Effect of Various Type of Nanoparticles on Mechanical and Tribological Properties of Wear-Resistant PEEK + PTFE-Based Composites

Article Effect of Various Type of Nanoparticles on Mechanical and Tribological Properties of Wear-Resistant PEEK + PTFE-Based Composites

Sergey V. Panin 1,2,* , Duc A. Nguyen 3, Dmitry G. Buslovich 1,2 , Vladislav O. Alexenko 1, Aleksander V. Pervikov 4 , Lyudmila A. Kornienko 1 and Filippo Berto 5

1 Laboratory of Mechanics of Polymer Composite Materials, Institute of Strength Physics and Materials Science SB RAS, 634055 Tomsk, Russia; [email protected] (D.G.B.); [email protected] (V.O.A.); [email protected] (L.A.K.) 2 Department of Materials Science, Engineering School of Advanced Manufacturing Technologies, National Research Tomsk Polytechnic University, 634030 Tomsk, Russia 3 Seaside Branch Russian-Vietnamese Tropical Center, Department of Tropical Endurance, Nha Trang 57106, Vietnam; [email protected] 4 Laboratory of Physical Chemistry of Ultraﬁne Materials, Institute of Strength Physics and Materials Science SB RAS, 634055 Tomsk, Russia; [email protected] 5 Faculty of Engineering, Department of Mechanical and Industrial Engineering, Norwegian University of Science and Technology, 7491 Trondheim, Norway; ﬁ[email protected] * Correspondence: [email protected]

Abstract: The mechanical and tribological properties of polyetheretherketone (PEEK)- and PEEK + PTFE (polytetrafluoroethylene)-based composites loaded with and four types of nanoparticles Citation: Panin, S.V.; Nguyen, D.A.; (carbonaceous, metallic, bimetal oxide, and ceramic) under metal- and ceramic-polymer tribological Buslovich, D.G.; Alexenko, V.O.; contact conditions were investigated. It was found that loading with the nanofillers in a small content Pervikov, A.V.; Kornienko, L.A.; (0.3 wt.%) enabled improvement of the elastic modulus of the PEEK-based composites by 10–15%. Berto, F. Effect of Various Type of In the metal–polymer tribological contact, wear resistance of all nanocomposites was increased by Nanoparticles on Mechanical and 1.5–2.3 times. In the ceramic-polymer tribological contact, loading PEEK with metal nanoparticles Tribological Properties of caused the intensification of oxidation processes, the microabrasive counterpart wear, and a multiple Wear-Resistant PEEK + PTFE-Based increase in the wear rate of the composites. The three component “PEEK/10PTFE/0.3 nanofillers” Composites. Materials 2021, 14, 1113. https://doi.org/10.3390/ma14051113 composites provided an increase in wear resistance, up to 22 times, for the metal–polymer tribological contact and up to 12 times for the ceramic-polymer one (with a slight decrease in the mechanical

Academic Editor: Debora Puglia properties) compared to that of neat PEEK. In all cases, this was achieved by the polymer transfer film formation and adherence on the counterparts. The various effects of the four types of nanoparticles Received: 27 January 2021 on wear resistance were determined by their ability to fix the PTFE-containing transfer film on the Accepted: 23 February 2021 counterpart surfaces. Published: 27 February 2021 Keywords: polyetheretherketone (PEEK); nanoparticles; solid lubricant fillers; wear factor; trans- Publisher’s Note: MDPI stays neutral fer film with regard to jurisdictional claims in published maps and institutional affil- iations. 1. Introduction Solid lubricants (primarily in the form of microparticles) are widely used for friction units, especially containing polymer composites. The most common are polytetrafluo- Copyright: © 2021 by the authors. roethylene (PTFE), graphite [1], molybdenum dioxide (MoS2)[2], boron nitride (BN), and Licensee MDPI, Basel, Switzerland. some others [3]. This article is an open access article PTFE is one of the best fillers to improve wear resistance of polyetheretherketone distributed under the terms and (PEEK)-based composites. Lu and Friedrich [4] investigated the tribological properties conditions of the Creative Commons of the PEEK-based composites loaded with PTFE. Dry sliding friction tests were carried Attribution (CC BY) license (https:// out according to the “pin-on-disk” scheme at a sliding speed of 1 m/s, and a contact creativecommons.org/licenses/by/ pressure of 1 MPa. It was shown that loading with PTFE enabled significant reduction 4.0/).

Materials 2021, 14, 1113. https://doi.org/10.3390/ma14051113 https://www.mdpi.com/journal/materials Materials 2021, 14, 1113 2 of 22

of the friction coefficient and wear rate. In this case, the optimum PTFE content ranged from 10 to 20 vol.% (23 to 45 wt.%). It was found that a multiple reduction in the friction coefficient and wear rate was achieved due to the formation of a transfer film on the surface of a steel counterpart. The effect of decreasing both the friction coefficient and wear rate of the PEEK-based composites loaded with PTFE is described in various papers [5–7]. However, the optimum PTFE content varied greatly. For example, Bijwe et al. [6] showed that the maximum wear resistance (30 times higher than that for neat PEEK) was achieved by loading with 7.5 wt.% PTFE. In another case [8], the “70 wt.% PEEK + 30 wt.% PTFE” composite possessed the highest tribological properties (its wear resistance increased by 900 times compared to that of neat PEEK, and 260,000 times compared to that of neat PTFE). It is possible that this distinction was related to the test conditions (namely reciprocating movement or the “pin-on-disk” scheme, the P·V (load sliding speed) factor, etc.), as well as the composite manufacturing methods. Hufenbach et al. [5] showed a decrease in the mechanical properties of the PEEK-based composites loaded with PTFE, which might limit their use under severe loading conditions. To compensate for this drawback, other fillers were added into the composites, for example, reinforcing fibers or particles, etc. This approach was applied in various studies [9–11]. Xie et al. [9] used potassium titanate (K2TiO3) to improve the mechanical and tribological properties of the PEEK-based composites loaded with PTFE. It was shown that loading with K2TiO3 enabled increase of both, microhardness and wear resistance. In the studied case, the optimal K2TiO3 content was 5 wt.%. In another case [10], loading with graphene nanoparticles in an amount of 2 wt.% resulted in wear resistance enhancing of the “PEEK + 10 wt.% carbon fibers (CF) + 10 wt.% PTFE” composite. Graphite possesses excellent lubricity under dry sliding friction conditions. Therefore, it is widely used to enhance the tribological properties of the PEEK-based composites. The results of the investigations [12–14] showed that graphite reduced the friction coefficient and increased wear resistance of the multicomponent PEEK-based composites. In addition, loading with graphite improved the heat dissipation characteristics of the polymer composites in some cases [15]. However, graphite is not efficient under the water lubrication conditions, since water adsorption reduces the binding energy between hexagonal planes, which resulted in its rapid abrasion due to friction [16]. The layered structure of molybdenum disulfide (MoS2) determines its unique antifriction properties and enables its efficient application as a solid lubricant [17–23]. Zalaznik et al. [18] studied the effect of types, sizes, and contents of solid lubricant particles on the tribological properties of the PEEK-based composites. Two types of solid lubricant nano- and microparticles were investigated—molybdenum disulfide (MoS2) and tungsten disulfide (WS2), with contents from 0.5 to 5 wt.%. It was shown that solid lubricant particles at their maximum content of 5 wt.% could reduce the friction coefficient of the PEEK-based composites from 0.6 down to 0.4, regardless of the type and size. However, wear resistance was improved only in the case of loading with MoS2 micro- or nanoparticles or WS2 nanoparticles, in amounts of 1 wt.%. A further increase in the filler content reduced wear resistance of the PEEK-based composites. Theiler and Gradt [24] compared the efficiency of MoS2 and graphite loaded in the antifriction “PEEK + PTFE + CF” composites for friction in vacuum. It was shown that loading with MoS2 particles provided better wear resistance compared to graphite, especially at low temperatures (−80 ◦C). The low friction coefficient and high wear resistance was achieved due to the formation of a thin transfer film on the surface of a steel counterpart. Many authors [25–29] also reported on the possibility of wear resistance improving of the polymer composites, by loading with solid microparticles like titanium dioxide (TiO2), zirconium dioxide (ZrO2), silicon carbide (SiC), copper oxide, and sulfide (CuO and CuS), etc. Bahadur et al. [25], the tribological properties of the PEEK-based composites loaded with CuO, CuS, and CuF2 microparticles in an amount of 35 vol.% was studied. It was shown that despite the high friction coefficient, the composites loaded with solid Materials 2021, 14, 1113 3 of 22

microparticles possessed better wear resistance compared to that of neat PEEK. Bahadur et al. [26–29] also confirmed the possibility of wear resistance increase of the polyphenylene sulfide (PPS)-based composites, by loading with solid microparticles. Progress in the field of nanotechnologies intensified research on the effect of nanoscale fillers, on the performance properties of polymers and their composites. Compared to the micro-scale ones, the nanofillers possess some advantages—(a) less abrasive effect on counterparts; (b) a large specific surface area, and therefore, a large adhesion to the polymer matrix; and (c) the possibility of improving a number of the composite characteristics with a relatively low filler content. In addition, a chance to increase the tribological properties of the composites based on the promising polymers by loading with nanoparticles is discussed [30–34]. Wang et al. [30] investigated the PEEK-based composites loaded with Si3N4 nanoparticles, obtained by compression sintering. Wear tests were carried out according to the “block-on-ring” scheme. It was found that the composite containing 2.8 vol.% Si3N4 possessed the lowest wear rate. The PEEK-based nanocomposites loaded with other nanoparticles (SiO2 and ZrO2) were also studied [31,32]. It was shown that the lowest wear rate was also obtained by loading with ~1.5–3.5 vol.% SiO2 and ZrO2 nanoparticles. Bahadur S. et al. [33,34] tested the PPS-based nanocomposites loaded with Al2O3, TiO2, ZnO, CuO, and SiC particles, according to the “pin-on-disk” scheme. It was shown that the nanoparticle content of 2 vol.% was the optimum for the designed wear-resistant PPS-based nanocomposites. Table1 presents compositions of some PEEK nanocomposites [ 35]. The maximum wear resistance was achieved by loading with ~7 wt.% nanoparticles.

Table 1. Wear resistance of some polymer nanocomposites [35].

The Optimum Content Type of Nanoparticles The Lowest Wear Rate Matrix of Nanoparticles (Particle Size) (10–6mm3/N·m) (vol./wt.%)

PEEK Si3N4 (<50 nm) 1.3 2.8/7.5

PEEK SiO2 (<100 nm) 1.4 3.4/7.5 PEEK SiC (80 nm) 3.4 1–3/2.5–10

PEEK ZrO2 (10 nm) 3.9 1.5/7.5

Recently, carbon nanofibers (CNF) and carbon nanotubes were widely applied to design PEEK-based nanocomposites [36–38]. Werner P. et al. [36,37] showed that loading with CNF in amounts up to 15 wt.% enabled improvement of the elastic modulus of the PEEK-based composites. At the optimal CNF content (~10 wt.%), the wear rate of the nanocomposites decreased significantly compared to that of the neat PEEK. In addition to the microfillers, the nano- ones were also loaded to design multicomponent composites, where each filler performed its own function [39–42]. Molazemhosseini et al. [39] studied the tribological properties of the PEEK-based composites loaded with CF and SiO2 nanoparticles. It was shown that at a content of 20 vol.% CF and 1.4 vol.% nano-SiO2, the friction coefficient and wear rate decreased under all tribological testing conditions. According to the authors, loading with nanoparticles could increase the strength at the interface between the CF and the polymer matrix, which improved wear resistance of the multicomponent PEEK-based nanocomposites. Guo L. et al. [40] showed that loading with SiO2 nanoparticles in amounts up to 4 vol.% resulted in significant wear resistance improvement and reduction in the friction coefficient of the multicomponent PEEK-based composites, in comparison with the cases of loading with CF or PTFE, separately. The formation of a transfer film containing SiO2 nanoparticles was revealed, which was the main reason for the nanocomposite wear resistance improvement. Thus, various types of fillers were applied to increase the wear resistance of the polymer-based composites. Each filler played its own role. By understanding their function Materials 2021, 14, x FOR PEER REVIEW 4 of 23

