load peak of 10 mN over 4 μm horizontal displacement for 30 seconds. SPM topographical 2D post scratch images were captured to ensure successful scratches. Lateral force versus normal force, lateral displacement verses normal force and lateral forces, and normal displacement versus time curves were extracted and analyzed to investigate film adhesion.

2.4. In-vitro cell free studies “surface apatite formation”

To compare the bioactivity of the SiON and SiONP-A-Ti coatings to the A-Ti, in-vitro cell free studies were performed at two time-points (12 hours and 7 days). For each time-point, three samples of each group (i.e., A-Ti as a control, SiON-Ti, and SiONP-Ti) were cleaned in 100 % pure ethanol followed by two rinses with phosphate buffered saline (PBS-1x) and finally immersed in α-MEM in sterile tissue culture plates. Then, samples were incubated in 37°C for 12 hours and 7 days. The α-MEM solution was refreshed every 48 hours to mimic the physiological turnover of body fluids. At each time-point, samples were washed twice with PBS-1x and dried overnight at 37°C. Then samples were used for comprehensive analysis. It is important to mention that α-MEM is commonly used in cell culture studies and its composition is more similar to the concentration of salts in human blood plasma, which makes the α-MEM a more suitable solution compared to the SBF solution [37]. As well as α-MEM is an important source for passive ion exchange with the bioactive material surfaces, thus it can be used to investigate the biomaterials response to the blood plasma and the cell culture media environment.

The HA and mineral deposition on the surfaces were confirmed and studied using various techniques, including SEM/EDX followed by ImageJ software analysis, high resolution transmission electron microscopy (HR-TEM), X-ray diffraction, optical profilometer, and FT-IR analysis. SEM/EDX analysis was used to scan the surface morphology after HA formation on the surface for 12 hours and 7 days in α-MEM. EDX data was used to investigate the chemical composition and calculate the calcium to phosphate ratio (Ca/P) to compare the formed HA on each sample to the Ca/P ratio of the standard hydroxyapatite. SEM images were further used to quantitively investigate the HA surface area coverage using ImageJ software [38]. The HA formation was further confirmed by FT-IR analysis. HR-TEM (H-9500 HR-TEM, Hitachi, Japan) with selected area electron diffraction (SAED) was used to image the HA crystal planes and diffraction patterns. This diffraction pattern from HR-TEM was further confirmed by XRD analysis.

2.5. In-vitro cytotoxicity studies

In-vitro cytotoxicity studies were performed according to ISO 10993-5 “biological evaluation of medical devices-part 5: tests for in-vitro cytotoxicity”. MSCs in-vitro cell culture studies were used to investigate the surface effect on cellular adhesion, cytotoxicity, and proliferation. Male human bone marrow MSCs were obtained from Lonza (Lonza Walkersville Inc., MD, USA). According to the provider, all cells are authenticated, performance assayed, and tested negative for mycoplasma. Cells were allowed to grow, and viability and morphology were measured after recovery from cryopreservation. For each experiment, cells were cultured in MSCGM BulletKit^™ (Lonza, PT-3238 & PT-4105) specific growth medium. Cells were maintained at 37°C and 5% CO₂ in a completely humidified incubator. Cells were seeded and allowed to grow on the SiON- and SiONP-Ti modified surfaces as well as the A-Ti surface as a control with n=3 samples for each group. Cytotoxicity tests were performed after 24 and 48 hours of cell seeding. The MTS assay CellTiter 96® AQueous One Solution Cell Proliferation Assay, (Promega, Madison, WI, USA) was used for quantitative analysis of cell growth and proliferation. A LIVE/DEAD^™ Viability/Cytotoxicity Stain Kit was purchased from Thermo Fischer Scientific Inc. (Waltham, MA, USA) and used for qualitative analysis of cytotoxicity on the tested surfaces. All cell culture studies and assays were performed according to our previously published protocols[25]. The colorimetric absorbance of the MTS assay was determined using a microplate reader (SpectraMax® i3, Molecular Devices, CA) at 490 nm. Live/dead fluorescent images were then taken using a DMi8 inverted Leica microscope (Leica Microsystems Inc., IL, USA), with green staining for live cells and red staining for dead cells.

2.6. Statistical analysis

OriginPro 8.5 software was used for all graphs and statistical analysis, bar graphs display group means and standard deviations. A one-way ANOVA followed by Tukey’s post hoc was used for between group comparisons. For significance level, P < 0.05 was considered as statistically significant, with * representing p < 0.05, ** representing p < 0.01, and *** representing p < 0.001. For in-vitro studies, a minimum of three replicates at two separate time points were used for each experiment per each group of samples according to ISO10993-5.

