Effect of Low-Friction Composite Polymer Coatings Fabricated by Electrophoretic Deposition and Heat Treatment on the Ti-6Al-4V Titanium Alloy’s Tribological Properties

In this work, polytetrafluoroethylene/polyetheretherketone (PTFE/PEEK 708) coatings were fabricated by electrophoretic deposition (EPD) and heat treatment to improve the tribological properties of the Ti-6Al-4V alloy. A stable suspension containing chitosan polyelectrolyte allowed the co-deposition of both polymer types. The effect of soaking temperature and cooling rate on the coating microstructure was determined. The homogeneous coatings were obtained by heating at 450 °C and cooling with a furnace or in water. The crystalline PTFE separate particles and their agglomerates were homogeneously distributed in a PEEK matrix. Interestingly, the PEEK structure was amorphous regardless of the cooling rate after heating. This new phenomenon is associated with the fluorination process. The coatings exhibited moderate scratch resistance but significantly decreased the coefficient of friction (COF) and enhanced the wear resistance of the alloy during the dry friction process in sliding contact with an Al2O3 ball. The COF of the PTFE/PEEK 708 coating equaled 0.10 at room temperature (RT) and was lower than that of the alloy and pure PEEK 708 coating investigated in the same conditions, which reached 0.70 and 0.27, respectively. The wear rate of the coated alloy at RT was lower than that of the uncoated and PEEK 708 coated alloy, around 1900 and 10 times, respectively. Both the COF and wear increased slightly during friction at 150 °C. The coating was not effective in improving the alloy’s tribological properties during friction at 260 °C.

In this work, polytetrafluoroethylene/polyetheretherketone (PTFE/PEEK 708) coatings were fabricated by electrophoretic deposition (EPD) and heat treatment to improve the tribological properties of the Ti-6Al-4V alloy. A stable suspension containing chitosan polyelectrolyte allowed the co-deposition of both polymer types. The effect of soaking temperature and cooling rate on the coating microstructure was determined. The homogeneous coatings were obtained by heating at 450°C and cooling with a furnace or in water. The crystalline PTFE separate particles and their agglomerates were homogeneously distributed in a PEEK matrix. Interestingly, the PEEK structure was amorphous regardless of the cooling rate after heating. This new phenomenon is associated with the fluorination process. The coatings exhibited moderate scratch resistance but significantly decreased the coefficient of friction (COF) and enhanced the wear resistance of the alloy during the dry friction process in sliding contact with an Al 2 O 3 ball. The COF of the PTFE/PEEK 708 coating equaled 0.10 at room temperature (RT) and was lower than that of the alloy and pure PEEK 708 coating investigated in the same conditions, which reached 0.70 and 0.27, respectively. The wear rate of the coated alloy at RT was lower than that of the uncoated and PEEK 708 coated alloy, around 1900 and 10 times, respectively. Both the COF and wear increased slightly during friction at 150°C. The coating was not effective in improving the alloy's tribological properties during friction at 260°C. https://doi.org/10.1007/s11661-020-05900-3 Ó The Author(s) 2020

I. INTRODUCTION
FRICTION processes are the reason for the tribological wear of materials. They cause huge energy, material, and economic losses. A large amount of the world's energy is consumed to overcome friction resistance and, thus, to improve the tribological properties of cooperating components. Titanium alloys are an important group of structural materials, but unfortunately, they are particularly subject to wear by friction due to the high coefficient of friction (COF) as well as poor wear resistance. [1] Despite numerous beneficial properties, such as low density and Young's modulus, good electrochemical corrosion resistance, and a favorable balance between strength and plasticity, [2,3] the use of these materials for components operating in friction and wear conditions is strongly limited.
