Li Jinlong*,
Zhong Huasheng and
Wang Yongxin
Key Laboratory of Marine Materials and Related Technologies, Zhejiang Key Laboratory of Marine Materials and Protective Technologies, Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, Ningbo, China 315201. E-mail: Lijl@nimte.ac.cn
First published on 27th October 2016
TiN/TiCN double coatings were deposited on Ti6Al4V alloy by arc ion plating to improve wear resistance of the titanium alloy in seawater. The TiN interface layer was incorporated between the top TiCN coating and Ti6Al4V substrate. The TiN/TiCN coatings have a typical columnar crystal of the inner TiN coating with a thickness of 1 μm and top TiCN coating with a thickness of 1 μm. The TiN/TiCN coatings have a hardness of about 30 GPa and a low friction coefficient of 0.15, which is far lower than 0.35 of untreated Ti6Al4V alloy. The Ti6Al4V with TiN/TiCN coatings have a minimum wear rate of 2.42 × 10−5 mm3 N−1 m−1, which is far lower than 23.3 × 10−5 mm3 N−1 m−1 for uncoated Ti6Al4V alloy. During a friction process in seawater, these coatings have a small fluctuation in the potential and show more stability. However, the untreated Ti6Al4V substrate shows notable fluctuation in potential and it is more affected by the interactions of corrosion and wear. Compared with the TiN coating and Ti6Al4V alloy, TiN/TiCN coatings have a small amount of the removable surface layer and little chemical transfer film due to a lower friction coefficient, excellent chemical stability, and high hardness. Seawater medium accelerates the initiation of cracks and induces formation of delamination failure of the coatings.
Tribocorrosion examples of titanium alloys can be seen in many applications, and it is an irreversible transformation of a material due to simultaneous actions of corrosion and wear taking place in a tribological contact. One example is where titanium biomaterials have been used widely as surgical implants in physiological fluids of the human body. Release and loss of metallic ions by corrosion and wear is of vital importance for reducing induced failure of implants.12 In recent years, sea exploitation has paid more attention to deficit and exhaustion of land resources. Some key friction components of marine equipment were designed and manufactured using titanium alloys due to their corrosion resistance in a marine environment. In a marine environment, some contact frictions even occur in seawater. The components are subjected to harsh environment coupling wear and corrosion, and moreover, lubricating oil also does not employ well in water media.13
Protective coatings with excellent lubrication properties are a promising way to improve the tribocorrosion property of titanium implants or titanium moving components.14–17 Duanjie Li et al. reported that SiN/SiC/a-C multilayer coatings reduced the wear rate and friction coefficient by a factor of 175 and 4 in 1 wt% NaCl solution using Ti6Al4V alloy.18 R. Bayon et al. studied influence of carbon content on corrosion and tribocorrosion performance of Ti-DLC coatings for titanium alloys.19 As a biomaterial, TiN coating has good biocompatibility,20 but it possesses a high friction coefficient. A little carbon was incorporated into TiN to design the TiCN coating, and that coating was an ideal choice to improve tribological properties of this titanium alloy. Using physical vapor deposition, the TiCN coatings were employed to improve wear resistance and achieve a good effect due to coupling of high hardness and good antifriction properties.21–23 Qianzhi Wang et al. revealed effects of counterparts on the tribological properties of TiCN coatings with low carbon concentration in water lubrication.24 Virginia Saenz de Viteri et al. performed tribocorrosion and fretting tests of the TiCN coating in a phosphate buffered saline solution.25 V. Saenz de Viteri et al. studied the properties of TiCN coatings deposited on Ti6Al4V for biomedical applications.26 However, there is a lack of information on tribological behavior in artificial seawater for the TiCN coated Ti6Al4V alloy.