3. Results and discussion

3.1. Surface modification and characterization

After surface modification, comprehensive surface analysis was performed to critically study the effect of PECVD coatings on the micro- and nano-structure, surface functional groups, surface wettability and surface energy, surface roughness, and micro- and nano-mechanical properties compared to the A-Ti. The physical appearance and microstructure of medical grade A-Ti locking plates, as received, and after surface modification with SiON- and SiONP-coatings, are shown in Figure 1. Due to the importance of visual identification of Ti plates and screws for orthopedic surgeons in the medical field, it was important to visually differentiate the A-Ti plates and the PECVD coated plates based on their colors. The A-Ti plates have a green color (Figure 1-A), while SiON-Ti plates have a gray color (Figure 1-B), and SiONP-Ti plates are dark green (Figure 1-C). The microstructure of each sample, as scanned by SEM at different magnifications, is shown in Figure 1(D–I). Low and high magnification SEM imaging of the coated plates revealed that SiON and SiONP PECVD coatings have no effect on the surface topography or morphology. Figure 1 (G–L) clearly shows the microstructure and surface roughness profile that can be attributed to the glass bead blasting prior to anodic oxidation of the surface as mentioned by the manufacturer. Glass bead blasting is an effective technique for increasing surface roughness and enhancing the coating-substrate interaction by reducing the native oxide layer on the substrate surface [39]. The surface roughness parameters were measured by scanning a large surface area of each sample using an optical profilometer as shown in Figure 1 (J–L). Comparison of the surface roughness parameters (Table 3) of A-Ti (S_a = 471± 18 and S_q = 627 ± 53) to the SiON surface (S_a = 495 ± 16 and S_q = 664 ± 21) and SiONP (S_a = 575 ± 16 and S_q= 742 ± 18) indicated that there was no significant difference in surface roughness after the SiON coatings, while significant increase in the surface roughness was observed after the SiONP coatings. This increase in the surface roughness on SiONP coated implants could facilitate and enhance the cell adhesion on the implant surface.

Fig. 1.

Visual identification (A-C) and HR-SEM images (D-I) compare the bare anodized-Ti to the SiONx and SiONPx PECVD coated plates. Microstructure confirms no topography changes after the coating. J-L) Optical profilometer 3D images show the surface roughness of the bare A-Ti (J) compared to the coated surfaces (K & L).

Table 3.

Surface roughness parameters of A-Ti, SiON-Ti, and SiONP-Ti before and after HA formation on the surface.

Sample	Time-point	3D Surface Roughness Parameters
		Height Parameters		Hybrid Parameters		Feature Parameters
		Sa (nm)	Sq (nm)	Sdq (°)	Sdr	Ssc/μm	Sds/μm2
A-Ti	0 h	471±18	627±53	57.45±1.1	89.3±5.6	9.09±0.3	0.0725±0.005
	12 h	481±30	632±22	57.42±1.3	93.12±8.5	9.367±0.39	0.0675±0.005
	7 days	589±22	732±21	64.82±1.1	154.1±17	11.75±1.2	0.0875±0.009
SiON-A-Ti	0 h	495±16	664±21	57.88±1.2	93.75±8	9.29±0.44	0.070±0.0
	12 h	592±31	768±23	63.83±1.6	151.1±18	12.50±10	0.0775±0.005
	7 days	720±15	910±14	72.05±0.17	339±16	20.56±0.2	0.085±0.005
SiONP-A-Ti	0 h	575±16	742±18	64±0.82	155±10	13.05±0.79	0.073±0.004
	12 h	619±54	821±63	67.99±0.88	219±16	16.31±0.8	0.08±0.008
	7 days	831±71	1019±87	68.25±1.2	217±22	15.1±0.84	0.0825±0.005

Ultra HR-SEM images in Figure 2 present the microstructure and surface topography of the A-Ti compared to the SiON and SiONP PECVD coatings at different magnifications. The A-Ti surface indicates the presence of an ATO layer with nanopore structure on the surface as shown at Figure 2 (A & D) and confirmed by EDX (Figure 2-G). This ATO layer was developed during the surface anodization and displays diversification of nanopores with sizes from 150 to 350 nm. This layer enables specific advantages on the Ti surface such as increasing surface area, large pore volume, and uniform pore size distribution [40]. Previous studies have also indicated that the formation of an active ATO layer on the surface of Ti improves the biocompatibility [41], enhances the HA formation, and strengthens the interfacial strength between Ti substrate and surface coatings [42]. Here, the microstructure investigation of the SiON (Figure 2-B &E) and SiONP (Figure 2-C &F) PECVD films on the A-Ti revealed the formation of thin film coatings with cell-like surface features. The formation of cell-like features of the deposited thin films indicates that the ATO layer nanopores could work as anchors and deposition sites to support thin film adhesion on the Ti substrate. Furthermore, these nanoporous features of ATO could facilitate the deposition and adhesion of thin films by working as pinholes, ensuring mechanical interlocking between the Ti substrate and the deposited thin films. This behavior has been confirmed in previous studies by Jin et al. [43] and Zhou et al. [40] which indicated that nanopores, with 200 nm pore diameters, were effective as bonding pinholes for nanostructured grain growth with strong interfacial adhesion. The chemical composition of each sample was investigated using EDX composition analysis (Figure 2 G–I) and the atomic percentage (at %) of each sample is reported at Table 4. The A-Ti compositional analysis revealed the presence of oxygen as well as phosphorus (P) at low % of 0.512±0.07 which can be attributed to the anodic oxidation of the surface. On the other hand, SiON and SiONP PECVD coated samples indicated the presence of Si, O, and N on both surfaces plus P with a % of 3.47±0.36 on the SiONP surface.