144 Thus, various types of fillers were applied to increase the wear resistance of the pol- 145 ymer-based composites. Each filler played its own role. By understanding their function 146 in the composites, it becomes possible to design ones possessing the optimal properties, 147 in accordance with specified operating conditions. 148 In these investigations, the task was to study step-by-step, the effect of several types 149Materials 2021, 14, 1113 of nanofillers (carbonaceous, metallic, bimetal oxide, ceramic) on the mechanical4 of 22 and 150 tribological properties of the PEEK-based composites, in order to design antifriction high- 151 strength three-component ones for the operations in the metal- and ceramic-polymer 152 tribological contacts of the friction units. Section 2 defines the used materials and the 153 in themethods composites, of their it becomes studies. possible The structure, to design the ones physical possessing and themechanical optimal properties, properties in of the 154 accordance“PEEK/0.3 with nanoparticles” specified operating composites conditions. are presented in Section 3.1. Section 3.2 succinctly 155 describesIn these investigations, the mechanical the and task tribological was to study properties step-by-step, of the the“PEEK/ effect7 of nanofiller” several types compo- 156 of nanofillerssites, since (carbonaceous, this number of metallic,nanoparticles bimetal was oxide,the most ceramic) widely on studied the mechanical previously. and Section 157 tribological3.3 was propertiesdevoted to of a thedetailed PEEK-based analysis composites, of the structure, in order and to the design physical antifriction and mec high-hanical 158 strengthproperties three-component of the “PEEK/10PTFE/0.3 ones for the nanoparticles” operations in thecomposites. metal- and At the ceramic-polymer end of the presen- 159 tribologicaltation of contacts the results, of thethe frictionreasons units.for the Sectionobserved2 defines processes the are used discussed materials and and conclusions the 160 methodsabout of the their prospects studies. for The the structure,use of nanocomposites the physical are and drawn. mechanical properties of the “PEEK/0.3 nanoparticles” composites are presented in Section 3.1. Section 3.2 succinctly 161 describes2. Materi the mechanicalals and Methods and tribological properties of the “PEEK/7 nanofiller” composites, since this number of nanoparticles was the most widely studied previously. Section 3.3 was 162 The “Victrex” PEEK powder (450 PF, Victrex PLC, Lancashire, UK) with an average devoted to a detailed analysis of the structure, and the physical and mechanical properties 163 particle size of 50 μm was used as a matrix resin, which was loaded onto PTFE polytetra- of the “PEEK/10PTFE/0.3 nanoparticles” composites. At the end of the presentation of 164 fluoroethylene (particle size of 6–20 μm, F4-PN20 grade, “Ruflon” LLC, Perm, Russia), the results, the reasons for the observed processes are discussed and conclusions about the 165 with four types of nanofillers. prospects for the use of nanocomposites are drawn. 166 1) Carbonaceous—the “Taunit” CNF (multiwall tubes) with an outer diameter of 60 nm and 167 2. Materialsa length of and 2– Methods3 μm obtained by the gas-phase chemical deposition (NanoTechCenter LLC, 168 Tambov,The “Victrex” Russia), PEEK Figure powder 1a. (450 PF, Victrex PLC, Lancashire, UK) with an average 169 particle2) Metallic size of— 50copperµm was (Cu) used nanoparticles as a matrix with resin, a whichsize of was 87 ± loaded 1 nm, ontoobtained PTFE by polytetraflu- the exploding 170 oroethylenewire method (particle (EWM) size [43 of 6–20], Figureµm, 1 F4-PN20b. grade, “Ruflon” LLC, Perm, Russia), with 171 four3) Bimetal types of oxide nanofillers.—copper ferrite (CuFe2O4) nanoparticles with a size of 34 ± 1 nm obtained 172 (1) byCarbonaceous—the the exploding wire “Taunit” method CNF (EWM), (multiwall Figure tubes)1c. with an outer diameter of 60 nm 173 4) Ceramicand a length—the of“Tarkosil” 2–3 µm obtainedsilicon dioxide by the (SiO gas-phase2) nanoparticles chemical with deposition sizes of (NanoTech- 25–35 nm fab- 174 ricatedCenter by LLC, evaporating Tambov, Russia),initial substances Figure1a. in an electron accelerator, Figure 1d.

(a) (b) (c) (d)

175 FigureFigure 1. The 1. TEMThe TEM micrographs micrographs of nanoﬁllers: of nanofillers (a) CNF;: (a) CNF; (b) Cu; (b) ( cCu;) Fe-Cu-O; (c) Fe-Cu (d-)O; SiO (d2.) SiO2.

176 (2) Metallic—copperCu nanoparticles (Cu) were nanoparticles produced with in a an size argon of 87 atmosphere± 1 nm, obtained from bythe the copper explod- wires. 177 Then,ing wire the nanoparticles method (EWM) were [43 ],passivated Figure1b. to prevent their spontaneous combustion by slow 178 (3) airBimetal inlet into oxide—copper a chamber for ferrite 48 h. Fe (CuFe-Cu-2OO nanoparticles4) nanoparticles were with produced a size ofby 34electric± 1 nmexplod- 179 ingobtained both iron by theand exploding copper wires wire in method an oxygen (EWM),-containing Figure1 c.atmosphere (Ar + 20 vol.% O2). 180 (4) TheCeramic—the iron (66 at.%) “Tarkosil” to copper silicon (34 at. dioxide%) ratio (SiO was2) determined nanoparticles by withthe wire sizes diameters. of 25–35 nm fabricated by evaporating initial substances in an electron accelerator, Figure1d. Cu nanoparticles were produced in an argon atmosphere from the copper wires. Then, the nanoparticles were passivated to prevent their spontaneous combustion by slow air inlet into a chamber for 48 h. Fe-Cu-O nanoparticles were produced by electric exploding both iron and copper wires in an oxygen-containing atmosphere (Ar + 20 vol.% O2). The iron (66 at.%) to copper (34 at.%) ratio was determined by the wire diameters. The average particle size was assessed by transmission electron microscopy (TEM). Figure1 shows the TEM micrographs of the nanoparticles. Their parameters are given in Table2. Materials 2021, 14, 1113 5 of 22

Table 2. Parameters of the nanoﬁllers used.

Content Specific Specific Heat Conductivity Nanofiller Type (wt/vol.%) Area, m2/g Wt/cm·K CNF 0.3/0.24 ≥160 4.5–10 Cu 0.3/0.044 8.0 ± 0.5 400

CuFe2O4 0.3/0.076 14.2 ± 0.5 1.8–2.0

SiO2 0.3/0.149 ≥130 1.3–1.5

Cylindrical workpieces were fabricated by compression sintering at a specific pressure of 15 MPa and a temperature of 400 ◦C. The subsequent cooling rate was 2 ◦C/min. The polymer powder and the fillers were mixed by dispersing the suspension components in alcohol using a “PSB-Gals 1335-05” ultrasonic cleaner (PSB-Gals, Ultrasonic equipment center, Moscow, Russia). The generator frequency and the processing duration were 22 kHz and 3 min, respectively. After mixing, the suspension was dried in an oven with forced ventilation at a temperature of 120 ◦C for 3 h. Using alcohol as a mixing medium suggested the absence of volatiles in the ready-made mixtures for hot pressing. Samples were cut out from cylindrical blanks with the help of computer controlled Purelogic PLRA4 milling machine. The cutting speed was 100 m/min. Sample surfaces were free from visible cracks, scratches, and cavities. In addition, the sample surface was additionally polished with a sandpaper (grit P2000). Shore D hardness was determined by an “Instron 902” facility (Instron, Norwood, MA, USA) in accordance with ASTMD 2240. The tensile properties of the PEEK-based composite specimens were measured using an “Instron 5582” electromechanical testing machine (Instron). The specimen shapes met the requirements of ASTMD 638. Wear tests were carried out according to the “pin-on-disk” scheme, under the dry sliding friction conditions, at a load of 10 N and a sliding speed of 0.3 m/s. A “CSEMCH- 2000” tribometer (CSEM, Neuchâtel, Switzerland) was employed in accordance with ASTMG99. The diameter of counterparts was 6 mm. They were made from bearing steel with a hardness of HRC60 and ZrO2 ceramics. The sliding distance was 3 km and the tribological track radius was 10 mm, i.e., the rotation speed was 286 rpm. The wear track profiles were determined using the data for at least 10 tracks. Then, the wear rate values were estimated on the basis of the experimental test data over at least four samples of each type. Mathematical statistics methods were used for the experimental results processing. Surface topography of the wear tracks was studied using a “Neophot-2” optical microscope (Carl Zeiss, Oberkochen, Germany), equipped with a “Canon EOS 550D” digital camera (Canon Inc., Tokyo, Japan), and an “Alpha-Step IQ” contact profiler (KLA- Tencor, Milpitas, CA, USA). A “Neophot 2” optical microscope (Carl Zeiss, Jenna, Germany) was used to examine the wear track surfaces after testing. The supermolecular structure of the composites was studied on the cleaved surfaces of the notched specimens, mechanically fractured after exposure in liquid nitrogen. A “LEO EVO50” scanning electron microscope (Carl Zeiss, Oberkochen, Germany) was employed at an accelerating voltage of 20 kV.

3. Results and Discussion 3.1. The PEEK-Based Composites Loaded with 0.3 wt.% Nanoparticles Initially, the possibility of improving the mechanical and tribological properties of the PEEK-based composites by loading with nanoparticles of various types in an amount of 0.3 wt.% was studied. In order to compare the efficiency of loading with the nanofillers, four types were applied—(i) Carbonaceous—carbon nanofibers (CNF); (ii) Metallic—copper (Cu); (iii) Bimetal oxide—copper ferrite (CuFe2O4); and (iv) Ceramic—silicon dioxide (SiO2). Materials 2021, 14, 1113 6 of 22

The physical and mechanical properties of neat PEEK and the PEEK-based composites are presented in Table3. Loading PEEK with CNF did not cause a noticeable increase in tensile strength and the elastic modulus. Shore D hardness decreased slightly after loading with Cu metal particles. However, this parameter increased by one unit, after loading with SiO2 ceramic nanoparticles, which was highly likely, due to their higher hardness. The elastic modulus also increased by 10–15% after loading with all studied types of nanoparticles. After loading with SiO2 and CuFe2O4 particles, tensile strength was maintained at the neat PEEK level. Finally, elongation at break decreased by 5–10% for all types of the used nanoparticles.

Table 3. The physical and mechanical properties of neat PEEK and the PEEK-based composites loaded with 0.3 wt.% nanoparticles.

Filler Content, Elastic Modulus Tensile Strength Elongation at Density ρ, g/cm3 Shore D Hardness wt. % E, MPa σ, MPa Break ε,% PEEK 1.308 80.1 ± 1.17 2840 ± 273 106.9 ± 4.7 25.6 ± 7.2 PEEK/0.3CNF 1.314 80.3 ± 0.2 3034 ± 91 107.8 ± 1.7 23.6 ± 4.3 PEEK/0.3Cu 1.324 79.5 ± 0.5 2981 ± 118 100.9 ± 4.4 17.2 ± 3.9

PEEK/0.3CuFe2O4 1.309 80.2 ± 0.9 3113 ± 35 108.4 ± 0.4 19.8 ± 1.3

PEEK/0.3%SiO2 1.317 81.0 ± 0.6 3155 ± 238 111.4 ± 1.6 14.7 ± 3.2

The tribological tests of the nanocomposites were carried out under the dry sliding friction conditions on the steel and ceramic counterparts. Figure2 shows the dependence of the friction coefficients from the sliding distance, as well as their average values for neat PEEK and the PEEK-based nanocomposites. The friction coefficients of all studied nanocomposites decreased from 0.34 (for neat PEEK) down to 0.23–0.28 (the distinction was 35%), when sliding on the metal counterpart (Figure2b). When sliding on the ceramic counterpart (Figure2d), the “PEEK/0.3CNF” nanocomposite possessed the lowest friction coefficient. Its value was 0.20 ± 0.02, which was dozen percent less than that of neat PEEK. The friction coefficients of the composites loaded with CuFe2O4 bimetal oxide particles decreased insignificantly with respect to that of neat PEEK. According to the authors, a noticeable decrease in the friction coefficient by loading with CNF was associated with the type of the nanocomposite supermolecular structure formed (similar to neat PEEK one). Another reason was the formation of a transfer film (it is discussed below in the analysis of the wear track surfaces of both the polymer composites and the counterparts). A diagram of the wear factor values for neat PEEK and the PEEK-based nanocomposites is shown in Figure3. When sliding on the steel counterpart, the wear factor of the nanocomposites decreased by 1.5–2.4 times. When sliding on the ceramic counterpart, the wear factor of all nanocomposites increased by 2–10 times, with the exception of one loaded with CNF (its wear rate decreased by 1.5 times). Accordingly, the studied nanofillers did not play the role of a solid lubricant medium in PEEK (unlike other matrices, for example, thermoplastic semi-crystalline antifriction ultra-high molecular weight polyethylene). In addition, most likely, this pattern of the change in two-component nanocomposites wear resistance was related to both the oxidation processes developing on the friction surfaces and an increase in hardness (as well as their strength, as compared to that of the neat PEEK). Moreover, nanoparticles affected the wearing intensity to varying extents, depending on their chemical nature. Materials 2021, 14, x FOR PEER REVIEW 6 of 23

of 0.3 wt.% was studied. In order to compare the efficiency of loading with the nanofillers, four types were applied—(i) Carbonaceous—carbon nanofibers (CNF); (ii) Metallic—copper (Cu); (iii) Bimetal oxide—copper ferrite (CuFe2O4); and (iv) Ceramic—silicon dioxide (SiO2). The physical and mechanical properties of neat PEEK and the PEEK-based composites are presented in Table 3. Loading PEEK with CNF did not cause a noticeable increase in tensile strength and the elastic modulus. Shore D hardness decreased slightly after loading with Cu metal particles. However, this parameter increased by one unit, after loading with SiO2 ceramic nanoparticles, which was highly likely, due to their higher hardness. The elastic modulus also increased by 10–15% after loading with all studied types of nanoparticles. After loading with SiO2 and CuFe2O4 particles, tensile strength was maintained at the neat PEEK level. Finally, elongation at break decreased by 5–10% for all types of the used nanoparticles.

