The improvement in the performance and durability of tissue- and bone-cutting devices is dependent on the structural, morphological, mechanical and biological properties of hard coatings applied on the surface.
Titanium nitride (TiN) is a well-known, extremely hard, biocompatible ceramic material [2
]; moreover, it possesses high electron conductivity and mobility, as well as a high melting point [3
]. Its excellent physical and chemical properties, which can be varied in a broad range [4
], have generated great interest and have been exploited in the field of hard and protective coatings [6
]. We note that TiN presents good wear and corrosion resistance properties when used in physiological environments [10
TiN films synthesized by ion beam deposition [12
], direct current magnetron sputtering [13
], cathodic arc evaporation [14
], pulsed laser deposition [15
], chemical precursor synthesis [17
], chemical vapor deposition (CVD) and plasma-assisted CVD techniques [18
] are used for a wide variety of applications, including biomedical ones [20
], with specific demands for improving the wear resistance, adhesion to the substrate and fatigue [22
]. After applying TiN coatings, the biocompatibility of implants manufactured from various metallic alloys (such as cobalt-chromium, chromium-nickel or titanium alloys) is improved, increasing the wear and corrosion resistance and avoiding allergic reactions that may occur when a metallic implant is introduced inside the human body [23
Stainless steel is a versatile class of material that has high strength and resistance to oxidation. As compared to Ti, it is easy to machine and, thus, commonly used for surgical instruments, bone screws and other medical equipment [25
]. Grade 410 stainless steel is the basic martensitic stainless steel. The surgical devices fabricated from martensitic stainless steel are currently used as a standard tool for soft and hard tissue surgery [26
In the field of high-quality thin film growth, pulsed laser deposition (PLD) has proven to be a quite versatile method [27
]. In this technique, a very intense pulsed laser beam is focused onto a target in order to ablate its surface under vacuum or different process gas atmospheres, and the vapors are collected on substrates in the form of thin films [27
]. One common modification to the basic PLD technique involves the introduction of ambient gases during the deposition process [29
]. In particular, nitrogen (N2
) presence in the deposition chamber has the role of varying the chemical composition of the films [27
]. This way, the properties of the coatings can be easily tailored for a wide variety of applications. The interaction of the plume with the environment, which takes place in the gas phase, but also on the target and substrate surface, plays an important role in generating the atomic and molecular precursors necessary for the growth of compound phases.
It is to be mention that currently, blade edges are covered with TiN films by different physical vapor deposition processes with the aim of obtaining cutting devices with important advantages, such as increased tool life and cutting speed rates [30
The aim of the present study was to tailor the thickness of TiN films grown by the PLD method in order to be applied as thin protective layers for medical devices (e.g., drills, burrs, scalpels or chisels) used for cutting purposes in soft and hard tissue surgery. We shall show that even small differences in films thickness affect the biological response of these surfaces. A thorough physical-chemical investigation was conducted in order to unveil the factors that vary with thickness, and that are responsible for the particular biological behavior of films.
2. Experimental Section
2.1. PLD Experiment
PLD experiments have been performed inside a stainless steel reaction chamber using a KrF* excimer laser source (λ = 248 nm, τFWHM ≤ 25 ns), running at a repetition rate of 10 Hz. The laser beam was incident at 45° with respect to the target surface.
TiN pellets from Plasmaterials (2.5 cm diameter × 0.6 cm thickness) were used as targets, while grade 410-L stainless steel plates (further denoted as 410SS), with dimensions of 2.3 × 1.8 × 0.2 cm3, or Si (100) wafers and soda-lime glasses of 1 × 1 cm2 from Thermo Scientific were used as deposition substrates. A study was performed prior to the deposition experiments, taking all cautions to assure similar characteristics for the films, irrespective of the nature of the substrates. We note that all of the substrates had a mirror-polished surface quality. The experimental conditions were identical for all substrates. The target and substrate were assembled in a frontal (on-axis) geometry, at a 5-cm separation distance. The laser fluence onto the target surface was set at ~5 J/cm2 (the corresponding pulse energy was 500 mJ).
Before deposition, the targets were submitted to a “cleaning” procedure with laser pulses. During this treatment, a shutter was interposed between the target and the substrate to collect the potential remnant impurities and defects on the target surface. The targets were continuously rotated with 0.4 Hz and translated along two orthogonal axes to avoid drilling and to ensure a uniform deposition.
