The competing wear mechanisms and wear maps for steels

WEAR Wear 181-183 ELSEVIER The competing (1995) 280-289 wear mechanisms and wear maps for steels L. Rapoport Department of Mechanics and Control, Center for Technological Education Holon, PO Box 305, Holon 58102, Israel Received 29 April 1994; accepted 25 August 1994 Abstract The development of wear maps is connected to the definition of dominant wear mechanisms and transition regions for pairs rubbing under different wear conditions. The well-known wear maps allow us to reveal the dominant contact conditions (mechanical approach) or the wear mechanisms (structural approach). The purpose of this work is to investigate the dominant and competive wear mechanisms of steel under different contact conditions. For comparison of results, the tests were carried out at the same normalized force (Lim’s and Ashby’s map). Oxidational wear is a dominant wear mechanism in the region of elastic and elastic-plastic contact and well-known parameters of the wear map may be used. The wear rate is defined by a number of competing wear mechanisms (oxidational wear, ploughing, delamination and adhesion) in the plastic region. The number of competing mechanisms is increased with a loading. There exists a definite critical depth of damage accumulation for each wear mechanism. The possibility of estimating a critical depth for damage accumulation under different contact conditions is discussed. Keyword.s: Competitive wear mechanisms; Transition regions; Elastic and plastic contacts; Normalized force; Mechanical and structural approaches 1. Introduction One of the important problem in the construction of a wear map is the search for boundaries of regions of predominant wear and transitions where the competing processes may essentialy change the wear rate. Two alternative approaches exist for estimating the wear modes. The mechanical approach is based on the contact mechanics with the coordinates (&,B) [1,2] or the ratio of contact pressure P, to hardness H, [3,4], or the degree of penetration D and the ratio r/k [5]. The wear maps represent the different contact regions in accordance with the severity of plastic flow. There are boundaries between elastic, elastic-plastic and plastic contact in the mechanics of sliding wear. The link between contact conditions and wear mechanisms is not quite clear. Further investigation of the influence of contact conditions on dominant or competing wear mechanisms is required. The structural approach is based on the definition of dominant wear mechanisms and transitions [6]. It allows us to account for the structure and a chemical state of the surface layer, and to estimate the wear rate for dominant mechanisms. In Ref. [6] it is assumed that the definite wear mechanisms predominate in all 0043-1648/95/$09.50 0 1995 Elsevier Science S.A. All rights reserved SSDI 0043-1648(94)07017-2 regions with the exception of the transitions, i.e. the wear rate is associated with one dominant mechanism. The map represents the mild and severe delamination and oxidational wear and two mild-severe transitions from one to the other. In comparison with the other maps, this one defines the wear behaviour of a steel pair over a wide range of loads and sliding velocities, taking into account the hardness of the rubbing pairs. There is, however, no the quantitative study of contact conditions and wear-mechanism modes for closely similar steels with different hardnesses in pin-disk pairs. The predominant wear mechanisms have been ploted for five different hardnesses as a function of the contact pressure and sliding velocity under lubricated sliding wear of steel [7]. Various wear mechanisms have been found to differ for steel with different hardnesses at the same set of axes. The wear rate of a pin appears to depend mainly on the hardness of the countersurface PIIn light of the above observation, the connection between contact conditions (mechanical approach) and dominant and competing wear mechanisms (structural approach), the quenching and tempering of medium carbon steel pairs will be studied. L.Rapopori I Wear 181-183 (1995) 280-289 281 Table 1 Heat treatment hardness and load conditions P and F (N) for disks Heat treatment Hardness (HV ) Normal force F (N) Q uenched 560 3.63 11.3 19.8 28.0 35.0 Q and T 440 2.76 8.80 15.5 22.1 27.6 22.7 Q and T 360 2.27 7.20 12.7 18.1 Q and T 300 1.90 6.00 10.6 15.1 18.8 A nnealed 200 1.26 3.72 6.97 9.95 12.6 5 x 10-S 1.6~ lo+ 2.8 x lo+ 4 x 1o-4 5x 1o-4 Normalized force F The purpose of this work is to study the dominant and competive wear mechanisms of steel under different contact conditions. 