In seawater, removal of the protective coatings is from mechanical wear and chemical corrosion. These two mechanisms do not proceed independently. There is an interaction between corrosion and wear, i.e. corrosion can affect wear and similarly wear will accelerate corrosion. For a marine environment, the reaction medium with high activity will accelerate crack initiation and growth. In seawater, chemical reactions and a film plays an important role in tribological behavior. However, there is an experimental difficulty in controlling the surface chemistry of contacts; thus, the interaction relevance between surface chemistry and tribological behavior is not well understood. The measurement of open circuit potential was introduced during a wear test, and this combination offers a possibility to follow chemical reactions in situ and in real time. Therefore, it is a good procedure to reveal tribocorrosion behavior from the complex interaction of corrosion and wear.
In this paper, the TiN/TiCN coatings were fabricated on Ti6Al4V by arc ion plating. The structure and tribocorrosion behavior of these coatings were studied in seawater.
The morphologies of the coatings were observed by a field emission scanning electron microscope (S4800) equipped with EDS (OXFORD X-Max). The phase structure was investigated by X-ray diffraction (Bruker D8 X-ray facility) using Cu Kα radiation (λ = 0.154 nm), which was operated at 40 kV and 40 mA. The scanning speed was 4° min−1 with 0.02° step size. Element chemical states were characterized by X-ray photoelectron spectroscopy using an Al Kα source with 12 kV and 10 mA. Survey spectra were collected from 0 to 1200 eV with pass energy of 160 eV. High resolution scans for the elements were performed with pass energy of 20 eV. The C1s peak at 284.6 eV was employed to calibrate the peak position of these core spectra.
Coating adhesion was measured by a scratch tester (CSM Revetest) with a conical diamond tip of 0.2 mm radius and 120° taper angle. The measurements parameters are as follows: table speed 6 mm min−1, loading rate 118 N min−1, loading scale 0–100 N, and scratch length 5 mm. Acoustic emission was detected when the coating was broken and the load at the point of breaking was called the adhesive critical load (LC). A nanoindentation test was carried out using a MTS Nano indenter@G200 system with the continuous stiffness measurement (CSM) option. The drift is 0.1 nm s−1 and Poisson ratio is 0.2. The eight indentations in each sample configured on different areas were performed to have reliable statistics.
Tribocorrosion behaviors were evaluated by a tribocorrosion evaluation apparatus (MFT-R4000). Fig. 1 shows the schematic diagram of the tribocorrosion apparatus. High purity graphite was employed as the counter cathode, samples are the working electrode and a saturated calomel electrode is the reference electrode. In this study, the tribocorrosion tests were carried out in artificial seawater. The artificial seawater was prepared according to Standard ASTMD 1141-98. A load of 5 N was applied to the sample through a Si3N4 ball with a diameter of 6 mm. The sliding distance was 5 mm and frequency was 0.1 Hz. Each test was repeated three times under the same conditions in order to check reproducibility. The morphologies of track were examined using a scanning electron microscope with the energy dispersive spectrometer.
![]() | ||
Fig. 2 SEM images of (a) top view, (b) cross section, and (c) scratch track with an acoustic emission–load curve for TiN/TiCN coatings. |
Fig. 3 shows XPS spectra for the TiN/TiCN coatings. The detected escape depth of the photoelectron is only several nanometers, thus XPS results only reveal element chemical states from the surface of the TiCN coating. The survey spectrum reveals the main elements of Ti, C, and N are in the TiCN coatings, and there also is oxygen introduced from the atmosphere and a little argon from the mixing gases of Ar and N2 during deposition. By deconvoluting the core level spectra of Ti2p, C1s, and N1s, the results reveal that Ti presents as predominant in TiN, TiC, and TiO; N was identified as TiN; C exists as TiC in the TiCN coating. A lot of contamination carbon and a small amount of adsorption nitrogen also was found on the surface of the TiCN coatings. Fig. 4 shows XRD patterns from the TiN/TiCN coatings. TiCN (200) and TiN (111) diffraction peaks were detected. The TiCN peak is particular broad and this implies its crystal size is very small. On the other hand, TiN has a strong diffraction peak and this reveals its excellent crystallization. These results also have good agreement with the cross section images, and the TiN layer has a more clearly columnar structure, but the TiCN layer has a more dense structure. From XRD patterns, the strong Ti diffraction peaks (100), (101), and (002) also have been found. The reason is that the coatings are not too thick, thus signals were detected from the Ti6Al4V substrate.