Table 3. The physical and mechanical properties of neat PEEK and the PEEK-based composites loaded with 0.3 wt.% nanoparticles.

Shore D Elastic Modulus Tensile Strength Elongation at Filler Content, wt. % Densityρ, g/cm3 Hardness E, MPa σ, MPa Break ε, % PEEK 1.308 80.1 ± 1.17 2840 ± 273 106.9 ± 4.7 25.6 ± 7.2 PEEK/0.3CNF 1.314 80.3 ± 0.2 3034 ± 91 107.8 ± 1.7 23.6 ± 4.3 PEEK/0.3Cu 1.324 79.5 ± 0.5 2981 ± 118 100.9 ± 4.4 17.2 ± 3.9 PEEK/0.3CuFe2O4 1.309 80.2 ± 0.9 3113 ± 35 108.4 ± 0.4 19.8 ± 1.3 PEEK/0.3%SiO2 1.317 81.0 ± 0.6 3155 ± 238 111.4 ± 1.6 14.7 ± 3.2

Materials 2021, 14, x FOR PEER REVIEW 7 of 23

(a) (c)

(b) (d)

Figure 2. The dependences of the friction coefficients from the sliding distance (а,c), as well as their average values (b,d) for neat PEEK (1) and the PEEK-based nanocomposites loaded with 0.3 wt.% nanoparticles—CNF (2); Cu (3); CuFe2O4 (4);

SiO2 (5); the metal (а,b) and ceramic (c,d) counterparts. (b) (d) A diagram of the wear factor values for neat PEEK and the PEEK-based nanocompo- Figure 2.FigureThe dependences 2. The dependences of the friction of the coefﬁcients friction coeffi fromcients the slidingfrom the distance sliding (a distance,c), as well (а, asc), theiras well average as their values average (b,d )values for (b,d) sites is shown in Figure 3. When sliding on the steel counterpart, the wear factor of the neat PEEKfor (1)neat and PEEK the (1) PEEK-based and the PEEK-based nanocomposites nanocomp loadedosites with loaded 0.3 wt.% with nanoparticles—CNF0.3 wt.% nanoparticles (2);— CuCNF (3); (2); CuFe Cu (3);2O4 CuFe(4); 2O4 (4); 2 nanocomposites decreased by 1.5–2.4 times. SiO2 (5);SiO the metal(5); the (a metal,b) and (а ceramic,b) and ceramic (c,d) counterparts. (c,d) counterparts.

A diagram of the wear factor values for neat PEEK and the PEEK-based nanocomposites is shown in Figure 3. When sliding on the steel counterpart, the wear factor of the nanocomposites decreased by 1.5–2.4 times.

Figure 3. Wear factors of neat PEEK (1) and the PEEK-based composites loaded with 0.3 wt.% Figure 3. Wear factors of neat PEEK (1) and the PEEK-based composites loaded with 0.3 wt.% na- nanoparticles—CNF (2); Cu (3); CuFe2O4 (4), and SiO2 (5) during the dry sliding friction on the metal noparticles—CNF (2); Cu (3); CuFe2O4 (4), and SiO2 (5) during the dry sliding friction on the metal and ceramic counterparts. and ceramic counterparts.

Thus, loading with CNF, copper (Cu), copper ferrite (CuFe O ), and silicon dioxide When sliding on the ceramic counterpart, the wear factor of2 all4 nanocomposites in- (SiO ) nanoﬁllersFigure 3. Wear in smallfactors contents of neat PEEK (0.3 wt.%)(1) and enabled the PEEK-b improvementased composites of the loaded strength with prop-0.3 wt.% na- creased2 by 2–10 times, with the exception of one loaded with CNF (its wear rate decreased erties ofnoparticles the PEEK-based—CNF (2); composites Cu (3); CuFe (elastic2O4 (4), and modulus SiO2 (5) increased during the bydry 10–15%) sliding friction through on the a metal by 1.5 times). Accordingly, the studied nanofillers did not play the role of a solid lubricant dispersedand hardeningceramic counterparts. mechanism. It was shown that the increase in wear resistance by medium in PEEK (unlike other matrices, for example, thermoplastic semi-crystalline antifriction ultra-highWhen slidingmolecular on theweight ceramic polyethylene). counterpart, In addition,the wear factormost likely, of all nanocompositesthis pattern in- of the changecreased in bytwo-component 2–10 times, with nanocomposites the exception ofwear one resistanceloaded with was CNF related (its wear to both rate the decreased oxidationby processes 1.5 times). developing Accordingly, on the the studied friction nanofillers surfaces and did annot increase play the in role hardness of a solid (as lubricant well as theirmedium strength, in PEEK as compared(unlike other to thatmatrices, of the for neat example, PEEK). thermoplastic Moreover, nanoparticles semi-crystalline an- affected tifrictionthe wearing ultra-high intensity molecular to varying weight extents, polyethylene). depending onIn addition,their chemical most nature.likely, this pattern Thus,of theloading change with in two-componentCNF, copper (Cu), nanocomposites copper ferrite wear(CuFe resistance2O4), and wassilicon related dioxide to both the (SiO2) nanofillersoxidation inprocesses small contents developing (0.3 wt.%) on the enabled friction improvement surfaces and of an the increase strength in prop- hardness (as erties ofwell the PEEK-basedas their strength, composites as compared (elastic to modulus that of theincreased neat PEEK). by 10–15%) Moreover, through nanoparticles a affected the wearing intensity to varying extents, depending on their chemical nature. Thus, loading with CNF, copper (Cu), copper ferrite (CuFe2O4), and silicon dioxide (SiO2) nanofillers in small contents (0.3 wt.%) enabled improvement of the strength properties of the PEEK-based composites (elastic modulus increased by 10–15%) through a

Materials 2021, 14, 1113 8 of 22

1.5–2.3 times in the metal–polymer tribological contact was observed for the composites loaded with 0.3 wt.% CNF and nanoparticles—copper (Cu), silicon dioxide (SiO2), and copper ferrite (CuFe2O4). This was due to the polymer transfer film formation on the steel counterpart, which had enabled protection of the surfaces of both the counterpart itself and the polymer composites from the micro-abrasive damaging and the oxidation. Loading PEEK with metal nanoparticles resulted in the intensification of the oxidation processes, the counterpart microabrasive wear, and the multiple rising of wear rate in the ceramic-polymer tribological contact. Loading with oxide and ceramic nanoparticles prevented the active development of the oxidation processes, in addition to the formation of the more homogeneous supermolecular structure. This caused the polymer transfer film formation on the counterpart, but could not improve wear resistance compared to that of neat PEEK.

3.2. The PEEK-Based Composites Loaded with 7 wt.% Nanoparticles Since it was not possible to significantly increase the strength and tribological properties of PEEK due to the rather low content of nanoparticles (0.3 wt.%), an attempt was made to enhance the filling degree by several times, similar to that of publications on similar topics (see Table1). The content of the nanofillers in the PEEK-based composites enabled the improvement of the tribological properties to about 7 wt.%, according to previous studies [44,45]. In order to compare with these data, the results of the studies of the composites loaded with the same nanoparticles, but in an amount of 7 wt.% are presented below. The physical and mechanical properties of the PEEK-based composites loaded with 7 wt.% nanofillers are shown in Table4. Their density increased by loading with all types of fillers. Shore D hardness changed in different ways. It decreased by one unit after loading with Cu particles; increased by the same amount after the addition of SiO2 particles, and was at the level of neat PEEK, after filling with CNF, as well as CuFe2O4 particles.

Table 4. The physical and mechanical properties of the PEEK-based composites loaded with 7 wt.% nanoparticles.

Filler Type and Elastic Modulus Tensile Strength Elongation at Density ρ, g/cm3 Shore D Hardness Content, wt.% E, MPa σ, MPa Break ε,% PEEK 1.308 80.1 ± 1.17 2840 ± 273 106.9 ± 4.7 25.6 ± 7.2 PEEK/7CNF 1.344 79.9 ± 0.4 3191 ± 43 91.3 ± 12.6 3.6 ± 0.6 PEEK/7Cu 1.375 78.9 ± 0.4 2937 ± 199 104.4 ± 1.8 14.0 ± 3.9

PEEK/7SiO2 1.354 81.4 ± 0.3 2860 ± 90 83.5 ± 20.5 4.0 ± 1.5

PEEK/7CuFe2O4 1.370 80.2 ± 0.9 3058 ± 257 102.4 ± 5.3 6.3 ± 1.7

The elastic modulus increased by 3–13%, depending on the filler type. However, it changed insignificantly, compared to the case of loading with 0.3 wt.% of nanofibers and nanoparticles (Table3). On the other hand, both tensile strength and elongation at break were reduced for all nanocomposites. At the same time, loading with Cu metal nanoparticles caused a decrease in elongation at break to a small extent (dozens percent). Consequently, the tensile strength of Cu nanocomposites also decreased slightly, compared to that of neat PEEK. Loading with carbonaceous CNF, as well as bimetal oxide CuFe2O4 and ceramic SiO2 nanoparticles, on the contrary, resulted in a significant decrease in both elongation at break and tensile strength of the PEEK-based nanocomposites (Table4 ). It is highly likely that this effect was associated with the formation of the composite supermolecular structures. Figure4 shows the average values of the friction coefficients for neat PEEK and the PEEK-based nanocomposites. When sliding on the metal counterpart (Figure4a), the friction coefficients of all composites decreased compared to that of neat PEEK. In this case, the lowest value was registered for the composite loaded with Cu nanoparticles (f = 0.22 ± 0.02), which was lower by 1.6 times than that of neat PEEK. Other patterns were Materials 2021, 14, x FOR PEER REVIEW 9 of 23 Materials 2021, 14, x FOR PEER REVIEW 9 of 23

Materials 2021, 14, 1113 9 of 22

± 0.02), which was lower by 1.6 times than that of neat PEEK. Other patterns were observed when± 0.02), sliding which on was the lowerceramic by counterpart. 1.6 times than The that friction of neat coefficient PEEK. Otherdecreased patterns when were ob- loadingobserved servedwith when CNF, when sliding as sliding well on theas on Cu ceramic the metal ceramic counterpart. nanoparticles, counterpart. The but friction The increased friction coefﬁcient due coefficient to decreased loading decreased with when when loadingloading with CNF, with as CNF, well as as well Cu metal as Cu nanoparticles, metal nanoparticles, but increased but increased due to loadingdue to loading with with SiO2 and CuFe2O4 oxide ones (Figure 4b). SiO2 andSiO CuFe2 and2 OCuFe4 oxide2O4 onesoxide (Figure ones (Figure4b). 4b).

