Hybrid coatings for orthopaedic implants formed by physical vapour deposition and microarc oxidation

Hybrid coatings for orthopaedic implants formed by physical vapour deposition and microarc oxidation Roman Gabor a , Ladislav Cvrc ˇek b , Martina Doubková c,d , Václav Nehasil e , Josef Hlinka f,g , Petr Unucka h , Mate ˇj Bur ˇil b , Adéla Podepr ˇelová g , Jana Seidlerová a , Lucie Bac ˇáková c a Nanotechnology Centre, CEET, VSB – Technical University of Ostrava, 17, listopadu 15/2172, 708 00 Ostrava-Poruba, Czech Republic b Department of Materials Engineering, Faculty of Mechanical Engineering, Czech Technical University in Prague, Karlovo náme ˇstí 293/13, 120 00 Prague 2, Czech Republic c Laboratory of Biomaterials and Tissue Engineering, Institute of Physiology of the Czech Academy of Sciences, v.v.i., Videnska 1083, 142 20 Prague 4, Czech Republic d Charles University, Second Faculty of Medicine, V Uvalu 84, 150 06 Prague 5, Czech Republic e Department of Surface and Plasma Science, Charles University, Prague, Czech Republic f Department of Materials Engineering, Faculty of Materials and Technology, VSB-Technical University of Ostrava, 17, listopadu 2172/15, 708 00 Ostrava-Poruba, Czech Republic g Centre for Advanced Innovation Technologies, VSB-Technical University of Ostrava, 17, listopadu 2172/15, 708 00 Ostrava-Poruba, Czech Republic h VUHZ, a.s., 739 51 Dobra, Czech Republic highlights Unique combination of PVD and MAO techniques was used to prepare hybrid coatings to improve Ti-6Al-4V alloy properties. Prepared hybrid coatings with ZrTi exhibited improved corrosion and abrasion resistance, and a lower coefficient of friction. Hydrophilic oxide structure of prepared hybrid coatings increased osteogenic differentiation of human bone marrow mesenchymal stem cells in vitro. Combined use of MAO nad PVD techniques provides new possibilities of fine tuning the material surface layer properties. graphical abstract article info Article history: Received 14 December 2021 Revised 29 May 2022 Accepted 31 May 2022 Available online 3 June 2022 Keywords: Ti-6Al-4V PVD coating MAO process Corrosion protection Tribology Mesenchymal stem cells abstract This study is focused on the preparation of new hybrid layers intended for surface modification of Ti-6Al4V alloys for potential orthopaedic and dental applications. Combination of the technology of physical vapour deposition (PVD) and subsequent micro-arc oxidation (MAO) was utilized for the deposition of Ti and ZrTi to form hybrid oxide layers. The oxide layers were prepared using an alkaline electrolyte with glycerol as an additive under micro-arc discharge conditions with different Si content on their surfaces. The hybrid ZrTi coatings with a Zr/Si structure achieved the best tribological properties described by a low friction coefficient of 0.3 and high abrasion resistance. There was also an increase in corrosion potential and polarization resistance of hybrid ZrTi coatings. Although the proliferation of human bone marrow mesenchymal stem cells was slower on these hydrophilic Ti and ZrTi coatings than both on uncoated Ti6Al-4V and the reference tissue culture polystyrene dishes, both types of hybrid coating promoted greater osteogenic differentiation of these cells, indicated by approx. twice as high activity of alkaline phosphatase. The hybrid oxide layers newly developed in this study – especially the layers with Zr – are therefore promising for coating metallic bone implants. Ó2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1016/j.matdes.2022.110811 0264-1275/Ó2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Materials & Design 219 (2022) 110811 Contents lists available at ScienceDirect Materials & Design journal homepage: www.elsevier.com/locate/matdes 1. Introduction The design and the uses of biomaterials for bone implantation are closely bound to their strength, modulus of elasticity, wear resistance and non-toxicity, osseointegration or bioinertness. High osseointegration potential, i.e. attractiveness for the adhesion, growth and phenotypic maturation of bone cells, is needed for long-term/permanent orthopedic implants, such as stems for replacements of large and small joints (e.g., hip, knee, shoulder, and trapeziometacarpal joints) and for stomatological implants, especially for parts which are integrated into the bone (e.g., the stems and cups of hip joint replacements). Firm osseointegration of these implants improves their stability, prolongs their lifetime and, at the same time, reduces the likelihood of failure and the need for reoperation. Bioinertness, i.e. repulsiveness of a biocompatible material for cell adhesion and growth, is needed for short-term/temporary implants, e.g. for traumatological implants such as splints, wires, screws or spikes, so that they can be removed easily after the bone damage has healed [1]. The excellent corrosive and mechanical properties of pure Ti and Ti alloys make them suitable for application in both types of bone implants mentioned above [2]. Significantly better mechanical properties were achieved via stabilisation of the a +bphases in Ti-6Al-4V alloy [3]. Titanium alloys with a stabilised bphase, containing elements such as Nb, Ta or Zr, are becoming more and more popular. The main benefits of these alloys are their lower modulus of elasticity and the absence of toxic elements such as Al and V [1]. Growing evidence of the release of these elements into the body is being gathered. This release is linked with neurodegenerative diseases like Alzheimer and Parkinson, peripheral neuropathy, and also can affect negatively the proliferation of adjacent bone cells, extracellular matrix synthesis and the formation of bone apatite, which an lead to osteomalacia [4]. The cytotoxicity of these elements is further potentiated or enhanced by the release of other metal ion species (particularly Ti), which is of increasing importance, especially during tribocorrosion phenomena. This type of synergistic adverse effect of the released ions is well-known both for pure Ti and for titanium alloys, and it is therefore crucial to address the material surface degradation process in the human body [5]. Titanium itself in the form of ions, and in the form of nanoparticles and microparticles, can cause cytotoxic, inflammatory, allergenic and even mutagenic and carcinogenic effects [6]. The well-known poor tribological properties of Ti and titanium alloys, accompanied by a high friction coefficient and insufficient abrasion resistance, place increased emphasis on the subsequent surface treatment ensuring sufficient tribological, corrosion, and biocompatible properties [7]. Commercial coatings for implants that are most commonly applied to improve these properties include diamond-like carbon (DLC) [8], nitride [9] and oxide ceramic coatings [10]. Other promising bioactive coatings include various inorganic substances, such as calcium phosphates or bioactive glass, and also organic substances, such as natural