Volume 18 - Friction, Lubrication, and Wear Technology Part 7 pps

discussion, the difference between RCW and RCF is that, in RCF, surface fatigue is the damage accumulation process that eventually results in wear particle formation.. Rolling contact fa

discussion, the difference between RCW and RCF is that, in RCF, surface fatigue is the damage accumulation process that eventually results in wear particle formation Rolling contact fatigue may continue for hundreds, thousands, or even millions of cycles before the first wear particles are removed Furthermore, the corners of pits or other RCW damage features may act as nucleation sites for additional fatigue cracks and spread the damage across the surface Because RCF and RCW are so closely related, the causes and effects of both processes will be discussed

The magnitude of the effects of RCW varies from one tribosystem to another Sometimes, a component can sustain appreciable RCW damage before its function is impaired; other times, loss of performance immediately results from the first spall For example, a guide roller in a hot metal bar-handling system may sustain considerable RCW, but this wear may be unimportant as long as the component continues to function adequately In the case of ultraprecision ball bearings for missile guidance systems, however, a very small spall may cause the center of mass of the rapidly spinning bearing to shift, resulting in significant guidance errors

Rolling contact is frequently accompanied by slip or sliding The complex motions experienced by tribocomponents in many types of rolling contact situations produce at least a small percentage of slip or sliding Pure rolling is probably the exception rather than the rule in the diverse applications of rolling components For example, Fig 1 is a schematic profile

of meshing spur gears When the gear teeth first touch, there is a measure of sliding plus rolling When the contact point coincides with the pitch circle, there is pure or nearly pure rolling (depending on the accuracy of gear alignment or lateral vibration) Past this point, slip again occurs between tooth surfaces; it reaches its second maximum just at the point where the surfaces separate Slip can result in scuffing or adhesive wear damage to the mating surfaces if the lubrication is inadequate Therefore, proper lubricant and surface treatment selection is important to minimize the deleterious effects of slip in many types of rolling contact arrangements Lubrication and wear of rolling-element bearings is discussed in the article "Friction and Wear of Rolling-Element Bearings" in this Volume

Fig 1 Engagement of gear teeth in a gear set

Gears. As previously mentioned, the wear modes experienced by gears usually involve both rolling and slip or sliding Dudley (Ref 2) has described the wear of gear teeth, noting that the relative amount of sliding increases as the number of teeth increases Furthermore, the wear is not generally a function of the relative sliding velocity along a tooth face; however, at very high speeds, scoring can occur, and the friction losses in gear sets are strongly affected by sliding

velocity Hypoid gears, worm gears, and spiroid gears can run relatively well even when appreciably worn; consequently, gear designers do not always use wear as a critical design parameter Major considerations in gear design are that:

• Gear tooth stress does not exceed a critical value for desired life

• Gear materials are of the right kind and quality

• The form and finish are adequate

• The lubricant and lubrication system are adequate

• The system is adequately protected from rust and contaminants

• Pitting fatigue life versus contact stress and elastohydrodynamic lubrication conditions

However, determination of the size of the gear unit that is, the pit

ch diameter, gear ratio, and the face width often must involve consideration of the Hertzian conditions necessary to avoid pitting failure of the gear face within the design life Several types of pitting are recognized with regard to gear surface fatigue failures:

• Initial pitting Surface fatigue that usually occurs as a narrow band just below or at the pitch line at the

beginning of component operation and that stops after the asperities have widened sufficiently (worn-in)

to carry the load adequately

• Destructive pitting Usually starts below the pitch line and spreads in number and size of pits until the

gear shape is rendered unusable

Other terminology has been used for describing surface damage in gears Terms include micropitting, surface origin pitting, subsurface origin spalling, subcase fatigue (also known as case-crushing), scuffing, and plastic flow

Spalling is another type of wear mode that occurs in RCW of gear teeth Spalling occurs sporadically, its frequency and exact location being statistical The relatively large particles produced by surface spalling can cause damage if they embed in the contact surface or find their way into clearance-critical portions of gears or bearings Dudley (Ref 2) states that both destructive pitting and spalling seldom happen at less than about 10,000 cycles, but notes that plastic flow (primarily in metallic or polymeric gears) can occur at a relatively low number of cycles if there is an overload on the contact surfaces In summary, there generally are three types of gear wear: (1) normal or polishing wear, (2) moderate wear that is not necessarily destructive to gear life, and (3) destructive wear For more details, see the article

"Metalworking Lubricants" in this Volume

Rolling-Element Bearings. As with gears, pitting, smearing, and spalling are important RCW manifestations in rolling-element bearings The design and wear characteristics of rolling-element bearings are discussed in the article

"Friction and Wear of Rolling-Element Bearings" in this Volume and in Ref 3 The depth of the spall tends to be related

to the location of the maximum Hertzian shear stress below the surface The fatigue spalling life of a bearing, L, is usually defined in terms of the first appearance of a spall and is generally based on the ratio of the equivalent dynamic load, P, to the load capacity, C, obtained from the manufacturer:

(Eq 1)

The exponent p depends on the type of bearing For ball bearings, p = 3; for roller bearings, p = 10/3

In rolling bearings, the design and operating conditions significantly affect the location and rate of RCW damage For example, a given bearing ball may experience a complex rotational path as it moves between the inner and outer bearing races A given point on the surface of the ball may not be stressed once per revolution of the inner race, but rather, because of compound motions, may bear a Hertzian contact stress only occasionally The center of the bearing race groove, on the other hand, is repeatedly stressed each time a ball rolls by Therefore, the race tends to accumulate localized RCF damage cycles more rapidly than the balls

Damage in rolling-element bearings can result from a complex combination of radial loads, axial loads, eccentric loads, thrust loads, and internal (cage/retainer) loading effects A thorough treatment of the failure of rolling-element bearings, including not only rolling contact wear but also fracture, may be found in the article “Failures of Rolling-Element

Bearings” in Failure Analysis and Prevention, Volume 11 of ASM Handbook, (Ref 4) The mechanisms of surface

distress are also discussed in a collection of papers published in 1985 (Ref 5) Widner and Littmann (Ref 6) also provide a comprehensive treatment of bearing damage analysis

Rolling Contact Fatigue Testing

Because RCF is a primary process for the production of RCW, some of the methods that investigators have developed to test the response of materials to RCF conditions similar to those they may experience in service are of interest A

particularly useful reference is Rolling Contact Fatigue Testing of Bearing Steels (Ref 7) In the introduction to this book,

the editor states, "In building any moving machine, it is of primary concern to obtain a long endurance life of the rolling bearing In designing a bearing life test, on the other hand, a long testing time should be avoided The test must be accelerated so that results can be obtained within a reasonably short period of time Unfortunately, the shorter the test becomes, the farther the simulation departs from real conditions of application."