Surface treatment involving the deposition of a protective coating on their surfaces is one of the solutions that allows the improvement of unfavorable tribological properties and, consequently, increases their potential applications. [4,5] Increasingly important in improving the tribological properties of titanium alloys are composite coatings with a polymer matrix. Polymers, especially thermoplastic ones, are a group of materials combining many favorable properties, such as low density, relatively high mechanical strength, and beneficial tribological properties. [6] In our previous work, [7] we proved that pure polymeric polyetheretherketone (PEEK) coatings significantly improved the tribological properties of the Ti-6Al-4V alloy at room temperature (RT) and elevated temperatures up to 260°C . PEEK exhibits excellent mechanical and tribological properties, especially high mechanical strength and elastic modulus, as well as very good wear resistance. [8][9][10] An additional increase in the load-bearing capacity and wear resistance is obtained by introducing a reinforcement phase into the PEEK matrix. These may be carbon fibers [11] or glass fibers [12] as well as various particles such as Si 3 N 4 , [13,14] ZrO 2 , [15] or Al 2 O 3 . [16] For example, Wang et al. [17] proved that the addition of SiC whiskers in a low content of 1.25 wt pct resulted in a decrease in the PEEK wear index. In terms of mixing and processing, reinforcing phases in the form of spherical or spherical-like particles are more favorable than fibers. [18] Unfortunately, virgin PEEK and some PEEK-based composites, despite excellent mechanical properties and wear resistance, have a relatively high COF. [7,14,16,19] In previous works, [7,14,16] the COF of coatings obtained in a ball-on-disc test during dry friction in cooperation with an Al 2 O 3 ball at RT was 0.27, 0.26, and 0.25 for the PEEK 708, Si 3 N 4 /PEEK 708, and Al 2 O 3 /PEEK 708 coatings, respectively. In another work, [19] the COF of PEEK in a ball-on-disc test during dry friction in cooperation with a bearing steel ball was around 0.25. This is a disadvantage because a high COF usually generates a high temperature at the contact of working materials so that the polymers are destroyed in a similar way to metallic materials during lubricated and nonlubricated contact. [19,20] Unfortunately, reducing the COF can often cause deterioration of the wear resistance and mechanical properties of PEEK. [21] Thus, coatings with a balance of mechanical properties, wear resistance, and COF are of great importance.
The reduction in PEEK's COF can be achieved by introducing solid lubricants into the matrix, e.g., graphite, molybdenum disulfide, or polytetrafluoroethylene (PTFE). [22] COF reduction may occur, e.g., by formation of a transfer film on the counterface. [23] PTFE is a well-known synthetic fluoropolymer, which exhibits unique properties, such as excellent chemical resistance, high-temperature stability, and corrosion resistance. [24][25][26] Moreover, PTFE with its excellent sliding properties is widely used in engineering applications as a self-lubricating material. [27,28] However, due to the poor wear resistance, poor adhesion to substrates, and low creep resistance, PTFE cannot be used alone in most applications. [29][30][31][32] The development of PEEK/ PTFE bulk composites for tribological applications is widely described in the literature. [18,20,22,33,34] However, there is only one article on the electrophoretic deposition (EPD) of PTFE/PEEK coatings. [35] In that work, the authors focused on the preparation of suspensions for EPD with appropriate chemical compositions by adding stabilizing agents. Then the direct current (DC) voltage values were selected and the efficiency of the deposition process was examined using differential scanning calorimetry measurements. Extensive literature exists on the EPD of different PEEK-based coatings. [35][36][37][38][39][40][41] These works and our previous experience [14,42] proved that EPD and post-EPD heat treatment is a suitable method to produce pure PEEK 708 and PEEK 708-based coatings on titanium alloy substrates for tribological applications. The previously obtained PEEK 708, [7] Si 3 N 4 /PEEK 708, [14] and Al 2 O 3 / PEEK 708 [16] coatings on the Ti-6Al-4V alloy, despite a significant improvement in wear resistance in cooperation with an Al 2 O 3 ball at RT, had a relatively high COF of 0.27, 0.26, and 0.25, respectively.
Thus, the main objectives of this work are to fabricate composite low-friction and wear-resistant PTFE/PEEK 708 coatings on a Ti-6Al-4V titanium alloy substrate with the use of EPD and post heat treatment, to characterize the coating microstructure, and to determine their influence on the micromechanical and tribological properties of the alloy at RT and elevated temperatures.