Fig. 2.

Ultra HR-SEM images (A-F) compare the anodized-Ti to the SiONx and SiONPx PECVD coated plates. A&D) Confirm the presence of an anodic titanium oxide (ATO) layer with nanopore structure. B-F) Show the cell-like nanostructure after the SiONx and SiONPx deposition on the anodized-Ti. G-I) EDX compositional analysis spectra showing the exact chemical composition of each surface.

Table 4.

Atomic percentage (At%) shows the elemental composition of A-Ti, SiON-Ti, and SiONP-Ti as reported from EDX analysis.

at (%)	N	O	Si	P	Ti
Pure-Ti	—	21.7±1.9	—	0.512±0.07	77.04±2.8
SiON	19.1±3	9.9±1.38	41.1±2	—	29.9±2
SiONP	14.48±1.8	6.035±2	43.59±2.5	3.47±0.36	32.41±2

The surface of A-Ti, SiON-Ti, and SiONP-Ti was further investigated to study the surface functional groups as presented from the FT-IR spectra (Figure 3). The A-Ti spectra indicates the presence of a broad strong band in range 550–900 cm⁻¹, which can be attributed to the vibrational modes of the Ti-O bond from the ATO surface layer [44,45]. For SiON-Ti and SiONP-A-Ti, the peaks near 800 cm⁻¹ represent the Si-O-Si [46], while the modes at 950 cm⁻¹ can be attributed to the Ti-O-Si vibration [47]. Bands at 842 cm⁻¹ represent the Si-N stretching mode [48], while bands at 1175 and 3330 cm⁻¹ are attributed to the bending and stretching vibration modes of N-H, respectively [49]. Bands at 953 cm⁻¹ on PECVD SiON-Ti and SiONP-Ti surfaces are attributed to the presence of Si-OH stretching modes [50]. These bands were observed on the surface of PECVD coatings due to the readily available silanol groups on the oxide surface, which are commonly observed for low temperature PECVD [28]. It is important to mention that deposition of amorphous silica based thin films brings the unique surface properties of Si₃N₄, SiON, and SiONP to the surface of reconstructive fixative materials (e.g., titanium plates). These bioactive amorphous coatings with surface multi-functional groups (i.e. Si-O, Si-N, Si-H, and N-H) are partially soluble under physiological conditions because N-H and Si-H bonding allow immediate dissolution/degradation in-vitro [19–21,26, 28]. As a result, these bioactive surfaces enhance cell adhesion, protein adsorption, and rapid mineralization leading to early bone regeneration on the implant surface [15,19,23]. Furthermore, these coatings also promoted sustainable/optimized release of Si⁺⁴ for enhancing osteogenesis, angiogenesis, and myogenesis [15,20,23,25,51,52].

Fig. 3.

FT-IR spectra of the anodized-Ti compared to SiONx-Ti and SiONPx-Ti coatings.

Compositional analysis of the SiON layer and the Ti-TiO₂-SiON interface was performed using HR-TEM and XPS by milling through the thickness of the surface coatings (100 nm) reaching the Ti substrate. Figure 4-a presents the cross-section micrographs of the interface which indicated amorphous SiON layer on the top of the TiO₂ layer and the crystalline Ti substrate. The elemental composition of the coating as a function of the deposition depth is presented in Figure 4-b. This identifies the composition in all layers including regions A (amorphous SiON layer), region B (TiO₂ layer), and region C (Ti crystalline substrate). Binding energy spectra of the individual elements present in these regions is presented in Figure 4-c. Si 2p analysis indicated the predominate of Si-O bonding at the coating layer (region A), N 1s analysis indicated the presence of N-H and Si-N in the coating layer. The O 1s analysis indicated the presence of both O-Ti and O-Si bonding at the interfacial layer (region B). Also, Ti 2p indicated the presence of Ti-O bonding at region B while Ti-Ti predominant at region C. This results indicate the presence of SiO₂ and TiO₂ at the interface layer providing cross-bonding between the coating and the substrate as presented in Rxn 4. This interfacial bonding leads to strong adherent coatings which agrees with previously published data [21,28]. These results also agree with the SEM/EDX, and FT-IR data presented above.

Fig. 4.

TEM and XPS analysis of the Ti-SiONx coating interface. A) TEM micrographs magnify the Ti-TiO₂-SiONx interface. B) XPS depth scale determined by sputter rate of 15 nm/min, sputter cycles were 30 sec and shaded-areas show region that had different chemical signals. (C) Individual XPS spectra from these regions are compared for Si, O, N, and Ti. Note that N 1s signal are magnified by a factor of 10.