Fig. 5 shows variation of hardness and modulus with indentation depth for the TiN/TiCN coatings, TiN coating, and substrate. Hardness was identified at the plateau of the hardness curves. The hardness of coatings is far higher than 5 GPa of Ti6Al4V substrate, and they are 30 GPa and 27 GPa for the TiN/TiCN coatings and TiN coating, respectively. In the near surface (the depth from 100 nm to 250 nm), hardness was not very high, this revealed that the out TiCN coating has a relatively low hardness. The hardness difference of the TiN/TiCN coatings and TiN coating also can be seen from the load–displacement curves. For the same indentation depth of 1000 nm, there needed to be a bigger load of 250 mN for the TiN/TiCN coatings compared with the TiN coating. This implies the TiN/TiCN coatings have a bigger resistance to plastic deformation and better load bearing capacity. From the load–displacement curves, the elastic recovery (R) can be calculated by the following formulation: R = (hmax − hf)/hmax, where hmax is the maximum indentation depth when the maximum load was applied and hf is the indentation depth after the unload. The TiN/TiCN coatings have a slightly high R of 27.3% compared with 26.3% of the TiN coating. The ratio of hardness to modulus (H3/E2) can predict the ability of a coating to resist mechanical failure, and the higher ratio implies the coating has a higher elastic recovery and excellent resistance to mechanical failure. The TiN/TiCN coatings have a H3/E2 of 0.138, which is higher than 0.096 of the TiN coating. This implies that TiN/TiCN coatings have an excellent toughness and resist mechanical failure.
![]() | ||
Fig. 5 Load–displacement curves and variation of hardness and modulus with indentation depth for TiN/TiCN coatings. |
Fig. 6 shows the wear rate of Ti6Al4V with and without coatings in seawater. The Ti6Al4V coated with TiN/TiCN coatings have a minimum wear rate of 2.42 × 10−5 mm3 N−1 m−1, which is far lower than 23.3 × 10−5 mm3 N−1 m−1 for uncoated Ti6Al4V alloy. For Ti6Al4V coated with TiN coating, the wear rate is 8.32 × 10−5 mm3 N−1 m−1.
In order to reveal tribochemical reaction and transfer film behavior, we monitored in situ the evolution of open circuit potential during tribocorrosion tests. Fig. 7 shows the evolution of friction coefficient and open circuit potential before, during, and after mechanical wear with the same parameters. Before the tribological process, all samples were immersed in artificial seawater for 20 min to make the surface show a stable potential and a passivity surface. Considering friction coefficients, the untreated Ti6Al4V alloy has a friction coefficient of 0.35, which has a big fluctuation during tribocorrosion tests. The coatings have a stable friction coefficient, and they are 0.4 and 0.15 for TiN and TiN/TiCN coatings. For the open circuit potential, there are three different stages along time for all samples. In stage 1, the stable open circuit potential exists in the first soaking process, and the coatings have a higher potential than untreated Ti6Al4V substrate. With stage 2, the potentials sharply decrease to more negative values in all cases while the load is applied to start wear tests. This behavior is related to total or partial removal of the passive layer and exposure of fresh metal substrate to the electrolyte. Untreated Ti6Al4V shows a sharp decline with a large scale, and the TiN/TiCN coatings have a slight decrease of the potential. During tribocorrosion tests, the potential remains a constant value for the coatings, but the untreated Ti6Al4V substrate shows a notable fluctuation in potential. This reveals the Ti6Al4V substrate seems to be more affected by the interaction of corrosion and wear and there is constant increase of the fresh metal area as consequence of the passive layer removed by the counter-ball. On the other hand, the coatings do not present notable fluctuation of potentials, and the reason is that the TiN or TiCN coatings are stable. And in absence of an accelerated corrosion, moreover, the coatings have a small removal of the top layer due to its high hardness (30 GPa is far higher than 5 GPa of Ti6Al4V substrate) and excellent electro-chemical stability in the seawater. The above factors also lead to a very lower wear rate of the TiN/TiCN coating compared with untreated Ti6Al4V and TiN coatings. In stage 3, the potentials quickly increase to a higher value, which is similar to potentials before friction, due to the formation of a stable passive layer on the sample surface while the wear process has stopped.