Prior to introduction inside the deposition chamber, in order to eliminate micro-impurities, the substrates were successively cleaned in an ultrasonic bath in acetone, ethanol and deionized water for 20 min and then blown dry with high purity N2. During deposition, the substrates were heated and maintained at a constant temperature of 500 °C using a PID-EXCEL temperature controller. A ramp of 25 °C/min was chosen for heating the substrates to reach the deposition temperature. Once the film growth was completed, cooling down to room temperature (RT) was performed in the same ambient gas atmosphere as used for the film growth, with a ramp of 10 °C/min.
Before each experiment, the reaction chamber was evacuated with a high vacuum installation down to a residual pressure of 10−5 Pa. The dynamic ambient gas pressure during the thin film growth was kept at around 0.2 Pa by feeding high-purity N2 into the chamber with the aid of a calibrated gas inlet. An MKS 4000 controller was used for monitoring the gas pressure.
Three sets of TiN samples with 5000 (further denoted as 5A), 10,000 (further denoted as 10B) and 20,000 (further denoted as 20C) subsequent laser pulses were deposited. The number of pulses corresponds to the following deposition times: 8 min and 20 s (5A), 16 min and 40 s (10B) and 33 min and 20 s (20C).
2.2. Physical-Chemical Characterization of Deposited Structures
2.2.1. Biological Assays
18.104.22.168. Cell Cultures
Fibroblasts, Hs27 cell-line from ATCC, were used to assess the cytotoxicity of TiN thin films. Cells were grown with Dulbecco’s Modified Eagle’s Medium with l-glutamine (DMEM), supplemented with 10% bovine fetal serum, penicillin (100 UI/mL) and streptomycin (100 µg/mL), in a humidified atmosphere incubator with 5% CO2, at 37 °C.
The square substrates with a surface of 1 cm2 were cleaned with 70% and absolute ethanol, sterilized with dry heat (180 °C/1 h) and transferred in sterile 24-well plates, in the cell culture laminar flow hood.
When cells reached confluence, they were detached with trypsin, collected and centrifuged at 250× g for 10 min after trypsin inhibition. Cells were re-suspended with complete growing medium, counted with a Bürker-Türk counting chamber, and the concentration was adjusted to 105 cells/mL. On each sample, 104 cells in 100 µL DMEM were seeded. The plates were put in the humidified atmosphere incubator and allowed to adhere for 5 h. After this interval, 400 µL were added, and the cells were allowed to grow in the incubator for another 24 h. After 24 h, the medium was collected in order to assess the cytotoxic effect with a lactate dehydrogenase (LDH) experiment, and cells were subjected to further analysis. All cell culture reagents were purchased from Sigma Aldrich (St. Louis, MO, USA).
22.214.171.124. Cell Morphology
Cells grown on different samples (410SS bare substrate and 5A, 10B and 20C TiN films) were examined by means of fluorescence imaging to observe any morphology modifications that can occur after their development on the films. Cells were fixed with 4% para-formaldehyde dissolved in phosphate-buffered saline (PBS) for 15 min at RT. The cells were then washed thrice with PBS and then incubated at RT, for 1 h, with 100 µL of phalloidin-AlexaFluor546 (Invitrogen, Carlsbad, CA, USA) diluted accordingly to the manufacturer's specifications. Samples were washed thrice with PBS for 15 min each and then incubated with 1 µg/mL 4',6-diamidino-2-phenylindole (DAPI), produced by Sigma Aldrich (St. Louis, MO, USA). After incubation, the cells were washed twice with PBS and once with double-distilled water for 15 min (each wash) and then mounted using 0.17-mm thin glass microscopy coverslips and fluorescence mounting medium (Invitrogen). Samples were examined using a Zeiss Axioplan fluorescence microscope with appropriate filters.
126.96.36.199. Proliferation Assay
Cells’ proliferation on the surface of the samples was investigated using an MTS kit (Promega, Madison, WI, USA). Briefly, after the cell culture medium was removed, 400 µL of fresh DMEM without phenol red were added, and plates were transferred it the incubator. After 30 min, in each well, 80 µL MTS ready-to-use reagent were added, and the plates were introduced back into the incubator. After 1 h of incubation, the plates were put on a gentle shaker (150 rpm/1 min), and then, 120 µL of medium were harvested from each well and transferred into 96 microplates. Absorption was read at 490 nm with a Zenyth 3100 multimode microplate reader (Anthos).