polymers (collagen, gelatin, cellulose, chitosan, alginate, hyaluronic acid) and synthetic polymers (polycaprolactone, polyether ether ketone, poly-Llactic acid, poly lactic-co-glycolic acid, polyurethane, polyvinyl alcohol), which can serve as carrier matrices for the delivery of therapeutic drugs, antimicrobial agents, growth factors and other biologically active compounds [11]. Steady progress in the application of protective coatings in the last two decades has achieved required properties such as wear resistance, good tribological properties, oxidation resistance, corrosion resistance, and a low coefficient of friction [12]. Coatings providing these properties can be applied using physical vapour deposition (PVD) technology [13], for example cathodic arc evaporation (CAE), direct current (DC) or pulsed DC magnetron sputtering [14], radio frequency (RF) magnetron sputtering [15], and high power impulse magnetron sputtering (HIPIMS), which is widely applied today [16]. Other important coating technologies include chemical vapor deposition (CVD), electrophoretic deposition, sol– gel deposition, layer-by-layer deposition, biomimetic deposition, dip coating, drop coating, plasma spraying and 3D printing [11]. A promising alternative to the methods mentioned above, microarc oxidation (MAO), also referred to as the plasma electrolytic oxidation (PEO) technique, can be used to prepare continuous ceramic oxide layers with excellently high adhesion to the substrate (e.g. Ti, Al, Nb, Zr, Ta, Hf), and with high resistance to wear and corrosion. In the preparation of hard ceramic layers, a pulsed BIor UNI-polar mode is used to produce the discharge observed during plasma electrolysis in a liquid electrolyte [10]. The oxide layer prepared by the MAO technique consists of a porous amorphous outer layer and a compact crystalline inner layer [17]. The growth of the ceramic layer (up to 20 l m) occurs at temperatures of 10 3 -10 4 K and at a pressure of 10 2 -10 3 Pa after exceeding the breakdown voltage while the formation of plasma channels with a micro-arc discharge is taking place [18]. The surface properties of these ceramic layers, including their porosity, roughness, and chemical composition, determine whether their successful application range is for short-term implants or for long-term implants [19]. The properties of ceramic coatings defined in this way are affected not only by the applied voltage, the pulse size, the pulse width and the process time but also by the composition of the electrolyte that is applied. The ions (Ca, P, Si) from the electrolyte are incorporated into the coating and affect its functional properties, including corrosion resistance, abrasion resistance, and biocompatibility [20]. The roughness and the porous structure of the outer porous layer of the MAO coating negatively affect the resulting coefficient of friction [21]. Although MAO coatings achieve high corrosion resistance, both layer failure and reduced corrosion resistance can occur under loading. In an effort to eliminate these shortcomings, many authors have investigated the possibility of doping the MAO electrolyte using nano particles TiO 2 [22], MoS 2 [23], ZrO 2 [24] or yttrium [25] and the use of secondary layer deposition in the form of DLC coating or graphene coating [26]. The Al/Ti-6Al-4V duplex system prepared by a combination of MAO and magnetron sputtering was also tested with the aim to improve the tribological properties of the MAO coatings [27]. The options available in the use of the hybrid system preparation technique allow a more detailed study of the PVD-deposited layer and its final properties, achieved by subsequent MAO. Application of this procedure makes it possible, using PVD with the required functional properties, to enrich the MAO layers with the oxides of deposited metals that are formed. To increase the osteointegration properties, porous Ca/P layers on Ti-6Al-4V and Ti-29Nb-xHf substrates were prepared using a combination of MAO/PVD techniques [28]. Subsequently, the RF magnetron sputtering technique was used on the prepared MAO porous layer for the deposition of an Mn layer [29], and a Zn layer [30]. Several works [31,32] have studied titanium alloys with Nb, Zr, Ta, which – after MAO treatment – achieved increased wear resistance and excellent corrosion properties due to the presence of stable oxides (e.g. Ta 2 O 5 , ZrTiO 4 ). Due to their high corrosion resistance, very good mechanical properties and excellent biocompatibility, zirconium and its alloys are promising biomaterials for orthopaedic implants [33].MAO applications - previously used for titanium and its alloys - can also be used for the preparation of oxide layers based on zirconium [34]. The zirconium oxide layers thus prepared can increase corrosion resistance and wear resistance, and can reduce the coefficient R. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 2 of friction [35], while eliminating the release of toxic Al and V ions from the Ti-6Al-4V alloy substrate. As proved by our earlier study [36], MAO enables the preparation of highly hydrophilic and bioinert surfaces that reduce cell adhesion and growth without affecting the cell viability, and further modification of the MAO-deposited layers can modulate the surface wettability and attractiveness for cell colonisation in a controllable manner. Options for readily modifying the process parameters of the formation and composition of a material coating are valuable for fine tuning the surface properties to modulate the desired cell response. For this purpose, we propose the application of a novel combination of surface treatment and deposition techniques of magnetron sputtering and MAO. Although each of these techniques is commonly used for modifying materials, they have not yet been systematically used and researched together. The combined use of these techniques could provide an interesting opportunity to study their possibilities with regard to their material properties and their performance in biological environments represented by cultures of osteogenic cells in vitro. In this study, a combined PVD and MAO technology was used for the development of Ti and ZrTi layers on a Ti-6Al-4V substrate. These layers were then characterized in terms of their physical, chemical and tribological properties and their interaction with human bone marrow mesenchymal stem cells in cultures on their surface. We found that the deposition of both types of coatings, particularly ZrTi, improved the resistance of the samples against friction, wear and corrosion, and improved the cell osteogenic differentiation. 