Twenty-one papers in this collection describe testing machines, methods, and effects of processing and microstructure on RCF damage Table 1 provides a guide to the RCF testing methods described in Ref 7 and 8 and Fig 2, 3, 4, 5, 6, 7, 8, 9,

10, 11, and 12 illustrate the testing arrangements listed in the table A variety of RCF machines have been developed, ranging from one-of-a-kind research instruments to commercially manufactured bearing test rigs

Table 1 Summary of RCF testing methods

NASA five-ball testing

apparatus (Fig 2)

Four lower balls, freely rotating 90° apart in a separator, simulates the kinematics of a thrust-loaded bearing; the contact angle can be varied; vibration sensor detects failure in unattended tests; low-(cryo) and high-temperature testing (to 1000 °C, or 1830 °F)

7(a)

Flat-washer testing

apparatus (Fig 3)

16 retained balls rolling in a circle on a flat washer with a 75 mm (3 in.) OD, 50 mm (2 in.) bore, and 6.4

mm ( in.) thickness; 4.17 GPa (605 ksi) contact stress; 1500 rev/min; filtered lubricant delivery system; piezo sensor detects vibration

7(b)

Unisteel testing

apparatus (Fig 4)

Flat washer on retained balls; hanging dead-weight load; contact stress approximately 4.5 GPa (650 ksi);

1500 rev/min; drip feed of lubricant; vibration detection system; thermocouples monitor temperature (typically 50-60 °C, or 120-140 °F)

Fig 3 Flat-washer RCF testing apparatus See Table 1

Fig 4 Unisteel RCF testing apparatus See Table 1

Fig 5 Rolling contact testing apparatus See Table 1

Fig 6 Ball-rod RCF testing apparatus See Table 1

Fig 7 Cylinder-to-ball RCF testing apparatus See Table 1

Fig 8 Cylinder-to-cylinder RCF testing apparatus See Table 1

Fig 9 Ring-on-ring RCF testing apparatus See Table 1

Fig 10 Four-bearing RCF testing apparatus See Table 1

Fig 11 Four-ball RCF testing apparatus See Table 1

Fig 12 Ball-on disk RCF testing apparatus See Table 1

Mechanisms of RCW

Dowson (Ref 9) describes some of the earliest studies of the mechanisms of bearing fatigue and wear failure performed

by Goodman in the late 1800s and early 1900s Goodman introduced the terms pitting, flaking, scratching, and peeling to the wear failure analysis vocabulary, and in a 1912 review paper introduced the concept of surface fatigue thusly: "The speed-effect on the balls has possibly some relation to the well-known effect of very rapid reversals of stress."

Hershey (Ref 10) describes the history of understanding the mechanisms of subsurface fatigue, beginning with groundbreaking work in the mid- to late 1930s that established the manner in which intersecting, subsurface fatigue cracks formed surface pits in rolling elements It is now known that the mechanisms of rolling contact wear involve far more than subsurface fatigue Rolling contact wear can involve mild abrasion from contaminants in lubricants, it can be affected by water or acids in lubricants, it may be exacerbated by transfer of cage material, and in addition, the contributing combination of mechanisms may change over time

Much of the research done on the mechanisms of RCF and RCW has been based on metals and alloys; consequently, relatively little is known about the detailed behavior of ceramics and composites in similar situations When new metal rolling components are first started up, a surface-conditioning process normally occurs In this process, subtle changes take place in both the surface roughness and the subsurface microstructures to reach the steady-state, running condition Initially, the surface material may be stressed beyond its elastic limit, causing some plastic deformation As repeated rolling contact occurs, a condition may eventually be reached wherein no further plastic deformation occurs This process

is called shakedown and has been analyzed in detail by Johnson (Ref 11)

The processes of RCF and RCW generally involve the following steps:

1 Accumulation of dislocations caused by repetitive stressing of the subsurface microstructure; a dislocation cell structure may be formed

2 Nucleation of voids or microcracks in regions of maximum Hertzian stress or nearby discontinuities in the microstructure, such as grain boundaries, preexisting porosity, or inclusion/matrix interfaces

3 Propagation of the microcracks in the subsurface

4 Linking of the cracks and movement of the crack tip toward the free surface

5 Creation of flakes, pits, and/or spalls

6 Spread of the damage to adjoining portions of the surface

7 Initiation of major fatigue cracks from surface or subsurface defects, sometimes causing catastrophic fracture

A bearing or rolling component is often considered to have "failed" at the first appearance of a spall (step 5) In some systems, RCW will go undetected until step 7 As mentioned above, lubricants may accelerate step 3 by helping wedge cracks to open The ceramic bearing industry has made a significant investment in materials development to improve the fracture toughness of their materials and to reduce the proclivity toward catastrophic failure Composite materials are under consideration because they contain internal crack stoppers or crack-branching, energy-dissipating internal interfaces

Associated damage processes derive from the movement of wear particles spalled from the surface of fatigued rolling elements or their counterfaces The particles may embed in the surfaces of the rollers, score the mating surfaces, or act as third-body abrasives to alter the surface roughness and thus the lubricant flow characteristics in the interface

The reader is referred to several reviews of RCF and RCW mechanisms Zwirlein and Schlicht (Ref 12) discussed RCF damage mechanisms in bearing steels, including the development of "white bands" in the near-surface regions of bearings

in conjunction with the decay of the martensitic structure For 52100 bearing steel, the morphology of the white bands

depended on the combination of contact pressure and number of loading cycles Olver et al (Ref 13) reviewed wear in

rolling contacts with small amounts of slip and drew comparisons with Suh's delamination theory of water (Ref 14) Several differences were noted between their observations and the predictions from the delamination theory In particular,

Olver et al found that cracks nucleated at or near the surface, rather than below the surface Also, cracks tended to run

obliquely to the surface in a direction opposite to that of the tractive stress, rather than parallel to the surface Finally, the wear was dependent on the relative hardness of the contacting bodies, not on sliding distances, as stated in the delamination theory Even with relatively small amounts of slip relative to rolling, unacceptable wear was obtained if the lambda ratio, (the ratio of the elastohydrodynamic lubrication film thickness to the composite surface roughness), was

small Improved wear life can be produced by ensuring that the lubricant has adequate load-bearing capacity, in order to maintain a higher ratio in operating bearings

The effect of retained austenite on the RCF of a steel containing 0.18% C, 1.5% Cr, 4.25% Ni, and 1.0% W was reported

by Zhu et al (Ref 15) Using carburizing followed by carbonitriding to increase the amount of retained austenite in

surface layers of the steel, they conducted RCF tests with crowned 50 mm (2 in.) diam rollers They found that the greater the retained austenite content, the greater the contact fatigue resistance, and reasoned that increased toughness from the austenite enhanced the microstructural resistance to crack propagation

Common ferrous materials for rolling-element bearings include AISI 52100 steel, 440C stainless steel, and M-50, M-50 Nil, M-1, M-2, and M-10 tool steels The temperature limit for 52100 and 440C is about 250 °C (480 °F), and molybdenum-containing tool steels such as M-1, M-2, and M-10 may be used up to about 500 °C (930 °F) High-reliability bearings are also made from vacuum degassed and vacuum induction melting/vacuum arc remelting (VIM/VAR) steels Advanced silicon nitride ceramics, such as Norton Company's NBD 200, offer higher temperature capabilities and excellent RCF properties, but are considerably more expensive than steel rolling-element bearings

A key to improving the RCF of ceramic bearings was to reduce the porosity and inclusion content of the materials so that crack-initiating microstructural features would be absent from the near-surface regions where Hertzian contact stresses are maximum High-performance aerospace and military applications will continue to drive the development of high-temperature ceramic materials for rolling-element bearings, and commercial applications of ceramics in advanced automotive and truck engines are gradually appearing Single-piece cam roller-followers made from silicon nitride ceramics are replacing more complicated steel parts that contain needle bearings on their inside diameter The cost of the parts, particularly the cost of machining, will affect the rate at which they are introduced into mass-produced automobiles and trucks The goal of producing truck engines that will run for 750,000 to 1,000,000 miles will also spur the use of ceramics, which have demonstrated outstanding reliability in engine tests

