CHAPTER 1 Tribological applications of polymers and their composites – past, present and future prospects Brian J. Briscoe*, Sujeet K. Sinha† *Department of Chemical Engineering & Chemical Technology, Imperial College, London, UK, †Department of Mechanical Engineering, National University of Singapore, Singapore CHAPTER OUTLINE HEAD 1.1 Introduction ...........................................................................................................1 1.2 Classical Works on Polymer Tribology .....................................................................3 1.2.1 Friction ...............................................................................................3 1.2.2 Wear...................................................................................................4 1.3 Tribology of Polymer Composites ............................................................................7 1.3.1 Bulk modification – “hard and strong” fillers in a “softer” matrix ..............8 1.3.2 Interface modification – “soft” and “lubricating” fillers in a “hard and strong” matrix ......................................................................8 1.4 Tribology of Polymer Nanocomposites .....................................................................9 1.5 Future Prospects .................................................................................................17 1.6 Final Remarks .....................................................................................................18 Acknowledgment .......................................................................................................19 Notations and Abbreviations .......................................................................................19 References ................................................................................................................20 1.1 INTRODUCTION Polymers play an important part in materials and mechanical engineering, not just for their ease in manufacturing and low unit cost, but also for their potentially excellent tribological performance in engineered forms [1]. In the pristine or bulk form, only a few of the polymers would satisfy most of the tribological requirements; however, in the composite and hybrid forms, polymers often have an advantage over other mate- rials such as metals and ceramics. Polymer tribology, as a research field, is now well mature given that roughly 50 plus years have seen publication of numerous research articles and reports dealing with a variety of tribological phenomena on a consider- ably large number of polymers, in bulk, composite and hybrid forms. Tribological applications of polymers include gears, a range of bearings, bearing cages, artificial Tribology of Polymeric Nanocomposites. http://dx.doi.org/10.1016/B978-0-444-59455-6.00001-5 1 Copyright © 2013 Elsevier B.V. All rights reserved. 2 CHAPTER 1 Tribological applications of polymers human joint bearing surfaces, bearing materials for space applications including coatings, tires, shoe soles, automobile brake pads, nonstick frying pans, floorings and various types of surfaces for optimum tactile properties such as fibers. The list is growing. For example, in the new area of microelectromechanical systems (MEMS), polymers (such as poly(methylemethacrylate) (PMMA) and poly(dimethylsiloxane) (PDMS)) are gaining popularity as structural materials over the widely used mate- rial, Si [2]. Often, Si is modified by a suitable polymeric film in order to enhance frictional, antiwear or antistiction properties [3]. Similar to the bulk mechanical responses, the tribological characteristics of poly- mers are greatly influenced by the effects of temperature, relative speed of the inter- acting surfaces, normal load and the environment. Therefore, to deal with these effects and for better control of the responses, polymers are modified by adding appropriate fillers to suit a particular application. Thus, they are invariably used in composite or, at best, blended form for an optimum combination of mainly friction and wear per- formances. In addition, pragmatically fillers may be less expensive than the polymer matrix. The composition of the filler materials, often a closely guarded secret of the manufacturer, is both science and art, for the final performance may depend upon the delicately balanced recipe of the matrix and filler materials. However, the last many years of research in the area of polymer tribology in various laboratories have shed much light into the mechanisms of friction and wear. This has somewhat eased the work of materials selection for any particular tribological