Tribology of Elastomers

LICENTIATE T H E S I S

Luleå University of TechnologyDepartment of Applied Physics and Mechanical Engineering

Division of Machine Elements

2007:62|: 02-757|: -c -- 07⁄62 --

2007:62

Tribology of Elastomers

Mohammadreza Mofidi

i

Tribology of Elastomers

Mohammadreza Mofidi

Luleå University of Technology

Department of Applied Physics and Mechanical Engineering

Division of Machine Elements

ii

Cover figure: The figure shows the smoke plume after the break up of the Space Shuttle Challenger. The Space Shuttle Challenger disaster occurred in the United States, over the Atlantic Ocean on January 28, 1986. The shuttle exploded due to the failure of an O-ring seal in its solid rocket booster (SRB).

From: http://news.bbc.co.uk/2/shared/spl/hi/pop_ups/06/sci_nat_1986_challenger_disaster/html/1.stm, accessed at: 2007-11-02

http://news.bbc.co.uk/2/shared/spl/hi/pop_ups/06/sci_nat_1986_challenger_disaster/html

iii

PrefaceThe work presented in this thesis has been carried out at the Division of Machine Elements, Department of Applied Physics and Mechanical Engineering at Luleå University of Technology (LTU) in Luleå, Sweden.

I would like to express my deep gratitude to my supervisors, Professor Braham Prakash and Dr. Elisabet Kassfeldt, for their wholehearted support and guidance throughout this work. I have learnt a lot from the various courses I attended at this university and would like to thank all my teachers, especially Prof. Braham Prakash and Prof. Roland Larsson.

I wish to thank the “Ministry of Science, Research and Technology of IRAN” for awarding me the scholarship to pursue research at Luleå University of Technology. This work would not have been possible without this support and I am really grateful to the Government of Iran for this.

I am also extremely grateful to Dr. Richard Schaake, Mr. Joop Vree and Professor Piet Lugt, (all from the SKF Engineering and Research Centre) and Dr. Marika Torbacke (from Statoil Lubricants) for their intellectual as well as material inputs to this work.

All my colleagues at the Division of Machine Elements, especially Jens Hardell and Donald McCarty have been very helpful whenever I had any difficulty and I sincerely acknowledged their support.

I would like to thank my friends at Luleå University of Technology and their families, Pourghahramani, Barabady, Ahmadi, Keramati, Toufighi, Akhavan, Ghodrati, Khatibi and Payman Roonasi and I shall always remember their supports and kindness.

A special thanks to my wife, Sedigheh and my son Aref, for their support and patience. My sincere gratitude goes to my mother-in-law for her affection and kindness who suddenly passed away last year. I would like to extend my appreciation to all my brothers and sisters for their profound kindness.

Finally, I am deeply indebted to my parents, Parviz and Soghra, and feel a tremendous sense of appreciation for their genuine support, care, encouragement, patience and eternal dedication.

Luleå, December 2007

Mohammadreza Mofidi

iv

v

AbstractElastomers are the most commonly used materials for various sealing applications owing to their low modulus of elasticity, large elongation-to-break, and high Poisson’s ratio. Most seals operate in the presence of lubricants; therefore, the sealing elastomer-oil interaction plays an important role in determining the tribological performance of elastomers. Furthermore, at times, such as starting up periods, the seals may also operate under dry condition and the seal material can be affected by high friction coefficient and wear.

In this work, the tribological behaviour of different elastomers has been studied. The influence of aging of a sealing elastomer in different lubricants on its tribological behaviour has been investigated. Further studies pertaining to the influence of lubrication on the abrasive wear of a sealing elastomer have also been carried out.

The results show that aging of the nitrile rubber in ester base fluids leads to reduction of friction coefficient. Dry abrasive wear of the aged rubber in ester base fluids and rapeseed oil are higher than that in the mineral oils.

The abrasive wear of elastomers may increase or decrease in presence of lubricants depending upon whether tearing or smearing is the dominant wear mechanism. Presence of the lubricant reduces the frictional forces resulting in a decrease in local mechanical rapture but it accelerates the decomposition of the molecular network to a low molecular weight. The overall effect of the lubricant on the abrasive wear depends on the elastomer-lubricant compatibility, abrasive coarseness, geometry of sliding contact area and contact pressure.

In unidirectional dry sliding of an elastomer against a counterface, the friction coefficient decreases during the running-in period. The longest running-in periods have been observed when the elastomers slide against relatively smooth surfaces. The running-in time of HNBR is significantly longer than those for other materials. The surface roughness has the minimum and maximum effects on the steady state friction coefficient of ACM and HNBR respectively. The results of the tests conducted at high contact pressure show that roll formation is the dominant wear mechanism in tribological pairs involving FKM and HNBR and results in high wear of these elastomers. The wear particles of the ACM were powdery in nature whereas those of FKM and HNBR were of roll shape.

vi

vii

Appended papers:

Paper A: M. Mofidi, E. Kassfeldt, B. Prakash, Tribological behaviour of an elastomer aged in different oils, Tribology International, accepted for publication

Paper B: M. Mofidi, B. Prakash, Two body abrasive wear and frictional characteristics of NBR elastomer under lubricated sliding conditions

Submitted for publication

Paper C: M. Mofidi, B. Prakash, Influence of counterface topography on sliding friction of some elastomers under dry sliding conditions

Submitted for publication

viii

ix

Contents Preface ................................................................................................................................. iiiAbstract ................................................................................................................................. v1. Introduction ...................................................................................................................... 1

1.1 Elastomeric seals .................................................................................................................... 1

1.2 Elastomers .............................................................................................................................. 2

1.3 Oil and heat resistance of elastomer..................................................................................... 2

1.4 Friction.................................................................................................................................... 3

1.5 Wear........................................................................................................................................ 5

1.6 Objectives and outline of the research ................................................................................. 7

1.7 Limitations.............................................................................................................................. 8

2 Lubricated sliding friction................................................................................................. 92.1. Experimental method ........................................................................................................... 9

2.2 Test materials ....................................................................................................................... 10

2.3 Results and discussion ......................................................................................................... 112.3.1 Lubricated sliding.......................................................................................................................... 112.3.2 Influence of aging the elastomer in different oils.......................................................................... 11

3 Two body abrasive wear in reciprocating sliding ........................................................... 133.1 Experimental method .......................................................................................................... 13

3.2 Test materials ....................................................................................................................... 13

3.3 Results and discussion ......................................................................................................... 133.3.1 Abrasive wear and the effect of lubrication .................................................................................. 143.3.2 Influence of aging the elastomer in different lubricating fluids .................................................... 14

4 Abrasive wear in unidirectional sliding.......................................................................... 174.1 Experimental method .......................................................................................................... 17

4.2 Test material......................................................................................................................... 18

4.3 Results and discussion ......................................................................................................... 18

5 Dry sliding friction .......................................................................................................... 235.1 Experimental method .......................................................................................................... 23

5.2 Test materials ....................................................................................................................... 24

5.3 Results and discussion ......................................................................................................... 24

6 Conclusions...................................................................................................................... 297 Future work ..................................................................................................................... 31References:.......................................................................................................................... 33

x

1

1. Introduction Seal is a component which prevents the leakage of fluid or gas from machine and prevents contamination entering the machine. Seals can be divided into static and dynamic seals. Static seals perform the sealing function between surfaces which do not have any motion relative to each other. Dynamic seals perform the sealing function between the surfaces in relative motion. Dynamic seals can be subdivided into rotary and reciprocating seals [1]. Tribological aspects are significant in dynamic seals owing to their sliding against sealing surfaces.

1.1 Elastomeric seals Depending on the application, metal, plastomer, elastomer and composite material can be used as a seal material, however elastomer is the most popular seal material in general application. The advantages of elastomer as a seal material are as follows:

Elastomers are reasonably inexpensive; even expensive special elastomer seals can give a low cost for total seal system in comparison with the seals which are designed and produced of other materials;

They have low module of elasticity and high elongation-to-break; they can be deflected largely to follow irregularities and vibration of the sealed surface without giving high contact stresses;

They have a high Poisson's ratio (close to 0.5, therefore the material behaves in a manner similar to a liquid under pressure, transferring any applied pressure hydrostatically), enabling an elastomeric seal to create its own sealing force automatically in proportion to the pressure;

They have low shear modulus G which in combination of its high poisson’s ratio enables them to change shape easily without change of volume; thus the elastomeric seal can conform to the shape of its housing.

Elastomeric seals also have certain disadvantages:

They change from rubber to glass as the temperature falls bellow a critical temperature, the glass transition temperature Tg (e.g. Challenger catastrophe);

They can have friction characteristics which are not always predictable;

Chemical and temperature resistance of elastomers is poor compared with many other engineering materials; and

Elastomer under pressure readily extrudes into even very smal clearances, owing to the high Poisson's ratio and low elastic modulus.