2. Experimental procedure The test was carried out on pin-disk machine with sliding velocity V= 0.67 m s-l. Friction coefficient, wear loss and temperature were measured. Each experiment lasted for 3.6 x lo4 s, corresponding to a sliding distance of 6.8~ 103 m. Then, the disks were weighed with presition of 0.1 pg. The results of wear loss are presented as the wear coefficients K, [9]. The bulk temperature was measured by placing a thermocouple at a distance of 1.5~ 10M3 m from the work edge of pin. The pin and the disk were of similar steel-AISI 1040 in all experiments. The pin was quenched (H,=650). The disks were quenched and tempered to different hardnesses. The data are shown in Table 1. A hard pinmostly soft disk combination was chosen to localize a fracture in the main in the near-subsurface layers of the disk. The wear loss of the pin was not controlled in this test. For comparison of results, tests were carried out at the same normalized force [6] to obtain comparatible results: unit area were determined using three-dimensional definite integrals F,, F, and F,, tabulated by McCool WI3. Results 3.1. Friction and w ear experimeni The effect of normalized force F on friction coefficients is shown in Fig. 1. (The curves for the specimens with H = 3000 MPa and H = 3600 MPa are not included.) Two typical dependences are observed. One is for the specimen with hardness (H= 4400 MPa, curve l), and the other for specimen with H=2000 MPa (curve 2). Note that the curve for the specimen with H=3000 MPa is similar to curve 2 and specimens with H=3600 and 5600 MPa are similar to curve 1. So, the specimens with hardnesses H= 2000 MPa and H= 3000 MPa will be known as “soft” specimens and these with Hr3600 MPa as “hard” ones. The bulk temperature for “hard” specimens is higher than for “soft” specimens under all the same magnitudes of F. The dependences of wear coefficients on F are shown in Fig. 2. Originally, K, is not changed for “soft” F ~:=-.-.- The values of F and F used in the work are listed in Table 1. Microhardnesses of the polished and weared pin and disk surfaces were estimated. In this case, 60 impressions were measured for each testing. The curves of the hardness probability distributions were represented for each specimen. The surface topography was measured before and after testing in two perpendicular directions. The changes in maximum height (AR,) and peak-to-valley height of the profile (AR,) have been determined as a measure of surface damage. The well-known Greenwood and Williamson model [lo] was used for estimation of geometrical contact parameters. The standard deviation of asperity heights a,, the radius of curvature of their summits R and number n of contacts in any TT 0.9 AnHo E - 70 O.R - .$ B -60 g & 7- 6 ‘E ‘e I& 0.6 - - 50 0.5 - -40 6 c! f 0.4 ’ I 10~4 I 2.104 I 3.111-4 N orm nlizc d I 4.10-s I 5.10-4 i: force Fig. 1. The effect of normalized force f on friction coefficient f and bulk temperature T: curves 1,2,f; (H=4400 curves 3,4, T; curves 1, 3, specimen M Pa); curves 2, 4, specimen (H= 2CKJOM Pa). L. Rapoport 282 10-s 1 0 I 1 I 2 N orm a lize d Fig. 2. The curve I 4 I 3 effect of normalized I 5 I Wear 181-183 * F.lW ~orcc force E on wear coefficient K,: 1, H = 2000 M Pa; curve 2, H= 3000 M Pa; curve 3, H= 3600 M Pa; curve 4, H=4400 M Pa. specimens whereas for the “hard” group it is decreased. For I’> 1.6 X lo-“) the difference between the wear coefficients is basically increased for two groups of specimens. For “hard” specimens, the wear coefficients are slightly changed (Fig. 2, curves 3 and 4), while for “soft” specimens, K, is considerably increased and the difference among the “soft” group also rose. It is seen that there is no definite correlation between the normalized force F and the wear coefficient for specimens with the different hardnesses. It is only among the “hard” group the a close relation is found between different specimens. (1995) 280-289 The surface profiles for specimens with different hardnesses under p= 5 X 10d5 are shown in Fig. 3. Under minimum pressure the