2. Materials and methods 2.1. Sample preparation Round samples of Ti-6Al-4V titanium alloy (15 mm diameter, 2.6 mm thickness) were subjected to mass finishing to unify the surface using an HV 20 vibratory finishing machine (OTEC, Germany) for 8 h with the use of KF 10 plastic bodies. Ti and ZrTi PVD layers were deposited on mechanically treated samples of Ti-6Al-4V alloy in the Hauzer Flexicoat 850 (Hauzer, Netherlands) PVD unit based on cathodic arc evaporation. The targets used for arc evaporation of Ti and ZrTi were 63 mm in diameter, and the pressure in the working chamber was 10 -3 mbar during the application. The process parameters of the PVD unit used for deposition of the Ti and ZrTi layers are given in Table 1. Samples with PVD layers (Ti-PVD, ZrTi-PVD) deposited on a Ti6Al-4V substrate were modified under MAO conditions using a pulse source at 93 Hz, 7 % duty cycle, voltage 500 V reached in 30 s, voltage 500 V for 15 min (Fig. 1). Samples with MAO coatings (Ti-MAO, ZrTi-MAO) were prepared in an electrolyte containing 15 g/L Na 2 SiO 3 9H 2 O; 8 g/L NaOH; 80 g/L glycerol (conductivity 28.8 mS/cm, pH 12.5). 2.2. Surface analysis of the coatings The surfaces and cross-sections of the coatings were studied using the JEOL JSM-7610F Plus (JEOL, Japan) Scanning Electron Microscope (SEM) equipped with an autoemission cathode as the electron source. The samples were scanned in BSE mode (20 keV accelerating voltage) with detection of the secondary electrons. The chemical composition of the coating was determined using an energy dispersive X-ray spectrometer (EDX, Oxford Instruments). The KSG 110 Calotest device (INOVAP, Germany) was used to determine the thickness of the PVD coatings. The MAO coating thickness was studied by SEM from polished cross-sections in back scatter emission mode. An atomic force microscope (AFM, LiteScope TM ) was used in semi-contact mode to measure the topography and the surface roughness of the sample surfaces. The correlative analysis also provides data from both AFM and SEM and allows a 3D correlative probe and electron microscopy view (CPEM, Nenovision s.r.o.). Surface roughness R a measurements (the arithmetic average of the absolute values of the profile heights), R t measurements (the vertical distance between the highest and lowest points of the profile) and R z measurements (the maximum height of the profile) were carried out in contact mode by AFM (AFM LiteScope TM ). The chemical states and the composition of the prepared layers were analysed using the method of photoelectron spectroscopy - XPS (X-ray photoelectron spectroscopy), which was carried out in an ultra-vacuum apparatus with a basic pressure of 1.10 -7 Pa. Photoelectrons were emitted using an X-ray tube with a double anode (Al, Mg). The Al K a 1,2 line with primary energy of 1486.6 eV was used in the experiments. XPS spectra were recorded using an Omicron EA125 hemispherical analyser. 2.3. Tribological and adhesion tests The friction coefficient and the wear rate were tested using a CSM THT pin-on-disc tribometer (CSM Instruments, Switzerland). The friction pair consisted of the tested planar sample and an Al 2 O 3 ball 6 mm in diameter. The tests were performed at room temperature in phosphate-buffered saline (PBS), which served as a simulation of the human body environment. The solution was prepared by dissolving one PBS tablet (Sigma-Aldrich, USA) in 200 mL of distilled water. The solution contained 10 mM phosphate buffer, 2.7 mM KCl and 137.0 mM NaCl with pH 7.4 at 25 °C. During the tribological test, a normal load of 1 N, linear sliding speed 50 mm s 1 and number of laps 5000 with a radius of 6 mm were used. Each measurement was repeated twice for each sample. The coefficient of friction ( l ) was calculated from the ratio of the tangential friction force and the normal force. The wear of the surface of the Al 2 O 3 ball and the width of the wear track on the planar sample were analysed using an Olympus DSX1000 (Olympus Corporation, Japan) digital microscope after the test. The wear rate was calculated from the following equation (Archard,1953): k=V/Fs where kis the wear rate, Vis the wear volume, Fis the normal load, and sis the sliding distance. The wear volume was obtained by multiplying the area of the wear track cross-section and the circumference of the wear track. The wear track cross-section area was calculated from the wear track profile, which was analysed by a Zygo NewView 7200 optical profilometer (Zygo Corporation, USA). The CSM Revetest Xpress+ (CSM Instruments, Switzerland) scratch tester was used for adhesion evalution. The scratch tester was equipped with a Rockwell diamond indenter (tip radius 200 l m). The scratch test load was set to increase linearly from Table 1 Process parameters. Parameters Ti ZrTi Target 99.98 % Ti 85 % Zr / 15 % Ti Coating pressure (mbar) 5.0 10 -3 8.0 10 -3 Temperature (°C) 430 430 Gas Ar Ar Gas Flow (sccm) 540 380 Deposition time (min) 180 150 Current on target (A) 60 80 Voltage bias (V) 80 80 R. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 3 1 N to 30 N along the 5 mm scratch path with linear speed 10 mm. min 1 . 2.4. Corrosion tests As the corrosion resistance of titanium and its alloys is significant, direct exposure corrosion tests would not be suitable; a potentiodynamic polarisation method was therefore used for determining the corrosion parameters of the tested samples. All electrochemical tests were performed using a Voltalab PGZ 100 potentiostat equipped with Voltamaster 10 software. The parameters of the test were used according to the ASTM F 2129 and ASTM F 746 standard test methods [37]. PTFE/PE corrosion cells with a lower exposure hole (0.5 cm 2 ) and a volume of 20 mL were used for testing. The samples were tested in an isotonic physiological solution (0.9 wt% NaCl in double-distilled H 2 O) in order to simulate the environment of a living tissue. No gas bubbling was applied during testing, and the temperature was set to 25 °C. A threeelectrode setup was used, consisting of a sample connected as a working electrode, a saturated Calomel electrode (SCE, +241 mV vs SHE) set as a reference electrode, and a high purity carbon rod connected as an auxiliary electrode. A 60-minute time lag was applied after filling the corrosion cells with the solution to stabilise partial corrosion processes. After the time gap, an electrochemical impedance spectroscopy (EIS) method was used to determine the electrochemical parameters of the tested surface layers. Before starting the potentiodynamic polarisation, the initial potential value was set to 70 mV vs. the open circuit potential (OCP) after stabilisation of the corrosion equilibrium, with the polarisation rate set to 60 mV.min 1 . The dependence between the current flowing through the sample and the potential applied to the test sample was recorded during the measurement. The potential was gradually applied to the tested sample, which increased over time with the value of the polarisation rate. 2.5. Contact angle The angle between the tested surface and the physiological solution was evaluated by the sessile drop method. The surface contact angle was measured by the SEE system, and the free surface energy was calculated by Advex Instrument software (Advex Instruments, s.r.o., Czech Republic). Two l L droplets of high purity water were applied to the tested surface, and the contact angle h was determined by the tangent to the drop profile at the point of contact of the three phases (liquid, solid and gas) with the line of the surface of the sample. 