Seals can fail through different mechanisms resulting in leakage contamination entering the lubricant. The most important mechanisms of seal failure are abrasion, thermal degradation, chemical degradation, compression set, plasma degradation, over compression, extrusion, extraction etc. Friction of seal sliding against sealing surface can affect the overall efficiency of the machine. Furthermore, high friction results in an

2

increase in temperature and accelerates the failure of seal through different mechanisms such as thermal degradation, chemical degradation, abrasion and so on. Most seals operate in lubricated conditions; however they may occasionally operate in dry conditions such as the running-in periods. Thus, understanding the wear mechanisms of elastomeric seals and frictional behaviour of sealing elastomers, particularly in presence of lubricants, is important in determining their performance and service life. Tribological behaviour of elastomers has been investigated extensively, however most of the previous works were pertaining to the dry condition and the tribological behaviour of elastomer in lubricated condition has investigated scantily.

1.2 Elastomers Elastomers are a class of polymeric materials that possess the quality of elasticity, i.e., the ability to regain shape after deformation. Elastomer comes from two terms, “elasto” which indicates the ability of material to return to its original shape and “mer” which comes from polymer. Polymer is a substance composed of molecules with large molecular mass composed of repeating structural units, or monomers, connected by covalent chemical bonds. Elastomer refers to all the polymeric materials with high elasticity including crosslinked rubber. However, a distinction is made between raw rubber and crosslinked rubber. The former is completely deformable in a plastic-like manner, particularly at high temperatures, because it does not have a rigid network structure. In contrast, the crosslinked rubber does not have a plastic transition zone due to their three dimensional networks, which restrain the movement macro-molecular chain molecules [2].

1.3 Oil and heat resistance of elastomer Oil and heat resistance of elastomers has an important role in sealing applications. When an elastomer and a base fluid are brought in contact with each other, the elastomer material may absorb the base fluid or the base fluid may extract soluble constituents of the elastomer. The base fluid may also react with the elastomer [3]. Presence of the polar side-groups in the backbone chain of elastomer increases the oil resistance of the polymer [4]. Crosslinking also limits the degree of polymer swelling by providing tie points (constrains) that limit the amount of solvent that can be absorbed into the polymer [4].

Elastomers may show progressive change in their physical properties due to exposure to heat. Three types of changes have been observed: additional crosslinking resulting in higher crosslink density and an increase in hardness, chain scission leading to reduction in chain length and average molecular weight and consequently softening of the elastomer, chemical alternation of the polymer chain by formation of polar or other groups [5]. Figure 1 shows the oil and heat resistance of different elastomers [4].

3

Figure 1: Oil resistance (%swell in #3 oil) [4]

1.4 Friction The coefficient of friction of a rubber surface against a hard surface can be expressed in terms of the contribution of adhesion, deformation (hysteresis), viscous and cohesion (tearing) [6, 7]. Adhesion (Figure 2) is generally recognized to consist in the making and breaking of junctions at a molecular level [8]. Hysteretic friction (Figure 2) is a consequence of energy loss associated with internal damping within the viscoelastic body [9]. The cohesive component of friction is the contribution of wear to the bulk losses and the viscous component is the viscous drag under wet condition [7]. Most texts have considered only two terms for friction components since the deformation component can represent both the hysteresis and tearing component whereas the viscous component of friction can be a subset of adhesion component [6]. Recent studies show that the independency of the adhesion and deformation components of friction is only a simplified assumption. It has been assumed that the adhesive force per unit area should be constant during any deformation while the surface free energy is a function of both internal energy and entropy, and so it should change if the internal energy and/or entropy change due to any bulk deformation [10].

Figure 2: Adhesion and hysteresis [6]

P

FHYST.FADH.

4

Contribution of adhesion and hysteretic friction depends on the temperature, sliding velocity, geometry and cleanliness of the mating surfaces [11, 12]. The adhesion component is more significant on very clean and smooth rubber surfaces [12, 13, and 14]. It can also be significant at low loads, even in lubricated conditions [15], because of the significance of the attractive Van der Waals’ forces in temporary bonds between the surfaces in comparison to the normal load [16].

The frictional force of rubber sliding at various velocities and temperatures on a given surface can be expressed by a single master curve and the glass temperature of material [17].This transform agrees closely with the William-Landel-Ferry (W.L.F.) transform [18] and thus shows that both friction mechanisms are viscoelastic in nature.

When a soft rubber slides against a hard track, or a hard slider slides against a soft rubber track, the relative motion between the two frictional members is often due to ‘waves of detachment’ crossing the contact area at high speed from front to rear. These waves, which move much faster than the two bodies in sliding [19], are known as ‘Schallamach waves’ named after the researcher. The Schallamch waves appear at a critical sliding speed whose value depends on the adhesive properties of the interface, the geometrical characteristics of the contact, elastic properties of the rubber-like material, load and the temperature [20]. Figure 3 shows the Shallamach waves generated on the surface of a rubber by a hard spherical slider.

Figure 3: Shallamach waves on the surface of a rubber generated by a hard sphere at a sliding speed of 0.43 mm/s (8 frames at 1/32 s intervals) [19]

The lubrication decreases the real contact area between the rubber and hard counterface resulting in a decrease in friction coefficient. Presence of fluid between rubber and hard substrate reduces not only the adhesion but also the hysteretic component of friction. On a lubricated substrate the valleys turn into fluid pools which are sealed off and effectively smooth out the substrate surface (Figure 4). This smoothing reduces the viscoelastic deformation caused by the surface asperities, and thus reduces rubber friction [21, 22].

Sliding direction of hard sphere

5

Figure 4: Smoothing the substrate in presence of lubricant [21]

1.5 Wear Three different mechanisms of wear can be identified when an elastomer slides against a hard substrate. During sliding against a hard countersurface with a sharp texture, abrasive wear takes place as a result of tearing of the sliding surface of the elastomer. Fatigue wear is another mechanism of wear which occurs on the surface of an elastomer sliding against blunt projections on the hard substrates. When a highly elastic elastomer slides against a smooth surface, roll formation occurs. In this type of wear the high frictional force shears a projection on the rubber surface, tears and then rolls the tongue along the direction of sliding [6]. A critical value of shear stress can be defined for each rubber such that if the shear stress is higher than the critical shear stress, roll formation occurs and for shear stresses lower than the critical value, wear is mainly due to fatigue. Thus the friction coefficient is one of the most important properties of rubber governing the type of wear [23]. Figure 5 shows the schematic diagram of the friction and wear mechanisms in elastomers.

Figure 5: Schematic diagram of the friction and wear mechanisms in rubber-like materials [6].

Rubber

Hard counterface

Fluid

6

In practice, a combination of three forms of wear occurs and it is difficult to separate the contribution of each mechanism to the overall wear [6].

When rubber is abraded without a change of direction, sets of parallel ridges are often found on the surface of the samples at right angles to the direction of motion which have been called “abrasion patterns” [24]. The surfaces of elastomers abraded by fatigue wear exhibit pitting marks and the surfaces of harder elastomers, sliding against sharp asperities, exhibit scratches parallel to the direction of sliding [25]. The scratches, parallel to the sliding direction, occur on the surface of elastomers sliding in point contacts with sharp asperities. The elastomer surface is pulled in the direction of sliding and fails in tension behind the contact perpendicular to the tensile stress field [26]; see Figure 6(a). The formation of ridges starts by initiation the cracks at the rear of the contact region, due to the high shearing stress, and continues by growing the cracks under repetitive loading [27]; see Figure 6(b).

Figure 6: Mechanism of scratch and ridge formation on the sliding surface of elastomer; (a) scratch formation [26], (b) ridge formation [27].

Grosch and Schallamach found that on sharp tracks, such as abrasive paper, linear wear rate as a result of tensile failure was proportional to the ratio between frictional energy dissipation and energy density at break [28]. Southern and Thomas studied abrasion of rubber surfaces by a razor blade in line contact and formulated a theory which describes the correlation between the wear rate and frictional force as well as the crack growth characteristics of the rubber. They also mentioned that the pattern spacing depends on the abrading force and test temperature [27]. Zhang and Yang have introduced a theoretical wear equation of rubber abrasion in a line contact from the viewpoint of energy on the basis of experimental results [29, 30].

Another classification introduces the wear of elastomers as a result of two processes; local mechanical rupture (tearing) and decomposition of the molecular network to a low molecular weight (smearing) [31]. The oily decomposition product which forms during smearing protects the underlying rubber from tearing and thus decreases the rate of wear [32]. Experiments show that the rate of wear during smearing decreases by introducing antioxidants [32, 33].