profiles are shown to be closely similar. The difference in the surface profiles is increased with further loading, especially on passing F= 1.6~ 10w4. The surface profiles at #=5X lop4 are represented in Fig. 4. The “soft” specimens are seen to be damaged more than “hard” specimens under equal values of F. The changes in AR, with hardness for 1”=5x10P5 and p=4x10P4 are shown in Fig. 5. If we assume that AR, indirectly characterizes the surface damage, it may be stated that the surface damage of “soft” specimens is essentially higher than for “hard” specimens under the same normalized force. The data for plasticity index ?P and other geometrical parameters are listed in Table 2. It is seen that for “soft” specimens P>l and in accordance with [lo] plastic flow should occur even under minimum pressure. For “hard” specimens ?P<l. Therefore, the plastic strain is hindered and localized in thin surface layer. Thus, essential differences in friction, wear and roughness parameters were observed for “soft” (!P> 1) and “hard” (?P< 1) specimens under equal values of F. 3.2. The sugace state and microhardness For annealed steel under low load (F=5 x 10P5), the oxide film covering the entire surface and ploughing tracks are observed. This is connected with the fracture of oxide films and an insignificant ploughing (Fig. 6). With load (F= 2.8 X lo-“) thicker oxide films are formed and the intensity of ploughing is increased. The places ., - Fig. 3. The surface profiles for specimens with different hardnesses under E=5 profile 3, H= 3600 M Pa; profile 4, H=4400 M Pa; profile 5, H= 5600 M Pa. X .(,, lo-? rc H = 3600 M Pa H = 4400 M Pa profile 1, H=2000 ME%; profile 2, H=3000 M Pa; L. Rapopoti I Wear 181483 (199.5) 280-289 283 H - 2000 MPa H - zyxwvutsrqponmlkjihgfedcba 3000 MPa H - 3600 MPa H = 4400 MPa H - 5600 MPa Fig. 4. The surface profiles for specimens with different hardnesses under P=5 of metal delamination are also observed. The results show that for “soft” specimens several competings wear mechanisms occur. There is oxidational and ploughing wear under low loads, while severe oxidational, ploughing and delamination wear occur under high loads. A further increase in load involves an additional wear mechanism-adhesion. The surface of a pin rubbing with force F on radius of contact R, standard deviation H=ZOOO MPa R (X10e6 5 x10-s 1.6~ lo+ 2.8X 10-d 120 60 55 m) D (X10+ 0.12 0.10 0.08 m) (the same designations as in Fig. 3). “soft” specimens is smooth without the oxide films and with small areas of ploughing at all loads. Under friction of “hard” specimens, the picture is drastically different. At minimum load (#= 5 X 10m5) the surface of disk (H=4400 MPa) is oxidized with appreciable fracture of oxide films and ploughing (Fig. 7). It is important to note that the pin’s surface was also oxidized (Fig. 8). At F= 1.6 X 10m4the surface of the pin became smooth without the oxide films. The dominant wear mechanism of the disk becomes the oxidational wear. The direction of the oxidized and destroyed areas conforms exactly to the sliding direction (Fig. 9). For quenched steel the dominant oxidational wear without visible tracks of plastic deformation and ploughing is observed at all loads (Fig. 10). The results of microhardness test show that for “soft” specimens, the microhardness did not exceed 4000 MPa at all loads (Fig. 11). Interesting results were obtained for steel with H=3000 MPa, Fig. 12. On curve 2 the region with H27000 MPa appears. This feature is thought to be associated with formation of the “white” Fig. 5. The change of maximum peak to valley height of profile AR, on hardness for 1’=5x lo-’ (curve 1) and f=4X lo-” (curve 2). Table 2 The effect of normalized X10e4 o and plasticity index H=5600 MPa ry R (X10e6 1.74 2.20 2.10 90 60 50 m) Q (E* = 110 GPa) u (X10m6 0.1 0.1 0.1 m) Y 0.65 0.80 0.88 284 L. Rapoport Fig. 6. The surface of annealed steel under low load (i= I Wear 181-183 5 X lo-‘. (1995) 280-289 Fig. 9. The oxidation and destruction of surface films (If= 4400 M Pa, P= 1.6~ 10-5). Fig. 7. The surface of “ hard” disk (H=4400 M Pa) under low load (F=sxlo-5. Fig. 10. The oxidational wear of quenched steel (F=2.8 32iM ) 2imo H. MPa Fig. Il. The hardness distribution for annealed steel (H = 2000 M Pa): curve 1, p= 5 areas (Hz7CKKl MPa). It was reasonable to assume that a further increase in load would raise the temperature and then increase the number of areas with H 2 7000 MPa. However, in our case, the microhardness was decreased to less than H=7000 MPa (curve 3). Obviously, with a load, when the depth of the damaged 4&O 10e4). Flardncss Fig. 8. The oxidated surface of a pin rubbing with “ hard” disk under I’=5x10-5. 