application. This chapter on the tribology of nanocomposite, an area which is still in its infancy, would endeavor to set the background of the research in polymer tribology. We will refer to the term “polymer” for synthetic organic solid in pristine form, with some additives but no fillers aimed at modifying mechanical properties. The word “composite” would be used when one or more than one fillers have been added to a base polymer with the aim of drastically changing mechanical and tribological prop- erties. “Nanocomposite” will mean a composite in which at least one filler material has one of its dimensions in the range of a few to several nanometers. The chapter would review some of the past, but now classical, works when much of the mechanisms of friction and wear for general polymers were studied and these explanations have stood the test of the time. The early works led to the area of polymer composites where polymers were reinforced with particles and/or short or long fibers. Often, the use of the filler materials has followed two trends that mainly reflect the actual function that the fillers are expected to perform. This type of work on the design of multiphase tribological materials continues mainly aimed at improving an existing formulation, or, using a new polymer matrix or novel filler. The present trend is expected to extend into the future but with much more refinement in materials and process selec- tions. For example, the use of nanosized particles or fibers coupled with chemical enhancement of the interactions between the filler and the matrix seems to produce better tribological performance. Also, there have been some very recent attempts on utilizing some unique properties of polymers, often mimicking the biological systems in one way or the other, which have opened up new possibility of using polymers in tribological applications. One example of this is polymer brush that can be used as a 1.2 Classical Works on Polymer Tribology 3 boundary lubricant. This trend will definitely continue into the future with great prom- ises for solving new tribological issues in micro, nano and bio systems. 1.2 CLASSICAL WORKS ON POLYMER TRIBOLOGY 1.2.1 Friction The earliest works on polymer tribology probably started with the sliding friction studies on rubbers and elastomers [4,5]. Further work on other polymers (thermosets and thermoplastics) led to the development of the two-term model of friction [6]. The two-term model proposes that the frictional force is a consequence of the interfacial and the cohesive works done on the surface of the polymer material. This is assuming that the counterface is sufficiently hard in comparison to the polymer mating surface and undergoes only mild or no elastic deformation. Figure 1.1 shows a schematic diagram of the energy dissipation processes in the two-term model [7]. The interfacial frictional work is the result of adhesive interactions and the extent of this component obviously depends upon factors such as the hardness of the poly- mer, molecular structure, glass transition temperature and crystallinity of the poly- mer, surface roughness of the counterface and chemical–electrostatic interactions between the counterface and the polymer. For example, an elastomeric solid, which has its glass transition temperature below the room temperature and hence is very Velocity Normal load Hard surface Polymer Interfacial zone Rigid asperity Cohesive zone FIGURE 1.1 Two-term model of friction and wear processes. Total friction force is the sum of the forces required for interfacial and cohesive energy dissipations. Likewise, the distinction between interfacial and cohesive wear processes arises from the extent of deformation in the softer material (usually polymer) by rigid asperity of the counterface. For interfacial wear, the frictional energy is dissipated mainly by adhesive interaction while for cohesive wear the energy is dissipated by adhesive and abrasive (subsurface) interactions [7] (with pub- lisher's permission). 