Surface mechanical properties of the modified surfaces were evaluated to study any change after the SiON and SiONP thin films deposition compared to the A-Ti surface. Figure 5 and table 5 present the micro- and nano-mechanical properties as determined by Vickers and nanoindentation tests, respectively. Based on the Vickers test, hardness of the A-Ti was 198.1±6.8 HV (~1.92±0.07 GPa), which was close to the hardness value (1.97±0.4 GPa) as estimated from load-displacement curves after nanoindentation testing (Figure 5-B). Vickers indents “diamond-like shape” as scanned by HR-SEM after the indentation test is shown in Figure 5-A. The Vickers hardness did not reveal any significant changes after SiON and SiONP deposition, while the nano-hardness of SiONP (6.7±0.99 GPa) was significantly higher compared to both A-Ti (1.97±0.4 GPa) and SiON (3.05±1.1 GPa) with a significance level of ***p<0.001. The difference in micro- and nano-hardness values is attributed to the indentation depth that exceeds the thin film depth and reaches the substrate with Vickers test compared to the nanoindentation that did not exceed the thin films thickness. The reduced elasticity modulus, calculated from load-displacement curves, of A-Ti was 139.7±26 GPa which slightly increased after SiON and SiONP thin films deposition to 147±15 and 160±24 GPa, respectively, with no significant difference compared to the A-Ti. The maximum indentation/penetration depth (h_Max) of each surface can be seen from the load-displacement curves in Figure 5-B and is reported in table 5, which is in a reasonable correlation with the hardness of each material. The reported H and E_r values in this study are in a reasonable agreement with previously reported values [34,53–55].

Fig. 5.

Mechanical properties of the anodized-Ti compared to SiONx-Ti and SiONPx-Ti coatings. A) HR-SEM images show the Vickers indents “diamond-like shape” after the indentation test. B) Load-displacement curves of nanoindentation for the A-Ti compared to PECVD coatings. C-F) Nano-scratch test analysis for 300 nm PECVD thin-film on the anodized-Ti surface. C) Lateral vs. normal forces curve with maximum scratch depth of 400 nm, image insert shows the 2D SPM topography after the scratch test. D) Normal displacement vs. time. Lateral displacement vs. normal force (E) and vs. lateral force (F).

Table 5.

Mechanical properties of A-Ti, SiON-Ti, and SiONP-Ti as calculated from the Vickers hardness and nano-indentation tests.

	Vickers hardness		Nano-indentation
	Hardness (HV)	Hardness (GPa)	Reduced modulus (GPa)	Hardness (GPa)	hMax (nm)
A-Ti	198.1±6.8	1.92±0.07	139.7±26	1.97±0.4	440±31
SiON-A-Ti	199.9±4.9	1.95±0.05	147±15	3.05±1.1	302±43
SiONP-A-Ti	206.1±4.2	2.01±0.04	160±24	6.7±0.99	224±30

Nano-scratch test results are shown in Figure 5 (C–F). In the load-controlled scratch tests, the normal force is gradually increased to the maximum applied load. Thus, at low force at the beginning, the deformation is considered elastic, which then turns to plastic deformation at the applied high force. This significant increase in the applied force may result in thin film cracking and delamination. Normal force versus lateral force or lateral displacement curves are usually used to investigate thin films interfacial adhesion, cracking, fracture, and delamination [40,56]. Here, there was no significant change in resistant “lateral force” with increasing vertical load or lateral displacement which confirms strong PECVD thin films adhesion with the underlying A-Ti as shown in Figure 5-C & F. The slight change in the resistance (Figure 5-C) can be attributed to the hardness change within the layers. This was further confirmed by normal displacement versus time curves that indicated slope changes without significant reductions in normal displacement, indicating no breakthrough of the thin films. The change around 300 nm can be explained by the indenter tip resistance present while transitioning from the thin film to the substrate. Also, the slope of the normal force-lateral displacement curve did not indicate any sudden change which means no cracking or significant fracture was observed as shown in Figure 5-E. The results from the nano-scratch tests indicate a strong interfacial adhesion between PECVD thin films and the A-Ti substrate which can be attributed to the ATO layer nanopores that work as anchors and deposition sites to support thin film adhesion on the Ti substrate. Moreover, the advantages of using low temperature PECVD to deposit thin films without thermal expansion mismatch and well-adherent layers are demonstrated by these results [28].