3&o x X lo-‘; curve 2, F= 1.6 x lo-$ curve 3, fi= 2.8 x 1O-4. layer exceeds the thickness of the transformed layer, the maximum hardness is decreased. The investigations of the surface state and microhardness results showed that the dominant oxidational wear is observed in wide load range for “hard” spec- L. Rapoport Fig. / Wear 181-183 12. The hardness distribution for a specimen with H=3000 M Pa: curve F=4X 10-4. 1, F=5X 10-s; curve 2, P=2.8X 10m4; curve 3, imens, while some competing wear mechanisms are operated under the same values of p for “soft” specimens. 3.3. Dominant and competing wear mechanisrm Different values of critical points h/R [12,13], attack angle 0 [14] and the degree of penetration D [S] have been calculated for annealed and quenched steels and listed in Table 3. For the boundary of ploughing (D < 0.06 for steel) [15] other parameters were calculated. Then, we compare these critical points with the experimental results, for example, at F= 2.8 X 10p4, when a marked increase_ in the wear coefficient K,,, is observed (Fig. 2). For F = 2.8 x 10p4, the displacement h and the radius of the contact spot a have been calculated (Table 4). Since intense plastic deformation is observed (V=2.1, (1995) 280-289 285 Table 2) for the specimen (H=2000 MPa) under F = 2.8 X 10e4, the results are shown only for the transition to the plastic region. For a quenched specimen the calculated values of h and a are shown for elasticplastic contact (V= 0.88). The experimental estimation is accomplished in accordance with Johnson’s dependences for full plasticity and elastic-plastic contact [16]. The radius of the contact spot was a =4.4 pm for annealed and quenched steel under the same p. This value corresponds to the transition to the plastic region for annealed steel while it is the region of elastic-plastic contact for quenched steel (Table 4). It is seen that, under one and the same values of F/H,,, different contact conditions are reached which result in the end to different wear rates for annealed and quenched steels. For the critical attack angle fP=7 (ploughing for annealed steel, Table 3) and different values of (r/k), the friction coefficient in accordance with [17] has been calculated, Table 5. It is seen that even under the maximum value of 7/k, the friction coefficient is found to be of 50-60% from the real value (f=O.7-0.8). This indicates that the ploughing is accompanied by other friction mechanisms. To compare wear mechanisms for rubbed pairs with different ratios of hardnesses, the results are presented in coordinates plasticity index @normalized force p (Fig. 13). For the region p<O.6 the data from Ref. [18] were used. At Y’> 1, several competing wear mechanisms operate. Under friction (tY< 1) the wear rate is associated with one dominant mechanism over the quite wide load range. Thus, the different contact conditions are observed for “hard” and “soft” specimens under any normalized forces. Obviously, in the plastic region some competing friction mechanisms operate. 4. Discussion It was established that the dominant wear mechanism is oxidational wear for specimens of the “hard” group while for “soft” specimens a transition from mild to severe wear is observed. The kink in the wear rate for “soft” specimens under fi = 1.6 X 10e4 may be associated according to [6] with a change in the dominant mechanism from mild oxidational wear to delamination wear Table 3 The dependence of the critical parameters h/R, 0 and D for different contact conditions H=2000 h/R Transition from elastic to elastic-plastic contact Transition to plastic contact Transition to ploughing M Pa H=5600 fP 0.001 2.56 0.002 < 0.007 3.62 6.8 D hlr 0.022 0.03 < 0.06 M Pa 00 D 0.007 6.8 0.06 0.016 - 10.3 0.135 - L. Rapoport 286 I Wear 181-183 (1995) 280-289 Table 4 The radius of contact R, the displacement h and the radius of the contact spot a under different contact conditions H=2000 H=5600 M Pa R (X10m6 m) h (X10d6 m) a (X10-‘m) 50 Transition from elastic to elastic-plastic contact Transition