A. EPD of Coatings
PEEK 708 and PTFE micro-particles have been used for the electrophoretic deposition of composite coatings. Basic information about the powders is presented in Table I. The coatings were deposited on the two-phase (a + b) Ti-6Al-4V titanium alloy substrates delivered in hot-rolled and annealed (750°C/2 h) condition by BÖ HLER Edelstahl GmbH (Du¨sseldorf, Germany). Samples used for deposition were in the form of disks with a diameter of 22 mm and thickness of 3 mm. The surface of the titanium alloy substrate was ground on sandpaper with a gradation from 200 to 3000. Subsequently, it was polished with the standard colloidal silica suspension (OP-S, 0.04 lm) of Struers (Ballerup, Denmark) to the mirror finish condition. The deposition of coatings was carried out from a suspension containing 30 g/L of PEEK 708 powder and 2 g/L of PTFE powder. A mixture consisting of 75 vol pct of ethanol and 25 vol pct of chitosan solution was used as the dispersion medium. The chitosan solution was prepared by adding 0.5 g/L medium molecular weight chitosan powder (Sigma-Aldrich) and 1 vol pct acetic acid to distilled water. The chitosan solution was used as a cationic polyelectrolyte. The zeta potential of the suspensions as a function of their pH was measured using a Zetasizer Nano ZS 90 from Malvern Instruments Ltd. (Malvern, United Kingdom), and the pH was controlled using an ELMETRON CPC-505 pH meter (Zabrze, Poland).
The deposition process was carried out in a standard two-electrode system, where the working electrode was a cathode and austenitic stainless steel was used as an anode. Directly before EPD, the suspension was ultrasonically dispersed for 15 minutes and magnetically stirred for 10 minutes. EPD was carried out using an EX752M PSU Multi-mode (United Kingdom) laboratory power supply. The coatings were deposited at a constant voltage in the range of 10 to 100 V with a change of 10 V and a constant time of 40 seconds. During deposition, the suspension was not mixed and the distance between the electrodes was constant at 15 mm. To determine the deposition rate, the coatings were deposited at a constant voltage of 90 V and a constant time in the range of 20 to 100 seconds with a change every 20 seconds. After each deposition, the samples were weighed with analytical scales from Ohaus Europe GmbH (Switzerland) with an accuracy of 0.1 mg.
Immediately after deposition and drying of the coatings, coated samples were subjected to heat treatment at a temperature of 400°C or 450°C for 20 minutes. After the heating process, they were cooled with a furnace (2°C/minute) or in water at RT (approximately 430°C/minute). The heat treatment was carried out in a Carbolite-Gero LHT 4/30 laboratory oven (United Kingdom) (temperature of 400°C) and in an LHT 08/18 furnace (Nabertherm GmbH, Germany) (temperature of 450°C).

B. Microstructure, Surface Topography, and Property Characterization
Phase identification of the as-deposited and heattreated coatings was performed by X-ray diffraction (XRD) in Bragg-Brentano geometry using a Panalytical Empyrean DY1061 diffractometer (the Netherlands).
The investigations of microstructure were carried out using a scanning electron microscope (SEM), FEI Nova NanoSEM 450 (the Netherlands), and a transmission electron microscope (TEM), JEOL JEM-2010 ARP. The chemical composition and element distribution maps were determined using scanning electron microscope energy dispersive X-ray spectroscopy (SEM-EDS) microanalysis. Thin foils for the TEM studies from the suspension used for EPD were prepared by placing a droplet of the suspension on the 300-mesh copper grid covered with carbon and then drying. The lamellae from the cross section of the coating after heat treatment were prepared by a focused ion beam (FIB) using an FEI QUANTA 3D 200i device (the Netherlands).
Middle infrared (MIR) structural investigations were performed with a Bruker Vertex 70v vacuum spectrometer. For the measurement of external reflectance spectroscopy, a Harrick Seagull attachment was used. A total of 128 scans with a resolution of 4 cm -1 were collected.
The surface topography of the substrate material used directly for the EPD of coatings and that of the coatings were investigated using a Filmetrics Profilm3D (San Diego, USA) noncontact optical profilometer. Several images of surface topography were acquired in areas of 1 mm, each in a different place on the sample. The surface topography was described by the parameters Sa (arithmetic mean height) and Sq (root-mean-square height), which were determined in accordance with ISO 25178.