2.6. Cell culture conditions Human mesenchymal stem cells derived from bone marrow (hMSCs; No. 7500, ScienCell Research Laboratories, USA) were used in all experiments. Prior to the experiments, the hMSCs were cultured in 75 cm 2 tissue culture flasks (TPP, Switzerland) in 15 mL of Mesenchymal Stem Cell Medium (MSCM medium; No.7501, ScienCell Research Laboratories, USA) to reach ca 70% confluence. The Ti-6Al-4V samples (Ti-6Al-4V, Ti-MAO, ZrTi-MAO) were disinfected with ethanol for 2 h, were rinsed in dH 2 O, were sterilised in an autoclave, and were then inserted into tissue culture polystyrene 24-well plates (TPP, Switzerland). The hMSCs were seeded on these samples at a density of 9,000 viable cells per well (ca 5,000 cells/cm 2 ) in 1.5 mL of a -MEM medium (No. 11900-016, Gibco, USA) supplemented with 10% of fetal bovine serum (FBS), L-glutamine (2 mmol/L; No. A2916801, Gibco, USA) and gentamicin (40 l g/mL). In addition to the samples, the cells were also seeded directly into the tissue culture plate wells (PS), to provide information about the standard cell behaviour. The cells were cultured at 37 °C in a humidified atmosphere with 5% CO 2 for 14 days. In order to support osteogenic differentiation of the cells, the culture medium was enriched with ascorbic acid (50 l mol/L; No. 49752-10G, Sigma-Aldrich, USA), b-glycerophosphate (20 mmol/L; No. G9422, Sigma-Aldrich) and dexamethasone (10 nmol/L; No. D2915, Sigma-Aldrich). The medium was replaced twice per week. 2.7. Fluorescence staining Fluorescence staining was used for visualising and analysing the cell spreading area and morphology (day1), and for evaluating the number and the viability of the cells growing on the samples (day 1, 3, 7 and 14). The samples were rinsed with PBS, then fixed with 4% paraformaldehyde (pH 7.4, 15 min), and were rinsed again twice with PBS. The fixed samples were treated with 1% bovine serum albumin in PBS containing 0.1% Triton X-100 (20 min) and with 1% Tween 20 in PBS (20 min) (both Sigma-Aldrich, USA). Subsequently, the cell nuclei were stained with DAPI (100 ng/mL, blue fluorescent signal; No. 32670, Sigma-Aldrich), the cell membrane and cytoplasm were stained with Texas Red C 2 -maleimide (20 ng/mL, red fluorescent signal; Molecular Probes, Invitrogen, USA), or the F-actin cytoskeleton was stained with TRITCconjugated phalloidin (100 ng/mL, red fluorescent signal; SigmaAldrich). All dyes were diluted in PBS and were added to the cells for 1 h at room temperature (protected from light). A parallel set of samples for each time interval was rinsed with PBS and was stained with a Live/Dead Viability/Cytotoxicity kit for mammalian cells (No. L3224, ThermoFisher Scientific, USA) according to the manufacturer’s instructions. Calcein AM stains the living cells in green, whereas ethidium homodimer 1 stains the dead cells with damaged cell membranes in red. Additional staining of the cell nuclei with DAPI was performed on day 7 and 14 for easier cell counting. Microphotographs (20 per samples/well) of randomly chosen fields were taken with an Olympus IX51 epifluorescence microscope, equipped with a DP70 camera (both from Olympus Corp., Japan). The microphotographs were analysed in ImageJ FIJI software (https://imagej.net/Fiji; 51). The initial cell spreading areas are presented in l m 2 . The cell viability was determined as the percentage of living cells on the sample. The cell population densities on the samples are presented as the cell number per cm 2 on a log scale to assess proliferation. The data from day 1 and day 3 were used for calculating the cell population doubling times, according to the following equation: Fig. 1. Schematic diagram of the deposition and layer modification process applied to the Ti-6Al-4V substrate material. R. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 4 DT ¼log2tt 0 logN t logN t 0 The time intervals are represented by t 0 (day 1) and t (day 3), and Nt 0 and Nt represent the number of cells in a particular time interval. 2.8. Resazurin assay A resazurin assay based on the activity of mitochondrial enzymes was used to investigate the metabolic activity of the hBMSC cells. On days 1, 3, 7 and 14 of cultivation, the samples were transferred to fresh 24-well culture plates with a fresh culture medium without phenol red supplemented with 10% of FBS. The samples were then incubated with resazurin (No. R7017, Sigma– Aldrich, USA) diluted to a final concentration of 40 l mol/L in a fresh culture medium without phenol red (1 mL/well) at 37 °Cin a humidified atmosphere containing 5% CO 2 , protected from light. After 4 h (day 1 and 3) or 2.5 h (day 7 and 14) of incubation, the fluorescence was measured (Ex/Em = 530/590 nm) by a Synergy HT Multi-Mode Microplate reader (BioTek, USA) in triplicate aliquots. A solution incubated in a well without cells served as a blank control. The results were corrected to the background control (solution without cells) and were then relativised to the estimated total number of cells growing on each sample. 2.9. ALP assay A quantitative analysis of alkaline phosphatase (ALP) enzymatic activity to assess early osteogenic differentiation of the cells on the samples was performed using 1-Step PNPP Substrate Solution (pnitrophenyl phosphate; No. 37621, ThermoFisher Scientific) on days 7 and 14 of culture. The samples were rinsed with PBS three times. After a total of 250 l L of PNPP had been added to each sample, the culture plate was incubated for 20 min at room temperature for the reaction to develop. The absorbance of the watersoluble yellow-colour product was measured at 405 nm using a Synergy HT Multi-Mode Microplate reader (BioTek, USA), in duplicates for each well. The absorbance results were corrected to a blank control (PNPP solution without cells), and were then relativised to the estimated total number of cells growing on each sample. 2.10. Statistical analysis All biological experiments were performed independently three times with samples in triplicate in each time interval (n = 3 for each experimental group). Statistical analyses and data visualization were performed in GraphPad Prism 8.3.0 (GraphPad Software, USA). One Way ANOVA with Tukey’s test or Kruskal-Wallis ANOVA with Dunn’s post hoc test were performed on the data from the biological experiments. The data are expressed in summary graphs as median with interquartile range (IQR). The results were considered statistically significant at p < 0.05. 