to plastic contact 55 0.11-0.12 3.5-3.8 Transition to ploughing 55 0.385 6.5 Table 5 Calculated values of the friction coefficient f for different values of r/k 1171 r/k f ‘5 0.1 0.3 0.6 0.8 0.15 0.2 0.29 0.45 ]~-___---_-__,-_-L__-_____~_~__~ O xitla lio rwl we a r + p lo ug hiq _ _ - - _ _ _ - - - - Oxitlalionel _-_---_----- wc:w 1 --- I I IO 4 1vs No rlllillid R (X10m6 P ~~)KCS Fig. 13. The effect of normalized force I’ and plasticity index q on the wear. under the slidingvelocity tested. The “plasticity-induced delamination” is assumed to be the basic wear mode at still lower sliding velocities for ductile materials [19]. However, some competing wear mechanisms are observed to operate under friction of “soft” specimens and the wear rate is defined by the interaction between several wear modes, oxidation, ploughing and delamination. Archard and Hirst [20] and later Welsh [18] established that the transition from mild to severe wear is associated with straight metal contact. It may probably be presented as plastic wear modes (the competing metal ploughing and delamination), according to the concept of wear-mechanism map. Friction of “ hard” specimens may emphasize the dominant wear mechanism-oxidational wear over a wide load range. The high hardness of subsurface layers provides localization of plastic deformation and their oxidation in thin layers. High hardness of oxide films and friction martensite protect the surface from severe wear. Under small load, the wear rate of “hard” disk was greater. This phenomenon is associated with a kinetic of oxidation under pin-disk interaction. To estimate the effect of oxidation we will discuss briefly the Arrhenius constant A, (prefactor) in the M Pa m) h (X10V 6 0.8 m) a (X10m6 m) 5.9 kinetic dependence for the oxidation rate. Quinn [21] showed that A, is basically different under static and tribological conditions. It may be assumed that under low loads (j= 5 X 10m5 and$= 1.6 x 10p4) the quenched surface of the pin is deformed elastically in contact with the “soft” disk, while more intense deformation occurs for the pair “hard” pin-“hard” disk. Obviously, we may use the static Arrhenius constant (A,=3.2 N2 m -4 s-l) [21] for a pin rubbed with a “soft” specimen. The time of pin oxidation is found to be closely equal to the time interval between two neighboring heights (T) and so the oxidation of the surface layer of the pin is hindered (the calculations are skipped). However, plastic deformation occurs and accelerates the oxidational reaction already at a minimal pressure under friction of a pin rubbed with a “hard” specimen. If under this condition, the Arrhenius tribological constant AO= 10 N* mV4 s-l [21] is used, the oxide has time to form and oxide films appear on the pin’s surface. The increase of wear rate is explained by the ploughing of a disk surface by the hard oxide films. The time interval (7) is decreased with increasing load, and oxidation of the pin is stopped. Thus, the change in the kinetics of oxidation on pin and disk surfaces reflects the change of wear rate. Clearly the simple parameters used as coordinates of the wear map do not allow such a complicated phenomenon as the change in the rate of oxidation reactions under different contact conditions to be described. Our experiment confirms the existence of the critical bulk hardness-similar to that found by Welsh [16]. In our case is H,,=3600 MPa. Analysis of the plasticity index qshowed that q= 1 atH= 3600 MPa. Apparently, the maintenance of q< 1 is one of the important conditions for the preservation of dominant oxidational wear. The results showed that contact parameters h/R, D and 8 allow one to estimate fundamental critical transition points from elastic, elastic-plastic, and plastic contacts to ploughing. The dominant mechanism for elastic and elastic-plastic contact of steel pairs is oxidational wear, i.e. the definite contact region suits the dominant wear mechanism. Inside the plastic region (for example, friction of “hard” pin-“soft” disk), several wear mechanisms operate simultaneously (oxidational wear, ploughing, delamination wear and adhesion), and L. Rapopoti / Wear 181-183 it is difficult to determine the dominant wear mechanisms with the aid of parameters of contact mechanics. At the same time several friction mechanisms compete in the plastic region and