Microhardness and elastic modulus were examined using a Vickers indenter with the instrumental indentation method. The indenter was pressed with a constant rate (200 mN/min) and held for 15 seconds at a maximum load of 100 mN. At least 10 measurements were made, every time in the new measuring area. To obtain the intrinsic mechanical properties of the coating, the indentation depth at the maximum load must be lower than 10 pct of the coating thickness, which is approximately 45 lm for all samples. The scratch resistance of the coatings was examined in the microscratch test using the Micro-Combi Tester (MCT, CSM Instruments, Switzerland). A Rockwell C diamond stylus with a tip radius of 0.2 mm was used. The samples moved the distance of 5 mm with a constant velocity of 5 mm/min when the stylus scratched their surface with a linearly increasing load in the range of 0.01 to 30 N. The critical loads L C1 (at which the first cohesive cracks occurred) and L C2 (at which the first adhesive cracks occurred) were determined during the tests, which were repeated 3 times for each sample. Indentation and scratch tests were carried out at RT (approximately 23°C) and a relative humidity of 55 to 60 pct.
The tribological properties of the uncoated and coated alloy were examined in a ball-on-disc test with a high-temperature tribometer (ITeE Radom, Poland). An Al 2 O 3 ball with a diameter of 6 mm was used as a counterbody. Tests were carried out at three different temperatures, i.e., 23 (RT), 150°C, and 260°C, and were repeated 3 times at each of them. Test parameters were as follows: normal load (F n ) of 5 N, sliding distance (s) of 2000 m, friction radius of 3 mm, and sliding speed of 0.05 m/s. Based on the results obtained, the COF and specific wear rate (W v ) were determined. W v was calculated from the expression W v = V coating /F n AEs, where the worn volume of the coating (V coating ) was obtained from a cross section of the coating's wear tracks.

A. Electrophoretic Co-deposition of PEEK and PTFE Particles
The first stage of the EPD route consisted of elaboration of a stable suspension for the co-deposition of both polymer particles, especially the selection of a suitable dispersing phase. It is well known that PEEK particles are negatively charged in pure ethanol [7] as are PTFE particles, as we have confirmed experimentally in this work. In the available literature, there are not many reports about the EPD of PTFE particles. Jang et al. [43] co-deposited PTFE with hydrous ruthenium oxide from a mixture of ethanol and water, while Hamagami et al. [44] used acetone solution with a small amount of water for the co-deposition of PTFE particles with titanium oxide. Liu et al. [45] used cholic acid sodium salt as a dispersant for the deposition of PTFE particles onto stainless steel substrates. In this case, the use of this dispersant facilitated the fabrication of a stable PTFE suspension. The deposition of PEEK particles is well known in the literature and was successfully carried out from pure ethanol on various substrates, including conductive carbon, stainless steel, and titanium alloy. [7,[46][47][48][49] However, the co-deposition of PEEK and PTFE from pure ethanol turned out to be impossible.
To the best of the authors' knowledge, there is only one article about the co-deposition of PEEK and PTFE particles. [35] The authors of that work decided to prepare separate suspensions for PEEK and PTFE in ethanol and then mix them together. As dispersants, they used a solution of triethylamine and citric acid for the PEEK suspension and polyvinylpyrrolidone for the PTFE suspension. In the present work, based on our previous experience, [14,42] we used chitosan as a cationic polyelectrolyte.
The success of the deposition process depends largely on the suspension stability, and a reliable indicator of this is the zeta potential. [50] Therefore, zeta potential was measured for PTFE and PEEK particles dispersed in (1) pure ethanol and (2) a mixture of ethanol and chitosan polyelectrolyte. In pure ethanol, both particle types showed negative potential over the entire measured pH range of 3 to 12 ( Figure 1). The addition of chitosan polyelectrolyte to the suspension changed the surface charge of the particles to positive ( Figure 1) and enabled effective polymeric particle deposition, despite generally lower zeta potential values compared to pure ethanol. The adsorption of the chitosan particles and chains to the surface of the PEEK and PTFE particles is clearly visible in the TEM images ( Figure 2). The effect of such interaction is a change in the surface potential and steric stabilization of PEEK and PTFE particles in the ethanol. The mechanism of interaction between chitosan and PEEK particles has been explained previously by Luo and Zhitomirsky. [51] According to their results, an electrochemical decomposition of water occurs at the cathode, resulting in a local increase in pH. Neutralization of the chitosan amino group charge, as well as the accumulation of PEEK particles with adsorbed protonated chitosan, causes the cathodic deposition.