Activation of the Ti surface by deposition of such bioactive coatings with its multi-surface functional groups is hypothesized to enhance the surface wettability and increase the surface free energy. Thus, surface wettability of the modified surfaces was tested by contact angle measurements followed by surface energy calculations. Figure 6 presents the contact angle and the corresponding surface free energy of each coating compared to the A-Ti. The water contact angle on the A-Ti surface was found to be 65.65 ± 3.6°, which agrees with previously reported data [11,57,58]. Deposition of SiON/SiONP thin films on the A-Ti surface significantly enhances the surface wettability as indicated by the significant decrease in the water contact angle compared to the A-Ti (***p<0.001). The water contact angle was found to be 43.45 ± 0.7° and 41.65 ± 0.95° for SiON and SiONP, respectively. The contact angle of the diiodomethane displayed the same behavior as DI water (Figure 6-A & B). The surface free energy of each coating as calculated according to OWK equation (1) broken into its polar and dispersive components is shown in Figure 6-C. The A-Ti surface energy was 37.92 mJ/m² which significantly increased after SiON and SiONP deposition to be 56.01 and 57.78 mJ/m², respectively. Surface wettability is a crucial aspect for all medical implants. Implants with enhanced hydrophilicity and low contact angles (<90°), will present high surface energy leading to rapid and improved cell adhesion and protein adsorption within the first 6 hours of cell culture in-vitro [20,59]. Furthermore, increasing surface free energy enhances the calcium phosphate deposition leading to faster HA nucleation in SBF [60], enhanced osteoprogenitor cell adhesion, and increased collagen and extracellular matrix production [61].

Fig. 6.

Surface wettability and energy of the anodized-Ti compared to SiONx-Ti and SiONPx-Ti coatings. A) Shows the droplet spreading on each surface as captured within 10 seconds. B) Shows the average values of the contact angles with standard deviations. C) The calculated surface free energy broken into its dispersive and polar components. (***) presents the significant level when p<0.001 compared to Ti.

3.2. In-vitro cell free “surface apatite formation”

Formation of bone-like apatite on the surface of the implantable materials is a vital requirement for successful bone growth on synthetic materials in vivo. This apatite layer adsorbs and activates signaling proteins and cells responsible for bone formation [62]. Testing the formation of CaP on the implant’s surface in-vitro gives beneficial insights on the in vivo interaction on the surface [63]. In this section, the surface bioactivity and its effect on CaP/HA formation was investigated by sample immersion for 12 hours and 7 days in α-MEM, with similar ionic composition to human blood plasma. Then, the effect of the SiON- and SiONP-Ti modified surfaces on CaP/HA formation compared to the A-Ti was comprehensively studied.

HR-SEM was used to capture the CaP formation on the tested samples after 12 hours (Figure 7) and 7 days (Figure 8) of immersion in the α-MEM. By comparing the SiON- and SiONP-Ti treated surfaces to the A-Ti, it is revealed that the treated surfaces have significantly higher surface coverage of CaP deposition as early as 12 hours and after long-term immersion for 7 days. The bright deposit on the samples’ surfaces, as shown from the SEM images Figs. 7 & 8 (A–F), present the CaP deposition as confirmed from the EDX compositional analysis Figs. 7 & 8 (J–L). Samples’ surface was again scanned by an optical profilometer after 12 hours and 7 days of immersion to reveal any surface profile changes as shown in Figs. 7 & 8 (G–I). Comparing each surface to its bare surface “before immersion” revealed that the surface topography/profile had significantly changed on the SiON- and SiONP-Ti modified surfaces as confirmed from the 3D surface roughness parameters in table 3. This significant change in the surfaces’ profile is attributed to the CaP deposition on each surface. The SEM images were further used for semi-quantitative analysis of the CaP/HA surface area coverage (%) on each sample after 12 hours and 7 days (Figure 8). Figure 8(A–F) presents the SEM images after ImageJ analysis to calculate CaP/HA surface area coverage, black color represents the background surface of each sample, and the white color indicates the HA deposition. The mean values with standard deviation of the calculated HA (%) are presented in Figure 8 (G). The modified surfaces of SiON-Ti and SiONP-Ti significantly increased the CaP/HA nucleation as early as 12 hour after immersion when compared to the A-Ti with a significant level of ***p<0.001. Furthermore, the SiONP-Ti modified surface displayed the highest HA surface coverage (~40 %) compared to the A-Ti (~10 %) and the SiON-Ti (~27 %) after 7 days as shown in Figure 8 (G). The Ca/P ratio of the formed CaP/HA was calculated from the EDX analysis and is shown in Figure 9 (H). The A-Ti presented Ca/P of 1.55±0.23 compared to 1.54±0.03 and 1.63±0.02 for the SiON- and SiONP-A-Ti, respectively, after 12 hours. These Ca/P ratios were increased to 1.6±0.03, 1.62±0.08, and 1.65±0.09 for A-Ti, SiON-, and SiONP-A-Ti, respectively, after 7 days of immersion. For both A-Ti and SiON-A-Ti, the Ca/P ratios at early stage of immersion, 12 hours, are close to the CaP ratio of the stoichiometric tricalcium phosphate (TCP, stoichiometric Ca/P ratio = 1.50), this Ca/P ratio increased after 7 days to fit more with the Ca/P ratio of stoichiometric HA. Most important is that SiONP-Ti surface indicated CaP formation with Ca/P ratio closer to the stoichiometric HA even at the early immersion time of 12 hours (1.63±0.02) and increased to (1.65±0.09) at 7 days as shown in Figure 9 (H). The results of SiONP-Ti was pronounced and revealed unique surface chemistry that significantly increased the HA surface coverage with highest Ca/P ratio closest to the stiochiometric HA at early time points. The A-Ti and SiON-Ti results are in reasonable agreement with results reported previously at early stages of in-vitro immersion [26] and in vivo implantation [64,65]. At early in-vitro and in-vivo stages, the thermodynamically metastable phase TCP is kinetically favorable and tends to form HA as immersion time increases. Previous studies indicated apatite formation with Ca/P ratio range from 1.28 to 1.46 after Ti surface immersion in SBF [66–68]. The higher Ca/P ratios reported here indicating rapid apatite nucleation is expected to lead to a faster osseointegration of the modified Ti implants.