(a) (b) (a) (b) Figure 4. The average values of the friction coefficients for neat PEEK (1) and the PEEK-based nanocomposites loaded Figure 4. The average values of the friction coefficients for neat PEEK (1) and the PEEK-based nanocomposites loaded withFigure 7 wt. 4. The% nanoparticles: average values CNF of the(2); frictionCu (3); coefficientsCuFe2O4 (4), for and neat SiO PEEK2 (5) metal (1) and (а the) and PEEK-based ceramic (b) nanocomposites counterparts. loaded with with 7 wt. % nanoparticles: CNF (2); Cu (3); CuFe2O4 (4), and SiO2 (5) metal (а) and ceramic (b) counterparts. 7 wt. % nanoparticles: CNF (2); Cu (3); CuFe2O4 (4), and SiO2 (5) metal (a) and ceramic (b) counterparts. The wear factor diagram for neat PEEK and the PEEK-based composites is shown in FigureThe 5. When wearThe factorsliding wear diagram factoron the diagram metal for neat counterpar for PEEK neat andPEEKt, the the wearan PEEK-basedd the rate PEEK-based decr compositeseased by comp 3.0–4.5 isosites shown times is shown in in dueFigure to loading5Figure. When with5. sliding When CNF onsliding and the non-metallic metalon the counterpart, metal nanoparticles. counterpar the weart, theFilling rate wear decreased with rate metal decr by easednanoparticles 3.0–4.5 by times3.0–4.5 times resulteddue toloadingdue in an to increase loading with CNF inwith wear and CNF non-metallicfactor and bynon-metallic up to nanoparticles. 1.4 times. nanoparticles. Thus, Filling the withFillingmost metal efficient with nanoparticles metal fillers nanoparticles for resultedresulted in an increase in an increase in wear in factor wear by factor up to by 1.4 up times. to 1.4 Thus,times. the Thus, most the efficient most efficient fillers for fillers for the composites in the metal–polymer tribological contact were CNF, as well as SiO2 and the compositesthe composites in the metal–polymerin the metal–polymer tribological tribological contact contact were CNF, were as CNF, well as as well SiO2 asand SiO2 and CuFe2O4 ceramic nanoparticles with their content of 7 wt.%. The increase in their content CuFe O ceramic2 4 nanoparticles with their content of 7 wt.%. The increase in their content from 0.32 CuFe to4 7 wt.%O ceramic caused nanoparticles a decrease in wearwith theirrate by content 1.3–2.7 of times, 7 wt.%. depending The increase on the in theirtype content from 0.3 to 7 wt.% caused a decrease in wear rate by 1.3–2.7 times, depending on the type of the fillersfrom (Figures 0.3 to 7 wt.%3 and caused 5). a decrease in wear rate by 1.3–2.7 times, depending on the type of the fillersof the (Figures fillers (Figures3 and5). 3 and 5).

Figure 5. Wear factor of neat PEEK (1) and the PEEK-based composites loaded with 7 wt.%

Figurenanoparticles—CNF 5. Wear factor of (2); neat Cu (3);PEEK CuFe (1)2 andO4 (4) the and PEEK-b SiO2 ased(5) during composites the dry loaded sliding with friction 7 wt.% on the nano- metal Figure 5. Wear factor of neat PEEK (1) and the PEEK-based composites loaded with 7 wt.% nano- particlesand ceramic—CNF counterparts. (2); Cu (3); CuFe2O4 (4) and SiO2 (5) during the dry sliding friction on the metal and ceramicparticles counterparts.—CNF (2); Cu (3); CuFe2O4 (4) and SiO2 (5) during the dry sliding friction on the metal Whenand ceramic sliding counterparts. on the ceramic counterpart, a decrease in wear rate was not evident. ThisWhen parameter sliding values on the for ceramic the composites counterpart, loaded a decrease with CNF in wear and rate CuFe was2O4 notremained evident. at When sliding on the ceramic counterpart, a decrease in wear rate was not evident. Thisthe parameter nearly neat values PEEK for level. the composites In other cases, loaded the with wear CNF rate and increased CuFe2O signiﬁcantly4 remained at (up the to nearly6 times). neatThis Thus,PEEK parameter loadinglevel. Invalues other with for thecases, the investigated thecomposites wear rate nanoﬁllers loaded increased with in significantly CNF the amount and CuFe (up of2 toO 74 wt.%6 remained times). did at the not enablenearly improvement neat PEEK level. in wear In other resistance cases, of the the wear PEEK-based rate increased composites significantly for use (up in to the 6 times). ceramic-polymer tribological contact.

Materials 2021, 14, 1113 10 of 22

Protection against the oxidation processes, as well as a solid lubricating effect could be achieved by loading with PTFE particles [46]. For this reason, further wear resistance improvement was supposed to be achieved by simultaneous loading with PTFE solid lubricant particles (including protecting the counterparts from wear by the adhered debris and oxidation products) and nanoparticles. Despite the fact that loading the two-component nanocomposites with 7 wt.% PFTE provided greater increase in wear resistance, SEM micrographs revealed clear signs of nanoparticle agglomeration. This was clearly reﬂected in the decrease in elongation at break. Since loading with 10 wt.% PTFE also contributed to the deterioration of the supermolecular structure, it was decided to load nanoparticles in the amount of 0.3 wt.% into three-component composites.

3.3. Three-Component PEEK-Based Composites Loaded with PTFE and the Nanofillers This section explores the possibility of further wear resistance improvement by loading with PTFE microparticles in an amount of 10 wt.% and four types of nanofillers in the amount of 0.3 wt.%, namely (i) carbonaceous CNF, (ii) metallic Cu, (iii) bimetal oxide CuFe2O4, and (iv) ceramic SiO2 nanoparticles. Table5 presents the physical and mechanical properties of the PEEK-based composites loaded with 10 wt.% PTFE and 0.3 wt.% nanofillers. It could be concluded that: (1) Density of the three-component composites increased by loading with the nanofillers. (2) Their Shore D hardness decreased by 3–4 units compared to that of neat PEEK and was at the level of the “PEEK/10PTFE” composite. (3) The elastic modulus decreased slightly (down to 10%) by loading with CNF, Cu, and SiO2, and increased by filling with CuFe2O4 (up to 10%), as compared to that of the “PEEK/10PTFE” composite. (4) Tensile strength increased by loading with the nanofillers—the “PEEK/10PTFE/0.3 CuFe2O4” composite possessed the maximum value of 95.5 MPa, which was 12 MPa higher, compared to that of the “PEEK/10PTFE” composite. (5) The values of elongation at break also increased by ∆ε = 3–5%, due to loading with the nanofillers compared to that of the “PEEK/10PTFE” composite.

Table 5. The physical and mechanical properties of the PEEK-based composites loaded with 10 wt.% PTFE and 0.3 wt.% nanoparticles.

Density Elastic Modulus Tensile Strength Elongation at Filler Content, wt.% Shore D Hardness ρ, g/cm3 E, MPa σ, MPa Break ε,% PEEK 1.308 80.1 ± 1.17 2840 ± 273 106.9 ± 4.7 25.6 ± 7.2 PEEK/10PTFE 1.324 77.3 ± 0.24 2620 ± 158 83.9 ± 2.4 5.0 ± 0.8 PEEK/10PTFE/0.3CNF 1.344 77.2 ± 0.3 2559 ± 71 86.4 ± 0.6 8.2 ± 1.7 PEEK/10PTFE/0.3Cu 1.356 77.9 ± 0.3 2566 ± 64 88.6 ± 0.9 10.1 ± 2.3

PEEK/10PTFE/0.3CuFe2O4 1.352 77.6 ± 0.2 2744 ± 102 95.5 ± 4.1 8.2 ± 1.1

PEEK/10PTFE/0.3SiO2 1.341 76.6 ± 0.2 2487 ± 47 91.8 ± 2.4 7.8 ± 1.2

Thus, increasing the elastic modulus of the three-component composites by loading with the nanofillers was not evident, with the exception of CuFe2O4 nanoparticles. Figure6 illustrates the supermolecular structure of the studied three-component composites. PTFE particles were distributed quasi-uniformly over the boundaries of the structural elements of the polymer supermolecular structure (Figure6a–d). It could be seen at higher magnification (Figure6e–h) that the nanofillers were segregated and agglomerated in the region of PTFE inclusions. Apparently, due to this fact, the dispersed hardening effect was not pronounced in the three-component composites loaded with the nanofillers, unlike the case of the two-component ones. Materials 2021, 14, x FOR PEER REVIEW 11 of 23

hardening effect was not pronounced in the three-component composites loaded with the nanofillers, unlike the case of the two-component ones. Materials 2021, 14, x FOR PEER REVIEW 11 of 23

Materials 2021, 14, 1113 11 of 22 hardening effect was not pronounced in the three-component composites loaded with the nanofillers, unlike the case of the two-component ones.

(a) (b) (c) (d)

(e) (f) (g) (h) Figure 6. The SEM micrographs of the supermolecular structure of the PEEK-based composites: “PEEK/10PTFE/0.3CNF” (а,e); “PEEK/10PTFE/0.3Cu” (b,f); “PEEK/10PTFE/0.3CuFe2O4“ (c,g); and “PEEK/10PTFE/0.3SiO2“ (d,h).

Figure 7 shows the dependences of the friction coefficients versus the sliding dis- (e) tance, as( fwell) as their average values(g )for neat PEEK and the three-component(h) PEEK-based composites. The friction coefficients of the three-component composites loaded with CNF FigureFigure 6. The 6. SEM The micrographsSEM micrographs of the of supermolecular the supermolecular structure structure of the PEEK-basedof the PEEK-based composites: compos “PEEK/10PTFE/0.3CNF”ites: “PEEK/10PTFE/0.3CNF” (а,e); “PEEK/10PTFE/0.3Cu” (andb,f); copper“PEEK/10PTFE/0.3CuFe gradually increased,2O4“ (c ,wheng); and sliding “PEEK/10PTFE/0.3SiO on the metal counterpart2“ (d,h). at a distance of up (a,e); “PEEK/10PTFE/0.3Cu” (b,f); “PEEK/10PTFE/0.3CuFe O “(c,g); and “PEEK/10PTFE/0.3SiO “(d,h). to 1.5 km. Then it reached2 4 a constant value and was at the2 level of the “PEEK/10PTFE” FigureoneFigure (f7 shows≈ 0.15–0.17; 7 shows the dependences Figure the dependences 7, curves of the 2, friction3, of and the 4). coefficientsfriction However, coefficients versus the minimum the versus sliding value the distance, slidingof the friction dis- as welltance,coefficient as theiras well average was as their 0.1 for valuesaverage the “PEEK/10PTFE/0.3CuFe for values neat for PEEK neat andPEEK the and2O three-component4 ”the and three-component “PEEK/10PTFE/0.3SiO PEEK-based PEEK-based2” com- composites.composites.posites, The which frictionThe frictionwas coefficients 1.7 timescoefficients lowerthan of the of three-component the that three-component of the “PEEK/10PTFE” composites composites loaded one, loaded while with itwith CNF was CNF lower and copperandby copper 3.4 gradually times gradually than increased, that increased, of whenthe neat when sliding PEEK. sliding on theNo onteworthy, metal the metal counterpart they counterpart remained at a distance at constanta distance of up over of up the to 1.5to km. entire1.5 Then km. testing itThen reached range. it reached a constant a constant value and value was and at the was level at the of the level “PEEK/10PTFE” of the “PEEK/10PTFE” one (f ≈ 0.15–0.17;one (f ≈Other 0.15–0.17; Figure patterns7 ,Figure curves were 7, 2, curves observed 3, and 2, 4). 3, when and However, 4). sliding However, the on minimum the the ceramic minimum value counterpart. ofvalue the of friction the The friction friction coefficientcoefficientcoefficients was was 0.1 varied for 0.1 thefor slightly “PEEK/10PTFE/0.3CuFethe “PEEK/10PTFE/0.3CuFe for all investigated 2three-componentO24O”4” and and “PEEK/10PTFE/0.3SiO “PEEK/10PTFE/0.3SiO nanocomposites22 ”over” com- the composites,posites,entire which whichsliding waswas distance 1.71.7 timestimes (Figure lowerthanlowerthan 7c) and that did of no thet “PEEK/10PTFE” exceed 0.1. Its lowest one, one, while whilevalue it it0.04 was was waslower ob- lowerby byserved 3.4 3.4 times times for thethan than “PEEK/10PTFE/0.3Cu” that that of of the neat PEEK. composite, Noteworthy,Noteworthy, which theythey was remainedremained two times constant constant lower than over over that the for the entireentirethe testing“PEEK/10PTFE” testing range. range. one, and seven times lower than that for neat PEEK (Figure 7d). Other patterns were observed when sliding on the ceramic counterpart. The friction coefficients varied slightly for all investigated three-component nanocomposites over the entire sliding distance (Figure 7c) and did not exceed 0.1. Its lowest value 0.04 was observed for the “PEEK/10PTFE/0.3Cu” composite, which was two times lower than that for the “PEEK/10PTFE” one, and seven times lower than that for neat PEEK (Figure 7d).