Hybrid bearings that contain ceramic rolling elements and metallic races and retainers are also under development for military and high-performance applications Lubrication is a major problem at elevated temperatures, and various types of solid lubricants have been developed Solid lubrication poses the problem of ensuring a continual supply of the lubricant See the article "Solid Lubricants" in this Volume for a more complete discussion

The most common gear material is steel, but a variety of other materials nonferrous, plastic, and ceramic have also been used The material must be able to be fabricated into the proper contour and finish for effective operation, yet possess the hardness necessary to resist wear and deformation Generally, hardness is a major factor in the selection of gear materials, but temperature and corrosion resistance may also be important in specific applications Surface treatments such as case hardening are used after gear profiles have been machined from softer stock Proper selection of a compatible counterface material is also important Table 2 provides a summary of common ferrous gear materials, their typical heat treatments, and range of surface hardnesses

Table 2 Typical ferrous gear materials

Minimum surface hardness Material Surface treatment

HRC HB

Through-hardened and tempered 180-400 Flame- or induction-hardened 50-54

A-1 through A-5

Carburized and case-hardened 55-60

Nodular iron Annealed, quenched and tempered 140-270

Malleable iron (pearlitic) 165-240

References

1 J.A Schey, Tribology in Metalworking, American Society for Metals, 1983, p 249

2 D.W Dudley, Gear Wear, Wear Control Handbook, M.B Peterson and W.O Winer, Ed., American Society

of Mechanical Engineers, 1980, p 755-830

3 L.B Sibley, Rolling Bearings, Wear Control Handbook, M.B Peterson and W.O Winer, Ed., American

Society of Mechanical Engineers, 1980, p 699-726

4 R.L Widner, Failures of Rolling Element Bearings, Failure Analysis and Prevention, Vol 11, 9th ed.,

Metals Handbook, American Society for Metals, 1986, p 490-513

5 D Dowson, C.M Taylor, M Godet, and D Berthe, Ed., Mechanisms and Surface Distress, Butterworths,

London, 1985

6 R.L Widner and W.E Littmann, Bearing Damage Analysis, Proceedings of the Mechanical Failures

Prevention Group, Publ 423, National Bureau of Standards, 1976, p 67-84

7 J.J.C Hoo, Ed., Rolling Contact Fatigue Testing of Bearing Steels, STP 771, ASTM, 1982

8 L.D Wedeven, R.A Pallini, and C.G Hingley, Selection and Use of Wear Tests for Ceramics, C.S Yust

and R.G Bayer, Ed., STP 1010, ASTM, 1988, p 58-73

9 D Dowson, History of Tribology, Longman, London, 1979, p 385

10 M.D Hershey, Theory and Research in Lubrication, John Wiley & Sons, 1966, p 337

11 K.L Johnson, Contact Mechanics, Cambridge University Press, 1985

12 O Zwirlein and H Schlicht, Rolling Contact Fatigue Mechanisms Accelerated Testing Versus Field

Performance, Rolling Contact Fatigue Testing of Bearing Steels, J.J.C Hoo, Ed., STP 771, ASTM, 1982, p

358-379

13 A.V Olver, H.A Spikes, and P.B McPherson, Wear in Rolling Contacts, Proceedings of ASME Wear of

Materials Conference, American Society of Mechanical Engineers, 1985, p 254-272

14 N.P Suh, The Delamination Theory of Wear, Wear, Vol 25, 1973, p 111

15 D Zhu, F Wang, Q Cai, M Zheng, and Y Cheng, Effect of Retained Austenite on Rolling Element

Fatigue and Its Mechanisms, Proceedings of ASME Wear of Materials Conference, American Society of

Many industries employ processes that lead to impact wear Machine components, cams, and gears mate with a certain dynamic component Typical applications occur in electromechanical printers; a prime example is that of typefaces, which are expected to hold definition, thus assuring high print quality, often for billions of cycles Separable electrical connectors must make repeated contact without excessive removal of highly conductive noble or seminoble films In drilling devices applied to a variety of media, ranging from computer boards to oil-shale, tool wear is a concern of economy In the chattering of tubes carrying liquid wastes from nuclear reactors, safety and reliability are most important

In most of the above applications, impact occurs with a component of sliding, compounding the relative normal approach Thus, the term "compound impact wear" has been coined Another consideration for the mode of contact involves the intensity, of frequency of participation, for mating surfaces "One-body wear" is generated on one repeatedly exposed surface, as in the case of a printer typeface that is regularly impacted against continually renewed paper However, in some devices, such as relays, the same neighboring surfaces are continually mated If one of the mating surfaces is prone

to more wear than the other, and if it is the surface of primary concern, then it could still be viewed in terms of one-body wear If the wear of both surfaces needs to be monitored, then the situation is termed two-body wear

The various percussive wear mechanisms include adhesive, abrasive, surface fatigue, corrosive, and thermal wear These pure forms correspond to those that occur in sliding and rolling contacts The fundamental differences have been sought

in the nature of pressures and the friction coefficients that arise in those contact modes (Ref 2) Hybrid wear mechanisms, which combine several of the above, are frequent An example is the fretting that occurs on the back-printer device of a hammer striking against a print band (Ref 3, 4)

The wear mechanism strongly depends on the combination of mated materials used For example, thermal wear can occur

in polymers The contact stress range can also influence the wear mechanism If impact speed is increased, severe adhesive wear can succeed the low-stressed fretting wear The wear mode, such as the relative size of the sliding versus normal impacting speed component, often makes an essential difference The worn surfaces of aluminum projectile specimens subjected to compound impact cycles, upon repeatedly impacting a steel plate rotating at different speeds, are shown in Fig 1

Fig 1 Impact-worn surfaces of spherical-headed aluminum 2024-T4 projectiles, 15×; the 0.5 g (0.018 oz)

projectiles (for which = 1.1 m, or 44 in., and R = 41 mm, or 1.6 in.) repetitively impacted an alloy steel target plate at v = 1.7 m/s (68 in./s), q/ 0 0.8 (a) v = 0, N = 1.5 × 107 (b) v = 250 mm/s (10 in./s), N =

1.5 × 10 7 (c) v = 1.25 m/s (50 in./s), N = 104 (d) v = 3.0 m/s (120 in./s), N = 103

In order to systematically study the subject, it is helpful to divide variable parameters into these categories:

• Loads (for example, mass and speed components)

• Materials (elastic and strength data)

• Surface definition (topography and friction coefficient)

• Wear coefficients

A rational, semi-empirical impact wear theory (Ref 1), which embodies all the above ingredients, is outlined in this article

Experimental Background

From a long line of experimental impact wear studies, several that underlie the variety of industrial needs are described

below Examining vibrational contact for nuclear engineering materials, De Gee et al (Ref 5) wore sintered aluminum

powder (SAP) hemispheres against SAP planes, and Ko (Ref 6) investigated the chattering of heat-exchange tubes against annular supports Wellinger and Breckel (Ref 7) studied the wear of the plastically deforming contacts of a great variety

of metal specimens subjected to repeated normal impact Montgomery (Ref 8) correlated and modeled wear that was caused by rock drill bits The work of Sorokin (Ref 9), one of several Soviet industrial wear investigators, is also worth citing