EPD parameters, voltage and time, were selected experimentally to obtain macroscopically homogeneous coatings. Deposition of both particle types was not observed at voltages of 10 and 20 V. Coatings obtained in the voltage range of 30 to 60 V were thin and did not cover the substrate uniformly. Macroscopic homogeneity of coatings was obtained in the voltage range of 70 to 100 V, with the coating deposited at 90 V during 40 seconds being selected for further investigation. Studies on the deposition rate vs deposition time have shown that the rate was almost constant in the range of 20 to 60 seconds; then it increased slightly at 80 seconds and decreased at 100 seconds ( Figure 3).
The macroscopic and SEM images as well as the XRD pattern of the as-deposited coating are shown in Figure 4. It turned out that the coating was homogeneous also in a microscopic scale (Figure 4(a)), and both coating components, crystalline PTFE and nearly amorphous PEEK, were present in the XRD pattern ( Figure 4(b)). Both particle types were relatively evenly distributed on the surface of the titanium alloy. The thickness of the as-deposited coatings was about 130 lm.

B. Microstructure and Surface Topography of Heat-Treated Coatings
Coatings directly after the deposition exhibited relatively poor adhesion to the substrate and were very susceptible to destruction, so it was necessary to carry out heat treatment. PEEK belongs to the group of thermoplastic polymers, which means that, under elevated temperatures, it changes from a solid state into a plastic state. Therefore, heat treatment was carried out above the PEEK melting point, at a different temperature of 400°C or 450°C, in order to change the morphology of the particles as well as to densify the coatings and, consequently, give them mechanical properties. Coated samples after heating were cooled slowly with a furnace or rapidly in water at RT. It was found that the coating heated at 400°C and cooled with a furnace or in water was highly porous, with a pore diameter up to 55 lm on the coating surface ( Figure 5(a)). Moreover, the PTFE particles were not completely integrated and embedded in the PEEK matrix ( Figure 5(b)). Interestingly, the coatings heated at the temperature of 450°C after both cooling routes were macroscopically smooth and uniform, without macroscopically visible defects. Figure 6(a) shows macroscopic and SEM images of the coating after heating at 450°C for 20 minutes and cooling with a furnace. It can be observed that the coating was compact and homogeneous. There was no presence of any voids or pores. Brighter and darker areas are clearly visible on the coating surface. The presence of small microcracks was observed in the bright areas of the coating. SEM-EDS microanalysis was performed on both of them (Figures 6(b) and (c)). The results showed that the bright areas were enriched with fluorine ( Figure 6(b)), which corresponds to PTFE particles. Both separate PTFE particles and their agglomerates with a size up to 100 lm are visible on the coating surface (Figures 6(a) and 7(a)). The SEM-EDS element distribution maps confirmed the relatively uniform distribution of PTFE in the coating (Figure 7). Microstructure investigations with the use of a TEM were carried out on a lamella cut from the cross section of the outer part of the coating. Figure 8 shows that the coating was dense and uniform. However, small pores with the diameter up to 0.2 lm were sporadically present. The thickness of the heat-treated coatings decreased to 45 lm independently of the cooling rate after the heating process at 450°C, due to the coatings being densified and the formation of a compact matrix by PEEK particles. To determine the influence of the cooling rate after heating at 450°C on the coating structure, XRD studies were performed. The results showed that the PEEK in both coatings had an amorphous structure regardless of the cooling rate ( Figure 9). XRD patterns show a strong (100) crystalline diffraction peak of PTFE and one large amorphous peak belonging to PEEK.
Because the PEEK structure after heat treatment was the same, amorphous, further studies were carried out only on the samples cooled with a furnace due to the lower thermal stresses, as expected.