3. Results and discussion 3.1. Morphology and chemical analysis Fig. 2a-c presents SEM images of the surface structure of the substrate and of the PVD coating. The images confirm the homogeneous and compact structure of the coatings regarding their thickness and roughness, which was also determined by a Calotest instrument (Table 2). The results of the CPEM analysis, presented in Fig. 2d-f, revealed microscopic defects on the surfaces (such as traces of mechanical treatment or droplets). The occurrence of these defects following PVD deposition can be reduced by lowering the arc target current [38]. The SEM and CPEM images in Fig. 3a-d show that the surface morphology of the MAO coating is a compact porous structure with an average pore size of about 1 l m. The formation of a micro-arc discharge was recorded after reaching 500 V, and it was followed by a decrease in current associated with the growth of the oxide layer. The presence of commonly known structural microscopic defects of the MAO coating involving cracks was not confirmed on the studied surfaces with the unique CPEM technology. The development of the coating is accompanied by the formation of so-called volcanic craters (Fig. 3) during micro-discharges, when the released melt of the substrate (PVD layer) is deposited via discharge channels on the surface. Well-visible changes in surface topography imaged using CPEM technology indicate through 3D imaging the presence of craters that affect the resulting roughness of the hybrid MAO layers compared to the surface roughness for the deposited PVD layers (Fig. 2). The resulting measured thickness and roughness parameters of the MAO coatings are given in Table 2. The layer thicknesses were determined from the cross-sections of individual MAO coatings on which the distribution of the elements present in the layer was evaluated. Figs. 4 and 5 show the presence of the outer and inner MAO layers. Element mapping from cross-sections of the Ti-MAO sample (Fig. 4) and of the ZrTi-MAO sample (Fig. 5) confirmed both the presence of elements from PVD deposition and the elements incorporated from the solution. The resulting values of the porosity structure of the MAO coatings (Table 2) confirmed the increase in roughness in all three monitored parameters (R a ,R z ,R t ) of the surfaces of the MAO coatings in comparison with the bare substrate and the PVD coatings (Ti-PVD, ZrTi-PVD). The resulting surface structure is due to the high temperature (10 5 -10 7 K/s) and the pressure in the discharge of the channels during the MAO process. The resulting melt passes through the discharge channel and, after leaving the channel, it solidifies in the electrolyte environment [18]. The growth of the surface roughness and of the oxide layer is mainly associated with the process conditions (voltage, current density, time, temperature, electrolyte composition) and the presence of additives. The effect of glycerol as an additive on the surface roughness and on the structure of the MAO coating was described in detail in an earlier study by Gabor et al.[39]. In an effort to reduce the surface roughness and to improve the tribological and corrosive properties of the coating, glycerol was used as an additive in this work. The different values of the achieved thickness and roughness of the MAO coating (Table 2) are caused by the different development of the spark discharge, while the distribution and the intensity of the spark discharge directly affect both the resulting surface morphology and its phase composition [40]. The course of the MAO process is influenced by the different melting temperatures of Ti (1660 °C), Zr (1852 °C) and of ZrTi, which may lead to more difficult MAO development of the ZrTi-PVD sample, subsequently affecting the resulting surface microstructure [41]. The higher roughness parameters of the Ti-MAO coating are caused by the greater intensity and the larger amount of microdischarges on the surface of the sample than on the ZrTi-MAO coating. High microdischarge temperatures lead to dissolution of the substrates and to subsequent reactions with the electrolyte elements that are present, including rapid solidification and the formation of oxides that affect the resulting surface morphology [42]. Cross-sections and elemental mapping confirm the phenomenon of MAO coatings, the presence of a compact inner layer and a porous outer layer (Figs. 4 and 5). The outer layer is highly porous, rough, amorphous and not corrosion-resistant. The dense inner layer, characteristic of the MAO process, is corrosionresistant, crystalline and its thickness increases with the processR. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 5 Fig. 3. Morphology of MAO coatings. SEM images of Ti-MAO (a) and ZrTi-MAO (b); AFM images of Ti-MAO (c) and ZrTi-MAO (d). Fig. 2. Morphology of PVD coatings. SEM images of the Ti-6Al-4V substrate (a), Ti-PVD (b) and ZrTi-PVD (c); AFM images of the Ti-6Al-4V substrate (d), Ti-PVD (e) and ZrTiPVD (f). Table 2 Surface roughness and thickness of samples. Sample Roughness ( l m) MAO-coating ( l m) PVD-coating ( l m) R a R z Rt inner layer outer layer layer Substrate 0.09 ± 0.02 0.31 ± 0.62 0.55 ± 0.10 – – – Ti-PVD 0.19 ± 0.09 0.64 ± 0.30 1.23 ± 0.30 – – 3.4 ± 0.50 ZrTi-PVD 0.24 ± 0.11 0.84 ± 0.38 1.55 ± 0.69 – – 2.8 ± 0.44 Ti-MAO 0.82 ± 0.16 2.37 ± 0.40 3.86 ± 0.64 0.60 ± 0.12 3.35 ± 0.49 – ZrTi-MAO 0.58 ± 0.22 1.66 ± 0.58 2.75 ± 0.84 0.94 ± 0.11 4.25 ± 0.35 – R. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 6 ing time [43]. The distribution of elements in the Ti-MAO coating is shown in Fig. 4, and EDX analysis confirmed the presence of Ti, Si and O elements (Table 3). Increased Si contents were determined mainly in the outer porous layer, according to Gabor et al.[36]. According to [44], the MAO process in alkaline conditions with the presence of silicates provides better wear resistance and corrosion resistance. The proportion of silicon in the coating volume differed significantly for MAO coatings - the ZrTi-MAO silicon content (Fig. 5) determined by EDX was lower in the coating volume (Table 3). On the contrary, the Si contents on the surfaces of the outer MAO layers about 20 nm in thickness were relatively close due to the sensitivity of the XPS method (Table 4). The reduction Fig. 4. Cross-section SEM images of the Ti-MAO coating (a) and elemental mappings of Ti, Si, O (b-d). Fig. 5. Cross-section SEM images of the ZrTi-MAO coating (a) and elemental mappings of Zr, Ti, Si, O (b-e). Table 3 EDX results analysis of the Ti-MAO and ZrTi-MAO coatings. Sample Element (in at.%) Ti Si Zr O Ti-MAO 9.4 ± 0.2 13.9 ± 0.2 – 76.7 ± 0.3 ZrTi-MAO 2.5 ± 0.1 3.2 ± 0.1 18.3 ± 0.6 76.0 ± 0.2 R. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 7 of the silicon content in the outer layer is probably associated with the migration of Zr ions towards the surface and the limited transfer of the silicon species from the electrolyte to the depth of the coating [45]. This fact is related to the development of so-called ‘‘soft” sparking, which is reflected in the high density of bluewhite micro-discharges after the overall reduction of discharges over the entire sample area and noise reduction [46]. The resulting ‘‘soft” sparking is probably related to the occurrence of small diameter discharge channels and the associated limited mass transfer and reduced silicon content in the volume of the outer porous layer [47]. Both the composition of the surface and its chemical state were monitored by the XPS method. The results of the semi-quantitative analysis determined by the XPS method on individual samples (TiMAO, ZrTi-MAO) are shown in Table 4. The contents of the individual elements were determined from the peak area by the method of relative sensitivity factors after subtracting the Shirley background. The presence of Ti, O, C, Si and Na was registered on the surface of both samples. The presence of Zr from PVD deposition (ZrTi) was also determined on the surface of the ZrTi-MAO sample. Elements Si, Na and C were incorporated into the surface of the MAO layer of the electrolyte. The chemical state of individual elements was determined from the peak shifts of individual elements. The axis of binding energies –E B – was calibrated under the assumption that the main component of the peak C 1s lies at E B = 282.5 eV and corresponds to the Si-C bond [48]. A second component was also determined at peak C 1s (Fig. 6), which has an intensity of approximately 15% of the total signal C and lies at E B = 285.7 eV. According to Wang et al.[48], this binding energy corresponds to the C-O groups. On the surface of the Ti-MAO sample, two components of the Si 2p peak at E B = 99.7 and 101.0 eV were detected. According to Wang et al. [49], this corresponds to the structure of Si-C and C-O-Si. On the ZrTi-MAO coating, three components of the Si 2p peak were registered, namely E B = 97.7, 98.7 and 100.00 eV. There is probably an interaction of Si with Zr (Fig. 6), which is evident from the shift of E B to lower values by approximately 1 eV. The peaks at E B = 98.7 and 100.0 eV can both be identified as for the Ti-MAO sample, and correspond to Si-C and C-O-Si structures. The spectrum also has a peak at E B = 97.7 eV. This value is comparable to E B = 99.3 eV, reported in a study by Ding et al.[50] for the Zr-Si structure. The overall structure of Si can be shifted to E B lower by 1 eV. This peak can also be attributed to the Si-Si structure. Shifts of the Si 2p peaks to lower binding energies can be caused by a charge transfer from Zr to Si and other elements, because Zr has lower electronegativity (Pauling electronegativity 1.33) than Si (1.90) and other elements. This causes a reduction in the measured binding energy of the element to which the charge shifts moves. The O 1s oxygen peak (Fig. 6) of the Ti-MAO sample surface is wide, and can be divided into two components at energies of 529.3 and 529.9 eV. On the surface of the ZrTi-MAO coating, another component at E B = 527.7 eV was identified. All these peaks can be explained by an oxidic bond, because the range of oxygen binding energies in oxides is relatively wide (528.1 – 531.1 eV) [51]. These are probably therefore oxides of Si, Ti and Zr. The Ti 2p peak on the Ti-MAO sample lies at E B = 456.9 eV. Due to the energy distance of the Ti 2p 3/2 and O 1s peaks of 72.8 eV, we conclude that Ti 2 O 3 is present [52]. In both samples (Ti-MAO, ZrTi-MAO), the extension of the Ti 2p and O 1s spectra is evident. This suggests the presence of a mixture of lower Ti oxides and, at the same time, it corresponds to the measured O 1s binding energies. In the spectrum of zirconium Zr 3d 5/2 (Fig. 6) on the surface of the ZrTi-MAO sample, we find two components at E B = 177.6 and 179.2 eV. The peaks are widespread and the saddle between components 3d 5/2 and 3d 3/2 is almost Table 4 XPS analysis results of MAO coatings. Sample Element (in at.%) Ti O C Si Na Zr Ti-MAO 5.7 65.9 8.2 18.9 1.3 0.0 ZrTi-MAO 1.6 65.2 8.0 15.3 1.6 8.4 Fig. 6. XPS spectra of MAO coatings. R. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 8 indistinguishable, indicating high dispersion of the Zr present in the layer. According to Ding et al.[50], who studied the effect of ZrSi 2 particles on phenol degradation, the determined binding energy E B = 179.2 eV corresponds to the Zr/Si structure. The component with a lower binding energy E B = 177.6 eV could not be identified according to the spectrum atlas [51]. The XPS spectrum of Zr 3d 3/2 and Zr 3d 5/2 of the ZrTi-MAO sample is probably influenced by the atomic dispersion of zirconium and/or by the change in the coordination number of zirconia to form a complex oxide phase, which appears to occur on the sample surface [53]. The Zr 3d 5/2 binding energy values that were determined are lower than for pure ZrO 2 (182.5 eV). The Na 1 s peak at 1070.2 and 1067.9 eV binding energies was also registered on the surfaces of the sample (Ti-MAO, ZrTi-MAO). The wettability of the surfaces was evaluated using the determined contact angles (CA) of the substrate and of the MAO coatings (Supplementary Fig. S1). Compared to the substrate, which had CA = 89.07°± 3.28, the determined CA of the modified MAO coatings showed hydrophilic properties corresponding to CA = 72.32°± 4.15 of the Ti-MAO coating and CA = 69.01°± 5.63 of the ZrTi-MAO coating. The results confirm the well-known fact that MAO coatings, with their porous structure and greater roughness, are more hydrophilic than unmodified substrates [54]. 3.2. Wear behavior and coating adhesion The course of the coefficient of friction was first tested on the Ti-6Al-4V base material without surface treatment. Subsequently, this was compared with the course of the coefficient of friction for samples with a coating of pure Ti-PVD and ZrTi-PVD without and with MAO treatment. Fig. 7 shows that, after running-in, the coefficient of friction for the MAO samples is very stable, and that adhesive friction predominates. In the base material and in pure Ti and ZrTi metals, however. the course of the coefficient of friction is less stable, and a combination of adhesive and abrasive friction prevails. The average coefficient of friction of the tested samples is very similar, and ranges from 0.4 to 0.5, except for the ZrTiMAO coating, for which the friction coefficient decreased to 0.3. The wear rate (Fig. 9b) was determined from the wear track profile (Fig. 8), and the width of the wear track was used as an additional parameter for wear comparison (Fig. 9a). For the TiPVD and ZrTi-PVD samples, the measured values were comparable to the values for the Ti-6Al-4V alloy. Both the Ti-PVD coatings and the ZrTi-PVD coatings were abraded to the Ti-6Al-4V base material. This is also confirmed by the analysis of the chemical composition of EDX after tribological tests in PBS presented in Table 6. The chemical composition in the middle of the trace is comparable to the chemical composition of the Ti-6Al-4V base material. In contrast, a significant decrease in the wear rate was confirmed for the MAO coatings. Only the outer layer (Fig. 8) of the Ti-MAO coating was abraded. This is confirmed in Table 6, where there is a clear decrease in Si in the middle of the friction trace. The amount of Si dropped from 21.0 wt% to 3.8 wt%. However, the ZrTi-MAO coating clearly showed the best wear resistance properties. The outer layer was minimally abraded at the point of friction (Fig. 8). This is confirmed by a comparison between the chemical composition in the middle of the wear trace and on the outside of the wear trace, which is unchanged (Table 5). The ZrTi-MAO sample confirmed low friction and very good wear resistance in comparison with the Ti-MAO sample (Fig. 9 and Table 5), and also in comparison with the MAO sample (basic Ti-6Al-4V material modified with MAO), which was tested in our previous work [36]. Because only a small amount of surface wear was detected on the ZrTi-MAO sample, the chemical and bonding composition analysed by XPS can be used to explain this significant difference in Fig. 7. A comparison of the friction coefficients of the Ti-6Al-4V, Ti-PVD and ZrTi-PVD samples (a), the Ti-MAO and ZrTi-MAO samples (b), and their running-in phases (c, d). R. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 9 [25] B. Zhang, B. Li, S. Gao, Y. Li, R. Cao, J. Cheng, R. Li, E. Wang, Y. Guo, K. Zhang, J. Liang, B. Liu, Y-doped TiO 2 coating with superior bioactivity and antibacterial property prepared via plasma electrolytic oxidation, Mater. Des. 192 (2020), https://doi.org/10.1016/j.matdes.2020.108758 108758. [26] M.J. Mazinani, R. Nine, G. Chiesa, P. Candiani, T.T. Tarsini, D. Tung, Losic, Graphene oxide (GO) decorated on multi-structured porous titania fabricated by plasma electrolytic oxidation (PEO) for enhanced antibacterial performance, Mater. Des. 200 (2021), https://doi.org/10.1016/ j.matdes.2020.109443 109443. [27] S.H. Kang, W.B. Tu, J.X. Han, Z. Li, Y.L. Cheng, A significant improvement of the wear resistance of Ti6Al4V alloy by a combined method of magnetron sputtering and plasma electrolytic oxidation (PEO), Surf. Coat. Technol. 358 (2019) 879–890, https://doi.org/10.1016/j.surfcoat.2018.12.025. [28] A. Kozelskaya, G. Dubinenko, A. Vorobyev, A. Fedotkin, N. Korotchenko, A. Gigilev, E. Shesterikov, Y. Zhukov, S. Tverdokhlebov, Porous CaP coatings formed by combination of plasma electrolytic oxidation and rf-magnetron sputtering, Coatings 10 (11) (2020) 1113. [29] S. Y. Park, H. Ch. Choe, Mn-coatings on the micro-pore formed Ti-29Nb-xHf alloys by RF-magnetron sputtering for dental applications, Appl Surf Sci., 432 (2018), pp. 432:278-284. DOI.org/10.1016/j.apsusc.2017.08.023. [30] J. Hwang, H.C. Choe, Surface morphology and cell behavior of Zn-coated Ti6Al-4V alloy by RF-sputtering after PEO-treatment, Surf. Coat. Technol. 361 (2019) 386–395, https://doi.org/10.1016/j.surfcoat.2019.01.061. [31] Y.L. Zhou, M. Niinomi, T. Akahori, H. Fukui, H. Toda, Corrosion resistance and biocompatibility of Ti–Ta alloys for biomedical applications, Mater. Sci. Eng., A 398 (2005) 28–36, https://doi.org/10.1016/j.msea.2005.03.032. [32] C. Wang, F. Ma, P. Liu, J. Chen, X. Liu, K.e. Zhang, W. Li, Q. Han, The influence of alloy elements in Ti6Al4V and Ti35Nb2Ta3Zr on the structure, morphology and properties of MAO coatings, Vacuum 157 (2018) 229–236, https://doi.org/ 10.1016/j.vacuum.2018.08.054. [33] S.L. Aktug, I. Kutbay, M. Usta, Characterization and formation of bioactive hydroxyapatite coating on commercially pure zirconium by micro arc oxidation, J. Alloy. Compd. 695 (2017) 998–1004, https://doi.org/10.1016/ j.jallcom.2016.10.217. [34] Y. Yan, Y. Han, D. Li, J. Huang, Q. Lian, Effect of NaAlO 2 concentrations on microstructure and corrosion resistance of Al 2 O 3 /Zr 2 O 3 coatings formed on zirconium by micro-arc oxidation, Appl. Surf. Sci. 256 (2010) 6359–6366, https://doi.org/10.1016/j.apsusc.2010.04.017. [35] L. Wang, X. Hu, X. Nie, Deposition and properties of zirconia coatings on a zirconium alloy produced by pulsed DC plasma electrolytic oxidation, Surf. Coat. Technol. 221 (2013) 150–157, https://doi.org/10.1016/ j.surfcoat.2013.01.040. [36] R. Gabor, M. Doubkova, S. Gorosova, K. Malanik, M. Vandrovcova, L. Cvrcek, K. Drobikova, K. Mamulova Kutlakova, L. Bacakova, Preparation of highly wettable coatings on Ti–6Al–4V ELI alloy for traumatological implants using micro-arc oxidation in an alkaline electrolyte, Sci. Rep. 10 (2020) 19780, https://doi.org/10.1038/s41598-020-76448-w. [37] S.N. Rosenbloom, R.A. Corbett, An assessment of ASTM F 2129 test results comparing nitinol to other implant alloys, In Proceedings of the SMST-2006, Proceedings of the International Conference on Shape Memory and Superelastic Technologies (ASM International), 2008. [38] G.Y. Du, D.C. Ba, Z. Tan, W. Sun, K. Liu, Q.K. Han, Vibration damping performance of ZrTiN coating deposited by arc ion plating on TC4 Titanium alloy, Surf. Coat. Technol. 229 (2013) 172–175, https://doi.org/10.1016/ j.surfcoat.2012.05.140. [39] R. Gabor, L. Cvrc ˇek, S. Causidu, K. Drobíková, M. Vec ˇer ˇ, K.M. Kutláková, M. Bur ˇil, J. Hlinka, J. Seidlerová, Effect of additive for preparation of reduced-porosity ceramic layer on Ti-6Al-4V alloy for orthopaedic and trauma implants, Surf. Interfaces 25 (2021), https://doi.org/10.1016/j.surfin.2021.101209 101209. [40] Y. Cheng, F. Wu, E. Matykina, P. Skeldon, G.E. Thompson, The influences of microdischarge types and silicate on the morphologies and phase compositions of plasma electrolytic oxidation coatings on Zircaloy-2, Corros. Sci. 59 (2012) 307–315, https://doi.org/10.1016/j.corsci.2012.03.017. [41] R. Wang, X. He, Y. Gao, X. Zhang, X. Yao, B. Tang, Antimicrobial property, cytocompatibility and corrosion resistance of Zn-doped ZrO 2 /TiO 2 coatings on Ti6Al4V implants, Mater. Sci. Eng., C 75 (2017) 7–15, https://doi.org/10.1016/j. msec.2017.02.036. [42] Y.C. Jung, K.R.Shin, Y.G.Ko, D.H. Shin, Surface characteristics and biological response of titanium oxide layer formed via micro-arc oxidation in K3PO4 and Na3PO4 electrolytes, J. Alloys Compd., 586 (2014), pp. S549-S552. DOI.org/ 10.1016/j.jallcom.2013.01.060. [43] S. Durdu, M. Usta, A.S. Berkem, Bioactive coatings on Ti6Al4V alloy formed by plasma electrolytic oxidation, Surf. Coat. Technol. 301 (2016) 85–93, https:// doi.org/10.1016/j.surfcoat.2015.07.053. [44] J.M. Ríos, D. Quintero, J.G. Castan ˇa, F. Echeverría, M.A. Gómez, Comparison among the lubricated and unlubricated tribological behaviour of coatings obtained by PEO on the Ti6Al4V alloy in alkaline solutions, Tribol. Int. 128 (2018) 1–8, https://doi.org/10.1016/j.triboint.2018.07.010. [45] E. Matykina, R. Arrabal, P. Skeldon, G.E. Thompson, P. Wang, P. Wood, Plasma electrolytic oxidation of a zirconium alloy under AC conditions, Surf. Coat. Technol. 204 (2010) 2142–2151, https://doi.org/10.1016/ j.surfcoat.2009.11.042. [46] F. Jaspard-Mécuson, T. Czerwiec, G. Henrion, T. Belmonte, L. Dujardin, A. Viola, J. Beauvir, Tailored aluminium oxide layers by bipolar current adjustment in the Plasma Electrolytic Oxidation (PEO) process, Surf. Coat. Technol. 