(а) (b)

Figure 7. Cont.

(а) (b)

MaterialsMaterials2021, 14 ,2021 1113, 14, x FOR PEER REVIEW 12 of 2212 of 23

Figure 7.FigureThe dependences 7. The dependences of the friction of the coefﬁcientsfriction coeffi versuscients the versus sliding the distance sliding (distancea,c), as well (а,c as), as their well average as their values average (b,d values) for (b,d) neat PEEKfor (1)neat and thePEEK PEEK-based (1) and composites: the PEEK-based “PEEK/10PTFE” composites (2); “PEEK/10PTFE/0.3CNF”: “PEEK/10PTFE” (2); (3); “PEEK/10PTFE/0.3Cu”“PEEK/10PTFE/0.3CNF” (3); “PEEK/10PTFE/0.3Cu” (4); “PEEK/10PTFE/0.3CuFe2O4” (5), and “PEEK/10PTFE/0.3SiO2“ (6); metal (а and b) and ceramic (4); “PEEK/10PTFE/0.3CuFe O ” (5), and “PEEK/10PTFE/0.3SiO “ (6); metal (a,b) and ceramic (c,d) counterparts. (c and d) counterparts.2 4 2 Other patterns were observed when sliding on the ceramic counterpart. The friction coefﬁcientsThe varied wear slightly rate diagram for all investigated for neat PEEK three-component and the three-component nanocomposites PEEK-based over the nano- entire slidingcomposites distance is shown (Figure in7 c)Figure and did 8. When not exceed slidin 0.1.g on Its the lowest steel value counterpart, 0.04 was observedthe lowest wear for therate “PEEK/10PTFE/0.3Cu” was registered for the composite, “PEEK/10PTFE/0.3CuFe which was two2O times4” composite. lower than Wear that factor for the was 0.53 ⋅ −6 3 ⋅ “PEEK/10PTFE”± 0.033 10 one, mm and/N m, seven which times was lower 1.75 thantimes that less for than neat that PEEK for (Figurethe “PEEK/10PTFE”7d). com- Theposite, wear and rate 22 diagram times less for neatthan PEEK that for and neat the three-componentPEEK. Thus, the PEEK-basedgreatest increase nanocom- in wear re- positessistance is shown for inthe Figure metal–polymer8. When sliding tribological on the contact steel counterpart,under the dry the sliding lowest friction wear condi- rate wastions registered was achieved for the for “PEEK/10PTFE/0.3CuFe the three-component PEEK-based2O4” composite. composites Wear loaded factor with was 0.3 wt.% 0.53 ±CuFe0.0332·O104 −bimetal6 mm3/N oxide·m, nanoparticles. which was 1.75 times less than that for the “PEEK/10PTFE” composite,However, and 22 times the lesslowest than wear that forfactor neat on PEEK. the ceramic Thus, the counterpart greatest increase was observed in wear for the Materials 2021, 14, x FOR PEER REVIEW 13 of 23 resistance“PEEK/10PTFE/0.3CNF” for the metal–polymer composite. tribological Wear contact factor under was the 0.25 dry ± sliding0.033⋅10 friction–6 mm3/N condi-⋅m, which tions waswas achieved two times for less the than three-component that for the “PEEK/10PTFE” PEEK-based composites composite loaded and 12 with times 0.3 less wt.% than that CuFe2Ofor4 bimetalneat PEEK. oxide However, nanoparticles. it was very close to the “PEEK/10PTFE/0.3Cu”; with a wear factor of 0.28 ± 0.065⋅10–6 mm3/N⋅m. When sliding on the harder ceramic counterpart, the wear resistance improvement was primarily associated with the transfer of film formation and adherence on its surface. This protected the polymer from the oxidation and periodic impacts of the sliding ceramic ball, as well as the ceramic counterpart from the abrasive debris wear (discussed in detail below in the analysis of the wear track surfaces of both the composites and the counterpart). In this case, the effect of nanoparticles should be discussed from the standpoint of the transfer film adhesion on the counterpart surface. Figures 9 and 10 show the optical micrographs characterizing the topography of the wear track surfaces of the studied three-component composites, the steel and ceramic counterpart, as well as the wear track profiles.

Figure 8. Wear rates of neat PEEK (1) and the PEEK-based composites: “PEEK/10PTFE”

Figure(2); “PEEK/10PTFE/0.3CNF” 8. Wear rates of neat PEEK (3); (1) and “PEEK/10PTFE/0.3Cu” the PEEK-based composites: (4); “PEEK/10PTFE/0.3CuFe “PEEK/10PTFE” (2); 2O4” “PEEK/10PTFE/0.3CNF”(5) and “PEEK/10PTFE/0.3SiO (3); “PEEK/10PTF2“ (6),E/0.3Cu” under dry (4); sliding“PEEK/10PTFE/0.3CuFe friction conditions2O4” (5) on and the metal “PEEK/10PTFE/0.3SiOand ceramic counterparts.2“ (6), under dry sliding friction conditions on the metal and ceramic counterparts. However, the lowest wear factor on the ceramic counterpart was observed for the “PEEK/10PTFE/0.3CNF”In the metal–polymer composite.tribological Wear contact, factor a similar was 0.25 pattern± 0.033 of· 10the− 6topographymm3/N·m, of which the wearwas twotrack times surfaces less than thatwas forobserved the “PEEK/10PTFE” for the composite“PEEK/10PTFE/0.3CNF” and 12 times less and than “PEEK/10PTFE/0.3Cu”that for neat PEEK. However, composites. it was Small very closeshallo tow thelongitudinal “PEEK/10PTFE/0.3Cu”; micro-grooves withformed a wear on the sliding surfaces of the composites (Figure 9a,d). Roughness of the composite surfaces was low and almost the same (0.058 and 0.059 μm). On the steel counterpart surface, individual small longitudinal micro-grooves were also visible, while a thin transfer film had also formed (Figure 9c,f). There were almost no damage traces (micro-scratching) of the counterparts.

(а), Ra = 0.058 μm (b) (c)

(d), Ra = 0.059 μm (e) (f)

Materials 2021, 14, 1113 13 of 22

factor of 0.28 ± 0.065·10−6 mm3/N·m. When sliding on the harder ceramic counterpart, the wear resistance improvement was primarily associated with the transfer of film formation and adherence on its surface. This protected the polymer from the oxidation and periodic impacts of the sliding ceramic ball, as well as the ceramic counterpart from the abrasive debris wear (discussed in detail below in the analysis of the wear track surfaces of both the composites and the counterpart). In this case, the effect of nanoparticles should be discussed from the standpoint of the transfer film adhesion on the counterpart surface. Figures9 and 10 show the optical micrographs characterizing the topography of the Materials 2021, 14, x FOR PEER REVIEW 14 of 24 wear track surfaces of the studied three-component composites, the steel and ceramic counterpart, as well as the wear track profiles.

(а), Ra = 0.058 m (b) (c)

(d), Ra = 0.059 m (e) (f)

(g), Ra =0.140 m (h) (i)

(l) (j), Ra =0.139 m (k)

Figure 9.FigureThe topography 9. The topography of the wear of the track wear surfaces track ofsurfaces the polymer of the samples polymer and samples the wear and scar the surfaces wear scar of steelsurfaces counterparts, of steel counter- 444 as well asparts, the wear as well track proﬁlesas the (the wear sliding track distance profiles of 3 (the km)—the sliding “PEEK/10PTFE/0.3CNF” distance of 3 km)— (thea–c ); “PEEK/10PTFE/0.3CNF” “PEEK/10PTFE/0.3Cu” (а–c); 445 2 4 2 446 (d–f); “PEEK/10PTFE/0.3CuFe“PEEK/10PTFE/0.3Cu” (2dO–4f”(); “PEEK/10PTFE/0.3CuFeg–i); and “PEEK/10PTFE/0.3SiOO ” (g–i)2; “(andj– l“PEEK/10PTFE/0.3SiO) composites. “ (j–l) composites.

447

MaterialsMaterials2021, 14 2021, 1113, 14, x FOR PEER REVIEW 14 of 2215 of 24

(b) (а), Ra = 0.046 m (c)

(e) (d), Ra = 0.085 m (f)

(h) (g), Ra = 0.135 m (i)

(k) (j), Ra = 0.122 m (l)

FigureFigure 10. The 10. topography The topography of the ofwear the track wear surfaces track surfaces of the ofpolymer the polymer samples samples and the and wear the scar wear surfaces scar surfaces of ceramic of ceramic 448 counterparts,counterparts, as well as as well the as wear the track wear proﬁles track profiles (the sliding (the sliding distance distance of 3 km)—the of 3 km) “PEEK/10PTFE/0.3CNF”—the “PEEK/10PTFE/0.3CNF” (a–c); (а–c); 449 “PEEK/10PTFE/0.3Cu” (d–f); “PEEK/10PTFE/0.3CuFe2O4” (g–i); and “PEEK/10PTFE/0.3SiO2“ (j–l) composites. 450 “PEEK/10PTFE/0.3Cu” (d–f); “PEEK/10PTFE/0.3CuFe2O4”(g–i); and “PEEK/10PTFE/0.3SiO2“(j–l) composites.

In theWear metal–polymer rate of the tribological “PEEK/10PTFE/0.3CuFe contact, a similar2O4” patterncomposite of the was topography lower than of that the of the 451 wear trackother surfaces three-component was observed ones, for although the “PEEK/10PTFE/0.3CNF” micro-grooves on its surface and “PEEK/10PTFE/ were deeper (Figure 452 0.3Cu”9g). composites. This was confirmed Small shallow by the longitudinalroughness measurement micro-grooves data, formed which onincreased the sliding by almost 453 surfacesthree of thetimes composites by up to (Figure 0.140 9μa,d).m. Despite Roughness this offact, the the composite metal counterpart surfaces was surface low and was also 454 almostprotected the same by (0.058 a highly and 0.059oxidizedµm). polymer On the steeltransfer counterpart film (Figure surface, 9i). The individual most probable small rea- 455 longitudinalson for micro-groovesthe formation of were such also micro visible,-grooves while was a thin the transfermicro-abrasive ﬁlm had debris also formeddamage with 456 (Figurethe9c,f). dispersion There were-hardened almost composite. no damage It traces should (micro-scratching) be noted that the of transfer the counterparts. film was securely 457 Wearfixed rate on of the the friction “PEEK/10PTFE/0.3CuFe surface of the steel2 O counterpart.4” composite In was general, lower a than similar that pattern of the was 458 other three-component ones, although micro-grooves on its surface were deeper (Figure9g ).