therefore it is necessary to include the friction coefficient as a coordinate in the wear map. Thus, the mechanical approach allows us to define the boundaries of contact regions but it is very difficult to use the parameters of contact mechanics in the plastic region when several competing mechanisms operate. Parameters of the structure approach allow us to estimate the wear in the dominant region but cannot be used for comparison of “soft” and “hard” pairs rubbing under different contact conditions. Clearly, such a complicated phenomenon as friction and wear cannot be described by simple mechanical and structural parameters. The analyses of wear modes are thought may be based on the concept of damage accumulation 1221 and on accounting for of competing kinetic reactions. Friction and wear are defined by some thermallyactivated kinetic processes such as plastic deformation, (1995) 280-289 287 diffusion of materials and heat, oxidation, phase transformation, and fracture. Driving forces of these processes are gradients of plastic deformation, concentration (oxygen, atoms), temperature and damage. It is important to note that all these gradients are interdependent. The wear rate is defined by the rate and depth of damage accumulation. Obviously, a crack is formed on the depth where the rate of accumulation is maximum. Possible distributions of flow stress (af), hardness (H) and plastic deformation (E) vs. depth for “soft” and “hard” specimens are shown on Fig. 14. Temperature gradient (AT) is similar to these distributions. In addition, a distribution of hydrostatic pressure (gH) has been shown. The damage rate is a minimum on the surface owing to the hydrostatic pressure and reaches a maximum at some depth where the curves of hydrostatic pressure and a flow yield stress intersect. It was established that the crack is formed at some critical depth in delamination [23] and oxidational wear [21,6]. The damage gradient is determined by competition between plastic deformation and hydrostatic pressure in delamination wear [24], while in oxidational delamination 400 200 (a) 200 600 400 Hv 600 lI TV =y,Hx I 200 (b) Fig. 14. The distribution of hydrostatic pressure q,, for (a) annealed and (b) quenched steels: hdslrh, hdl, high deformed layer; ldl, low deformed layer. temperature I 400 I 600 I I 800 TV gradient A T, flow stress o, hardness H and plastic deformation and h,,, the thicknesses of the delaminated, oxidised and transformed c on depth layers, respectively; 288 L. Rapopori I Wear 181-183 wear it is probably controlled by gradients of the oxygen atom concentration and hydrostatic pressure. It is seen that, similar to the critical displacement h,, in elastic and elastic-plastic regions, the critical depth of damage accumulation is revealed in the plastic region. Ploughing starts under h_z 0.8 x lop6 m for annealed steel. The critical depths at the oxidational wear h,,, = 5-10~ low6 m [6,21] and at the delamination wear hcrpdel= 10-50X 10m6 m [23] are schematically shown in Fig. 14. Apparently, the definite depth of martensite phase transformation h,, also exists (Fig. 14). So, a definite succession is observed: hcr,de,> h cr,ox> hqp > hw Obviously, the critical depth (displacement) is h,,> 0.5-l X lop6 m for elastic-plastic steel contact. For the plastic region, the critical depth h,,z 3-5 x lo-” m, so that the ploughing and oxidational wear occur. Since the regions of oxidational wear and delamination overlap ( = 10 x 10e6 m), a transition between these competing mechanisms was observed (especially for “soft” specimens). Thus, the region near h = 10 X lop6 m is probably a transition between the competed ploughing, oxidational and delamination wear mechanisms. Obviously, after h > 10 X 10e6 m the predominant wear mechanism is the metal delamination. However, the high rate of oxidation, after contact, clouds the real picture and we quite often observe oxidised places on the surface. Unfortunately, it is very difficult to define the critical depth under delamination, ploughing and oxidational wear with accounting for loading, deformation and chemical parameters under real contact conditions. We consider that the study of surface gradients and competing kinetic reactions can produce a further step in the understanding of wear modes and their interaction. (1995) 280-289 ing and delamination). The number