Fig. 7.

Calcium-phosphate/hydroxyapatite formation on the surfaces after 12 hours of immersion in α-MEM. A-F) HR-SEM images show the HA formation on the anodized-Ti (A&D), SiONx-Ti (B & E), and SiONPx-Ti (C & F) at different magnifications. G-I) Surface profile and roughness as scanned by an optical profilometer after 12 hours of immersion. J-L) EDX spectra indicating the compositional analysis of the HA formed on each surface, respectively.

Fig. 8.

Hydroxyapatite formation on the surfaces after 7 days of immersion in α-MEM. A-F) HR-SEM images show the HA formation on the anodized-Ti (A&D), SiONx-Ti (B & E), and SiONPx-Ti (C & F) at different magnifications. G-I) Surface profile and roughness as scanned by an optical profilometer after 7 days of immersion. J-L) EDX spectra indicating the compositional analysis of the HA formed on each surface, respectively.

Fig. 9.

Hydroxyapatite coverage and Ca/P ratio after 12 hours and 7 days of immersion in α-MEM. A-F) ImageJ processed HR-SEM images show the HA formation on the anodized-Ti (A&D), SiONx-Ti (B & E), and SiONPx-Ti (C & F) at different magnifications, blue presents the background, and the white presents the HA deposition. G) Surface area coverage of HA (%) on each surface as calculated from the SEM images. H) Ca/P ratio as calculated from the EDX compositional analysis data.

The enhanced CaP coverage with higher Ca/P ratio on the modified PECVD SiON/SiONP-Ti surfaces can be attributed to the enhanced surface activity (i.e., wettability and surface energy) with readily available silanol groups on the silica based-PECVD thin films surfaces. In physiological environment (i.e., in-vitro or in-vivo) these silanol groups are immediately available for dissolution, polymerization, and reprecipitation to form a silica gel network for early carbonated HA formation [28]. The enhanced HA area coverage on SiONP compared to the SiON can be attributed to the incorporation of phosphorus in the structure. This phosphorus is presumed to be rapidly dissolved from the amorphous SiONP coatings concomitant with Si-ion release, leading to a relatively open structure layer of silica gel. This environment allows for ion exchange and formation of Ca-PO₄ within 6 hours, which becomes a more stable carbonated hydroxyapatite structure by consuming the Ca and P from the media/solution within 18 hours [37].

The structure of the formed CaP or hydroxycarbonate-apatite (HCA) was further investigated using FT-IR analysis as shown in Figure 10. FT-IR spectra indicate the presence of phosphate groups (PO₄³⁻) on the samples surface appeared as vibration mode at 603 cm⁻¹ and strong stretching band range from 1050–1100 cm⁻¹. Weak carbonate absorption (v3 CO₃²⁻) bands appeared at range of 1400–1550 cm⁻¹. A water absorption band was observed around 1650 cm⁻¹, while the broad band extended from 2500–3700 cm⁻¹ was assigned to the OH⁻ absorption bands. The strong peak around 870 cm⁻¹ appeared on the SiON-Ti surface can be attributed to the joint contribution of carbonate and HPO₄²⁻ ions, this peak was significantly weaker on the other surfaces. These FT-IR spectra are in a reasonable agreement with previously reported data [69,70] of HCA formation on CaSiO₃ surfaces immersed in simulated body fluid solution [70].

Fig. 10.

FT-IR spectra of HCA formation on the anodized-Ti compared to SiONx-Ti and SiONPx-Ti coatings after 7 days of immersion.