Materials 2021, 14, 1113 15 of 22

This was confirmed by the roughness measurement data, which increased by almost three times by up to 0.140 µm. Despite this fact, the metal counterpart surface was also protected by a highly oxidized polymer transfer film (Figure9i). The most probable reason for the formation of such micro-grooves was the micro-abrasive debris damage with the dispersion-hardened composite. It should be noted that the transfer film was securely fixed on the friction surface of the steel counterpart. In general, a similar pattern was observed for the “PEEK/10PTFE/0.3SiO2“composite, although the wear rate was almost twice as high in this case. In the ceramic-polymer tribological contact, the “PEEK/10PTFE/0.3CNF” composite possessed the lowest wear intensity. There were almost no longitudinal micro-grooves on its wear track surface. Roughness of 0.046 µm was very low (Figure 10a). A transfer film was adhered on the counterpart surface, which was similar to the “PEEK/10PTFE” composite. A close value of wear factor and a similar wear pattern was typical for the “PEEK/ 10PTFE/0.3Cu” composite. The smooth wear surface with a low roughness of 0.085 µm was also observed (Figure 10d). A transfer film was almost not visible on the ceramic counterpart surface (Figure 10f). In contrast, longitudinal micro-grooves were on the surfaces of both “PEEK/10PTFE/ 0.3CuFe2O4” and “PEEK/10PTFE/0.3SiO2“composite wear tracks (Figure 10g,h,j,k) and the counterparts (Figure 10i,l). Roughness increased up to 0.122–0.135 µm. At the same time, wear factor doubled in comparison with that of the “PEEK/10PTFE/0.3CNF” composite. Wear debris was found adhered on the ceramic counterpart surface (Figure 10i,l), which protected both the counterpart and the polymer composite from intensive wearing. Thus, when sliding on the metal counterpart, the “PEEK/10PTFE/0.3CuFe2O4” composite possessed wear resistance, which improved by 1.75 times compared to that of “PEEK/10PTFE” (which was 22 times higher than that of neat PEEK). When sliding on the ceramic counterpart, the lowest wear was achieved for the “PEEK/10PTFE/0.3CNF” composite. Its wear resistance was two times higher than that of “PEEK/10PTFE” (which was 12 times higher than that of neat PEEK). However, it was very close to “PEEK/10PTFE/0.3Cu” one. Since the increase in wear resistance was determined by the formation and the adhesion of the transfer film to the counterpart surface, further analysis was carried out for three-component nanocomposites loaded with PTFE. Their microanalysis (EDS) was Materials 2021, 14, x FOR PEER REVIEWcarried out for the both metal- and ceramic-polymer tribological contacts, after loading16 withof 23

the metal (Cu) and ceramic (CuFe2O4) nanofillers. They possessed nearly the highest wear resistance, while their volumetric content and specific surface area were approximately equivalent (Table2). Figure 11 shows the SEM micrographs of the surfaces of the metal and equivalent (Table 2). Figure 11 shows the SEM micrographs of the surfaces of the metal ceramic counterparts after the tribological tests. The chemical composition of the transfer and ceramic counterparts after the tribological tests. The chemical composition of the film is presented in Table6. transfer film is presented in Table 6.

(a) (b)

Figure 11. SEM micrographs of thethe steelsteel ((aа)) andand ceramicceramic ((bb)) counterpartcounterpart surfacessurfaces afterafter thethe tribologicaltribological tests.tests.

Table 6. The results of EDS analysis of the transfer film and debris on the steel and ceramic counterparts in accordance with the labels in Figure 11.

Ceramic Counterpart Zr 26.37/4.51 - - Fe - - - C 73.63/95.49- 67.52/76.68 99.35/99.88 F - 32.48/23.32 - Si - 0.25/0.12 - Cu - - 0.65/0.12

The data in Figure 11 indicate that the Cu (metal) and CuFe2O4 (bimetal oxide) nanofillers, in addition to the PTFE, formed the transfer film on the counterparts and adhered to their surface (the spectra at point 3). The spectra at point 2 corresponded to debris that did not adhered to the surface in the form of a thin film. It was characterized by an increased content of PTFE (Table 6). It should be noted that the data on the oxygen content were consciously excluded, since its presence could be associated not only with the tribological polymer oxidation, but also followed from its presence in the SEM chamber. A more pronounced (contrasting) transfer film formed from the nanofiller (Figure 11a) on the metal counterpart, compared to the ceramic one (Figure 11b). This was due to the higher activity of the counterpart material (steel compared to ceramics), which determined the higher wear resistance of the metal–polymer contact (Figures 9,10). Due to the predominant agglomeration of nanoparticles within PTFE inclusions, they were easier separated and transferred to the counterpart surface. Then, the transfer film was fixed on

Materials 2021, 14, 1113 16 of 22

Table 6. The results of EDS analysis of the transfer ﬁlm and debris on the steel and ceramic counterparts in accordance with the labels in Figure 11.

Element Spectrum 1 Spectrum 2 Spectrum 3 wt%/ at.% wt%/ at.% wt%/ at.% Steel Counterpart Cr 1.72/1.84 - 1.31/0.58 Fe 98.28/98.16 0.69/0.17 58.57/23.76 C - 75.19/83.05 40.08/75.64 F - 23.87/16.66 - Si - 0.25/0.12 - Cu - - 0.04/0.12 Ceramic Counterpart Zr 26.37/4.51 - - Fe - - - C 73.63/95.49- 67.52/76.68 99.35/99.88 F - 32.48/23.32 - Si - 0.25/0.12 - Cu - - 0.65/0.12

The data in Figure 11 indicate that the Cu (metal) and CuFe2O4 (bimetal oxide) nanofillers, in addition to the PTFE, formed the transfer film on the counterparts and adhered to their surface (the spectra at point 3). The spectra at point 2 corresponded to debris that did not adhered to the surface in the form of a thin film. It was characterized by an increased content of PTFE (Table6). It should be noted that the data on the oxygen content were consciously excluded, since its presence could be associated not only with the tribological polymer oxidation, but also followed from its presence in the SEM chamber. A more pronounced (contrasting) transfer film formed from the nanofiller (Figure 11a) on the metal counterpart, compared to the ceramic one (Figure 11b). This was due to the higher activity of the counterpart material (steel compared to ceramics), which determined the higher wear resistance of the metal–polymer contact (Figures9 and 10). Due to the predominant agglomeration of nanoparticles within PTFE inclusions, they were easier separated and transferred to the counterpart surface. Then, the transfer film was fixed on the counterpart surface due to the PFTE tribological oxidation, which resulted in the improved wear resistance, as compared to the “PEEK + 10 wt.% PFTE” composite. Different types of nanoparticles affected wear resistance in many ways. due to the following reasons—(i) various specific surface area; (ii) distinctive agglomeration degrees (for example, according to the manufacturer’s data, the initial Cu nanoparticles were highly agglomerated compared to CuFe2O4); (iii) divergent volumetric contents (and not the same agglomeration degrees as a result); (iv) unequable chemical activity; and (v) miscellaneous specific surface. Accordingly, nanoparticles in the transfer film on the counterpart surface determined the duration of the “polymer-transfer film” conditions in the tribological contact, providing the low stable friction coefficient values and improved wear resistance. It should be noted as a conclusion that the obtained results showed the dual function of nanoparticles in the “PEEK/PFTE/nanofiller” composite—(i) adhesion of the transfer film to the counterpart, and (ii) dispersed hardening, which increased the deformation and strength properties (tensile strength, elongation at break), as compared to the “PEEK/PTFE” composite. The nanofiller type (its composition) determined the tribological oxidation level and, as a consequence, the formation and retention of the transfer film on the counterpart. The volumetric content of the nanofillers affected their ability to form a transfer film on the counterpart, to a much lesser extent, both in the metal- and ceramic-polymer tribological Materials 2021, 14, 1113 17 of 22

contacts. However, it could facilitate the separation of debris from the friction surface of the composite, characterized by the presence of the agglomerated nanoﬁller.

3.4. Interpritation of Results In this research, the aim of loading the PEEK-based composites with the nanofiller was to improve their tribological characteristics due to the formation of adhesive layers on the counterpart and, as a result, to fix the transfer film that contained the “PEEK/PTFE” debris (but not to enhance their physical and mechanical properties, since it is now undeniable that improving wear resistance while maintaining the strength characteristics at the level of the neat polymer is the main challenge of designing advanced PEEK-based materials). The studied nanofillers did not play the role of a solid lubricating medium in the composites (unlike other matrices, for example, thermoplastic semi-crystalline antifriction UHMWPE). After loading, this character of change in wear resistance was likely based on both the tribological oxidation processes developed on the wear surfaces and the increase in their hardness. In this case, nanoparticles affected the intensity to varying degrees, depending on their chemical nature. The achieved low wear resistance values (in comparison to that of neat PEEK), the observed high wear of the metal and ceramic counterparts, as well as the noticeable decrease in the mechanical properties, enabled us to conclude that the loading level of 7 wt.% was too high for the studied nanoparticles. Figure 12 shows optical micrographs characterizing the topography of the wear surfaces of all studied composites on the steel counterpart, as well as the wear track profiles. Deep longitudinal micro-grooves were observed on the wear surfaces of the “PEEK/7Cu” composites, loaded with metal nanoparticles (Figure 12a,b). At the same time, roughness increased by several times, up to Ra = 0.6 µm. There was a lot of debris on the counterpart surface, due to the tribological oxidation of copper and polymer (Figure 12b). The micro-abrasive effect of these solid particles (they had actually plowed, as evidenced by the wear track profilograms with substantially non-smooth “ragged” profiles, Figure 12c) was the main reason for the multiple increase in the wear rate values. For the “PEEK/7SiO2“ and “PEEK/7CuFe2O4” composites, the formed thin transfer films (of the rainbow glow) were observed on the steel counterpart surface (Figure 12e,h). They protected the counterpart and the composites from intense micro-abrasion wear. It should be noted that a structural inhomogeneity associated with excessive PEEK filling was found on the wear surface of the “PEEK/7SiO2“composite (Figure 12e). For this reason, the debris fixation on the counterpart surface was observed in the form of irregularly shaped formations (Figure 12e). On the other hand, protection against the tribological oxidation processes and providing the solid lubricating effect was achieved by the PEEK loading with PTFE particles. It was proven that the mechanism of the lubricating action of the neat PTFE and PTFE-based composites was similar [47], due to the specific fluoroplastic structure. The PTFE transfer film was not continuous, due to poor adhesion. In addition, it consisted of separate fragments, which were removed from the friction zone when the critical thickness of 10–40 nm was reached [48]. Wear of the antifriction composites that contained fluoroplastic occurred according to the adhesive mechanism, the basis of which was both the frictional transfer and the weak adhesive interaction with the counterparts. The low friction coefficient value was provided by the specific supermolecular fluoroplastic structure, as well as the mobility of its molecular chains, and as a result, there was a low resistance to deformation and poor adhesion to the counterparts [49]. Figure 13 shows a scheme of the transfer film formation for the PEEK-based composites. Materials 2021, 14, x FOR PEER REVIEW 18 of 23

Materials 2021, 14, 1113 18 of 22 was found on the wear surface of the “PEEK/7SiO2“composite (Figure 12e). For this reason, the debris fixation on the counterpart surface was observed in the form of irregularly shaped formations (Figure 12e).

(a), Ra = 0.629 μm (b) (c)

(d), Ra = 0.096 μm (e) (f)

(g), Ra = 0.063 μm (h) (i)

Materials 2021, 14, x FigureFOR PEER 12. The REVIEW topography of the wear surfaces on the polymer composites and the steel counterpart, as well as wear track19 of 23 Figure 12. The topography of the wear surfaces on the polymer composites and the steel counterpart, as well as wear track profiles after a test distance of 3 km— “PEEK/7Cu” (a–c); “PEEK/7SiO2” (d–f); and “PEEK/7CuFe2O4” (g–i). proﬁles after a test distance of 3 km— “PEEK/7Cu” (a–c); “PEEK/7SiO2”(d–f); and “PEEK/7CuFe2O4”(g–i). On the other hand, protection against the tribological oxidation processes and providing the solid lubricating effect was achieved by the PEEK loading with PTFE particles. It was proven that the mechanism of the lubricating action of the neat PTFE and PTFE-based composites was similar [47], due to the specific fluoroplastic structure. The PTFE transfer film was not continuous, due to poor adhesion. In addition, it consisted of separate fragments, which were removed from the friction zone when the critical thickness of 10–40 nm was reached [48]. Wear of the antifriction composites that contained fluoroplastic occurred according to the adhesive mechanism, the basis of which was both the frictional transfer and the weak adhesive interaction with the counterparts. The low friction coefficient value was provided by the specific supermolecular fluoroplastic structure, as well as the mobility of its molecular chains, and as a result, there was a low resistance to deformation and poor adhesion to the counterparts [49]. Figure 13 shows a scheme of the transfer film formation for the PEEK-based composites.

Figure 13. The scheme of the transfer ﬁlm formation for the “PEEK/PTFE” (1), “PEEK/nano” (2), and “PEEK/PTFE/nano”

(3) composites. Figure 13. The scheme of the transfer film formation for the “PEEK/PTFE” (1), “PEEK/nano” (2), and “PEEK/PTFE/nano” (3) composites.

Figure 14 shows the dependence of the elastic modulus values from the wear rates for neat PEEK, as well as for the two- and three-component PEEK-based composites.