of competing mechanisms was increased with a load growing. Surface gradients have been considered. It was shown that a definite critical depth exists where the damage rate is maximum for each wear mechanism. Appendix A: Nomenclature r/k f H D h F A_” F ?I’ the ratio of the interfacial shear strength at the contact surface to the shear flow coefficient of friction hardness degree of penetration displacement load nominal area of contact normalized force plasticity index References 111 J.M . Challen, P.L.B. Oxley and B.S. Hockenhull, Prediction of A rchard wear coefficient for metallic sliding friction assuming a low cycle fatigue wear mechanism, Wear, 111 (1986) 275-288. 121 T.H.C. Childs, The mapping of metallic sliding wear, Proc. Inst. Mech. Erg., 202 (1988) 379-395. 131 K.L. Johnson, M apping of metallic sliding wear, Discussion, Proc. Inst. M ech. Eng., 202 (1988) steels; J. Tribal., I13 (1991) 65-72. [41 B. Samuels and M .N. Richards, The transition between mild and severe wear for boundary-lubricated steels, L Tribal., 113 (1991) 65-72. [51 K. Hokkirigawa and K. Kato, A n experimental and theoretical investigation of ploughing, cutting and wedge formation during abrasive wear, Ttibol. Int., 21 (1988) 51-57. 5. Conclusion Mechanical and structural approaches for wear map development were considered. The theoretical and experimental results showed that different contact conditions are realized for “soft” and “ hard” specimens under equal values of normalized force p. The parameter p or similar simple parameters may be used to describe the wear behavior of steel in dominant regions of elastic and elastic-plastic contact (q< l), for example, oxidational wear of “hard” specimens. The dominant process is oxidational wear in the elastic and elasticplastic regions. In the plastic region several competing wear mechanisms operate simultaneously. The critical bulk hardness of the substrate H23600 (q= 1) has been established. For “soft” specimens (?I’> 1) a transition from mild to severe wear was observed. The wear rate is defined by the several competing wear mechanisms (oxidational wear, plough- 161 S.C. Lim and M .F. A shby, W ear mechanism maps, A cta M etaN., 35 (1987) l-24. [71 T. A kagaki and K. Kato, Effects of hardness on the wear mode diagram in lubricated sliding friction of carbon steels, Wear, 141 (1990) PI 1-15. S. Bian, S. M aj and D.W . Borland, The unlubricated sliding wear of steels: the role of the hardness of the friction pair, Wear, 166 (1993) l-5. [91 E. Rabinowicz, New coefficients predict wear of metal parts, Product Eng., 19 (1958) 31-70. WI J.A. Greenwood and J.B.P. W illiamson, Contact of nominally flat surfaces, Pmt. Pll R. Sot. London, Ser. A, 295 (1966) 30&319. J.I. M cCool, Comparison of models for the contact of rough surfaces, Wear, 107 (1986) 37-60. P21 D. Tabor, The Hardness of Metals, Clarendon Press, Oxford, 1951, 175 pp. P31 I.V. Kragelsky, Friction and Wear, Butterworths, London, 1965. 1141 A .A . Torrance, A new approach to the mechanics of abrasion, W ear, 67 (1981) 233-257. u51 H. Kitsunai, K. Kate, K Hokkirigawa and H. Inoe, The transition between microscopicwear modes during repeated sliding friction observed by a scanning electron microscope tribosystem, Wear, 135 (1990) 237-249. K.L. Johnson, Contact Mechanics, Cambridge Univ. Press, 1985. WI L. Rapoport / Wear 181-183 P71 J.M. Chalien and P.L.B. Oxley, An explanation of the different regimes of friction and wear using asperity deformation models, WI Wear, 53 (1979) 229-243. N.C. Welsh, The dry wear of steels, Phil. Trans. R. Sot., A.257 Ser. (1965) 31-70. 1191 M.F. Ashby and S.C. Lim, Wear-mechanism maps, Scripta Mefall., 24 (1990) 805-810. WI J.F. Archard and W. Hirst, The wear of metals under unlubricated conditions, Proc R. Sm. London, Ser A., 236 (1956) 397410. (1995) 280-289 289 T.F.J. Quinn, J.L. Sulivan and D.M. Rowsen, Origins and development of oxidational wear at low ambient temperature, Wear, 94 (1984) 175-191. G. Le Roy, J.D. Embury, G. Edwards and M.F. Ashby, A PI model of ductile fracture based on the nucleation and growth of voids, Acru Metall., 29 (1981) 1509-1522. [231 N.P. Suh, An overview of the delamination theory of wear, Wear, 44 (1977) l-16. v41 A.T. Alpas, H. Hu and J. Zhang, Plastic deformation and damage accumulation below the worn surfaces, Wear, 162-164 (1993) 188-195. WI