Bayer, Engel, and their coworkers (Ref 10, 11) constructed impact wear test equipment, of both the ballistic projectile (Fig 2a) and pivotal hammering type (Fig 2b), out of electromechanical printer mechanisms They used various material combinations, stress range, lubrication conditions, and other parameters to optimize various machine contacts Rice (Ref 12) built a reciprocating impact wear tester and investigated, with his coworkers, a range of wear relationships (Ref 13,

14) that were dependent on metallurgy, such as grain structure Sugita et al (Ref 15) dealt with effects of impact wear on

MgO single crystals, a typical brittle material Fretting wear that was due to repeated impacting was studied by Levy and Morri (Ref 16) Sare (Ref 17) investigated wear-resistant materials subjected to abrasion in ore-crushing hammers Bayer (Ref 18) found that the impact wear of elastomers typically exhibits two stages: a rapid initial are of deformation, with ever-decreasing rates following it He found similar results in thermoplastic polyester urethane, polyester, and polyamide materials

Fig 2 Impact wear testing apparatuses (a) Ballistic impact wear tester with three projectile bays located 120°

apart (b) Pivotal hammering tester, where three hammers impact a polymer sheet

Instrumentation for an experimental impact wear study (Ref 1) includes impact force measurement devices (such as piezoelectric transducers) that calibrate force versus speed relations The contact stress for either a projectile or hammer that strikes a known material at known speed can be calculated by analytical or numerical (finite-element) methods This

information, which is related to the Hertz contact theory, turns out to be important in terms of the wear theory outlined in this article

The wear scar can be measured by removing the worn specimen and topographically scanning it It can then be repositioned for more load cycles if an accurate method is available (Ref 19) Scanning electron microscopy (Ref 20) is commonly used to check the wear mechanism, the wear scar shape, and any idiosyncrasies of the process The worn surface must be characterized for hardness (for example, by a Rockwell, Vickers, or Knoop microhardness test) (Ref 1, 21)

Model for Compound Impact

To model compound impact (Ref 22), Fig 3 shows a projectile of mass, m, approaching a target plane with a normal relative speed, V Subsequent to striking (at time, t, equated to 0) the plane that is moving in the tangential x-direction at a speed, v, the projectile is accelerated by friction in the x-direction Because friction is impending, the equation of motion

of the slipping projectile includes the Coulomb friction term:

Fig 3 Model of compound impact, where projectile strikes a tangentially moving plate (a) Projectile approach

(b) Impact pulse and sinusoidal approximation

If the impacting mass is to catch up with the plane at some time, < t*, then the slipping time, , can be calculated from

Eq 4, by setting = v at t = , obtaining:

(Eq 5)

It is understood that after t = , the projectile is adhered to the plane and no longer slips against it The negative

component of the arc cos function is characteristic of the slipping properties of the system, and is called the slip factor,

• f > 2 is high sliding speed; slipping persists throughout the impact contact time, and is imaginary in Eq 5

If a spring, k, is restraining the mass in the x-direction (Fig 4a), then a force k · x will act on the projectile, tending to

increase the slipping time The latter is calculated from a transcendental equation corresponding to Eq 5:

(Eq 7)

where = and = /t* Figure 4(b) shows the effect of spring restraint

Fig 4 Spring-restrained projectile model (a) Impact (b) Corrected slip factors for restrained projectiles

If rotation of the impacting projectile is considered in addition (Fig 5), then an inertial couple, I , must be added to the

free-body diagram of the projectile The slipping time is obviously shortened by this additional degree of freedom, because adherence to the target is easier to achieve Now is calculated from the equation

(Eq 8)

which corresponds to Eq 5 of the unrestrained, unrotating case of the impacting projectile The sliding restraint and rotational capacity can be used to more accurately model pivotal hammering devices

Fig 5 Spring-restrained projectile rotating during impact (a) Impact schematic (b) Corrected slip factors for

restrained rotating projectiles (c) Plots of versus f and

Linear Impact Wear

Two cases were found to be of great practical importance in repetitively impacted contacts Discussed first is a linear wear mechanism that occurs in print heads

Consider the linear wear mechanism that is characteristic of an abrasive sliding process:

(Eq 9)

where K is the wear constant for the material pair, P is the normal contact force, X is the sliding distance, H is the hardness of the wearing body, and W is the volume worn away When a contact, such as a print head that strikes, paper, undergoes a cycle of compound impact, the one-body wear of the print head per differential time, dt, during a single impact cycle, at time, t < , can be evaluated After adherence takes place, no more wear generation is assumed (Ref 1)

Fig 6 Function C(f) relating abrasive impact wear to slip factor

The above wear process is termed "linear," because the wear per pulse, W, is independent of the number of cycles, or of

the geometry of the wear scar as consecutive impacts are incurred A similar analysis was also applicable to computer print wires, which were observed to wear in a "pencil-sharpener" mode (Ref 1)

Abrasive wear constants, K, were taken from the wheel tester invented by Roshon (Ref 23), and adopted as an ASTM standard to grade computer paper for abrasivity A wide ribbon is stretched on the cylindrical surface of a d = 1.2 m (4 ft) diameter steel drum, against which a cantilevered bob made of the print head material of known hardness, H, is squeezed, producing a controllable normal force, P, on the contact The bob is given a slight tangential speed perpendicular to the

sliding direction, so that the paper surface tends to be renewed with subsequent wheel rotations Wear volume

measurements at the contact point of the bob after a known number, n, of wheel revolutions at constant speed finally yield

the wear constant from Eq 9:

(Eq 16)

Impact Wear of Machine Contacts

The repetitive impacts that take place in metallic machine contacts tend to create a wear scar, which appears after a

certain incubation period in the wear life This means that within a period of initial cycles, N0, called the zero-wear limit, the contact spot maintains the original surface median or root mean square (rms) height,

Figure 7 shows a set of wear curves, h versus N, for carbon steel ballistic impact wear projectiles (m = 1.27 g, or 0.05 oz, Rockwell hardness (HRC) = 20, R = 140 mm, or 5.5 in., head radius) impacted against an alloy steel plate at the common normal approach speed, V = 1.73 m/s (68 in./s) The target disk rotation was varied, however, producing the relative sliding speeds, v = 0, 0.25 m/s (10 in./s), 1.27 m/s (50 in.s/s), 3.81 m/s (150 in./s), and 7.62 m/s (300 in./s)

Fig 7 Compound impact wear of carbon-steel projectiles tested in ballistic impact wear apparatus Projectiles

are carbon steel, HRC = 20, V8 ( 0.5 m, or 20 in.), m = 1.27 g (0.045 oz), R = 140 mm (5.5 in.), V =

1.7 m/s (68 in./s); target disk is alloy steel 4140, HRC = 40 to 45, V16, = 0.62

Projectiles were inspected for wear at logarithmic intervals It was observed from the experimental measurement points

that the wear process became "measurable" at earlier cycle numbers as the v/V ratio was increased, that is, the sliding

contribution was increased with respect to the normal impact contribution

Zero-Wear Limit. In order to establish the zero-wear limit as a quantitative, semi-empirical relation, an extension of the