PEEK PEEK PEEK/10PTFE 3500 3200 PEEK/10PTFE PEEK/0.3CuFe O 2 PEEK/10PTFE/0.3CuFe O 3400 PEEK/0.3CNF 2 3100 PEEK/10PTFE/0.3CNF PEEK/0.3Cu 3300 PEEK/10PTFE/0.3Cu PEEK/0.3SiO 3000 2 PEEK/10PTFE/0.3SiO 3200 2 3100 2900 3000 2800

2900 2700 2800 2600 Elasticmodulus, MPа 2700 Elasticmodulus, MPа 2600 2500 2500 2400 024681012 012381012 Wear factor, 10-6mm3/N×m Wear factor, 10-6mm3/N×m

(a) (b)

Materials 2021, 14, x FOR PEER REVIEW 19 of 23

Materials 2021, 14, 1113 19 of 22 Figure 13. The scheme of the transfer film formation for the “PEEK/PTFE” (1), “PEEK/nano” (2), and “PEEK/PTFE/nano” (3) composites.

FigureFigure 14 14 shows shows the the dependence dependence of of the the elastic elastic modulus modulus values values from from the wearthe wear rates rates for neatfor neat PEEK, PEEK, as well as well as for as the for two- the two- and three-componentand three-component PEEK-based PEEK-based composites. composites.

2900 2700 2800 2600 Elasticmodulus, MPа 2700 Elasticmodulus, MPа 2600 2500 2500 2400 MaterialsMaterials 2021 2021, ,14 14, ,x x FOR FOR024681012 PEER PEER REVIEW REVIEW 012381012 2020 ofof 2323 Wear factor, 10-6mm3/N×m Wear factor, 10-6mm3/N×m

(a) (b)

PEEK PEEK PEEK PEEK PEEK/10PTFE PEEK/10PTFE 35003500 32003200 PEEK/10PTFE PEEK/10PTFE PEEK/0.3CuFe PEEK/0.3CuFeOO 2 2 PEEK/10PTFE/0.3CuFe O PEEK/10PTFE/0.3CuFe2 2O 34003400 PEEK/0.3CNF PEEK/0.3CNF 31003100 PEEK/10PTFE/0.3CNF PEEK/10PTFE/0.3CNF PEEK/0.3Cu PEEK/0.3Cu 33003300 PEEK/10PTFE/0.3Cu PEEK/10PTFE/0.3Cu PEEK/0.3SiO PEEK/0.3SiO 30003000 2 2 PEEK/10PTFE/0.3SiO PEEK/10PTFE/0.3SiO 32003200 2 2 31003100 29002900 30003000 28002800

29002900 27002700 28002800 26002600 Elasticmodulus, MPа Elasticmodulus, MPа 2700Elasticmodulus, MPа 2700 Elasticmodulus, MPа 26002600 25002500 25002500 24002400 0246810121416182002468101214161820 0.00.0 0.5 0.5 2.5 2.5 3.0 3.0 3.5 3.5 4.0 4.0 WearWear factor, factor, 10 10-6-6mmmm3/N3/N××mm WearWear factor, factor, 10 10-6-6mmmm3/N3/N××mm

((cc)) ((dd))

FigureFigure 14.14. The The elastic elastic modulus modulus valuesvalues vs.vs. wear wear factor factor on on the the metalmetal (a(a,b,b)) and and ceramic ceramic ( c(c,d,d)) counterparts. counterparts.

FigureFigure 15 15 shows shows thethe dependencedependence ofof thethe frictionfriction coefficients coefﬁcientscoefficients values values from from the the wear wear ratesrates forfor neat neat PEEK, PEEK, as as well well as as for for thethe two-two- andand three-componentthree-component PEEK-basedPEEK-based PEEK-based composites.composites. composites.

PEEK PEEK PEEK PEEK PEEK/10PTFE PEEK/10PTFE PEEK/10PTFE PEEK/10PTFE PEEK/0.3CuFe O 0.50.5 PEEK/0.3CuFe2 2O 0.50.5 PEEK/10PTFE/0.3CuFe O PEEK/10PTFE/0.3CuFe2 2O PEEK/0.3CNF PEEK/0.3CNF PEEK/10PTFE/0.3CNF PEEK/10PTFE/0.3CNF PEEK/0.3Cu PEEK/0.3Cu PEEK/10PTFE/0.3Cu PEEK/10PTFE/0.3Cu PEEK/0.3SiO PEEK/0.3SiO PEEK/10PTFE/0.3SiO PEEK/10PTFE/0.3SiO 0.40.4 2 2 0.40.4 2 2

0.30.3 0.30.3

0.20.2 0.20.2 Coefficientoffriction Coefficient of friction Coefficientoffriction 0.1Coefficient of friction 0.1 0.10.1

0.00.0 0.00.0 024681012024681012 012381012012381012 WearWear factor, factor, 10 10-6-6mmmm3/N3/N××mm WearWear factor, factor, 10 10-6-6mmmm3/N3/N××mm

((aa)) ((bb))

PEEK PEEK PEEK PEEK Figure 15. Cont. PEEK/10PTFE PEEK/10PTFE 0.4 PEEK/10PTFE PEEK/10PTFE 0.4 0.500.50 PEEK/10/PTFE/0.3CuFe PEEK/10/PTFE/0.3CuFe2OO PEEK/0.3CuFe PEEK/0.3CuFeOO 2 2 2 PEEK/10PTFE/0.3CNF PEEK/10PTFE/0.3CNF PEEK/0.3CNF PEEK/0.3CNF 0.450.45 PEEK/10PTFE/0.3Cu PEEK/10PTFE/0.3Cu PEEK/10PTFE/0.3SiO PEEK/10PTFE/0.3SiO PEEK/0.3Cu PEEK/0.3Cu 0.400.40 2 2 PEEK/0.3SiO 0.30.3 PEEK/0.3SiO2 2 0.350.35 0.300.30 0.20.2 0.250.25 0.200.20 0.150.15 Coefficient of friction Coefficient of friction

Coefficient of friction of Coefficient 0.1

Coefficient of friction of Coefficient 0.1 0.100.10 0.050.05 0.00.0 0.000.00 02468141618200246814161820 0.00.0 0.5 0.5 2.5 2.5 3.0 3.0 3.5 3.5 4.0 4.0 -6-6 3 3 WearWear factor, factor, 10 10-6-6mmmm3/N3/N××mm WearWear factor, factor, 10 10mmmm/N/N××mm

((cc)) ((dd))

FigureFigure 15. 15. The The friction friction coefficients coefficients vs. vs. wear wear factor factor on on the the metal metal ( (aa,b,b)) and and ceramic ceramic ( (cc,d,d)) counterparts. counterparts.

4.4. Conclusions Conclusions

Materials 2021, 14, x FOR PEER REVIEW 20 of 23

PEEK PEEK PEEK/10PTFE 3500 3200 PEEK/10PTFE PEEK/0.3CuFe2O PEEK/10PTFE/0.3CuFe2O 3400 PEEK/0.3CNF 3100 PEEK/10PTFE/0.3CNF PEEK/0.3Cu 3300 PEEK/10PTFE/0.3Cu PEEK/0.3SiO 3000 2 PEEK/10PTFE/0.3SiO 3200 2 3100 2900 3000 2800

2900 2700 2800 2600 Elasticmodulus, MPа 2700 Elasticmodulus, MPа 2600 2500 2500 2400 02468101214161820 0.0 0.5 2.5 3.0 3.5 4.0 Wear factor, 10-6mm3/N×m Wear factor, 10-6mm3/N×m

Figure 14. The elastic modulus values vs. wear factor on the metal (a,b) and ceramic (c,d) counterparts.

Figure 15 shows the dependence of the friction coefficients values from the wear rates for neat PEEK, as well as for the two- and three-component PEEK-based composites.

PEEK PEEK PEEK/10PTFE PEEK/10PTFE 0.5 PEEK/0.3CuFe2O 0.5 PEEK/10PTFE/0.3CuFe2O PEEK/0.3CNF PEEK/10PTFE/0.3CNF PEEK/0.3Cu PEEK/10PTFE/0.3Cu PEEK/0.3SiO PEEK/10PTFE/0.3SiO 0.4 2 0.4 2

0.3 0.3

0.2 0.2 Coefficientoffriction Coefficient of friction 0.1 0.1

0.0 0.0 Materials 2021, 14, 1113024681012 012381012 20 of 22 Wear factor, 10-6mm3/N×m Wear factor, 10-6mm3/N×m

(a) (b)

PEEK PEEK PEEK/10PTFE PEEK/10PTFE 0.4 0.50 PEEK/10/PTFE/0.3CuFe O PEEK/0.3CuFe O 2 2 PEEK/10PTFE/0.3CNF PEEK/0.3CNF 0.45 PEEK/10PTFE/0.3Cu PEEK/10PTFE/0.3SiO PEEK/0.3Cu 0.40 2 0.3 PEEK/0.3SiO2 0.35 0.30 0.2 0.25 0.20 0.15 Coefficient of friction

Coefficient of friction of Coefficient 0.1 0.10 0.05 0.0 0.00 0246814161820 0.0 0.5 2.5 3.0 3.5 4.0 -6 3 Wear factor, 10-6mm3/N×m Wear factor, 10 mm /N×m

FigureFigure 15.15. TheThe frictionfriction coefﬁcientscoefficients vs.vs. wearwear factorfactor onon thethe metalmetal ((aa,,bb)) andand ceramicceramic ((cc,,dd)) counterparts.counterparts.

4.4. Conclusions The mechanical and tribological properties of the composites, based on PEEK and PEEK+PTFE, with various types of nanoﬁllers (carbonaceous, metallic, bimetal oxide, ceramic), under the conditions of the metal- and ceramic-polymer tribological contacts, were investigated. Based on the obtained results, the following conclusions were drawn.

1. It was shown that loading with CNF, Cu, SiO2, and CuFe2O4 nanoparticles in the small content (0.3 wt.%) enabled improvement of the elastic modulus of the PEEK- based composites by 10–15%. Wear resistance of the composites loaded with 0.3 wt.% of the nanofillers increased by 1.5–2.3 times in the metal–polymer tribological contact. This was due to the polymer transfer film formation on the steel counterpart. In the ceramic–polymer tribological contact, loading PEEK with metal nanoparticles caused the intensification of the oxidation processes, the abrasive counterpart wear, and the multiple increases in wear rate. This was accompanied by the polymer transfer film formation on the counterpart, but was not able to improve the wear resistance compared to that of neat PEEK. 2. The formation of the transfer film from the PEEK-based nanocomposite debris on the steel counterpart surface was determined by its supermolecular structure, which is susceptible to destruction, while the ability to fix it depended on the activity of nanoparticles. In the case of CNF and Cu, the transfer film on the counterpart was less oxidized, which reduced the wear rate of the polymer composite. In the tribological tests of the PEEK-based composites loaded with SiO2 and CuFe2O4 nanoparticles, the transfer film was more oxidized. This caused more intense damages and wear of the polymer nanocomposite friction surface. 3. The three-component PEEK-based composites loaded with PTFE and nanoparticles, with the slight decrease in the mechanical properties, provided an increase in wear resistance under the dry sliding friction conditions by up to 22 times in the metal– polymer tribological contact, and up to 12 times (at the wear-free level) in the ceramic- polymer one, compared to that of the neat PEEK. In all cases, this was achieved by the PTFE containing transfer film formation and adhering to the counterpart. 4. The dispersed hardening effect was not pronounced in the three-component PEEK- based composites loaded with nanofillers, unlike the case of the two-component ones. Due to the predominant agglomeration of nanoparticles within PTFE inclusions, they were easier separated and transferred to the counterpart surface. Then, the transfer film adhered on the counterpart surface due to the PEEK tribological oxidation resulted in improved wear resistance, compared to the “PEEK/10PTFE” composite. 5. In the "PEEK/10PTFE/0.3 nanofiller” composite, the nanoparticles served a dual function—(i) adhesion of the transfer film to the counterpart and (ii) dispersed hard- Materials 2021, 14, 1113 21 of 22

ening, which increased the deformation and strength properties (tensile strength, elongation), as compared to the “PEEK/10PTFE” composite. The nanoﬁller type (its composition) determined the tribological oxidation level and, as a consequence, also determined the formation and adherence of the transfer ﬁlm on the counterpart.