Polymers are soluble in many organic fluids and there can be a synergistic effect between an aggressive solvent and the polymer resulting in significant wear. If the solvent can penetrate the surface of the polymer it will have a detrimental effect on its wear behaviour.

(a)(b)

7

The rapid wear which results is believed to occur by aggravated cracking of the solvent weakened polymer during contact with the counterface [34]. This is schematically illustrated in Figure 7.

Figure 7: Synergism between wear of polymer and damage by a solvent [34].

Muhr et al. have studied the influence of lubrication on the abrasion of rubber by a blade in line contact. They observed that when a lubricant is applied, a much finer pattern develops and the rate of abrasion is much lower but the horizontal force on the blade does not decrease as dramatically [35, 36]. Chandrasekaran and Batchelor have studied the friction and wear of butyl rubber sliding on abrasive paper as a function of temperature and load. They conducted dry and lubricated unidirectional sliding tests and reported that the presence of lubricant reduced the coefficient of friction but accelerated wear due to chemical degradation of rubber [37].

1.6 Objectives and outline of the research The purpose of this reasech is to study and develop knowledge pertaining to the tribological behaviour of sealing elastomers especially in lubricated condition. The specific objectives of this research are to study:

Lubricated sliding friction: The influence of different lubricants and the influence of aging of sealing elastomers in lubricants on its lubricated frictional behaviour.

Lubricated abrasive wear: Understanding the mechanisms of abrasive wear, the effects of lubrication and the influence of aging of sealing elastomers in lubricants on its abrasive wear.

Dry sliding of different sealing elastomers: The tribological behaviour of elastomers sliding against sealing surfaces in dry condition during run-in periods.

The outline of this research has been shown in Figure 8.

Penetration and softening of polymer surface by solvent

Aggravated cracking and wear in softened layer

Solvent

Polymer

Sliding

Counterface

8

Figure 8: Outline of this thesis

1.7 Limitations Although the elastomeric compounds are normally referred to by the name of the base polymer, this does not fully define the material. The details of compounding and processing affect the material properties very significantly, but these details are generally not revealed by the manufacturers.

9

2 Lubricated sliding friction The friction coefficient of an elastomers and the influence of aging in different oils on its frictional behaviour has been studied.

2.1. Experimental method An Optimol SRV machine (Figure 9) has been used to measure the friction coefficient of elastomeric discs against a steel cylinder in lubricated reciprocating sliding conditions. The effect of lubricants and also the influence of aging the elastomer in lubricating fluids have been investigated by using this machine. The machine reciprocates an upper cylindrical specimen loaded against the lower specimen. The sliding of cylinder is along its axis. The friction force is measured by piezoelectric force sensors. Temperature, normal force, frequency of motion and stroke length can be controlled during the tests. The diameter of the cylinder is 15 mm and its length is 22 mm. The edges of the slider are chamfered/rounded off with a view to minimise the edge effect.

Figure 9: Optimol SRV reciprocating friction tester

1. Drive axle 2. Upper specimen holder 3. Load axle 4. Specimen (upper)5. Specimen (lower)6. Temperature sensor 7. Heating8. Lower specimen holder 9. Piezo measuring element 10. Receiving block

The geometry of the edges of slider

Elastomeric sample

Upper specimen (Steel cylinder)

Reciprocating holder

3.3mm

0.4 mm

10

2.2 Test materials The elastomer which has been studied is acrylonitrile butadiene rubber (NBR). The content of acrylonitrile in the tested elastomer is 28%, which is common for oil applications. This elastomeric material is vulcanized by sulphur. The polymeric content is 44% and the remaining part consists of different types of additives. The rubber samples used in these studies were in the form of sheets of 4mm thickness. The initial hardness of the elastomer was 75±5 IRHD (international rubber hardness degrees). The surface was examined in a Wyko 3D optical surface profilometer. The elastomer surface was characterised by parallel grooves (Figure 10) which are caused during moulding of elastomeric sheets in steel mould. The test specimens were discs (ø25 mm and 4 mm thickness), cut out from the elastomeric sheets.

The aging of the elastomer samples was done by immersing the fresh samples in the oils at 120 °C for one week (168 hours).

Figure 10: Surface topography of the fresh elastomer sample (Ra=0.344 m, Rq=0.438 m)

The lubricating fluids used for aging the elastomer were paraffinic, naphthenic, 2 PAOs, 2 VHVIs, monoester, diester, polyolester, complex ester and rapeseed oils. The lubricants and their properties have been shown in Table 1.

Table 1: The properties of lubricants

Base fluid Density (kg/m3) Viscosity@40ºC (cSt) NPI Naphthenic base oil 896 30.0 — PAO2 830 28.5 — PAO1 790 5.5 — VHVI2 830 26.0 — VHVI1 822 12.0 — Paraffinic base oil 870 34.1 — Diester 910 (20°C) 26.1 82 Complex ester 980 46.0 80 Polyol ester 900 35.5 170 Monoester 864 (20 °C) 8.5 102 Rapeseed oil 910 (30 °C) 34.0 190

11

2.3 Results and discussion The experiments have been performed on non-aged elastomer at different loads and where as the tests on aged samples were done at one specified load only.

2.3.1 Lubricated sliding The effects of load, temperature, and orientation of sliding with respect to the grooves on the elastomer sample have been investigated. Figure 11 shows the average value of friction coefficient as a function of load at 25ºC, 40ºC and 80ºC. The results show that as the temperature increases, the friction coefficient decreases. As the temperature increases, the elastomer becomes more elastic (less viscous) and the hysteresis component of friction (which is dominant component in lubricated sliding) decreases. As shown in the figure, at high load where the hydrodynamic effect in film formation is negligible, the surface roughness of elastomer and the orientation of grooves have insignificant effect on the friction coefficient but at low loads, the friction coefficient is affected by the orientation of grooves. When the cylinder reciprocates parallel to the direction of lay on the elastomer surface, the friction coefficient is higher than that in perpendicular sliding. In perpendicular sliding, the oil can be trapped within the grooves which may result in some hydrodynamic effects and subsequently lower the friction coefficient. It confirms the Patir and Cheng’s [38] results for elastomers.

0

0,2

0,4

0,6

0,8

20 50 100 150 300Normal load (N)

Coe

ffici

ent o

f fric

tion

20 50 100 150 300Normal load (N)

20 50 100 150 300Normal load (N)

Figure 11: Friction coefficient as a function of load, Frequency: 50 Hz, Stroke: 1 mm, Temperature: 25ºC, 40ºC, 80ºC

2.3.2 Influence of aging the elastomer in different oils Figures 12 and 13 show the average values of friction coefficients on non-aged and aged elastomer samples in different base fluids respectively. The results show that there is no correlation between the friction coefficient and viscosity. It can be concluded that the lubrication is in boundary or mixed regime. Comparison of friction results in Figures 12 and 13 shows that aging the rubber in synthetic esters reduces friction coefficient.

Investigation of the surfaces of the tested specimens showed that the worn surface areas of the aged specimens are considerably less than those of non aged specimens.

T = 25 °C T = 40 °C T = 80 °C

Perpendicular sliding Parallel sliding

12

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Naphthen

icPAO2

PAO1VHVI2

VHVI1

Paraffi

nic

Rapese

ed oil

Monoester

Diester

Complex es

ter

Polyol este

r

Coe

ficie

nt o

f fric

tion


Figure 12: Coefficients of friction of non-aged samples in different base oils (Temperature: 40 ºC, Load: 100 N, Frequency: 50 Hz, Stroke: 1 mm, Test duration: 15 min)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Naphthenic

PAO2PAO1

VHVI2VHVI1

Paraffin

ic

Rapesee

d oil

Monoester

Diester

Complex este

r

Polyol es

ter

Coe

ficie

nt o

f fric

tion


Figure 13: Coefficients of friction of aged samples in different base oils (Temperature: 40 ºC, Load: 100 N, Frequency: 50 Hz, Stroke: 1 mm, Test duration: 15 min.)

13

3 Two body abrasive wear in reciprocating sliding The two body abrasive wear of an elastomer in dry and lubricated condition and the influence of aging the elastomer in different lubricating fluids on its abrasive wear have been investigated.