Bayer-Ku zero-sliding-wear theory (Ref 24) was made The latter has, for the number of passes, N, denoting zero wear:

(Eq 17)

In impact wear, it is sensible to consider subsurface damage, D1, which is caused by the maximum shear stress that occurs

in the depth of a contact without any shear tractions attending, and surface damage, D2, which is due to shear traction caused by superposed sliding effects that occur during the slipping time of compound impact cycles

The damage contributions (Ref 1) are defined as integrals of the representative maximum subsurface and surface stresses over an appropriate interval of the impact pulse Thus,

(Eq 21)

which has a coefficient that is to be determined via experiments

From many sets of experiments conducted on various steel and aluminum projectiles, was found to be close to 1.1 Note that an essentially constant r is another feature of the Bayer-Ku sliding wear theory The relationship of these two constants is discussed in Ref 1

Approximate expressions (Ref 25) of for common metals (Fig 5c) are:

(Eq 22)

(Eq 23)

Measurable Wear. Past the zero-wear limit, a wear scar becomes evident, and its dimensions undergo a growth process Only symmetric quadratic (spherical or cylindrical) contact surfaces subjected to repetitive impacts are considered here It is assumed that throughout the wear history, either the impact momentum or peak impact force remains constant However, it is recognized that because of changes in contact geometry that are due to wear, the contact area and, thus, the Hertz contact stress state per impact, will vary Figure 8 illustrates the descriptors relevant to measurable wear

Fig 8 Concepts relevant to measurable wear (a) Wear formation (b) Wear history (c) Impact history (pulse)

(d) Loading history

It is useful to establish the concept of the instantaneous wear state, which includes the geometry of the worn contact, and the Hertzian contact stresses arising in this contact, which are due to the prescribed impact cycles The instantaneous wear

state at the peak of the Nth impact would include the instantaneous peak contact radius, a, peak maximum pressure, q, and

others

The common experience with moderately stressed machine contacts is the tendency of wear scars to evolve toward conformance of the two mating surfaces Figure 9 shows an originally plane spring steel target gradually conforming to the outline of a tool steel pivotal hammer In one-body wear processes, the "softer" partner progressively wears toward the geometry of the nonwearing, "harder" partner (a greater resistance to wear is meant here by hardness because it is

usually proportional to H) Wear is dependent on the instantaneous contact area at any stage of the wear history The initial condition to this phase is the zero-wear geometry, defined by the quadratic surface belonging to N = N0, h = /2

Fig 9 Process of changing curvature on flat blued spring steel target (AISI-C1095, 48 TO 51 HRC) surface

impacted by hard cylindrical hammer (Carpenter extra, air-hardened tool steel, HRC = 62 to 64, r = 35 mm, or

1.38 in.) In terms of scale, each vertical division is 0.508 m (20 in.) and each small horizontal division is 50.8 m (0.002 in.)

In the measurable wear stage of the wear history, the growth of the wear scar can be followed by observing the deepening

of the wear scar, h, and the change of the radius of curvature, R It can also be described by the instantaneous peak contact radius, a, or peak contact pressure, q, because a and q are uniquely related by Hertz theory to the shape of the contact area defined by h It is also argued (Ref 1) that a coincides with the instantaneous radius of the wear crater

Let the mathematical expression of the wear mechanism be written in the following form for the volume of wear:

The differential form of the above, in terms of both the number of cycles and the instantaneous peak pressure:

Solution Methods for Measurable Wear

Example 1: Soft Ball Worn by Hard Plane

The ball wears from the original radius R1 to a radius R > R1 With reference to Fig 10, the wear is the difference between two paraboloids

with

(Eq 28)

where the wear scar radius a was equated with the edge of the peak contact radius at the stage, N, of wear history For the

depth of wear from this geometry:

(Eq 29)

Fig 10 Percussive wear schemes (a) Wear of soft ball versus hard plane (b) Wear of soft plane versus hard

In order to express the initial condition at the zero-wear limit, Eq 40 can be solved for = 0 Figure 11 facilitates the

conversion from dimensional to nondimensional variables, h to , and vice versa Note that Fig 11 also includes cylindrical geometries, for which the wear equations can be found in Ref 1

Fig 11 (a)h/C variation versus and for spherical and cylindrical contacts (b) Hard sphere or cylinder

wearing down soft plane (c) Soft sphere or cylinder wearing against hard plane

Example 2: Soft Plane Worn by Hard Ball

Figure 10(b) shows the geometry of the deepening wear scar of the plane, attending a continuously sharpening negative

radius, R The nondimensional curvature is taken as

(Eq 42)

All of the mathematical treatment given above for the wearing ball will apply if = + 1 is inserted

Computational Procedures. The differential equations of dimensional percussive wear (for example, Eq 38 and 39) for the cases treated here, and for several others (such as two curved bodies, cylindrical impact), admit the "master curve" solution technique (Ref 1) (Fig 12) A master curve (Fig 13) is first constructed for a fixed initial condition, such as 0

= 1, = 0.99 (or = -0.01, corresponding to - 1)

Fig 12 Master curve shift procedure

Fig 13 Solution for impact wear in terms of nondimensional curvature parameters with initial conditions of N0

= 1, 0 = -0.01 or 0 = 0.99

In order to obtain the desired solution belonging to the given set of initial conditions (N0, 0), we intercept the master curve at 0, obtaining the abscissa N' Now, the master curve can be shifted from N' to N0, and on this shifted portion of

the master curve, the solution for any N desired can be found

Besides the master curve method, an exact solution of Eq 39 can also be obtained for N:

(Eq 43)

where for a constant-impulse process, T = (8 + 9c)/5, U = (3 + 9c)/5

Note that the compound nature of the impact (sliding component) is included in the zero-wear limit, and no further amendment of the computational apparatus is required Several geometries that are more complex, and the case of two-body impact wear, treated in Ref 1 The calculation of hardness for a composite multilayered surface is treated in Ref 26

Plotting a Wear Curve

By way of example, consider the wear of an originally plane, massive carbon steel machine platen, subjected to repetitive

compound impact by a hard, nonwearing spherical-ended (R = 140 mm, or 5.5 in.) steel alloy (62 HRC) component of m

= 1.2 g (0.04 oz) mass, at an approach speed of V = 2.0 m/s (78 in./s) and sliding speed of v = 1 m/s (40 in./s) The

wearing surface has a roughness of /2 = 0.25 m, and hardness of 20 HRC, with a yield stress, y, of 500 MPa (75 ksi)

The modulus of elasticity, E, of steel is 207 GPa (30 × 106 psi), and its Poisson's ratio, v, is 0.3 The friction coefficient,

, for the unlubricated surfaces is 0.6

Calculating the wear curve for nonlinear wear, the zero-wear limit is first established The Hertz impact yields

P0 = 929.4N

t* = 9.494 s K1 = 245.9 MPa The slip factor is

for a moderate range

The surface contribution factor is, from Fig 5(c), = 320, so that for the zero-wear limit, using y = H/3 and q = K1:

and

h0 = 0.25 m

Solving Eq 41 for C, 3.345 m (130 in.) is obtained, so that from Fig 11, for h0/C = 0.0747, the values obtained are 0