4 CHAPTER 1 Tribological applications of polymers soft, would have very high adhesive component leading to high friction. Beyond interfacial work is the contribution of the cohesive term, which is a result of the plow- ing actions of the asperities of the harder counterface into the polymer. The energy required for the plowing action will depend primarily upon the tensile strength and the elongation before fracture (or toughness) of the polymer and, the geometric param- eters (height and the cutting angle) of the asperities on the counterface. The elastic hysteresis is another factor generally associated with the cohesive term for polymers that show large viscoelastic strains such as in the case of rubbers and elastomers [8]. Further, both the interfacial and the cohesive works would be dependent upon the prevailing interface and ambient temperatures, and the rate of relative velocity as these factors would in turn modify the polymer's other materials parameters. Pres- sure has some effect on the interfacial friction as normal contact pressure tends to modify the shear strength of the interface layer by a relation given as [9] τ=τo+αp (1.1) The implication of the above relation is that as the contact pressure increases, the shear stress would increase linearly leading to high friction. Equation (1.1), as simple as it may look in the form, hides the very complex nature of polymer. Also, it does not include the temperature and the shear rate effects on the shear stress. In a normal sliding experiment, it is nontrivial to separate the two terms (inter- facial and cohesive) and therefore most of the data available in the literature gener- ally include a combined effect. Often, the practice among experimentalists is to fix all other parameters and vary one parameter to study its effect on the overall fric- tion coefficient for a polymer. Looking at the published data one can easily deduce that depending upon other factors, the friction is greatly influenced by the class of polymers viz. elastomers, thermosets and thermoplastics (semicrystalline and amor- phous). Semicrystalline linear thermoplastic would give the lowest coefficient of friction, whereas elastomers and rubbers show large values. This is because of the molecular architecture of the linear polymers that helps molecules stretch easily in the direction of shear giving least frictional resistance. Table 1.1 provides some typi- cal values of the coefficient of friction for pristine or virgin polymers. 1.2.2 Wear The inevitable consequence of friction in a sliding contact is wear. Wear of poly- mers is a complex process and the explanation of the wear mechanism can be most efficiently given if we follow one of the three systems of classification. Depending upon which classification we are following, wear of a polymer sliding against a hard counterface may be termed as interfacial, cohesive, abrasive, adhesive, chem- ical wear etc. Figure 1.2 describes the classification of polymer wear [7]. It is to be noted that, similar to the case of friction, polymer wear is also greatly influenced by the type (elastomer, amorphous, and semicrystalline) of the polymer. Of partic- ular importance are the properties such as the elastic modulus, tensile strength and the percentage elongation at failure (toughness), which changes drastically as we 1.2 Classical Works on Polymer Tribology 5 Table 1.1 Friction Coefficient of Few Polymers When Slid against a Steel Disk Counterface (Surface Roughness, Ra = 1.34 μm) with the Corresponding Specific Wear Rates and the Pressure (P) × Velocity (v) Values Coefficient Specific Wear Rate PV Value Polymer of Friction (×10−6 mm3 Nm−1) (Pa m s−1) PMMA 0.48 1315.9 145,560 PEEK 0.32 31.7 149,690 UHMWPE 0.19 15.5 187,138 POM 0.32 168.2 149,690 Epoxy 0.45 3506.6 153,997 POM, poly(oxymethylene). Wearclassificationforpolymers Generic scaling Phenomenological Material approach approach response approach Origin of wear Polymerclassmodel: processmodel: Two-term Elastomers interacting Abrasivewear Thermosets model: Adhesive wear Glassy polymers Transfer wear Cohesivewear Semicrystalline polymers Chemicalwear Interfacial wear Fatigue wear Fretting wear Erosion Delamination wear FIGURE 1.2 Simplified approach to classification of the wear of polymers [7] (with publisher's permission). move from one type of polymer to another. Usually, high tensile strength coupled with high elongation at failure promotes wear resistance in a polymer. Therefore, given all other factors remain constant, some of the linear thermoplastic polymers with semicrystalline microstructure perform far better in wear resistance than ther- mosets or amorphous thermoplastics. These observations are in line with the idea that for polymers, surface hardness is not a controlling factor for wear resistance. In fact high hardness of a polymer may be harmful for wear resistance in dry 6 CHAPTER 1 Tribological applications of polymers sliding against hard counterface as hardness normally comes with low toughness for polymers. High extents of elongation at failure of a polymer means that the shear stress in a sliding event can be drastically reduced due to extensive plastic deformation of the polymer within a very thin layer close to the interface. This interfacial layer