3.1 Experimental method An abrasive wear tester (Figure 14) has been used to study the dry and lubricated abrasive wear. This tester consists of a platform on which the sample is mounted and reciprocated against an abrasive paper wrapped around the circumferential surface of a wheel (ø50 mm × 12 mm thick). The wheel is turned by a small angle at the end of each stroke so as to enable the rubbing of elastomer against fresh abrasive surface. The frequency of reciprocation of the test specimen was 100 cpm and the stroke length was 30 mm. All abrasive wear tests have been done at room temperature (22 ± 2 ºC). In lubricated tests, the oil was injected into the counterface by using a syringe. The rubber test specimens were washed in industrial petroleum for 3 minutes by an ultrasonic cleaner, dried in an oven for 20 minutes at 45 ºC and then weighed. The same procedure was repeated after running the test for each specimen to quantify abrasive wear.

Figure 14: Abrasive wear tester

3.2 Test materials The elastomer and lubricants used in this study were the same as used in section 2. In this study, rectangular elastomeric sheet specimens of 40 mm × 20 mm and 4 mm thickness were used.

3.3 Results and discussion The two body abrasive wear tests have been performed on non-aged elastomer at different loads and by using different abrasive grit size. On aged samples, the tests were conducted by using only one grit size abrasive paper and at one load.

Backing support plate


Elastomersurface

Abrasive tape on wheel

14

3.3.1 Abrasive wear and the effect of lubrication The results show that when the elastomer slides against fine abrasives, the lubrication increases the abrasive wear but in sliding against coarse abrasives, lubrication decreases abrasive wear. Figure 15 shows that the decrease in abrasive wear is more pronounced at the beginning of sliding. As shown in Figure 16 the effect of lubricant on abrasive wear is more significant at lower loads.

0

10

20

30

40

50

60

40 80 120 160Cycles

Abr

asiv

e w

ear

(mg)

40 80 120 160Cycles

40 80 120 160Cycles

Figure 15: Abrasive wear of elastomer as a function of total cycles (Load: 1000 gr)

0

10

20

30

40

50

60

400 1000 1600Load (gr)

Abr

asiv

e w

ear

(mg) Dry Lubricated

Figure 16: Abrasive wear of the elastomer as a function of load under dry and lubricated sliding condition (Abrasive grit size: #320, Total cycles: 160)

3.3.2 Influence of aging the elastomer in different lubricating fluids The two body abrasive wear results of aged and non-aged rubbers in dry and lubricated conditions have been shown in Figure 17. It shows that the abrasive wear of both aged and non-aged nitrile rubber in lubricated condition is higher than that in dry condition. Aging the rubber in any base fluid, especially in ester base fluid and rapeseed oil leads to more abrasive wear. However, the influence of presence of oil is more significant than that of aging especially for mineral oils. Use of the monoester base fluid resulted in the highest abrasive wear of non-aged and aged rubber samples.

Dry Lubricated

Grit size: #120 Grit size: #320 Grit size: #500

15

Figure 17: Abrasive wear of aged and non-aged elastomer by using different base fluids (Abrasive grit size: # 500, Load: 500 gr, Speed: 100 c.p.m, Total cycles: 160)

The changes in the mechanical properties of the same compound of nitrile rubber in the same base fluids were investigated previously by Torbacke and Johansson [3]. Our results have not shown any clear correlation of friction coefficient and abrasive wear changes with the changes in mechanical properties due to aging in lubricating fluids. It seems that the chemical characteristics of the base oils rather than the mechanical properties may have greater influence on the friction coefficients.

16

17

4 Abrasive wear in unidirectional sliding Two body abrasive wear behaviour of the NBR elstomer has been studied under dry as well as lubricated unidirectional sliding conditions.

4.1 Experimental method The rubber specimen, attached on a metal backing plate, was pressed against an abrasive tape glued on a rotating steel ring and the normal and frictional forces were recorded by piezoelectric sensors. The test configuration is shown in Figure 18. The rubber specimens’ dimensions were 16mm×4mm×2mm (the width of contact area was 4 mm). The steel rings were of ø 35 and 8 mm thick. The rubber specimens were washed in industrial petroleum for 3 minutes using an ultrasonic cleaner and dried in an oven for 10 minutes at 40 ºC and then weighed. The same procedure was repeated after running the test for each specimen to quantify the abrasive wear. All the tests were performed at room temperature (22 ± 2 ºC). In lubricated tests, the oil was injected into rubbing interface by a syringe. The worn rubber surfaces were examined using an optical microscope.

Figure 18: The test configuration for abrasive wear tests

Two sets of abrasive tapes with different grit sizes (#120 and #500) have been used in these experiments and the effect of lubricant on the abrasion of the rubber has been studied. The speed of the ring was 1 r.p.m. and the tests were run for 1, 2 and 3 minutes. For longer tests, the abrasive tape was changed after 1 revolution of the ring, so that, the rubber was sliding against a fresh abrasive tape during the entire test duration. The tests have been run at two different loads. All the experiments were performed twice to check for repeatability of results.

Elastomer block (16 mm×4 mm×2 mm)

Abrasive tape glued on steel ring

35 mm

LoadBacking support plate

18

4.2 Test materialThe elastomeric material studied is acrylonitrile butadiene rubber (NBR). The rubber samples were cut out from the elastomeric sheets of 2 mm thickness. The nominal hardness, tensile strength, elongation at break and density of NBR elastomer are given in Table 2.

Table 2: The properties of NBR

Hardness, Shore A

Tensile strength, MPa

Elongation at break, %

Density, g/cm3

76.1 25.4 466 1.31

The lubricant used in these experiments was a mineral oil (paraffinic oil) with a density of 870 kg/m3 and a viscosity of 34.1 cSt at 40 °C.

4.3 Results and discussion To understand the effect of a lubricant on the wear mechanism, the lost mass and friction coefficient were measured and the worn surfaces were examined by using an optical microscope.

The experiments have been carried out at two levels of contact pressure. It can be observed that a combination of both the ridges (perpendicular to the direction of sliding) and scratches (along the direction of sliding) has been formed on the contact surfaces. The ridges have been concentrated close to the highest contact pressure zone. Figure 19 shows the typical worn surfaces and the position of ridges.

Figure 19: The worn surfaces of rubber samples sliding against (a) coarse abrasives (left) and (b) fine abrasives (right) in dry condition, Load = 1500 g, sliding distance = 2 rev.

Figure 20 shows the worn surfaces of rubber samples sliding against coarse abrasives in dry condition, after 1 and 2 revolutions. As the sliding distance increases, the worn surface is characterized by an increase in ridges.

19

Figure 20: The worn surfaces of rubber samples sliding against coarse abrasives in dry condition, (a) after 1 revolution, (b) after 2 revolutions, (Load = 150 g)

The formation of the ridges on the surfaces sliding against fine abrasives is more pronounced than that on the surfaces sliding against coarse abrasives (Figure 21). The ridges formed on the contact surfaces in the dry condition more rapidly than those in the lubricated condition. The presence of lubricant results in a decrease in the real contact area between the asperities of abrasives and the rubber surface. Therefore, the shearing stress is concentrated in the areas of contact of elastomer and abrasive tips and consequently the surface of the rubber is pulled in the direction of sliding locally. In dry condition, the shear stress is distributed more uniformly on the apparent contact area resulting in more uniform tensile stress at the rear of the apparent contact area and consequently the formation of continuous ridges; see Figure 6.

Figure 21: The worn surfaces of rubber samples after 2 revolutions, Load = 1500 g

20

Figure 22 shows the worn mass of the rubber as a function of numbers of revolutions at low load. The results show a mass gain of the rubber sample sliding against a coarse abrasive paper after one revolution of sliding distance. This may be due to the absorption of the oil into the rubber specimen. For short sliding distance with coarse abrasives, cuts are produced on the surface of rubber and wear of the rubber may be very small (in comparison to the absorbed oil). The oil is absorbed in the tears on the surface of the rubber and is not completely removed during washing process. As shown in the Figures 22 and 23, the presence of the lubricant decreases the lost mass in most test conditions but it does not mean that the wear of rubber in lubricated condition is less than that in dry condition. These results appear somewhat conflicting vis a vis the results earlier presented in section 3. It may however be noted that the elastomer used in these experiments as well as the test configuration were different from those used in section 3. In the experiments in section 3, a broader area of the surface of elastomer was sliding against the abrasives and the contribution of the surface of elastomer in abrasive wear is more than that in these experiments. Figure 15 of section 3 also shows that the influence of lubricant on increase in abrasive wear is more significant only for a low sliding distance where a thin layer of the surface of elastomer contributes to abrasive wear.

Figure 22: Lost mass of the elastomer as a function of number of revolutions (Normal load = 150 g)

Figure 23: Lost mass of the elastomer as a function of number of revolutions (Normal load = 1500 g)

As shown in Figures 24 and 25, the friction coefficient in the dry condition is higher than those in the lubricated condition. The reduction in friction coefficient is due to the

21

reduction of friction components including the adhesive and hysteretic. The tearing component may also decreases due to reduction in strength of the rubber in presence of oil. The difference between dry and lubricated friction coefficients at high load is more significant than that at low load, especially with fine abrasives. The apparent contact area at high load is larger than that at low load and the sealing of the oil in the voids between asperities is more effective in decreasing friction coefficient.