201 (2007) 8677, https://doi.org/10.1016/j.surfcoat.2006.09.005. [47] E. Matykina, R. Arrabal, P. Skeldon, G.E. Thompson, Investigation of the growth processes of coatings formed by AC plasma electrolytic oxidation of aluminium, Electrochim. Acta 54 (2009) 6767–6778, https://doi.org/10.1016/ j.electacta.2009.06.088. [48] Y.Y. Wang, K. Kusumoto, C.J. Li, XPS analysis of SiC films prepared by radio frequency plasma sputtering, Phys Procedia 32 (2012) 95–102, https://doi.org/ 10.1016/j.phpro.2012.03.524. [49] Y. Wang, J. Lou, L. Zeng, J. Xiang, S. Zhang, J. Wang, F. Xiong, C.h. Li, Y. Zhao, R. Zhang, Osteogenic potential of a novel microarc oxidized coating formed on Ti6Al4V alloys, Appl. Surf. Sci. 412 (2017) 29–36, https://doi.org/10.1016/j. apsusc.2017.03.191. [50] J. Ding, Z. Huang, Y. Qin, M. Shi, C.h. Huang, J. Mao, Improved ablation resistance of carbon-phenolic composites by introducing zirconium silicide particles, Compos. B Eng. 82 (2015) 100–107, https://doi.org/10.1016/ j.compositesb.2015.08.023. [51] Handbook of Photoelectron Spectroscopy, Edited by Jill Chastain, PerkinElmer Corporation, Minesota, USA, 1992. [52] V.V. Atuchin, V.G. Kesler, N.V. Pervukhina, Z. Zhang, Ti 2p and O 1s core levels and chemical bonding in titanium-bearing oxides, J Electron Spectros Relat Phenomena 152 (2006) 18–24, https://doi.org/10.1016/j.elspec.2006.02.004. [53] B.M. Reddy, B. Chowdhury, P.G. Smirniotis, An XPS study of the dispersion of MoO 3 on TiO 2 –ZrO 2 , TiO 2 –SiO 2 , TiO 2 –Al 2 O 3 , SiO 2 –ZrO 2 , and SiO 2 –TiO 2 –ZrO 2 mixed oxides, Appl Catal A Gen 211 (2001) 19–30, https://doi.org/10.1016/ S0926-860X(00)00834-6. [54] K. Wang, D. Xiong, Construction of lubricant composite coating on Ti6Al4V alloy using micro-arc oxidation and grafting hydrophilic polymer, Mater. Sci. Eng., C 90 (2018) 219–226, https://doi.org/10.1016/j.msec.2018.04.057. [55] J. Song, Z. Huang, Y. Qin, H. Wang, M. Shi, Effects of zirconium silicide on the vulcanization, mechanical and ablation resistance properties of ceramifiable silicone rubber composites, Polymers 12 (2020), https://doi.org/10.3390/ polym12020496. [56] J. Li, X. He, G. Zhang, R. Hang, X. Huang, B. Tang, X. Zhang, Electrochemical corrosion, wear and cell behavior of ZrO 2 /TiO 2 alloyed layer on Ti-6Al-4V, Bioelectrochemistry 121 (2018) 105–114, https://doi.org/10.1016/j. bioelechem.2018.01.011. [57] M. Itagaki, S. Suzuki, I. Shitanda, K. Watanabe, Electrochemical impedance and complex capacitance to interpret electrochemical capacitor, Electrochemistry 75 (2007) 649–655, https://doi.org/10.5796/electrochemistry.75.649. [58] T.K. Yeh, P.I. Wu, C.H. Tsai, Corrosion of ZrO 2 treated type 304 stainless steels in high temperature pure water with various amounts of hydrogen peroxide, Prog. Nucl. Energy 57 (2012) 62–70, https://doi.org/10.1016/j. pnucene.2011.12.014. [59] U. Angst, M. Büchler, On the applicability of the Stern-Geary relationship to determine instantaneous corrosion rates in macro-cell corrosion, Mater. Corros. 27 (2014), https://doi.org/10.1002/maco.201407997. [60] S. Papavinasam, in: Trends in Oil and Gas Corrosion Research and Technologies, Elsevier, 2017, pp. 663–688. [61] W.Q. Yu, J. Qiu, F.Q. Zhang, In vitro corrosion study of different TiO 2 nanotube layers on titanium in solution with serum proteins, Colloids Surf. B: Biointerfaces 84 (2011) 400–405, https://doi.org/10.1016/ j.colsurfb.2011.01.033. [62] Q. Zhang, M. Zheng, Y. Huang, H.J. Kunte, X. Wang, Y. Liu, C. Zheng, Long term corrosion estimation of carbon steel, titanium and its alloy in backfill material of compacted bentonite for nuclear waste repository, Sci. Rep. 9 (2019), https://doi.org/10.1038/s41598-019-39751-9. [63] F.E.T. Heakal, K.A. Awad, Electrochemical corrosion and passivation behavior of titanium and its Ti-6AL-4V alloy in low and highly concentrated HBr solutions, Int. J. Electrochem. Sci. 6 (2011) 6483–6502. [64] A. Yilmazbayhan, A.T. Motta, R.J. Comstock, G.P. Sabol, B. Lai, Z. Cai, Structure of zirconium alloy oxides formed in pure water studied with synchrotron radiation and optical microscopy: Relation to corrosion rate, J. Nucl. Mater. 324 (1) (2004) 6–22. [65] E. Mujanovic ´, B. Zajec, T. Kosec, A. Legat, S. Hönig, G. Zehethofer, G. Mori, Activation and repassivation of stainless steels in artificial brines as a function of pH, Materials 12 (2019) 3811, https://doi.org/10.3390/ma12233811. [66] S.N. Rosenbloom, R.A. Corbett, An Assessment of Astm F 2129 Electrochemical Testing of Small Medical Implants - Lessons Learned, Corros. Conf. Expo (CORROSION 2007) 2007, pp. 105–114. [67] M. Rahmati, E.A. Silva, J.E. Reseland, C.A. Heyward, H.J. Haugen, Biological responses to physicochemical properties of biomaterial surface, Chem. Soc. Rev. 49 (2020) 5178–5224, https://doi.org/10.1039/D0CS00103A. [68] Z. Wang, M. Zhang, Z. Liu, Y. Wang, W. Dong, S. Zhao, D. Sun, Biomimetic design strategy of complex porous structure based on 3D printing Ti-6Al-4V scaffolds for enhanced osseointegration, Mater. Des. 218 (2022), https://doi. org/10.1016/j.matdes.2022.110721 110721. [69] W. Zhang, G. Wang, Y. Liu, X. Zhao, D. Zou, C.h. Zhu, Y. Jin, Q. Huang, J. Sun, X. Liu, X. Jiang, H. Zreiqat, The synergistic effect of hierarchical micro/nanotopography and bioactive ions for enhanced osseointegration, Biomaterials 34 (2013) 3184–3195, https://doi.org/10.1016/j.biomaterials.2013.01.008. [70] X. Li, M. Liu, F. Chen, Y. Wang, M. Wang, X. Chen, Y. Xiao, X. Zhang, Design of hydroxyapatite bioceramics with micro-/nano-topographies to regulate the osteogenic activities of bone morphogenetic protein-2 and bone marrow stromal cells, Nanoscale 12 (2020) 7284–7300, https://doi.org/10.1039/ C9NR10561A. [71] Y. Liu, Z. Rui, W. Cheng, L. Song, Y. Xu, R. Li, X. Zhang, Characterization and evaluation of a femtosecond laser-induced osseointegration and an antiR. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 16 inflammatory structure generated on a titanium alloy, Regen Biomater. 8 (2021) rbab006, https://doi.org/10.1093/rb/rbab006. [72] J. Barberi, S. Ferraris, A.M. Giovannozzi, L. Mandrile, E. Piatti, A.M. Rossi, S. Spriano, Advanced characterization of albumin adsorption on a chemically treated surface for osseointegration: An innovative experimental approach, Mater. Des. 218 (2022), https://doi.org/10.1016/j.matdes.2022.110712 110712. [73] M.T. Tsai, Y.Y. Chang, H.L. Huang, Y.H. Wu, T.M. Shieh, Micro-arc oxidation treatment enhanced the biological performance of human osteosarcoma cell line and human skin fibroblasts cultured on titanium–zirconium films, Surf. Coat. Technol. 303 (2016) 268–276, https://doi.org/10.1016/ j.surfcoat.2016.03.001. [74] B.A. Lee, H.J. Kim, Y.Z. Xuan, Y.J. Park, H.J. Chung, Y.J. Kim, Osteoblastic behavior to zirconium coating on Ti-6Al-4V alloy, J Adv Prosthodont 6 (2014) 512–520, https://doi.org/10.4047/jap.2014.6.6.512. [75] S. Acharya, A.G. Panicker, D.V. Laxmi, S. Suwas, K. Chatterjee, Study of the influence of Zr on the mechanical properties and functional response of Ti-NbTa-Zr-O alloy for orthopedic applications, Mater. Des. 164 (2019), https://doi. org/10.1016/j.matdes.2018.107555 107555. R. Gabor, L. Cvrc ˇek, M. Doubková et al. Materials & Design 219 (2022) 110811 17