accommodates almost all the energy dissipation processes and thus the bulk of the polymer undergoes minimal deformation or wear. Frictional heat generated at the interface is the major impediment to high wear life of the polymer. Many classical and recent works suggest that the wear rate of polymers slid against metal counterface in abrasive wear condition may be given by a simple proportionality as [10,11] W 1/Se (1.2) sp ∝ The above equation, often described as the Ratner–Lancaster correlation, is supported by data obtained by several researchers (Ref [1] provides a summary of published data for the above relation) (Figure 1.3). Equation (1.2) is applicable across one type of polymer class such as semicrystalline thermoplastics. We do not have data to compare this rule for other types of polymers. Because of the low friction and high wear resistance, many of the thermoplastics can be used in tribological applications without any reinforcement and notable among them are the ultra-high molecular weight poly(ethylene) (UHMWPE) and poly(ether ether 3.0E-02 M 1) 2.5E-02 – m N K 3 2.0E-02 m L m e ( 11.55EE-02 at J r r 1.0E-02 GH a I e W F 5.0E-03 E D A C 0.0E+00 B 0.0 0.1 0.2 0.3 0.4 1/(Se) (MPa–1) FIGURE 1.3 A plot of wear rate (cubic millimeters per millimeter per kilogram) as a function of the reciprocal of the product of ultimate tensile stress and elongation to fracture [1]. The data are taken from the literature. A – poly(ethylene); B – Nylon 66; C – PTFE; D – poly(propene); E − high-density poly(ethylene); F – acetal; G – poly(carbonate); H – poly(propylene); I – poly(ethyleneterephthalate glycol); J – poly(vinyl chloride); K – PMMA; L – poly(styrene); M – PMMA (refer to [1] for the sources of the data) (with publisher's permission). 1.3 Tribology of Polymer Composites 7 ketone) (PEEK). UHMWPE has found extensive usage as bearing material for arti- ficial human joints because of its excellent biocompatibility and wear resistance. PEEK, which is a high-temperature polymer, tends to show low wear rate but the coefficient of friction can be relatively high (~0.3). PEEK is now a popular poly- mer as matrix for some new composites with the aim of formulating wear-resistant materials. Nylons are other tribological materials that show low friction and low wear. Poly(tetrafluoroether) (PTFE), a linear fluorocarbon, normally shows very low friction coefficient but, relative to many other thermoplastics, high wear rate due to its unique characteristics of slippage in the crystalline formation of the molecular bond structure. Due to low friction property, PTFE is a good solid lubri- cant if used in composite form. Amorphous thermoplastics such as PMMA and poly(styrene) do not perform very well in a wear test. They show high coefficients of friction and high wear rates. Thermosetting polymers, although they posses high hardness and strength among polymers, show very high wear rate and high coefficient of friction because of very low elongation at failure values. Thermosets are normally used in the form of composites as fiber strengthening can drastically reduce wear. Fiber strengthen- ing sometimes improves the material's resistance to subsurface crack initiation and propagation giving reduced plowing by the counterface asperities or fatigue crack- ing. Interface friction can also be optimized by adding a suitable percentage of a solid lubricant. This trend has led to much research in recent times on producing composite or hybrid materials for optimum wear and friction control using epoxy or phenolic resins as the matrix [12]. 1.3 TRIBOLOGY OF POLYMER COMPOSITES Except for probably only UHMWPE and to some extent nylons, no other polymer is currently being used in its pristine form for a tribological application. The reason is that no polymer can provide a reasonable low working wear rate with optimum coefficient of friction required. Hence, there is a need to modify most polymers by a suitable filler that can reduce the wear rate and, depending upon the design require- ment, either increase or decrease the coefficient of friction. Such a need was realized quite early on [13] and this trend has continued. The second component or the filler can perform a variety of roles depend- ing upon the choice of the matrix and the filler materials. Some of these roles are, strengthening of the matrix (high load carrying capacity), improvement in the subsurface crack-arresting ability (better toughness), lubricating effect at the interface by decreased shear stress and the enhancement of the thermal conduc- tivity of the polymer. The entire aspects of the tribology of polymer composites can be quite complex so as to defy any economic classification. Therefore, a simple but efficient way to handle this topic is to classify the composites accord- ing to the role of the filler material in the composite, modifying the bulk or the interface [14]. 