Figure 24: Friction coefficient as a function of number of revolutions (Normal load = 150 g)

Figure 25: Friction coefficient as a function of number of revolutions (Normal load = 1500 g)

22

23

5 Dry sliding friction Friction and wear behaviour of acrylonitrile butadiene rubber (NBR), hydrogenated acrylonitrile butadiene rubber (HNBR), acrylate rubber (ACM) and fluoroelastomer (FKM) against steel surfaces under unidirectional dry sliding conditions have been studied.

5.1 Experimental method The friction and wear behaviour of four sealing elastomers in dry sliding conditions have been studied using the Micro-Tribometer UMT-2 (Figure 26). In these studies, a rubber specimen, attached on a metal backing plate, was pressed against a rotating ring and the normal and frictional forces were recorded by piezoelectric sensors. Three sets of bearing steel rings with different surface roughness values were used with a view to study the effect of surface roughness on friction and wear. The surface topographies of the used rings have been shown in Figure 27. Each test was run for duration of 12 hours (43200 sec). The rubber specimens’ dimensions were 16 mm×4 mm×2 mm (the width of contact area was 4 mm). The counterface bearing steel rings were of ø35mm and 8 mm thick. The rubber specimens were washed in industrial petroleum for 3 minutes using an ultrasonic cleaner and dried in an oven for 10 minutes at 40 ºC and then weighed. The same procedure was repeated after running the test for each specimen in order to quantify wear. Each ring was washed in industrial petroleum for 3 minutes using anultrasonic cleaner and dried before the test and used only in one test. All the tests have been performed at room temperature (22 ± 2 ºC).

Figure 26: Micro-Tribometer UMT-2

Elastomer block (16 mm×4 mm×2 mm)

Steel ring 35 mm

LoadBacking support plate

24

Figure 27: Surface topographies of steel rings

5.2 Test materials The elastomers studied during this work are commonly used seal materials, acrylonitrile butadiene rubber (NBR), hydrogenated acrylonitrile butadiene rubber (HNBR), acrylate rubber (ACM) and fluoroelastomer (FKM). All the elastomers have a module of elasticity of about 10 MPa at very low speed and room temperature. The nominal hardness, tensile strength, elongation at break and material densities of these elastomers are given in Table 3.

Table 3: Tested elastomers and their properties

Elastomeric materials Hardness (Shore A)

Tensilestrength (MPa)

Elongation at break (%)

Density (g/cm3)

Nitrile rubber , (NBR) 76.1 25.4 466 1.31

Hydrogenated nitrile rubber, (HNBR) 71.3 17.5 303 1.24

Acrylate rubber, (ACM) 73.4 7.8 171 1.49

Fluoro rubber, (FKM) 72.8 - - 2.03

5.3 Results and discussion The tests have been run by using three sets of rings with different ranges of surface roughness on four elastomeric materials. The normal load was 150 gr. Using the Hertz contact theory, the contact pressure at low load is estimated to be about 240 KPa. At low contact pressure, the lost mass of the tested elastomers was very small (less than 0.5 mg). Figure 28 shows the friction coefficients of the elastomers versus time. As shown in the figure, except for FKM on fine surface, the friction coefficients drop during running-in periods to steady state values and the longest running-in periods have been observed during sliding against fine surfaces. The running-in time for sliding friction of HNBR is significantly longer than those with other materials. The decrease in friction coefficient during running-in periods may be due to the material transfer from the rubber onto the surface of the ring. Smearing as a result of the decomposition of the molecular network may be another reason for the decrease in friction coefficient.

Fine surface (0.15<Ra<0.3 μm) Fine surface (0.35<Ra<0.55 μm) Fine surface (0.5<Ra<0.7 μm)

25

Apart from FKM, the steady state values of friction coefficients of tested materials increase as the surface roughness decreases. The surface roughness has the minimum and maximum effects on the steady state friction coefficient of ACM and HNBR respectively.

It seems that a low friction layer, including fine particles, has been formed on the sliding surface of ACM and the steady state friction coefficient is more affected by the properties and size of the particles in the layer than the surface roughness of the ring.

The steady state friction coefficient of FKM against the rough surface is higher than that on the surface with medium surface roughness. It may be due to the higher hysteretic component of friction which increases with roughness. However, more investigation on the properties of the material and the surface roughness of the ring at nanoscale are required in order to clearly explain this effect.

Figure 28: Friction coefficient vs. time, (Normal load: 150 gr, Speed: 10 r.p.m. )

The sliding friction tests have been run against the bearing steel rings with medium surface roughness (0.35-0.55 m) at higher normal load (1000 gr). Using the Hertz contact theory, a contact pressure of 750 KPa is estimated at this normal load. Figure 29 shows the worn surfaces of different elastomers from tests at high normal load. As shown in the figure, roll formation has occurred on the surface of FKM and HNBR which was not observed in the tests at low contact pressure. Furthermore, a white powdery layer has been formed on the surface of ACM. The surface of NBR has been torn locally but the roll formation observed in other tests has not occurred.

26

Figure 29: Worn surfaces of the tested rubbers (Normal load: 1000 gr, speed: 10 r.p.m, duration: 12 hours, surface roughness of rings: Ra = 0.35-0.55 m)

Figure 30 shows the worn particles of FKM and HNBR. As shown in the Figures 29 and 30, severe wear occurred on the surface of FKM which resulted from the roll formation. Although, the roll formation occurred on the surface of HNBR, the amount and size of wear particles are much smaller than those on FKM. Consequently the worn mass of HNBR is much lower than that of FKM. Among the tested materials, ACM and FKM have the minimum and maximum elongation at break respectively and the size of worn particles shows a correlation with this property.

Figure 30: Worn particles of FKM and HNBR (Normal load: 1000 gr, speed: 10 r.p.m, duration: 12 hours, surface roughness of rings: Ra = 0.35-0.55 m)

Figure 31 shows the friction coefficient of the tested rubbers against the bearing steel rings of medium roughness. The highest friction coefficient has been observed in tribological pairs involving FKM followed by HNBR which may, in part, be due to the energy dissipated in the tearing and roll formation. The friction coefficient in case of ACM

27

increases gradually during the test which may be due to changes in the properties and/or the dimensions of the particles in the powdery layer between the surfaces.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 5000 10000 15000 20000 25000 30000 35000 40000 45000time (sec)

Fric

tion

coef

ficie

nt

ACM

FKM

HNBRNBR

Figure 31: Friction coefficient vs. time, (Normal load: 1000 gr, Speed: 10 r.p.m, Surface roughness of rings: Ra = 0.35 - 0.55 m)

28

29

6 Conclusions In this work, the friction and wear characteristics of some elastomeric materials have been studied under different operating conditions. The influence of aging of an elastomer (NBR) on friction and abrasive wear has also been investigated.

Some of the salient conclusions from this work are as follows:

Friction coefficient of an elastomer sliding against hard counterface at low contact pressure, where the hydrodynamic effects are significant, is affected by the surface topography of elastomer, but at high contact pressure, the friction coefficient is not affected by the surface topography of elastomer.

Aging the nitrile rubber in ester base fluids reduces the friction coefficient. The ester base oils can diffuse in the nitrile rubber more rapidly than the mineral oils resulting in a decrease in internal friction of elastomer. Therefore, the friction coefficients of a hard slider against aged elastomer samples in ester base oils are lower than those against aged rubber in mineral oils.

Depending on the mechanism of wear and the oil-elastomer compatibility, the presence of lubricant may decrease or increase the abrasive wear of elastomers. If the oil can penetrate the surface of the elastomer, it will have a detrimental effect on its wear behaviour. However, lubricating fluids decrease the frictional force which has an important role in the tearing action resulting in abrasive wear. Decreasing the frictional force also decreases the heat generation in the counterface which is another factor in weakening the elastomer and higher abrasive wear.

Aging the NBR in any of tested fluids increases the abrasive wear. Aging the NBR in ester base fluids and rapeseed oil leads to the highest increase in abrasive wear.

In unidirectional abrasive wear, both scratches (parallel to the direction of sliding) and ridges (perpendicular to the direction of sliding) have been observed on the worn surface of elastomer.

The ridges on the worn surfaces are formed close to the zone of maximum contact pressure where the elastomer penetrates in more voids and the contact pressure is more distributed on the surface of elastomer. Increasing the contact pressure and/or using finer abrasives accelerate but the presence of lubricant in the contact decelerates the formation of ridges.