= 0.93 or 0 = -0.07

For the measurable wear range, the wear coefficient c = 3 will be used Because the wear of a "soft plane" is being

calculated, the nondimensional geometric wear parameter, = + 1 will be computed Equation 43 is now written:

to be solved for = 0.9, 0.8, 0.7, 0.6, and so on, delineating the wear curve, Fig 14 Conversion from nondimensional to

dimensional wear depth, h, is made by use of the wear severity factor C, once again

Fig 14 Impact wear of platen by hard projectile; analytical prediction based on engineering data of example

References

1 P.A Engel, Impact Wear of Materials, Elsevier, 1976

2 S.B Ratner and E.E Styller, Characteristics of Impact Friction and Wear of Polymeric Materials, Wear,

5 A.W.J DeGee, C.P.L Commissaris, and J.H Zaat, The Wear of Sintered Aluminum Powder (SAP) under

Conditions of Vibrational Contact, Wear, Vol 7, 1964, p 535-559

6 P.L Ko, The Significance of Shear and Normal Force Components on Tube Wear due to Fretting and

Periodic Impacting, Wear, Vol 106, 1985, p 261-281

7 K Wellinger and H Breckel, Kenngroessen und Verschleiss beim Stoss Metallischer Werkstoffe, Wear,

10 R.G Bayer, P.A Engel, and J.L Sirico, Impact Wear Testing Machine, Wear, Vol 19, 1972, p 343-354

11 P.A Engel and J.L Sirico, Impact Wear Study of Lubricated Contacts, J Lubr (ASLE Trans.), Vol 18,

1975, p 279-289

12 S.L Rice, Reciprocating Impact Wear Testing Apparatus, Wear, Vol 45, 1977, p 85-95

13 S.L Rice, The Role of Microstructure in the Impact Wear of Two Aluminum Alloys, Wear, Vol 54, 1979, p

291-301

14 S.L Rice, H Novotny, and S.F Wayne, Characteristics of Metallic Subsurface Zones in Sliding and Impact

Wear, Wear, Vol 74, 1981-1982, p 131-142

15 T Sugita, K Suzuki, and Y Nakata, Impact Wear of MgO Single Crystals, Wear, Vol 58, 1980, p 283-299

16 G Levy and J Morri, Impact Fretting Wear in CO Based Environments, Wear, Vol 106, 1985, p 97-138

17 I.R Sare, Repeated Impact-Abrasion of Ore-Crushing Hammers, Wear, Vol 87, 1983, p 207-225

18 R.G Bayer, Impact Wear of Elastomers, Wear, Vol 112, 1986, p 105-120

19 P.A Engel and D.L Millis, Surface Topography Study in Impact Wear, Wear, Vol 75, 1982, p 423-442

20 V Veronesi, Wear Phenomena in Impact Printers A Scanning Electron Microscopy Study, Wear, Vol 55,

1979, p 265-276

21 D Tabor, The Hardness of Metals, Oxford University Press, 1951

22 P.A Engel, J.L Sirico, and T.H Lyons, Impact Wear Theory for Steel Specimens, Wear, Vol 23, 1973, p

185-201

23 D.D Roshon, Testing Machine for Evaluating Wear by Paper, Wear, Vol 30, 1974, p 93-103

24 T.C Ku, R.G Bayer, et al., Handbook of Analytical Design for Wear, Plenum, 1964

25 P.A Engel, Predicting Impact Wear, Mach Des., Vol 49 (No 12), 1977, p 100-105

26 P.A Engel, A.R Chitsaz, and E.Y Hsue, Interpretation of Superficial Hardness for Multilayer, Thin Solid

articles can be found in this Volume The article “Forms of Mechanically Assisted Degradation” in Corrosion:

Fundamentals, Testing, and Protection, Volume 13A of ASM Handbook, is also relevant

Occurrences in Practice

Corrosion and wear damage to materials, both directly and indirectly, costs the nation hundreds of billions of dollars annually Although corrosion can often occur in the absence of mechanical wear, the converse is rarely true Corrosion accompanies the wear process to some extent in all environments, except in a vacuum and inert atmospheres Corrosion and wear often combine to cause aggressive damage in a number of industries, such as mining, mineral processing, chemical processing, and energy production Corrosion and wear processes involve many mechanisms, the combined actions of which lead to the mutual reinforcement of their effectiveness Seventeen synergistic relationships between abrasion, impact, and corrosion that could significantly increase wear damage in wet and aqueous environments have been identified (Ref 1)

Slurry Handling. The movement of materials as a slurry is an efficient means of transportation that is used in many industries However, the movement of these slurries can cause significant corrosive wear Pumps, elbows, tee junctions, valves, flotation cells, and hydrocyclones are parts of slurry handling systems that are subject to corrosive wear The periodic replacement of worn parts often results in costly process downtime Pipe rotation has been considered good practice for those pipes that are subject to localized wear, but careful monitoring is necessary to prevent failure of the worn parts because of corrosion (Ref 2) Turbulence and eddies are promoted either where fittings or valves project into the mainstream flow or when the diameter of a pipe changes The momentum of the particles is sufficient to cause

abrasive damage to the side walls of the pipe in its curves or elbows, where the change in direction does not produce eddies Slurry particles can impart damage to the bottom of the pipe by tumbling or sliding in straight pipe sections

Sliding wear can occur, for example, in an ore chute in a minerals processing plant, where moist ore particles slide on

an inclined platform, chute, or screen Figure 1 shows a worn wire cloth screen in a minerals processing plant Another example is a submerged valve that opens and closes to control the flow of material Here, the parts are subjected to metal rubbing, as well as to the abrasive and corrosive nature of a slurry Other, more common corrosive wear problems occur

in equipment such as automobile brakes an engines (Ref 3), fillings used in dental applications, and tapes rubbed by recording heads in electronic video and audio equipment Each of these wear examples has a corrosive component that results from exposure to reactive environments

Fig 1 Worn wire screen in minerals processing plant showing effects of sliding wear in moist environment

Crushing and Grinding. Corrosive wear is prevalent in the grinding of mineral ores The annual cost of grinding media frequently approaches or exceeds the cost of energy in some processes (Ref 4) Figure 2 shows ball mill liner plates

in both new and worn conditions Figure 3 shows a cone crusher liner that wore completely through after only three months of use

Fig 2 Ball mill liner plates (a) New condition, before installation (b) Worn condition, with partially worn

grinding belt

Fig 3 Worn cone crusher liner after 3 months of use

Corrosion and wear damage arising from comminution is equivalent to 12% of more than 30 billion kW · h of energy consumed annually by U.S industry for crushing and grinding purposes, whereas contributions from minerals account for 50%; cement, 25%; coal, 13%; and agricultural products, 12% (Ref 5) The development of more efficient mining, transportation, and processing systems has allowed exploitation of many large, low-grade ore bodies, which involves the crushing, grinding, and treatment of huge volumes of abrasive material to satisfy modern demand for materials (Ref 6) This trend will increase the future consumption of grinding media in an industry where over 0.9 Mg (1 × 109 tons) of ore are crushed annually (Ref 5)

Wear debris and corrosion products that are formed during comminution affect product quality and can adversely affect subsequent beneficiation by altering the chemical and electrochemical properties of the mineral system (Ref 4, 7, 8, 9) Electrochemical interactions between minerals and grinding media can occur, causing galvanic coupling that leads to increased corrosive wear Corrosion and wear are also important in nonmining industries, such as the pulp and paper industry, where physical processing steps such as the grinding of wood chips are increasingly being substituted for chemical processing steps in order to reduce water-treatment requirements and fulfill regulatory standards