HR-TEM coupled with SAED was used to further investigate the crystal structure of the formed HCA on the surfaces after 7 days of immersion in α-MEM. Figure 11 illustrates the HR-SEM and HR-TEM images with SAED diffraction patterns as well as XRD patterns. The TEM micrographs revealed a high degree of aggregation of the HCA particles as shown in Figure 11 (B). High magnification HR-TEM images (Figure 11-C) revealed the polycrystalline nature of the formed HCA as indicated by linear lattices arranged in random directions. This was further confirmed by the analysis of regions D, E, and F that show the (002), (211), and (222) planes marked with the corresponding d-spacing. The SAED pattern (Figure 11-G) shows the Debey rings which indicates a polycrystal HCA with irregularly shaped polycrystals in an isotropic orientation. The interplanar spacing of the presented Debey rings are in a reasonable agreement with the characteristic spacing of apatite-like structure [71–74]. This results were further confirmed with XRD analysis that presented exact agreement as shown in Figure 11-H. The XRD patterns indicate the presence of (002), (211), and (222) crystal planes at 2ϴ ₌ 25.5°,31.77°, and 45.5°, respectively, which exactly matches the nanocrystalline HA planes (JCPDS #09-0432) [75, 76]. These presented data from the HR-TEM coupled with SAED and XRD analysis confirmed the formation of a nanocrystalline HCA on the surface of the studied materials in this study. These results confirm and clarify what has been reported by our group in previous studies. Prior studies on a PECVD Ti/TiO₂-SiON revealed the formation of hydroxycarbonate apatite within 6 hours as confirmed by XPS and XANES analysis, and SiON surface chemistry induced osteogenic gene expression of human periosteal cells and led to a rapid bone-like biomineral formation within 4 weeks [21,28].

Fig. 11.

Crystal structure analysis of the HCA. A) HR-SEM image shows the HCA nucleation after 7 days of immersion. B-F) HR-TEM images at different magnifications show the crystallographic planes of HCA. G) SAED analysis indicating Debye rings of the polycrystalline HCA confirmed with and XRD (H).

3.3. MSCs in-vitro cytotoxicity studies

In this section, the surfaces’ effect on cellular behavior (i.e., cytotoxicity, cell viability, and proliferation) was investigated using MSCs cell culture studies for 24 and 48 hours. Cell viability was assessed by live/dead staining (qualitative cytotoxicity assay) while cells proliferation was further quantified by an MTS assay (quantitative cytotoxicity assay). MSCs are considered the main cell source for tissue-engineering applications due to their ability for self-renewal, differentiation into different cell lineages, and immunomodulatory properties[77]. MSCs are capable of proliferation, then differentiation to produce bone[78, 79], ligament[80], adipose[81], cartilage[82,83], and muscle[84] tissues based on the type of conditioned cell culture media being used. MSCs have shown proven osteogenic differentiation when co-cultured with osteoblast cells[85] or directly seeded within porous scaffolds implanted in an animal model for bone defect repair[86,87]. Figure 12 illustrates MSC viability and proliferation after 24 and 48 hours of culture on the SiON-A-Ti, SiONP-A-Ti, and the A-Ti. Fluorescence images show the live cells (green) and dead cells (red) on the different tested surfaces which indicated good cell adhesion and spread on the surfaces. Almost no red stained cells were observed meaning no cytotoxic effect on the modified surfaces as well as the A-Ti after 24 and 48 hours. This was further quantified using the cell proliferation MTS assay as indicated in Figure 10-B. MTS cell viability results confirmed that SiON- and SiONP-Ti have higher optical density (OD) values indicating higher cell viability/proliferation compared to the A-Ti at both time points, but no significant difference was observed. These data add more evidence for the surface bioactivity and biocompatibility of the amorphous silica-based PECVD as potential coatings for enhanced HA formation as well as cellular behavior without any cytotoxic effect.

Fig. 12.

Illustrates the MSCs viability and proliferation after 24 and 48 hours of culture on SiONx-Ti and SiONPx-Ti compared to the anodized-Ti. A) Fluorescence images show the live cells (green) and dead cells (red) on the tested surfaces which indicated a good cells adhesion and spread on the surfaces. B) MTS cell’s viability/proliferation assay presented in optical density (OD) indicates higher cells proliferation on the PECVD coatings compared to the anodized-Ti, but no significant difference was observed.

Finally, this study aimed to uncover the interfacial adhesion and surface bioactivity of anodized titanium modified with SiON and SiONP PECVD surface coating. This study indicated that a strongly adherent thin film coatings can be successfully deposited on any medical device or implantable material that can fit in the reactor geometry. PECVD is a well-known method that has been used in the semiconductor industries for decades. Using the PECVD reactor allows for uniform plasma deposition of various thin film coatings such as silicon dioxide [19], silicon nitride [88], silicon oxynitride [21], phosphosilicate glass [20,23], and borophosphsilicate glass at low temperatures from 250–400°C. Thus, the PECVD reactions (Rxn) can be summarized as presented below (Rxn 1–3). The N liberated from the NH₃ works as a network former to substitute the O in the silica network, while the P from PH₃ works as a network modifier in a small percent (i.e., < 4%) in the silica structure.