It is particularly interesting to obtain an amorphous structure of PEEK after heat treatment consisting of heating at 450°C and cooling with a furnace. Usually,   after heating above the melting point of PEEK and cooling with a furnace, the PEEK structure is semicrystalline. [7,14,16,49] A similar effect of PEEK amorphization after heating at 380°C for 20 minutes and cooling with a furnace was observed in our previous work on PEEK coatings reinforced with MoS 2 particles. [42] This was explained by the thermal sulfonation of PEEK, which caused its amorphization. [52] However, the amorphization of PEEK in the presence of PTFE is unexpected. Xie et al. [53] showed that PEEK/ PTFE composites prepared in the compression molding process and after sintering at 389°C for 1 hour  contained 37.8 pct of the crystalline phase in the PEEK matrix. Zhu et al. [54] fabricated composite PEEK/PTFE coatings with a various PTFE content of 1, 3, 5, and 10 wt pct using the electrostatic powder spraying technique. Coatings after quenching in water and annealing at a temperature of 260°C for 30 minutes also exhibited a semi-crystalline PEEK structure. Stuart and Briscoe [55] produced PEEK/PTFE composites by compaction and sintering. They showed that the presence of PTFE causes the PEEK molecules to become more ordered. In addition, an increase in PTFE concentration with a reduction in the intensity of the C-O-C stretching mode increases the crystallinity of PEEK.
Due to that, FTIR investigations were performed to explain this phenomenon in our coatings. Figure 10 shows a set of MIR spectra of (1) PEEK 708 and (2) PTFE powders used for EPD, (3) the as-deposited PTFE/PEEK 708 coating, and (4) PEEK 708 and (5) PTFE/PEEK 708 coatings heated at 380°C and 450°C for 20 minutes, respectively, and cooled with a furnace.
The complex nature of the spectra makes it very difficult to determine the effect of PTFE on the structure of the coatings obtained. Therefore, Figure 10(b) presents a set of MIR spectra in the range of 650 to 400 cm -1 , in which there are bands characteristic for C-F bonds not overlapping with bands characteristic for PEEK. On the MIR spectra of the PTFE/PEEK 708    (5)), the band at 557 cm -1 , characteristic of CF 3 bending vibrations, is visible. The presence of C-F bonds is also evidenced by the clear asymmetry of the band at 634 cm -1 , resulting from the overlap of this band with the bands at 639 and 637 cm -1 characteristic of deformation vibrations of C-F bonds. [56] Comparing the MIR spectra of PTFE/PEEK 708 coatings (3) before and (5) after heat treatment, a clear increase in the half width of the bands is visible, which indicates an increase in the amorphous degree of heat-treated coatings; this is in agreement with the XRD results ( Figure 9). As mentioned earlier, an identical effect of PEEK amorphization was observed in the case of MoS 2 /PEEK 708 coatings, which were combined with the sulfonation process. [42] Thus, based on the presented research results, it was found that, in the case of PTFE/PEEK 708 coatings, a fluorination process of PEEK occurred, which is responsible for inhibiting its crystallization.
Based on the similarity of MIR spectra (Figure 10) of the obtained polymer coatings (3) before and (4 and 5) after the thermal treatment process, it can be concluded that this process does not cause the degradation of the PEEK polymer. Similarly, the XRD patterns (Figure 9) of the as-deposited and heat-treated coatings clearly showed that the PTFE did not degrade during the two-step coating process. This is in line with the literature data, which shows that the thermal degradation of PEEK begins at a temperature of about 575°C , [57] while PTFE starts at a temperature of about 510°C . [58] It is well known [59] that, in addition to surface physical and chemical states, the tribological properties of materials are also directly affected by surface topography. Thus, in the present work, the surface roughness of the coating was investigated in relation to that of the substrate material. In comparison to the surface roughness of the polished, very smooth surface of the sample substrate used for coating deposition, the surface roughness of the PTFE/PEEK 708 coating heated at 450°C and cooled with a furnace was much higher (Figure 11). Roughness measurement results showed that the average values of root-mean-square height and arithmetic mean height for the coating were 2.76 ± 0.68 and 2.21 ± 0.53 lm and for the substrate 0.027 ± 0.002 and 0.021 ± 0.002 lm, respectively.  Taking such results of the microstructure investigation of the PTFE/PEEK 708 coatings into consideration, it was interesting to examine their micromechanical and tribological properties, especially during friction at elevated temperatures. The microhardness (H) of the coating heated at 450°C and cooled with a furnace was 0.2 ± 0.08 GPa, and the elastic modulus (E) was 4.3 ± 1.6 GPa. Attention should be paid to the large spread of results from the average value, which confirms the presence in the coating microstructure of areas enriched with or depleted of PTFE particles or their agglomerates differing in mechanical properties. Due to the amorphous structure of PEEK in this composite, the PTFE/ PEEK 708 coating exhibited significantly lower microhardness and modulus of elasticity compared to H = 0.32 ± 0.02 GPa and E = 5.9 ± 0.2 GPa, respectively, for the pure PEEK 708 coating with a semi-crystalline structure. [7] However, the microhardness and elastic modulus of amorphous PEEK 708 coatings were similar, which clearly indicates that the amorphous structure of PEEK is less resistant to pressing of the indenter.