188.8.131.52. Cell Toxicity Assay
Cell toxicity was investigated using an LDH activity Kit (Thermo Scientific, Waltham, MA, USA). After 24 h, the supernatant medium was harvested, and 50 µL were transferred in 96 micro-well plates. Fifty microliters of LDH substrate solution, prepared according to the producer’s specification, were added, and the plates were transferred to the incubator. After 30 min, the reaction was stopped by the addition of 50 µL of stop buffer, and absorptions were read at 490 and 620 nm with a Zenyth 3100 multimode microplate reader (Anthos). For each situation, the values were calculated by subtracting the 620-nm values from the 490 ones. The control of the experiment for LDH activity was represented by the fresh complete medium.
The thickness of the TiN thin films was determined by profilometry using an Ambios Technology XP-2 Stylus Profiler (Santa Cruz, CA, USA). The profilometry measurements were performed at RT, on films deposited onto glass substrates.
2.2.3. Scanning Electron Microscopy
The general morphology of the deposited films was investigated by scanning electron microscopy (SEM) with an FEI Inspect S electron microscope. The measurements were carried out at 5-kV acceleration voltages, in high vacuum, under secondary electron acquisition mode. Cross-section SEM images were recorded on samples deposited on Si (100) wafers, in order to evaluate film thickness.
2.2.4. Atomic Force Microscopy
Further, more insightful morphology analyses were carried out by atomic force microscopy (AFM) with a MultiView 4000 Nanonics system (Jerusalem, Israel) working in phase feedback. For AFM investigations, glass substrates were used. For each TiN-deposited film, the root mean square (RRMS) and average (Ra) roughness values were inferred. In order to check the uniformity of films, different zones (having areas of 2 × 2 μm2) on the same sample were scanned, and the corresponding images were acquired. All measurements were performed at RT.
2.2.5. X-ray Photoelectron Spectroscopy
X-ray photoelectron spectroscopy (XPS) analyses were performed in a dedicated chamber (Specs GmbH, Berlin, Germany), under ultra-high vacuum (base pressure ~10−9 mbar), using Al Kα = 1486.71 eV monochromatized radiation. The electrons were collected using a hemispherical electron energy analyzer (Phoibos 150) operated at a 20-eV pass energy. Resolution (in terms of full width at half maximum) of 0.45 eV was achieved. During measurements, a flood gun operating at a 1-eV acceleration energy and a 1-mA electron current was used in order to ensure sample neutralization. The TiN samples were investigated in (i) normal incidence (N.E.) with respect to the electron analyzer and (ii) at a 60° take-off angle (T.O.) in order to increase the surface sensitivity.
The experimental data were fitted with Voigt profiles (i.e., a Gaussian line convoluted with a Lorentzian one). The Gaussian components account mainly for the instrumental resolution, while the Lorentzian line is connected to the finite core-hole lifetime associated with the photoionization process.
2.2.6. X-ray Diffraction
The crystalline phase identification of films was carried out by X-ray diffraction (XRD) using a Bruker AXS D8 Advance diffractometer (Karlsruhe, Germany), with CuKα radiation, with parallel beam optics in both symmetric (θ–θ) and grazing incidence (α = 2°) geometries. XRD patterns were recorded in the 2θ = 30°–80° angular range, with a step size of 0.04° and 25 s per step.
The wetting properties of the samples were studied by measuring the static contact angle (CA) with a Drop Shape Analysis system, Model DSA100, from Kruss GmbH (Hamburg, Germany). The samples (glass and 410SS) were placed on a planar stage, under the tip of a water-dispensing disposable blunt-end stainless steel needle with an outer diameter of 0.05 cm. For CA measurements, two water droplets were poured on each sample. The volume of one water droplet was approximately 2 μL. The needle was attached to a syringe pump controlled by a PC for the delivery of the water droplet to the test surface. The drop size and the drip distance were kept constant in all cases. The dispensing of the droplet and analysis of the CA and of other drop parameters were carried out by the PC using the DSA3®
software supplied with the instrument. CA was measured by fitting a polynomial equation of second degree or a circle equation to the shape of the sessile drop and then calculating the slope of the tangent to the drop at the liquid-solid-vapor interface line. The camera was positioned so as to observe the droplet under an angle of about 2°–3° with respect to the plane of the sample surface supporting the droplet. The tests were carried out at RT [31
Solid surface free energy (SFE) calculations were performed based on CA measurements, using ethylene glycol and deionized water as standard wetting solvents. The concept of polar and dispersion components (Owens-Wendt approximation) was applied [36
]. For each investigated surface, five experiments were carried out. The mean value and standard deviation (SD) were computed.