$$O-Ti-O+wSi{H}_4+x{N}_{2}O\to O-Ti-O-S{i}_w{O}_x+2w{H}_2+x{N}_2$$ (Rxn.1)

$$O-Ti-O+wSi{H}_4+x{N}_{2}O+yN{H}_3\to O-Ti-O-S{i}_w{O}_x{N}_y+(2w+1.5y)H_2+x{N}_2$$ (Rxn.2)

$$O-Ti-O+wSi{H}_4+x{N}_{2}O+yN{H}_3+zP{H}_3\to O-Ti-O-S{i}_w{O}_x{N}_y{P}_z+(2w+1.5y+1.5z)H_2+x{N}_2$$ (Rxn.3)

The PECVD layer in the above reactions assume RBM, for the RMM interpretation of the PECVD films, the final films structure will be [Si-Si]_w−[Si-O]_x for amorphous silica, [Si−Si]_w−[Si−O]_x−[Si−N]_y for amorphous silicon oxynitride, and [Si−Si]_w−[Si−O]_x−[Si−N]_y−[O-P]_z for amorphous silicon oxynitrophosphide.

The surface characterization revealed that SiON-PECVD thin films coatings did not alter the surface topography or microstructure of the substrate while SiONP increase the surface roughness that could enhance its activity. Mechanical properties as studied by Vickers hardness and nano-indentation/scratch tests indicate a strong interfacial adhesion between the PECVD thin films and the A-Ti substrate due to the chemical bonding between the coating layer and Ti-O that anchors and supports the thin films on the surface as indicated in the reactions above. Deposition of an amorphous layer of SiON or SiONP prevents the mismatching at the interfacial layer and provide a hydrophilic surface with high surface free energy and readily available functional groups (e.g., Si-H, N-H, Si-OH, and P-OH) once immersed/implanted in a physiological environment. The presence of such a surface with its unique chemical structure will initiate the formation of apatite-like bone structure (HCA) according to the mechanism known for bioactive bioglass-bone formation [89,90], but with a faster rate due to the availability of the surface silanol groups from the PECVD coatings. The bioactive bioglass bonds with bone through the formation of a HCA layer as proposed by Hench et al., [90]. Here, HCA formation was confirmed on the surface as shown from the HR-TEM with SAED, XRD patterns, FT-IR, and SEM/EDX that indicated Ca/P ratio in the range of 1.5–1.6 that exactly matches the poly crystalline carbonate-substituted apatite [89]. As presented above, the SiONP showed a significant enhancement of HCA formation as confirmed by the highest surface coverage and Ca/P ratio compared to the SiON and the A-Ti. This pronounced effect on the SiONP surface can be attributed to the presence of phosphorus in the structure which rapidly dissolve leading to a relatively open structure layer of silica gel with Si-ions release. This environment allows for ion exchange and formation of Ca-PO₄ within 6 hours, which becomes a more stable HCA structure at a faster rate than other surfaces. These findings confirm the previously proposed role of phosphate which is lowering the energy barrier to appetite nucleation on the bioglass surface [91,92]. Furthermore, the release of Si-ions from these coatings serves a vital role to enhance osteoblast markers in-vitro [93], endothelial cell angiogenic properties [20], and enhance antioxidant expression, angiogenic marker expression, and reduce ROS levels needed for accelerating vascular tissue regeneration [15,19,20,23,24,94]. Our future studies will concentrate on the osteogenic effects and the complete healing process in-vivo.

4. Conclusions

The present study investigated the potential application of SiON and SiONP as bioactive amorphous thin film coatings for medical devices compared to the A-Ti surface. The surface characterization revealed that low temperature PECVD thin film coatings enhance the surface activity through increasing the wettability and surface energy, and the surface bioactivity by providing surface functional groups that work as nucleation sites for HCA formation at a faster rate and higher surface coverage compared to the A-Ti. Analysis of mechanical properties indicated no change in the microhardness while enhancing the nanohardness with SiON and SiONP coatings. SiON and SiONP thin films slightly increased the reduced elasticity modulus of the A-Ti, but no significant difference was observed. Nano-scratch tests indicated that PECVD coatings have a strong interfacial adhesion with the A-Ti substrate due to the chemical bonding between the coating layer and the nanoporous ATO structure that anchors and supports the thin films on the surface. In-vitro-cell free-studies performed only in cell culture media indicated the formation of HCA on the surface after 12 hours of immersion. The SiONP-Ti surface presented the highest surface coverage of HCA after 7 days of immersion with a Ca/P ratio close to the stoichiometric HA. The HCA structure and crystallinity were confirmed using various techniques such as HR-TEM, XRD, FT-IR, and HR-SEM/EDX. Finally, MSCs in-vitro studies indicated cells adhesion and proliferation on the modified surfaces without any cytotoxic effects. This study presents sufficient evidence for the surface bioactivity and biocompatibility of the SiON and SiONP amorphous silica-based PECVD coatings as potential coatings for enhanced HA formation and cellular behavior without any cytotoxic effect compared to the currently available medical grade A-Ti.