Author Contributions: D.A.N., D.G.B., and V.O.A. performed the experiments; S.V.P., D.A.N., D.G.B., L.A.K., and A.V.P. analyzed the data; S.V.P., D.A.N., L.A.K. and F.B. wrote the paper. All authors have read and agreed to the published version of the manuscript. Funding: The work was performed according to the government research assignment for ISPMS SB RAS, project FWRW-2021-0010 and RFBR grant number 20-58-00032 Bel_a. The work was also sup- ported by the RF President Council Grant for the support of leading research schools NSh-2718.2020.8. Tribological tests were carried out at the National Research Tomsk Polytechnic University, within the framework of the Competitiveness Enhancement Program of Tomsk Polytechnic University. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Conﬂicts of Interest: The authors declare no conﬂict of interest. The founding sponsors had no role in the design of the study; in the collection, in the writing of the manuscript, and in the decision to publish the results.

References 1. Bajpai, A.; Saxena, P.; Kunze, K. Tribo-Mechanical Characterization of Carbon Fiber-Reinforced Cyanate Ester Resins Modified With Fillers. Polymers 2020, 12, 1725. [CrossRef] 2. Vazirisereshk, M.R.; Martini, A.; Strubbe, D.A.; Baykara, M.Z. Solid Lubrication with MoS2: A Review. Lubricants 2019, 7, 57. [CrossRef] 3. Feng, Q.; Zou, S.; Li, H.; Dou, M.; Huang, F. Review of Polymer Self-lubricating Coatings. IOP Conf. Ser. Earth Environ. Sci. 2020, 526, 012077. [CrossRef] 4. Lu, Z.P.; Friedrich, K. On sliding friction and wear of PEEK and its composites. Wear 1995, 181–183, 624–631. [CrossRef] 5. Hufenbach, W.; Kunze, K.; Bijie, J. Sliding wear behaviour of PEEK-PTFE blends. J. Synth. Lubr. 2003, 20, 227–240. [CrossRef] 6. Bijwe, J.; Sen, S.; Ghosh, A. Influence of PTFE content in PEEK–PTFE blends on mechanical properties and tribo-performance in various wear modes. Wear 2005, 258, 1536–1542. [CrossRef] 7. Vail, J.R.; Krick, B.A.; Marchman, K.R.; Sawyer, W.G. Polytetrafluoroethylene (PTFE) fiber reinforced polyetheretherketone (PEEK) composites. Wear 2011, 270, 737–741. [CrossRef] 8. Burris, D.L.; Sawyer, W.G. Tribological behavior of PEEK components with compositionally graded PEEK/PTFE surfaces. Wear 2007, 262, 220–224. [CrossRef] 9. Xie, G.Y.; Zhuang, G.S.; Sui, G.X.; Yang, R. Tribological behavior of PEEK/PTFE composites reinforced with potassium titanate whiskers. Wear 2010, 268, 424–430. [CrossRef] 10. Liu, L.; Yan, F.; Gai, F.; Xiao, L.; Shang, L.; Li, M.; Ao, Y. Enhanced tribological performance of PEEK/SCF/PTFE hybrid composites by graphene. RSC Adv. 2017, 7, 33450–33458. [CrossRef] 11. Vande Voort, J.; Bahadur, S. The growth and bonding of transfer film and the role of CuS and PTFE in the tribological behavior of PEEK. Wear 1995, 181–183, 212–221. [CrossRef] 12. Rodriguez, V.; Sukumaran, J.; Schlarb, A.K.; De Baets, P. Influence of solid lubricants on tribological properties of polyetheretherketone (PEEK). Tribol. Int. 2016, 103, 45–57. [CrossRef] 13. Lin, L.; Schlarb, A.K. Tribological response of the PEEK/SCF/graphite composite by releasing rigid particles into the tribosystem. Tribol. Int. 2019, 137, 173–179. [CrossRef] 14. Zhang, Z.; Breidt, C.; Chang, L.; Friedrich, K. Wear of PEEK composites related to their mechanical performances. Tribol. Int. 2004, 37, 271–277. [CrossRef] 15. Buckley, D.H.; Johnson, R.L. Friction, Wear and Decomposition Mechanisms for Various Polymer Compositions in Vacuum to 10-9 Millimeter of Mercury; NASA-TN-D-2073; NASA Technical Note; Lewis Research Center: Cleveland, OH, USA, 1963. 16. Yen, B.K.; Schwickert, B.E.; Toney, M.F. Origin of low-friction behavior in graphite investigated by surface x-ray diffraction. Appl. Phys. Lett. 2004, 84, 4702–4704. [CrossRef] 17. Rouhi, M.; Moazami-goudarzi, M.; Ardestani, M. Comparison of effect of SiC and MoS2 on wear behavior of Al matrix composites. Trans. Nonferr. Met. Soc. China 2019, 29, 1169–1183. [CrossRef] 18. Zalaznik, M.; Kalin, M.; Novak, S.; Jakša, G. Effect of the type, size and concentration of solid lubricants on the tribological properties of the polymer PEEK. Wear 2016, 364–365, 31–39. [CrossRef] 19. Wang, A.H.; Xia, J.; Yang, Z.X.; Xiong, D.H. A novel assembly of MoS2-PTFE solid lubricants into wear-resistant micro-hole array template and corresponding tribological performance. Opt. Laser Technol. 2019, 116, 171–179. [CrossRef] Materials 2021, 14, 1113 22 of 22

20. Kato, H.; Takama, M.; Iwai, Y.; Washida, K.; Sasaki, Y. Wear and mechanical properties of sintered copper–tin composites containing graphite or molybdenum disulfide. Wear 2003, 255, 573–578. [CrossRef] 21. Rapoport, L.; Moshkovich, A.; Perfilyev, V.; Lapsker, I.; Halperin, G.; Itovich, Y.; Etsion, I. Friction and wear of MoS2 films on laser textured steel surfaces. Surf. Coat. Technol. 2008, 202, 3332–3340. [CrossRef] 22. Cho, M.H.; Bahadur, S.; Pogosian, A.K. Friction and wear studies using Taguchi method on polyphenylene sulfide filled with a complex mixture of MoS2, Al2O3, and other compounds. Wear 2005, 258, 1825–1835. [CrossRef] 23. Yang, Z.; Guo, Z.; Yuan, C. Effects of MoS2 microencapsulation on the tribological properties of a composite material in a water-lubricated condition. Wear 2019, 432–433, 102919. [CrossRef] 24. Theiler, G.; Gradt, T. Friction and wear of PEEK composites in vacuum environment. Wear 2010, 269, 278–284. [CrossRef] 25. Bahadur, S.; Gong, D. The role of copper compounds as fillers in the transfer and wear behavior of polyetheretherketone. Wear 1992, 154, 151–165. [CrossRef] 26. Bahadur, S.; Tabor, D. The wear of filled polytetrafluoroethylene. Wear 1984, 98, 1–13. [CrossRef] 27. Zhao, Q.; Bahadur, S. The mechanism of filler action and the criterion of filler selection for reducing wear. Wear 1999, 225–229, 660–668. [CrossRef] 28. Zhao, Q.; Bahadur, S. A study of the modification of the friction and wear behavior of polyphenylene sulfide by particulate Ag2S and PbTe fillers. Wear 1998, 217, 62–72. [CrossRef] 29. Schwartz, C.; Bahadur, S. The role of filler deformability, filler–polymer bonding, and counterface material on the tribological behavior of polyphenylene sulfide (PPS). Wear 2001, 251, 1532–1540. [CrossRef] 30. Wang, Q.; Xu, J.; Shen, W.; Liu, W. An investigation of the friction and wear properties of nanometer Si3N4 filled PEEK. Wear 1996, 196, 82–86. [CrossRef] 31. Wang, Q.H.; Xue, Q.; Shen, W. The friction and wear properties of nanonometre SiO2 filled polyetheretherketone. Tribol. Int. 1997, 30, 193–197. [CrossRef] 32. Wang, Q.-H.; Xu, J.; Shen, W.; Xue, Q. The effect of nanometer SiC filler on the tribological behavior of PEEK. Wear 1997, 209, 316–321. [CrossRef] 33. Schwartz, C.J.; Bahadur, S. Studies on the tribological behaviour and transfer film-counterface bond strength for polyphenylene sulfide filled with nanoscale alumina particles. Wear 2000, 237, 261–273. [CrossRef] 34. Bahadur, S.; Sunkara, C. Effect of transfer film structure, composition and bonding on thetribological behaviour of polyphenylene sulfide filled with nano particles of TiO2, ZnO, CuO and SiC. Wear 2005, 258, 1411–1421. [CrossRef] 35. Ruckdäschel, H.; Sandler, J.K.W.; Altstädt, V. On the friction and wear of carbon nanofiber–reinforced PEEK–based polymer composites. Tribol. Interface Eng. Ser. 2008, 55, 149–208. [CrossRef] 36. Sandler, J.; Werner, P.; Shaffer MS, P.; Demchuk, V.; Altstadt, V.; Windle, A.H. Carbon-nanofibre-reinforced poly(ether ether ketone) composites. Compos. Part A Appl. Sci. Manuf. 2002, 33, 1033–1039. [CrossRef] 37. Werner, P.; Altstadt, V.; Jaskulka, R.; Jacobs, O.; Sandler JK, W.; Shaffer MS, P.; Windle, A.H. Tribological behaviour of carbon- nanofibre-reinforced poly(ether ether ketone). Wear 2004, 257, 1006–1014. [CrossRef] 38. Modi, S.H.; Dikovics, K.B.; Gevgilili, H.; Mago, G.; Bartolucci, S.F.; Fisher, F.T.; Kalyon, D.M. Nanocomposites of poly(ether ether ketone) with carbon nanofibers: Effects of dispersion and thermo-oxidative degradation on development of linear viscoelasticity and crystallinity. Polymer 2010, 51, 5236–5244. [CrossRef] 39. Molazemhosseini, A.; Tourami, H.; Khavandi, A.; Yekta, B.E. Tribological performance of PEEK based hybrid composites reinforced with short carbon fibers and nano-silica. Wear 2013, 303, 397–404. [CrossRef] 40. Guo, L.; Zang, G.; Wang, D.; Zhao, F.; Wang, T.; Wang, Q. Significance of combined functional nanoparticles for enhancing tribological performance of PEEK reinforced with carbon fibers. Compos. Part A Appl. Sci. Manuf. 2017, 102, 400–413. [CrossRef] 41. Papageorgiou, D.G.; Lui, M.; Li, Z.; Valles, C.; Young, R.J.; Kinloch, I.A. Hybrid poly(ether ether ketone) composites reinforced with a combination of carbon fibers and graphene nanoplatelets. Compos. Sci. Technol. 2019, 175, 60–68. [CrossRef] 42. Zhang, G.; Wetzel, B.; Jim, B.; Oesterle, W. Impact of counterface topography on the formation mechanisms of nanostructured tribofilm of PEEK hybrid nanocomposites. Tribol. Int. 2015, 83, 156–165. [CrossRef] 43. Kotov, Y.A. The electrical explosion of wire: A method for the synthesis of weakly aggregated nanopowders. Nanotechnol. Russ. 2009, 4, 415–424. [CrossRef] 44. Mahmoodi, M.J.; Vakilifard, M. Interfacial effects on the damping properties of general carbon nanofiber reinforced nanocomposites—A multi-stage micromechanical analysis. Compos. Struct. 2018, 192, 397–421. [CrossRef] 45. Puhan, D.; Wong, J.S.S. Properties of Polyetheretherketone (PEEK) transferred materials in a PEEK-steel contact. Tribol. Int. 2019, 135, 189–199. [CrossRef] 46. Panin, S.V.; Nguyen, D.A.; Kornienko, L.A.; Ivanova, L.R.; Ovechkin, B.B. Comparison on efficiency of solid-lubricant fillers for polyetheretherketone-based composites. AIP Conf. Proc. 2018, 2051, 020232. [CrossRef] 47. Lancaster, J.K.; Play, P.; Godet, M. Third body formation and the wear of PTFE fibrebased dry bearings. J. Lubr. Tech. 1980, 102, 236–246. [CrossRef] 48. Belyy, V.A.; Sviridenok, A.I.; Petrakovets, N.I. Treniye i iznos materialov na osnove polimerov; Nauka i Tekhnika: Minsk, Belarus, 1976; p. 442. (In Russian) 49. Friedrich, K. Polymer composites for tribological applications. Adv. Ind. Eng. Polym. Res. 2018, 1, 3–39. [CrossRef]