In dry sliding of rubber against steel surface, the results show that the friction coefficients drop during running-in periods to steady state values which can be due to the transfer of a thin layer on to the hard counterface.

The longest running-in periods have been observed during sliding against fine surfaces. As the roughness of hard sliding surface increases, the mechanical action exerting on the elastomer surface increases and consequently the formation of transferred layer on to the counterface is accelerated leading to rapid decrease in friction.

30

31

7 Future work The existing theoretical knowledge to estimate the friction coefficient of elastomers is focussed on the hysteresis component of friction which is the dominant component of friction when an elastomer slides against a rough surface (e.g. road surfaces). The significance of adhesive component of friction coefficient of elastomers sliding against sealing surfaces, which are relatively smooth, has not been investigated sufficiently. The investigation of friction of elastomers against sealing surfaces and the comparison with the existing theories is to be performed in future.

Further, the influence of different parameters, such as contact pressure, abrasive coarseness and test periods on the lubricated abrasive wear of different sealing elastomers will also be studied.

In spite of the important roll of three-body abrasive wear of sealing elastomers on the seal performance, this subject has been studied very scantily. In view of this, three body abrasive wear of elastomers will also be another part of future work.

32

33

References:1) Muller, H., K., Fluid sealing technology principles and applications, Marcel

Dekker, Inc., Newyork, 1998 2) Hofmann, W., Rubber Technology Handbook, Hanser, Munich, 2001 3) Torbacke, M., Johansson, A., Seal Material and Base Fluid compatibility: An

overview, J. Synthetic Lubrication, 22-2, (2005), p 123-142 4) Patil, A. O., Coolbaugh, T. S., A LITERATURE REVIEW WITH EMPHASIS ON

OIL RESISTANCE, Rubber Chemistry and Technology, 78- 3, (Jul/Aug 2005), p 516

5) Tao, Z., Viriyabanthorn, N., Ghumman, B., Barry, C., Mead, J., HEAT RESISTANT ELASTOMERS, Rubber Chemistry and Technology, 78- 3, (Jul/Aug 2005), p 489

6) Moore, D.F., The Friction and Lubrication of Elastomers, Pergamon Press, 1972 7) Ludema, K. C., PHYSICAL FACTORS IN TYRE TRACTION, Physics in

Technology, January 1975 8) Moore, D.F., A review of adhesion theories for elastomers, Wear, 22, (1972), p

113-141.9) Moore, D.F., A review of hysteresis theories for elastomers, Wear, 30, (1974), p 1-

3410) Maeda, K., Bismark, A., Briscoe, B., Effect of bulk deformation on rubber

adhesion, Wear, 263, (2007), p 1016-1022 11) Roberts, Theories of dry rubber friction, Tribology International, 9-2, (April 1976),

p 75-81 12) Persson, B.N.J., On the theory of rubber friction, Surface Science, 401, (1998), p

445-45413) Persson, B.N.J., Volokitin, A.I., Rubber friction on smooth surfaces, European

Physical Journal E, 21-1, (Sep.2006), p 69-80 14) Fuller, K. N. G., Tabor, D., The Effect of Surface Roughness on the Adhesion of

Elastic Solids, Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, Vol. 345, No. 1642, (Sep. 1975), p 327-342

15) Greenwood, J.A., Tabor, D., The Friction of Hard Sliders on Lubricated Rubber: The Importance of Deformation Losses, Proc. Phys. Soc., 71, (1958) p 989-1001

16) Johnson, K. L., Kendall, K., Roberts, A. D., Surface energy and the contact of Elastic Solids, Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, Vol. 324, No.1558, (Sep. 1971), p 301-313

17) Grosh, K.A., The Relation between the Friction and Visco-Elastic Properties of Rubber, Proceedings of the Royal Society of London, Series A, Mathematical and Physical Sciences, 274 -1356. (Jun. 1963), p 21-39

18) Williams, M.L., Landel, R.F., Ferry, J.D., The Temperature Dependence of Relaxation Mechanisms in Amorphous Polymers and Other Glass-forming Liquids, Journal of the American Chemical Society, 77-14,(July 1955), p 3701

19) Shallamach, A., How does rubber slide?, Wear, 17, (1971), p 301-312

34

20) Barquins, M., friction and wear of rubber-like materials, Wear, 160, (1993), p1-11 21) Persson, B.N.J., Tartaglino, U., Albohr, O., Tosatti, E., Sealing is at the origin of

rubber slipping on wet roads, Nature Materials, 3-12, December, (2004), p 882-885

22) B. N. J. Persson, U. Tartaglino, O. Albohr, and E. Tosatti, Rubber friction on wet and dry road surfaces: The sealing effect, PHYSICAL REVIEW B 71, 035428 (2005)

23) James, D. I., Jolley, M., E., Abrasion of rubber, MacLaren & Sons Ltd., London, 1964

24) Schallamach, A., Friction and abrasion of rubber, Wear, 1, (1958), p 384-417 25) Thavamani, P., Khastgir, D., Bhowmick, A. K., Microscopic studies on the

mechanisms of wear of NR, SBR and HNBR vulcanizates under different conditions, Journal of Materials Science, 28, (1993), 6318-6322

26) Schallamach, A., Abrasion of rubber by a needle, J. Polymer Sci. 9 (1952) 385-404 27) Southern, E., Thomas, A. G., Studies of rubber abrasion, Rubber Chemistry and

Technology, 52-4, (Nov/Dec, 1979), p 1008-1018 28) Grosch, K. A., Schallamach, A., Relation between abrasion and strength of rubber,

Rubber Chemistry and Technology, 39, (1966), p 287-305 29) Zhang, S. W., Tribology of Elastomers, Elsevier, Netherlands, 2004 30) Zhang, S. W., Yang, Z., Energy theory of rubber abrasion by a line contact,

Tribology International, 30-12, (1997), p 839-843 31) Gent, A.N., Pulford, mechanisms of rubber abrasion, Journal of Applied Polymer

Science, Vol. 28, (1983), 943 32) Pulford, C.T.R., Antioxidant Effect during Blade Abrasion of Natural Rubber,

Journal of Applied Polymer Science, Vol. 28, (1983), 709 33) Schallamach, A., Abrasion, Fatigue, and Smearing of Rubber, Journal of Applied

Polymer Science, Vol. 12, (1968), 281 34) Stachowiak, G. W., A., W., Batchelor, ENGINEERING TRIBOLOGY

(2nd edition.), Butterworth-Heinemann, Boston, 2001 35) Muhr, A. H., Pond, T. J., Thomas, A. G., Abrasion of rubber and the effect of

lubricants, J. Chim. Phys., 84, (1987), p 331 36) Muhr, A. H., Roberts, A. D., Rubber abrasion and wear, Wear, 158, (1992), p 213-

22837) Chandrasekaran, M., Batchelor, A.W., In situ observation of sliding wear tests of

butyl rubber in the presence of lubricants in an X-ray microfocus instrument, Wear, 211, (1997), p 35-43

38) Patir, N., Cheng, H. S., An average Flow Model for Determining Effects of Tree- Dimensional Roughness on Partial Hydrodynamics Lubrication, J. Lubrication Technology, 100,(1978), p 12-17.

Papar A

1

Tribological behaviour of an elastomer aged in different oils

Mohammad Reza Mofidi*, Elisabet Kassfeldt, Braham Prakash

* Luleå University of Technology, Department of Applied Physics and Mechanical Engineering, Luleå SE- 971 87 Sweden

Tel: +46 (0)920 491038, Fax: +46 (0)920 491047; E-mail Address: [email protected]

_________________________________________________________________________

ABSTRACTThis paper presents the influence of aging the nitrile rubber, the most popular seal material, in various base fluids on sliding friction and abrasive wear. The lubricants used are synthetic esters, natural esters, different types of mineral base oils, poly- -olefins and very high viscosity index oils. Friction has been studied for two directions of motion with respect to lay on the elastomer sample by using the SRV Optimol test machine. These findings show that as compared to all other lubricant formulations, ageing the elastomer in polyol ester leads to the maximum reduction of friction coefficient especially in perpendicular sliding to the initial lay on the surface. The abrasive wear studies were carried out by using a two body abrasive wear tester against dry and lubricated elastomer. It was interesting to note that two body abrasive wear of elastomeric material was higher during rubbing in presence of the fluids as compared to that in dry condition. Further, aging the elastomer in these base fluids especially in ester base fluids, results in more abrasive wear.