The results of the microscratch tests showed that the PTFE/PEEK 708 coating had moderate scratch resistance and was smaller in comparison with the semicrystalline PEEK 708 coating. Complete failure of the slowly cooled coating occurred suddenly under the average load L C2 = 18 N, when large fragments of the coating were removed and easily detached from the substrate (Figure 12(a)). This process was accompanied by a clear signal of acoustic emission (Figure 12(b)). Such behavior of the polymeric coating during its destruction is rather characteristic for brittle materials, e.g., ceramic coatings. In spite of such a destruction mechanism of the coating, no cohesive cracks before L C2 were observed in the coating, only its abrasion.
The investigations of tribological properties focused mainly on the ball-on-disc tests of the alloy samples with coatings heated at a temperature of 450°C cooled with a furnace, due to the homogeneous microstructure of the coatings, without open porosity. The results of the tests indicate that the addition of PTFE to the PEEK matrix significantly improved the tribological behavior of the coatings. In sliding contact, the coating featured a very low COF at RT, as expected, and its average value was 0.10. This result is much lower than for the pure PEEK 708 coatings, with both a semi-crystalline and amorphous structure, [7] and the uncoated Ti-6Al-4V alloy, for which the mean coefficients of friction were 0.27, 0.33, and 0.70, respectively ( Figure 13).
The average wear rate of the PTFE/PEEK 708 coated alloy after friction at RT, 150°C, and 260°C is shown in Table II. It was found that the coated alloy exhibited a very low wear rate at RT, equal to 0.38 9 10 À6 mm 3 / Nm. In comparison, the wear rate of the pure semicrystalline PEEK 708 coating was higher and equaled 2.61 9 10 À6 mm 3 /Nm. [7] The PTFE/PEEK 708 coating also significantly reduced the wear rate of the coated Ti-6Al-4V alloy, which was incomparably higher and amounted to 720 9 10 -6 mm 3 /Nm. In Figure 14(a), it can be observed that the coating after friction at RT did not show any signs of severe damage in the wear track and only its normal abrasion occurred. The coating  exhibits very low resistance to motion, due to the presence of sliding PTFE particles and high wear resistance, despite its brittleness. Similar results were obtained by Vail et al. [34] They investigated PEEK composites with the addition of 10 vol pct of PTFE filaments, which had a COF of 0.11 and wear rate of 0.07 9 10 -6 mm 3 /Nm due to the fact that the filaments were aligned. Burris and Sawyer [33] also tested PEEK/ PTFE composites and obtained a COF equal to 0.111 (for PTFE content at the level of 50 vol pct) and an extremely low wear rate of 2 9 10 À9 mm 3 /Nm, but in this case, the PTFE concentration was almost 70 vol pct.
In the present work, the tribological properties of the coated alloy were also investigated at elevated temperatures of 150°C and 260°C. The COF during the friction process at 150°C progressively increased from a boundary low value of 0.06 up to 0.17 and was higher than the COF at RT. The wear rate at this temperature was also higher that that obtained at RT and reached the value of 1.42 9 10 -6 mm 3 /Nm. SEM observation of the wear track showed that it was not as uniform as that obtained at RT (Figure 14(b)). In the wear track, slight defects of the coating were visible and small fragments of material were detached and moved further by the Al 2 O 3 ball. In addition, a lot of the abrasive wear products were observed outside the wear track.