![]() | ||
Fig. 7 Variation of friction coefficient and open circuit potential with tribocorrosion time in seawater for TiN/TiCN coatings, (a) Ti6Al4V; (b) TiN coating; (c) TiN/TiCN coating. |
Fig. 8 shows the SEM morphologies and composition on the wear track. Fig. 9 presents the element chemical states from the track surface for Ti6Al4V with and without the coating. For the Ti6Al4V substrate, there are apparent abrasive characteristics with deep furrow and serious substrate materials pile-up. As shown in Fig. 9, the element chemical states reveal a lot of TiO2 and SiO2 exist on the track surface. The TiO2 film was formed by oxidation of Ti and SiO2 from the transfer film from friction pairs of Si3N4. The SiO2 films on the wear track were formed by tribochemical reactions. At the friction interface, Si3N4 was first decomposed into Si and N by friction force and friction heating from the wear process. And then Si and O combined to form SiO2.
![]() | (1-1) |
Si + 2O = SiO2 | (1-2) |
![]() | ||
Fig. 8 SEM morphologies and composition on track surface for Ti6Al4V, TiN coating, TiN/TiCN coatings. |
The SiO2 films on the wear track were formed by tribochemical reaction of Si from the Si3N4 ball and O from the seawater during the wear test. This can be explained in that the variation of open circuit potentials with time before and after the tribocorrosion process. When wear started on the Ti6Al4V substrate, the natural formation of oxide film was removed by wear, thus the potential decreased. The titanium alloy is very easy to passively form an oxide film on the surface, especial in seawater media. During a wear process, there is repeated processing of formation and stripping of the oxide film, and this can be reflected by a regular fluctuation of potential. While finishing wear, the potential will increase due to the stable hold of the oxide film. The TiN and TiCN coatings have a good wear resistance and oxide film is difficult to form and remove, thus there is little variation of the potential compared with the Ti6Al4V substrate. Thus, TiN and TiCN coatings have a higher potential and better corrosion resistance compared with Ti6Al4V substrate.
For the TiN coatings, there is a lot of SiO2 on the track surface. The reason is that the TiN coating has higher hardness and thus can lead to much transfer of film from friction pairs of Si3N4. The TiN/TiCN coatings have a smooth track surface, and little local delamination of coating appears on the track. The TiCN coating has an excellent lubrication effect due to carbon incorporated into TiN. The wear test shows TiCN coating has a lower friction coefficient of about 0.15. The XPS results also revealed there is small amount of TiO2 and SiO2 transfer film due to the low friction coefficient and excellent chemical stability during wear tests in seawater. This is why the open circuit potential doesn't have very obvious change before and after the tribocorrosion test.
Based on the results described above, a schematic diagram is shown in Fig. 10 for possible failure mechanisms of the TiN/TiCN coating on Ti6Al4V alloy. The TiN/TiCN coatings have the following failure modes that involve delamination of the top TiCN coating: first, the initiation of micro-cracks induced from the corrosion factor and Hertz contact stress on the coating surface. And then, applied force and water pressure promote propagation of the cracks in corrosion media with high reaction activity. When a crack is extended to the interface between TiCN and TiN coatings it will propagate along the interface due to weak bonding. Lastly, the outer TiCN coating delaminates from the TiN interlayer; this can be identified by composition from the bottom of delamination where there are only Ti and N elements, and C becomes very small compared with surface of the TiN/TiCN coatings. Another failure mode is when the large particles were pulled out of the coatings. These big particles are inevitable for the coatings made by arc plating. Thus, the big particles were pulled out by shear stress from the applied load during a tribocorrosion test. The failure modes can be found from the track morphologies shown in Fig. 7d.
This journal is © The Royal Society of Chemistry 2016 |