Key words: Elastomers; Lubricating fluids; Friction; Abrasive wear

1 INTRODUCTION Elastomers have some very useful properties such as low Young’s modulus, large elongation- to- break and high value of Poisson’s ratio which make them suitable for many sealing applications. Seal is a component which prevents the leakage of fluids or gas from the machine and contamination entering the machine. Most seals operate in presence of oils during their service life. Friction and wear are two important factors in seal performance and the overall efficiency of the machine. Therefore, the interaction between oils and elastomer and its influence on friction and wear behaviour of elastomer has an important role in seal performance. The aim of this study is to investigate the influence of aging nitrile rubber in different oils on its tribological behaviour.

mailto:[email protected]

2

1.1 Friction The coefficient of friction of a rubber surface during sliding against a hard surface in lubricated conditions can be expressed in terms of the contribution of liquid, adhesion, and deformation (hysteresis) components. The contribution of adhesion component is relatedto the asperity peaks where the fluid film is extremely thin and has properties distinct from the bulk lubricant in the voids. The lubricant film at asperities peaks has some of the properties of draped elastomer and shear strength at these areas is considerably higher than those of other areas [1]. Adhesion is generally recognized to consist in the making and breaking of junctions at a molecular level. Several theories have been proposed to describe the adhesion component of friction. These studies have confirmed that the adhesion friction of elastomer against hard surface decreases with the decrease in the Young’s modulus and it is a function of viscoelastic properties of elastomer which in turn depend on temperature and sliding velocity [2], [3].

In contrast to the ploughing action of metal-on-metal friction, the sliding elastomer flows readily over the rigid asperities of the mating counterface and conform to their contours. The deformation component of friction produced by such flowing action is called hysteresis. The existence of hysteresis friction is a consequence of energy loss associated with internal damping within the viscoelastic body [4]. Like the adhesion component of friction, the hysteresis component of friction is also a function of viscoelastic properties of the elastomer, but unlike the adhesion friction, the hysteresis friction increases with decrease in Young’s modulus of elastomer [1].

The fraction of contribution of adhesion and hysteresis friction depends on the geometry and cleanliness of the mating surfaces. The adhesion component is important only for very clean and smooth rubber surfaces [5, 6]. The main source of friction in well lubricated sliding arises from deformation [7, 8]. Greenwood and Tabor performed some tests with spherical and conical specimens sliding against rubber. They stated that the sliding friction of spherical specimen, in a well lubricated condition at high loads, is the same as rolling friction but at low loads the sliding friction is larger than rolling friction. They have concluded that as the load is reduced, the shearing term becomes more important [7]. The significance of adhesion friction at low load can be interpreted as the attractive forces as demonstrated by Johnson et al. [9]. Presence of fluid between rubber and hard substrate reduces not only the adhesion but also the hysteresis component of friction. On a lubricated substrate the valleys turn into fluid pools which are sealed off and effectively smoothen the substrate surface. Smoothening reduces the viscoelastic deformation from the surface asperities, and thus reduces rubber friction [10, 11].

1.2 Abrasion Abrasion occurs as a result of local mechanical rupture (tearing) and/or general decomposition of the molecular network to a low-molecular-weight material (smearing) [12]. If the hard sliding specimen against rubber is sharp, the abrasion results from tensile failure and if it is blunt, the abrasion results from fatigue failure. Schallamach has studied the rubber abrasion in dry condition and reported that the abrasion is proportional to normal load and proportional to the mean radius of asperity curvature if they can be approximated to hemisphere and independent of particle size if the particles are polyhedral [13]. The spacing of the abrasion pattern is proportional to the cube root of

3

the normal load, proportional to the two-thirds power of the particle size of the abrasive with polyhedral particles, and directly proportional to the size of abrasive hemispherical particles [13]. Later on Grosch and Schallamach found that on sharp tracks such as abrasive paper linear wear rate as a result of tensile failure, was proportional to the normal stress, friction coefficient and inversely proportional to the energy density at break [12]. Southern and Thomas studied abrasion of rubber surface by a razor blade in line contact and formulated a theory relating the rate of abrasion is to the crack growth characteristics of the rubber, the angle of crack growth and the frictional force on the blade [14]. Zhang and Yang have introduced a theoretical wear equation of rubber abrasion in a line contact from the viewpoint of energy on the basis of experimental results [15]. Muhr et al. have studied the influence of lubrication on the abrasion of rubber by blade in line contact. They observed that when a lubricant is applied, a much finer pattern develops and the rate of abrasion is much lower but the horizontal force on the blade does not decrease so dramatically [16, 17]. Chandrasekaran and Batchelor have studied the friction and wear of butyl rubber sliding on abrasive paper as a function of temperature and load. They conducted dry and lubricated unidirectional sliding tests and reported that the presence of lubricant reduced the coefficient of friction but accelerated wear due to chemical degradation of rubber [18].

1.3 Seal-oil compatibility Elastomers can swell and/or degrade in chemical seal environments through reactions with the polymer backbone and cross-link system, or by reactions with the filler system [19]. Presence of the polar side-groups in the backbone chain increases the oil resistance of the polymer [20]. Crosslinking also limits the degree of polymer swelling by providing tie-points (constrains) that limit the amount of solvent that can be absorbed into the polymer [20]. Nitrile rubber is a copolymer of acrylonitrile and butadiene. NBR is a low-cost elastomer with good mechanical properties. The concentration of acrylonitrile in the copolymer has a considerable influence on the polarity and swell resistance of the volcanizates in non-polar solvents. The greater the acrylonitrile content, the less the swell in motor fuels, oils, fats, etc [21]. However the elasticity and low temperature flexibility also become poorer. The mechanical properties of elastomers are affected by oils. Van der Waal [22] and Torbacke and Johansson [23] have studied the influence of different base fluids on the changes in mechanical properties of elastomers. Generally the influence of ester base fluids on deterioration of nitrile rubber is more significant in comparison to that of mineral oils and PAOs [23]. Some elastomers are sensitive to polar compounds. The difference between the chemical structure of ester base fluids and mineral oils and PAOs is the existence of more carboxylic groups in ester base fluids which are polar groups.

2 EXPERIMENTAL WORK Two series of experiments, including friction and two body abrasion tests have been performed. The influences of ageing the elastomer in different types of base oils on the friction coefficient and abrasive wear have been investigated.

2.1 Friction tests The friction tests have been carried out by using Optimol SRV machine. The machine reciprocates an upper cylindrical specimen loaded against lower specimen. The motion of

4

cylinder is parallel to its axis. The friction force is measured by piezoelectric force sensors. Temperature, normal force, frequency of motion and stroke length can be controlled during the tests. The diameter of the cylinder is 15 mm and its length is 22 mm. The edges of the slider are chamfered/rounded off with a view to minimise the edge effect. Figure 1 shows the test configuration and the surface topography of the slider.

Figure 1: Test configuration for friction studies under reciprocating sliding conditions by using Optimol SRV machine (left); Surface topography of the slider, Ra = 0.80 nm, Rq =107 nm (right).

The rubber specimens used in friction tests were washed in industrial petroleum for 3 minutes by using an ultrasonic cleaner and then dried for 10 minutes.

2.2 Abrasion tests The two body abrasion tests were conducted by using an abrasive wear tester (Figure 2). It consists of a table that holds the specimen and reciprocates it against an abrasive paper wrapped around the circumferential surface of a wheel (ø50 mm × 12 mm thick). The wheel is turned by a fraction of one rotation at the end of each stroke so as to enable the rubbing of elastomer against fresh abrasive surface. The frequency of reciprocation of the test specimen was 100 cpm and the stroke length was 30 mm. All abrasive wear tests were run for a total of 160 cycles. All these tests have been done at room temperature (22 ± 2 ºC).

Parallel sliding Perpendicular sliding

Surface topography of the slider

Elastomeric sample

Upper specimen (Steel cylinder)


The geometry of the edges of slider

3.3mm

0.4 mm

5

Figure 2: Test configuration for abrasive wear studies under reciprocating sliding

The rubber test specimens were washed in industrial petroleum for 3 minutes by an ultrasonic cleaner, dried in an oven for 20 minutes at 45 ºC and then weighed. The same procedure was repeated after running the test for each specimen to quantify abrasive wear.

2.3 Test materials and lubricants The elastomer which has been studied is acrylonitrile butadiene rubber (NBR). The content of acrylonitrile in the tested elastomer is 28%, which is common for oil applications. This elastomeric material is vulcanized by sulphur. The polymeric content is 44% and the remaining part consists of different types of additives. The rubber samples used in these studies were in the form of sheets of 4mm thickness. The initial hardness of the elastomer was 75±5 IRHD (international rubber hardness degrees). The surface was examined in a Wyko 3D optical surface profilometer. The elastomer surface was characterised by parallel grooves (Figure 3) and are caused during moulding of elastomeric sheets in steel mould.