8 CHAPTER 1 Tribological applications of polymers 1.3.1 Bulk modification – “hard and strong” fillers in a “softer” matrix A self-lubricating polymer, such as PTFE, can be made wear resistant by strengthen- ing the bulk with hard or strong filler material such as particles of ceramics/metals or a suitable strong fiber (carbon, aramid or glass fibers). The function of the filler here is to strengthen the polymer matrix and thus increase the load bearing capacity of the composite. The coefficient of friction remains low or increases marginally but the wear resistance can be increased up to an order of magnitude. The disadvantage of using fillers (especially the particulate type) is that the composite material may become somewhat less tough in comparison to the pristine polymer and thus encour- age wear by fatigue; however, this can be avoided by proper optimization of the mechanical and tribological properties. Strengthening by fibers, usually oriented nor- mal to the sliding interface, has shown better result in terms of load-bearing capacity and toughness. A fiber that is nonabrading to the counterface, such as aramid and carbon fibers, is even more beneficial as it promotes the formation of a tenacious and thin transfer film on the counterface that can help in reducing the wear of the com- posite after a short running-in period. Examples of a composite where a hard and strong filler has been added to a softer matrix are the PTFE/GF and Nylon 11/GF systems as shown in Figure 1.4 [1]. As we can see, the coefficient of friction for each case has increased slightly and there is considerable gain in the wear resistance as a result of fiber strengthening. 1.3.2 Interface modification – “soft” and “lubricating” fillers in a “hard and strong” matrix This type of composite utilizes the low shear strength and self-lubricating properties of the filler to reduce the coefficient of friction and, as a result, wear and frictional heating is drastically reduced. The main requirement is the availability of the filler at the interface in sufficient amount such that a reduction in the coefficient of friction and an increase in the wear resistance can be realized. The disadvantage of this type of composite is obviously the reduction in the strength and load carrying capacity of the material in the composite form. Hence, adding this type of fillers beyond a cer- tain percentage by volume or by weight would be counterproductive for tribological performance due to a drastic decrease in the bulk strength. Several researches have focused on finding an optimum ratio of the filler and the matrix to achieve maximum wear resistance [1]. PTFE and graphite in a variety of polymer matrices have been tried with good results; popular among matrices are epoxy, phenolic and PEEK. For both types of composites discussed above, the properties of the transfer film formed on the counterface will define whether or not the composite can have low wear rate. A strongly adhering and tenacious, yet lubricating, transfer film would reduce wear after the formation of the film during the running-in period. Bulky and thick film has the tendency to detach itself from the counterface, which may increase the wear rate due to a continuous film formation and detachment mechanism (trans- fer wear) as well as promoting thermal effects. 1.4 Tribology of Polymer Nanocomposites 9 1) 1.0E-07 – m N 1.0E-06 0.6 C 3 o m 1.0E-05 0.5 e e (m 1.0E-04 0.4 fficie Specific wear rat 1111....0000EEEE+---00001320 (i)PTFE +15%GF(ii)PTFE E+25%GFE(iii)PTF E+25%GF(iv)PTFE E+15%CF(v)PTFE n11(vi)Nylon Fon11+5.6GF(vii)Nylo GFon11+20.7G(viii)Nylo 000...231 nt of friction FIGURE 1.4 The effect of fiber addition on the specific wear rates of a few polymers [1]. The rectangu- lar bar chart indicates specific wear rate (units on the left of the graph) and vertical arrows indicate the coefficient of friction (units on the right of the graph). Test conditions are given below: (i) 440C steel ball (diameter = 9 mm) sliding on polymer specimen, Normal load = 5 N, v = 0.1 m s−1, roughness of polymer surface R = 400 nm, 30% humidity; (ii) test a conditions same as for (i); (iii) test conditions same as for (i); (iv) reciprocating – pin – steel plate apparatus, counterface roughness Ra = 0.051 μm, N2 environment; (v) test conditions same as for (iv); (vi) pin – on – steel (AISI02 quench hardened) disk apparatus, counterface roughness Ra = 0.11 μm, p = 0.66 MPa, v = 1 m s−1; (vii) test conditions same as for (vi); (viii) test conditions same as for (vi) (refer to [1] for the sources of the data; data for (iii) and (iv) are for the same formulation of the composite but the values are different as they have been taken from different research works) (with publisher's permission), GF - Glass Fiber. 