The adherence at the 410SS substrate-TiN films’ interface was measured by the pull-out method. This is a quantitative test that uses a certified adhesion tester, Model PAT MICRO AT101, from DFD Instruments (Kristiansand, Norway), equipped with 0.28-cm diameter stainless steel test elements. They were affixed onto the coating’s surface by a cyanoacrylate one-component epoxy adhesive, E1100S. The stub surface was first polished, ultrasonically degreased in acetone and ethanol and then dried in N2
flow. After gluing, the samples were placed in an oven for thermal curing (130 °C/1 h). By use of a portable pull-out adherence tester, a load was increasingly applied to the surface until the dolly was pulled off. Failure will occur along the weakest plane within the system comprised of the dolly, adhesive, coating system and substrate. The experimental procedure was conducted in accordance with the ASTM D4541 [37
] and ISO 4624 [38
] standards. Batches of five samples have been tested for each type of material, and a statistical estimation is given.
2.2.9. Statistical Analysis
All experiments were carried out in triplicate in order to achieve statistical significance. The statistical analyses were performed using the unpaired Student’s t-test, and the differences were considered significant when p < 0.05.
We note that, during our studies, no differences between films deposited on various substrates were observed.
Biological tests on TiN films deposited by PLD in different conditions revealed a strong connection between their thickness and biological response. The main question to be answered was: “which is the factor that is thickness dependent and highly influential for biological response?”. The first hypothesis was to check the surface morphology. It is known that crystallite size increases with films thickness [48
], and this could affect the overall structure, arrangement and, likely, the surface morphology of the films. However, microscopic investigations of the surface revealed defect-free and dense structures with roughness values less than 1 nm. Bollen, et al.
] and Größner-Schreiber, et al.
] showed that a surface roughness less than 200 nm has no effect on bacterial adhesion, because most of the bacteria have larger sizes, while Katsikogianni, et al.
] and Tsang, et al.
] showed that an increase in the surface roughness (superior to the aforementioned value) can encourage cell adhesion by creating more favorable sites for colonization inside surface irregularities [54
]. Therefore, it is highly unlikely that the small differences in roughness between our TiN films of different thicknesses could affect the biological behavior of films. Consequently, our attention was focused on film structure and surface chemistry.
XRD investigations demonstrated the existence of B1-structure TiN films. These types of structures have been found to possess key properties for technological applications, such as: hardness, excellent adhesion, high strength and rigidity, wear-corrosion resistance, high thermal stability, low friction coefficient and good chemical stability [55
]. Crystallite size increased with film thicknesses, being ~6 nm larger in the case of the 20C samples as compared to the 5A ones.
The adhesion strength at the coating-surface interface represents a critical parameter for the long-term stability of medical devices. The high values of adherence obtained in the case of TiN coatings should be therefore emphasized.
When a cell adheres and spreads, between the substrate surface and its membrane remains a nanometric space containing water molecules, ions, cell adhesion proteins and substrate-adsorbed proteins and organic substances. Therefore, the spreading of a cell on a surface is promoted by a sum of factors, like the hydrophilic behavior of the surface (SFE), the presence of some moieties, protein adsorption on the surface, surface roughness, texture, etc. In order to predict the cell adhesion and spreading, there is no simple answer, the outcome being determined by all of these factors that are partially linked to one another.
Cell spreading on a surface is a complicated and dynamic process that requires an equilibrium between tension and elastic forces on the membrane surface, cellular internal pressures, cytoskeleton rearrangement, molecule interactions, etc. In our case, the factors are: the polar component of the SFE and the surface composition (TiNxOy concentration and TiN stoichiometry).
Wettability is a critical factor that can directly influence the cells’ adhesion and growth. It is worth noting that in the case of our TiN samples, the water contact angles were between 80° and 90°. Tamada, et al.