The wear resistance of the coated alloy tested at the temperature of 260°C turned out to be the smallest. Its mean COF was 0.09 at the beginning, but, unfortunately, the coating did not withstand the test and was completely removed from the substrate before the assumed sliding distance of 2000 meters; then the COF reached 0.54. Friction at this temperature caused the formation of longitudinal cracks in the coating inside the wear track and finally the coating was completely worn down to the substrate. For this reason, as a result of several experiments, the cooperation time at 260°C was shortened so that it corresponded to a sliding distance of not more than 450 meters when the coating was still continuous and retained a low COF of 0.07. Under these conditions, when coating continuity was ensured, the wear rate reached the value of 6.5 9 10 -6 mm 3 /Nm.
The excellent sliding properties of the PTFE/PEEK 708 composite coating, especially when the temperature around the friction pairs was close to RT or 150°C, are the result of the formation of a self-lubricating polymer film in the friction process. The presence of PTFE, which is well known as a solid lubricant, promotes the formation of a sliding layer on the counterpart, which reduces friction and wear. [33,60] SEM-EDS line microanalysis of chemical composition performed for the Al 2 O 3 ball's surface after friction with the PTFE/PEEK 708 coated alloy, both at RT and at the elevated temperature of 150°C, indicated that a polymer layer was formed on the ceramic ball in a friction spot (Figures 15(a) and (b)). In this film, fluorine was found (Figures 15(c) and (d)), which may indicate the formation of a self-lubricating strip of PTFE, which reduces friction.
However, this film was very thin and unstable, probably due to its poor adhesion to the ceramic ball. This effect was also discussed by Nunez et al., [61] where different levels of adhesive interaction of the polymer film to the steel and ceramic counterparts were found and even the impossibility of polymer film transfer to  the ceramic surface was established. Hence, the COF reaches very low values even below 0.05 in the sliding contact of steel with PEEK composites containing PTFE. [18] However, the COF as well as the wear of the PTFE-PEEK composite may be higher during sliding with a ceramic material.

IV. CONCLUSIONS
Research carried out in this work allowed the following conclusions to be formulated.
1. The addition of chitosan polyelectrolyte to ethanol changed the zeta potential of both PEEK and PTFE particles from negative to positive. Chitosan provided steric stabilization of the suspension due to its adsorption to PEEK and PTFE particles. 2. Appropriate selection of EPD parameters allowed the fabrication of composite coatings that were homogeneous on both macro-and microscopic scales. The voltage of 90 V and deposition time of 40 seconds turned out to be optimal parameters for coating deposition. 3. Heating of coatings above the PEEK 708 melting point and cooling with a furnace or in water caused the transformation of PEEK from particles into a continuous coating matrix. Individual PTFE particles and their agglomerates were relatively homogeneously distributed in the PEEK matrix. 4. XRD studies have shown the crystalline PTFE structure and amorphous PEEK structure regardless of the cooling rate after heating. The fluorination process of PEEK occurred in the coatings during heating, which inhibits the formation of spherulites and crystallization of the PEEK. 5. The microhardness and elasticity modulus of the PTFE/PEEK 708 coating heated at 450°C and cooled with a furnace equaled 0.2 ± 0.08 and 4.3 ± 1.60 GPa, respectively. The coating was characterized by moderate scratch resistance and was completely removed from the substrate under a critical load of 18 N. 6. The PEEK 708-based coating incorporated with PTFE particles significantly decreased the COF and the wear rate of the uncoated titanium alloy in a ball-on-disc test during the dry friction process in cooperation with an alumina ball at RT and 150°C. Unfortunately, during the friction process performed at the temperature of 260°C, the coating was completely removed from the substrate without withstanding the entire test.
This work is important for understanding the development of PEEK-based composites with low friction performance. New knowledge about the thermal fluorination of PEEK in the presence of PTFE was obtained. The results presented can be useful for extending the use of titanium alloys in mechanical and biomedical engineering applications, in which resistance to wear and friction is required.

OPEN ACCESS
This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat ivecommons.org/licenses/by/4.0/.