1.4 TRIBOLOGY OF POLYMER NANOCOMPOSITES The use of nanoparticles in polymers for tribology performance enhancement started around mid-1990s and this area has become quite promising for the future as newer nanomaterials are being economically and routinely fabricated. In most of the cases, a polymer nanocomposite relies for its better mechanical properties on the extremely high interface area between the filler (nanoparticles or nanofibers) and the matrix (a polymer). High interface leads to a better bonding between the two phases and hence better strength and toughness properties over unfilled polymer or traditional polymer composites. For all polymer/nanoparticle systems, there will be an optimum amount of the nanoparticles beyond which there will be a reduction in the toughness as the stiffness and strength increase. Table 1.2 summarizes friction and wear results of polymer nanocomposites with data taken from the published literature [15–49]. There are mainly two types of polymer nanocomposites that have been tested for tribologi- cal performance. One type is where ceramic nanoparticles, mainly metal and some nonmetal oxides, have been added with the aim to improve load-bearing capacity and wear resistance of the material against the counterface. Examples of polymer nano- composite systems of this type include fillers such as SiO , SiC, ZnO, TiO , Al O , 2 2 2 3 Si N , CuO in polymer matrices such as epoxy, PEEK, PTFE and poly(phenylene 3 4 10 CHAPTER 1 Tribological applications of polymers Table 1.2 A Summary of the Tribological Results on Polymer Nanocomposites Optimum Filler Coefficient Content of Friction Specific Wear Rate (×10−6 mm3 Nm−1) With With Equipment Used, Counterface Size of Filler Lowest Lowest Without With Change Without With Change Material, Load/Pressure Used, S. No Matrix Filler Material Material (nm) COF Wear Filler Filler (%) Filler Filler (%) Sliding Velocity, Remarks (if any) Reference 1 PTFE ZnO 50 15 wt% 15 wt% 0.202 0.209 +3.4 1125.3 13 −98.8 Block-on-ring tribometer, stainless steel, [15] 200 N, 0.431 m s−1 2 PTFE Al O 40 20 wt% 20 wt% 0.152 0.219 +44.1 715 1.2 −99.8 Reciprocating tribometer, stainless steel, [16] 2 3 260 N, 50 mm s−1 3 PTFE CNT 20–30 30 vol.% 20 vol.% 0.2 0.17 −15.0 800 2–3 −99.6 Block-on-ring tribometer, stainless steel, [17] 200 N 4 PTFE Nanoattapulgite 10–25 5 wt% 5 wt% 0.22 0.2 −9.1 625.8 31.2 −95 Block-on-ring tribometer, steel, 200 N, [18] 0.42 m s−1 PTFE 2 M acid-treated 10–25 5 wt% 5 wt% 0.22 0.2 −9.1 625.8 4.9 −99.2 Block-on-ring tribometer, steel, 200 N, [18] attapulgite 0.42 m s−1, nanoattapulgite was treated with hydrochloric acid 5 Epoxy TiO 10 7 wt% 3 wt% 0.54 0.4 −25.9 26×103 1.63×103 −93.7 Pin-on-ring tribometer, carbon steel, 1 MPa, [19] 2 0.4 m s−1 6 Epoxy TiO 300 – 4 vol.% – – −25.9 40 14 −65 Pin-on-ring tribometer, carbon steel, 1 MPa, [20] 2 1 m s−1 7 Epoxy Al O 13 – 2 vol.% – – −25.9 5.9 3.9 −33.9 Pin-on-ring tribometer, carbon steel, 1 MPa, [21] 2 3 0.4 m s−1 8 Epoxy Si N <20 nm 0.8 vol.% 0.8 vol.% 0.57 0.38 −33.3 38 2 −94.7 Pin-on-ring tribometer, carbon steel, 3 MPa, [22] 3 4 0.4 m s−1 9 Epoxy SiO 9 2.2 vol.% 2.2 vol.% 0.58 0.45 −22.4 200 45 −77.5 Pin-on-ring tribometer, carbon steel, 3 MPa, [23,24] 2 0.4 m s−1 SiO -g-PAAM 9 2.2 vol.% 2.2 vol.% 0.58 0.35 −39.7 200 11 −94.5 Pin-on-ring tribometer, carbon steel, 3 MPa, [23,24] 2 0.4 m s−1, SiO nanoparticles were modified 2 with PAAM 10 Epoxy MWCNT-untreated 10–30 – – – – 9 12.5 +38.8 Ball-on-prism tribometer, 30 N, [25] 28.2 mm s−1, mixed with four-blade stirrer Epoxy MWCNT-acid 1 wt% – – – 9 4.5 −50 Ball-on-prism tribometer, 30 N, [25] treated 28.2 mm s−1, mixed with four-blade stirrer, CNTs were treated with nitric acid Epoxy MWCNT-acid 1 wt% – – – 31 3 −90.3 Ball-on-prism tribometer, 30 N, [25] treated 28.2 mm s−1, mixed with four-blade stir- rer, mixed using speed mixer, CNTs were treated with nitric acid 11 Epoxy SiC 61 – – 0.6 0.44 −26.7 290 6 −98 Pin-on-ring tribometer, 3 MPa, 0.4 m s−1 [26] Epoxy SiC-g-PGMA 61 0.6 0.41 −31.7 290 2.9 −99 Pin-on-ring tribometer, 3 MPa, 0.4 m s−1, [26] dry mixing and SiC nanoparticles were modified with PGMA Epoxy SiC-g-PGMA 61 – – 0.6 0.39 −35 290 0.8 −99.7 Pin-on-ring tribometer, 3 MPa, 0.4 m s−1, [26] wet mixing and SiC nanoparticles were modified with PGMA