] and Chang, et al.
] reported that angles around 80° provide maximum adhesion for fibroblast cell cultures. SFE results have shown that the polar free energy decreases in the case of deposited films as compared to bare substrates. This is probably due to the lack of polar groups on the TiN-finished surfaces, which determines a weakening of the bond strength connected to the polar elements of the liquid. The polar component of surface energy plays a key role in determining the hydrophobic behavior of the surface [58
]. Therefore, in the case of deposited films, the higher the value obtained for CA, the lower is the value for the polar component of SFE. The particular growth of fibroblast cells on the surface of the 10B sample (more elongated compared to the polygonal aspect of the cells seeded onto the other two types of samples (5A and 20C)) could be related to the higher SFE value recorded in the case of this type of surface (26 vs.
The XPS analyses showed differences in terms of the surface and bulk composition of films, depending on their thickness. As the cells “see” only the surface and not the volume of the sample, the attention was mainly focused on the evolution of the surface composition. The films consisted of a mix of titanium nitride (with various stoichiometry) and titanium oxynitride (TiNx
) (Figure 9
c). Further, even though all films exhibit good cytocompatibility, a difference between the biological responses of samples was detected, depending on the TiNx
concentration and TiN stoichiometry. The samples with lower TiNx
surface content induce more cellular death, ~45% with respect to the stainless steel control material (p
< 0.05) vs.
12% in the case of a TiNx
-rich surface. However, Banakh, et al.
] showed that for a wide range of TiNx
ratios, the structures remain biocompatible, without major viability and proliferation differences as a function of the atomic ratios of films. Therefore, it is suggested that in our case, the TiN surface stoichiometry is the key compositional factor that influences the cytotoxicity. The closer the films’ stoichiometry is to the theoretical one, the higher is the biocompatibility. This could be the reason why the thinnest sample (5A), having a Ti:N ratio of 1:1.1, is slightly more biocompatible than the other two (10B and 20C), which contain TiN compounds more enriched in nitrogen.
Our hypothesis is that the TiN stoichiometry variation with thickness (Table 4
) comes from the difference in the atomic weight of titanium and nitrogen. Considering that TiN is ablated congruent from the target, the plasma plume will consist of an approximately equal amount of titanium and nitrogen. Because titanium is heavier, it tends to go deeper into the film, a process that is stimulated by the multi-pulse deposition leading to a higher thickness of the film. Nitrogen would have a tendency to accumulate closer to the surface of the film. When the films are thinner (e.g., 5A), these accumulation areas are very close to each other and even overlap. The reconstitution of TiN occurs under continuous substrate heating. In this case, a Ti/N ratio closer to unity is obtained. In the case of thicker films, there is a wider gap between titanium and nitrogen accumulation areas. Therefore, close to the surface, there are zones richer in nitrogen.
Even under high vacuum, there are still traces of oxygen in the residual gas [28
]. The oxygen becomes more reactive with temperature, and the deposited layer always includes oxide impurities. The phenomenon is more visible in the case of thin films and stands at the origin of the formation of TiNx
compounds (Figure 9
Complex physical-chemical, mechanical and biological studies have been carried out on titanium nitride (TiN) thin films of various thicknesses, synthesized by pulsed laser deposition in low nitrogen pressure in order to explain a better biocompatibility observed in the case of thinner films. By comparing the samples to the control glass substrate, a fibroblast mortality higher with 45% was observed in the case of films of ~130 nm as compared to 12% obtained in the case of ~60 nm-thick films. As all of the films were extremely smooth, with roughness values of under 1 nm and crystallites of comparable sizes, the only main difference between films was found in their surface composition. XPS data revealed that all films’ surfaces consisted of a mix of non-stoichiometric TiN and TiNxOy. The decisive factor for the biological response was found to be the TiN stoichiometry: the closer to unity, the more biocompatible the film is.
A thin TiN film (i.e., 5A ≈ 60 nm) should be considered as the optimum solution. This assumption is further sustained by experimental physical-chemical, mechanical and biological evidence and further allows for minimum production costs and reduced environmental pollution. These support the potential use of very thin TiN films for improving the characteristics of medical instruments (e.g., drills, burrs, scalpels or chisels), where sharpness and edge retention are important.