High-Temperature Processes. Many chemical processes take place at elevated temperatures and involve corrosive wear A process such as coal gasification involves hot gases with entrained solid particles that impinge on the containment vessel surface Several studies have been conducted to describe the wear-corrosion synergism that takes place during the oxide formation and subsequent removal by abrasive particles (Ref 10, 11, 12)

Power-generation plants use processes that occur without any particles present and involve only the transport of solutions or steam In these cases, liquid forces on the solid surface, which are due to turbulence or droplet impacts, mechanically remove protective layers of corrosion products, thus exposing more base material to corrosive action An example of this type of process is the erosion-corrosion of steam turbines Other systems may involve the dissolution of the protective oxide layer, because of a continual flow of liquid past the surface Cavitation is also a problem where the high-speed flow of liquid is present This subject is discussed in the article "Cavitation Erosion" in this Volume

Effect of Environmental Factors on Corrosive Wear

Environmental factors affect corrosive wear in materials handling systems Some of the more important factors that influence the material losses in slurry transport and grinding operations in aqueous media and ambient conditions are discussed below

Slurry Particle Impingement on Two-Body Corrosive Wear

Slurry particles that strike the target material under the influence of their own momentum and that of the carrier fluid impart two-body wear damage to component parts They are not constrained by another solid, as distinguished from grinding, which is discussed later

Dependence on abrasive particle shape, density, size, and hardness, as related to wear rate, is discussed below

Particle Shape. Until recently, no investigation had directly reported the effect of particle shape on wear in slurry handling systems However, it was commonly agreed that the angularity of the slurry abrasives was important in determining the wear rate of materials where corrosion was not dominant When slurries are recycled in a closed slurry

pot system, the wear rate can decrease dramatically with time (Ref 13, 14, 15) This is due to the microscopic rounding of the abrasive particles The difference in wear rates for recycled and flow-through slurries is shown in Fig 4 for type A514 steel in a water/silica sand slurry containing 50 × 70 mesh particles

Fig 4 Comparison of slurry wear for the flow-through and recycled slurry systems using a low-alloy steel in a

2% silica sand slurry Source: Ref 13

Figure 5 shows the worn surfaces of the steel that resulted from the two tests The specimen exposed for 1 h in the through test (top photograph) had many fine grooves over its entire surface, typical of cutting wear The bottom photograph shows the smooth, wave-type pattern that developed after 1.67 h with a recycled slurry, indicating a deformation wear mechanism

flow-Fig 5 Comparison of wear surfaces for low-alloy steel specimens worn in flow-through (top) and recycled

(bottom) slurry tests for 1 h and 1.67 h, respectively Source: Ref 13

The angularity of a particle depends on the radius of each protruding point on its surface, relative to its average radius Bitter (Ref 16) has shown that the more angular a particle, the higher is its apparent density (Fig 6) This effect results in

a lower maximum velocity at which impact between the particles is purely elastic If the predominant mode of material loss is corrosion, the angularity of the particles may not be such an important consideration

Fig 6 Relationship between angularity and apparent density of a particle Apparent density is (R/r)3 × real reactive density Source: Ref 16

Particle density is an important parameter that influences the corrosive wear caused by slurries For a particle of a given angularity, the denser the particle, the more likely it is to cause either deformation or cutting wear on impact This

is a consequence of the dissipation of more energy in the same volume

Particle Size. Because of their greater kinetic energy, particles of the same density and angularity in slurries with the same nominal velocity impart greater wear losses to materials as their size is increased A recent study (Ref 17) has shown that larger particles have a higher collision efficiency and a higher impact velocity than smaller particles These factors contribute to increased wear rates as slurry particle size increases

Particle Hardness. The wear of materials has been shown to dramatically increase when the slurry particles are harder than the material being worn However, increases in the hardness of the abrasive, above a critical hardness value, do not appreciably increase the wear rate Likewise, when extremely hard abrasive particles are present in the slurry, moderate increases in the hardness of the worn material do little to retard the wear rate

Dependence on slurry velocity, angle of attack, and solids concentration is discussed below in terms of wear rate

Velocity of the slurry is the single most important factor that controls the rate of wear The wear rate is an exponential function of velocity; exponents are reported to range from 1.6 to 4.8, but are generally between 2 and 3 (Ref 18) The velocity of the slurry not only affects the rate of mechanical damage of a material, but the corrosion rate, as well Above a threshold velocity, corrosion products can effectively be stripped from an alloy, thus making available a new surface that

is susceptible to corrosive attack

The angle of attack of slurry particles on a material determines, to a large extent, the role that the abrasives will play

in corrosive wear In straight pipelines where the flow of slurry particles is basically parallel to the walls of the pipe, little damage to the underlying metal results from direct contact between the abrasive and alloy (Ref 19) The abrasives, along with the solution, serve to slowly remove the corrosion products and enhance the corrosion rate The extent of this removal process is controlled by the particle size, angularity, and slurry density

On the other hand, when a curvature in the pipe is present, the slurry particles tend to wear the pipe as depicted in Fig 7

In this condition, the slurry particles are more likely to not only remove any corrosion products, but to directly impart mechanical wear damage to the alloy If the alloy is one that normally passivates (such as a stainless steel), the abrasive

can continually remove the protective film and allow corrosion to proceed at a much higher rate than if no abrasive particles were present The removal of the base metal depends on the type of material being worn

Fig 7 Changing pattern of wear and, hence, changing angle of impingement, at bend in pipe Source: Ref 16

Generally, ductile alloys wear greatest at shallow angles of attack, from 10 to 30°, whereas brittle materials wear greatest

at 90° The reason for these differences is the type of wear that occurs Cutting wear is dominant for low angles of attack with ductile material, whereas deformation wear is greatest at larger angles of attack

Solids Concentration. Increasing the solids concentration in a slurry generally increases the wear rate The increase is only proportional to the solids concentration for dilute slurries For denser slurries, particle-particle interaction tends to decrease the dependence of the wear rate on slurry density (Ref 14, 20)

Dependence on Hydrodynamics. Recent work (Ref 21, 22) on the trajectories and impact velocities of particles during slurry erosion has shown that not all of the particles directed at a target will impact it The fraction of slurry particles that impact the surface of the target material is controlled by fluid velocity and viscosity, as well as the size, shape, and density of the slurry particles

Flow-dependent corrosive wear typically occurs at geometrical irregularities, such as fittings, valves, and weld beads, where the flow separates from the wall of the containment vessel Flow separation and reattachment produces high turbulence intensity and particle-wall interactions that lead to high corrosive wear rates (Ref 23, 24, 25) Figure 8 shows streamlines in a pipe for flow contraction and flow expansion

Fig 8 Streamlines for water in pipe of variable diameter Source: Ref 24

Blatt et al (Ref 23) and Nesic et al (Ref 24) have correlated material losses with the local, near-wall intensity of

turbulence in single-phase (liquid only) flows It was hypothesized that intensive near-wall turbulence disrupted the protective corrosion product and disturbed the mass transfer layer, thus enhancing oxygen transport to the metal surface and corrosion rates

Corrosion products and the mass transfer of oxygen play an important role in the total material losses

encountered in straight sections of a carbon steel pipeline Work done by Postlethwaite et al (Ref 19 and 26) shows that

the material loss rate is under oxygen mass transfer control, with corrosion being the dominant mode of metal loss The magnitude of the erosion-corrosion can be estimated using well-established mass transfer correlations for oxygen diffusion

In aqueous solutions of near-neutral pH, mechanical wear prevents the formation of a rust film that completely covers the interior of a pipe By providing a barrier to oxygen, this rust film in carbon steel pipes is responsible for corrosion rates of

<1 mm/year (<0.04 in./year) When abrasive solids are present, coverage of the metal by the rust film is incomplete, and the islands of bare metal that are present can act as efficient cathodes and result in substantially increased rates, such as

>10 mm/year (>0.4 in./year)

Grinding Wear: Impact and Three-Body Abrasive-Corrosive Wear

The wear of grinding media and crushers in mineral processing systems is caused by the combination of abrasion, corrosion, and high-energy impact of ore and metal components The increased size of modern crushing and grinding equipment greatly increases the kinetic energy at metal/ore interfaces, which results in high wear rates The crushed ore particles participate in three-body abrasion by virtue of being confined between two solid surfaces

Dependence on Force. The strong effect of impact severity on the grinding media wear rate was reported by Dunn (Ref 1) for dynamic loading conditions He showed that the wear rate was four times greater in an 8.5 m (28 ft) semiautogenous grinding (SAG) mill (8.30 kgf, or 18.3 lbf, balls) than that attained in a 2.9 m (9.5 ft) ball mill (1.81 kgf,

or 4 lbf, balls) The ball velocities in the two mills were calculated to be 9.14 m/s (30 ft/s) and 5.33 m/s (17.5 ft/s),

respectively However, under conditions of a static load, Kotlyar et al (Ref 27) found that the total wear rate of a metal

alloy was approximately linear with load They used a rotating metal alloy specimen between two ceramic anvils submerged in an abrasive slurry

Dependence on Abrasive Type. Abrasive wear rates for steels and irons generally increase with an increase in mineral hardness When the metal-to-mineral hardness ratio is greater than 0.6, marked improvement in the ability to resist abrasive wear is shown (Ref 28) In order to achieve this favorable hardness ratio, hard (Fe,Cr)3C carbides found in low-alloy nickel-chromium irons and harder (Fe,Cr)7C3 carbides found in high-chromium white cast irons are used

Dependence on Galvanic Interaction between Minerals and Metal Alloys. Electrochemical interactions between ore materials and grinding media can occur Mineral particles with potentials that are nobler than those of steel grinding media become galvanically coupled and result in accelerated corrosive wear (Ref 6) In recent studies (Ref 29, 30), it has been reported that the relative contributions from corrosive and abrasive wear are highly variable from one mineral-metal system to another, and are largely dependent on the mineral slurry characteristics, as well as the properties

of the grinding media In the presence of surface abrasion, and corrosion resistance, hardness of the grinding media, pH and conductivity of the pulp, and oxygen content influence the magnitude of galvanic currents between minerals and

grinding media Jang et al (Ref 31) also found that galvanic currents exist between ferritic and martensitic phases in a

microstructure of steel grinding media, as well as between sulfide minerals and these two phases

Role of Localized Corrosion in Grinding. In the study of high-carbon low-alloy (HCLA) steel grinding media, Kotlyar and Wadsworth (Ref 27) found that localized electrochemical cells existed between strained and unstrained regions of the alloy This resulted in preferential anodic dissolution in microgrooves, which promoted pitting The overall corrosive wear rates of the steels were found to be strongly dependent on the localized corrosion

Experimental Measurement of Corrosion-Wear Synergism

Quantitative measurements of corrosion, mechanical wear, and wear-corrosion synergism have been made using various

experiments The combined effects of wear, W, and corrosion, C, often result in total damage rates that are much greater

than the additive effects of each process taken alone, indicating a strong synergism between wear and corrosion The

synergistic component, S(C,W) is that part of the total damage that results from the interaction of corrosion and wear

processes The total material loss, T(C,W), is related to the synergism by

usually higher than the corrosion rate C0 because of mechanical wear interaction, and is designated as Cw Using this notation, an equation similar to Eq 1 can be written as:

T(C,W) = W0 + Cw + S'(C,W) (Eq 2)

where S'(C,W) is the increase in the mechanical wear losses when corrosion is present Changes in the corrosion rate that

are due to wear-corrosion synergism are included in Cw, which is the sum of all electrochemical corrosion processes

occurring on the material during wear The term W0 can be obtained either by conducting the experiment under cathodic protection conditions or by using corrosion inhibitors

The value Cw can be measured by performing polarization measurements with the specimen of interest as the working electrode The electrochemical corrosion rate can be determined by making measurements of the current density within

±20 mV of the open-circuit corrosion potential The current density versus polarization potential curves were fitted to the following theoretical equation that characterizes a reaction which contains only two processes under activation energy control:

(Eq 3)

where I is the measured current density, Ic is the corrosion current density, is the polarization potential, and a and c

are the anodic and cathodic Tafel slopes, respectively

Equation 3 is suitable for describing the behavior of many alloys in aqueous solutions A nonlinear regression technique

can be used to determine Ic, a, and c, and values for the corrosion current can be converted to units of volume loss per

area per hour to give Cw in Eq 2

Another method used to obtain the corrosion rate is the polarization resistance method, in which the slope of the potential

versus the current density (that is, the polarization resistance, Rp) is measured at potentials that are very close (±5 mV) to

the open-circuit potential The corrosion current, i, is calculated according to:

(Eq 4)

The value Ic can then be calculated by dividing ic by the electrode area Cw (in units of material volume loss per area per

unit time) is obtained after multiplying Ic by the appropriate conversion factors In this case, a and c can either be

measured graphically from Tafel plots of the potential versus log i or estimated from tabulated values for similar alloys

Examples that show how these measurements have been made in slurry particle impingement systems and experiments that simulate ore-grinding environments are described below Mechanisms for corrosive wear have been proposed, based

on the cited studies

Slurry Particle Impingement Systems

Various types of slurry wear tests designed to measure corrosive wear are reported in the literature and include slurry pot, jet impingement, and pipeline configurations The experiments described in this section involve the wear condition known

as low-stress, two-body abrasive wear Madsen (Ref 13, 14, 32, 33) utilized a slurry pot that was capable of using either recycled or flow-through slurry Pitt and Chang (Ref 34, 35) conducted experiments with a jet-impingement apparatus,

and Postlethwaite et al (Ref 26) used electrodes in a closed-loop pipeline to measure both wear and corrosion rates

Slurry Pot Experiments. A slurry wear test developed by the Bureau of Mines (Ref 32, 33) was used to measure both wear and corrosion rates during slurry wear in order to establish the relationship between the wear and corrosion of metal-alloy specimens Figure 9 depicts the slurry wear apparatus

Fig 9 Flow-through slurry wear apparatus Source: Ref 13

Three types of tests were conducted to determine the terms in Eq 2 The total material loss, T(C,W), was determined from weight losses and densities during an experiment conducted at open circuit The wear rate that was due to mechanical wear only was determined while cathodically protecting the specimens to eliminate corrosion