Figure 3: Surface topography of the fresh elastomer sample (Ra=0.344 m, Rq=0.438 m)

The test specimens for tribological studies were cut out from the elastomeric sheets. For friction studies, discs of ø25 mm and 4 mm thickness were used. In abrasive wear tests, rectangular sheet specimens of 40 mm × 20 mm and 4 mm thickness were used.

Backing support plate


Elastomersurface

Abrasive tape on wheel

6

The lubricating fluids used were paraffinic, naphthenic, 2 PAOs, 2 VHVIs, monoester, diester, polyolester, complex ester and rapeseed oils. The base fluids used in these studies and their properties are listed in Table 1.

Table 1: The properties of lubricants

Base fluid Density (kg/m3) Viscosity@40ºC (cSt) NPI Naphthenic base oil 896 30.0 —PAO2 830 28.5 —PAO1 790 5.5 —VHVI2 830 26.0 —VHVI1 822 12.0 —Paraffinic base oil 870 34.1 —Diester 910 (20°C) 26.1 82 Complex ester 980 46.0 80 Polyol ester 900 35.5 170 Monoester 864 (20 °C) 8.5 102 Rapeseed oil 910 (30 °C) 34.0 190

The aged elastomer samples were prepared by immersing them in different base fluids at 125 ºC for one week.

3 RESULTS AND DISCUSSION Figure 4 shows the average values of friction coefficients during sliding of bearing steel cylinder against the non-aged rubber sample in the presence of different base fluids.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Naphthen

icPAO2

PAO1VHVI2

VHVI1

Paraffi

nic

Rapese

ed oil

Monoester

Diester

Complexes

ter

Polyol este

r

Coe

ficie

nt o

f fric

tion


Figure 4: Coefficients of friction of non-aged samples in different base oils (Temperature: 40 ºC, Load: 100 N, Frequency: 50 Hz, Stroke: 1 mm, Test duration: 15 min)

7

It shows that there is no correlation between the friction coefficient and viscosity. It means that the sliding occuring mainly is in boundary or mixed lubrication regime. When the slider reciprocates parallel to the direction of lay on the rubber surface, the friction coefficient is marginally higher than that in perpendicular sliding. In perpendicular sliding, the oil can be trapped within the grooves which may result in some hydrodynamic effects and subsequently lower the friction coefficient. It confirms the Patir and Cheng’s [24] results for rubber.

Figure 5 shows the average values of friction coefficients on aged rubbers in different base fluids. Comparison of friction results in Figures 4 and 5 shows that aging the rubber in synthetic esters leads to decrease in friction coefficient.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Naphthen

icPAO2

PAO1VHVI2

VHVI1

Paraffin

ic

Rapes

eed oil

Monoester

Diester

Complex es

ter

Polyol e

ster

Coe

ficie

nt o

f fric

tion


Figure 5: Coefficients of friction of aged samples in different base oils (Temperature: 40 ºC, Load: 100 N, Frequency: 50 Hz, Stroke: 1 mm, Test duration: 15 min.)

Investigation of the surfaces of the tested specimens shows that the worn surface areas of the aged specimens are considerably less than those of non aged specimens.

The two body abrasive wear results of aged and non-aged rubbers in dry and lubricated conditions have been shown in Figure 6. It shows that the abrasive wear of both aged and non-aged nitrile rubber in lubricated condition is higher than that in dry condition.

Aging the rubber in any base fluid, especially in ester base fluids and rapeseed oil leads to more abrasive wear as well, but the influence of presence of oil is more significant than that of aging especially for mineral oils. Use of the monoester base fluid has resulted in highest abrasive wear of non-aged and aged rubber samples.

8

Figure 6: Abrasive wear of aged and non-aged elastomer by using different base fluids (Abrasive grit size: # 500, Load: 500 gr, Speed: 100 c.p.m, Abrasive wheel rotation: 1/200, Total cycles: 160)

The changes in the mechanical properties of the same compound of nitrile rubber in the same base fluids were investigated previously by Torbacke and Johansson [23]. Our results have not shown any clear correlation between the changes in friction coefficient and changes in mechanical properties due to aging in different lubricating fluids (Figure 7).

9

Figure 7: Friction coefficient vs. the changes in mechanical properties of aged rubber in different base fluids

It seems that the changes in the physico-chemical properties of the surface of exposed rubber, rather than changes in mechanical properties may play a more significant role in determining the frictional behaviour. Figure 8 shows the correlation between the abrasive wear and the changes in mechanical properties. The decrease in tensile strength of rubber due to aging in monoester fluid seems to cause relatively higher abrasive wear.

10

Figure 8: Abrasive wear vs. the changes in mechanical properties of aged rubber in different base fluids

Figure 9: Abrasive wear and friction coefficient vs. the non polarity index (NPI) of oils

Figure 9 shows the abrasive wear and friction coefficient as a function of the non-polarity index of oils (NPI). It can be seen that in both aged and non-aged samples abrasive wear is maximum when the monoester with NPI 102 is used. However, non-polarity index does not seem to have any influence on frictional behaviour.

11

4 CONCLUSION Ageing the nitrile rubber in the synthetic ester base fluids leads to reduction of friction coefficient. This effect in reducing the friction coefficient, especially in perpendicular sliding to the initial lay on the surface, is more considerable for the sample aged in polyol ester. The presence of the base fluids increases the abrasive wear of tested nitrile rubber. Ageing the nitrile rubber in the lubricating fluids increases the abrasive wear in both dry and lubricated conditions. The maximum change in abrasive wear due to aging is observed for the samples aged in monoester. Dry abrasive wear of the aged nitrile rubber in ester base fluids is higher than that in the mineral oils. The influence of presence of oil on increasing the abrasive wear is more significant than that of aging.

REFERENCES1) Moore, D.F., The Friction and Lubrication of Elastomers, Pergamon Press, 1972. 2) Moore, D.F., A review of adhesion theories for elastomers, Wear, 22(1972), p 113-

141.3) Persson, B.N.J., On the theory of rubber friction, Surface science, 401, (1998), p

445-4544) Moore, D.F., A review of hysteresis theories for elastomers, Wear, 30(1974), p 1-

34.5) Persson, B.N.J., Volokitin, A.I., Rubber friction on smooth surfaces, European

Physical Journal E, 21-1, (Sep.2006), p 69-80 6) Fuller, K. N. G., Tabor, D., The Effect of Surface Roughness on the Adhesion of

Elastic Solids, Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, Vol. 345, No. 1642. (Sep. 1975), p 327-342.

7) Greenwood, J.A., Tabor, D., The Friction of Hard Sliders on Lubricated Rubber: The Importance of Deformation Losses, Proc. Phys. Soc., 71, (1958) p 989-1001.

8) Trachman, E.G., Williams, R., Ping Sheng, Orientation effects in the friction of a hard ellipsoid sliding on rubber, J. applied physics, 48-8 (1977), p 3270-3273.

9) Johnson, K. L., Kendall, K., Roberts, A. D., Surface energy and the contact of Elastic Solids, Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, Vol.324, No.1558. (Sep. 1971), p 301-313.

10) Persson, B.N.J., Albohr, O., Tartaglino, U., Volokitin, A.I., Tosatti, E., On the nature of surface roughness with application to contact mechanics, sealing, rubber friction and adhesion, Journal of Physics Condensed Matter, 17-1, Jan 12, (2005), p R1-R62

11) Persson, B.N.J., Tartaglino, U., Albohr, O., Tosatti, E., Sealing is at the origin of rubber slipping on wet roads, Nature Materials, 3-12, December, (2004), p 882-885

12) GENT, A. N., PULFORD, C. T. R., Mechanisms of Rubber Abrasion, Journal of Applied Polymer Science, 28 (1983), p 943-960.

13) Schallamach, A., On the Abrasion of Rubber, Proc. Phys. Soc., B 67, (1954), p 883-891.

14) Southern, E., Thomas, A. G., Studies of rubber abrasion, Rubber Chemistry and Technology, 52-4, (Nov/Dec, 1979), p 1008-1018.

12

15) Zhang, S. W., Yang, Z., Energy theory of rubber abrasion by a line contact, Tribology International, 30-12, (1997) p 839-843.

16) Muhr, A. H., Roberts, A. D., Rubber abrasion and wear, Wear, 158 (1992) p 213-228

17) Muhr, A. H., Pond, T. J., Thomas, A. G., Abrasion of rubber and the effect of lubricants, J. Chim. Phys., 84 (1987) p 331.

18) Chandrasekaran, M., Batchelor, A.W., In situ observation of sliding wear tests of butyl rubber in the presence of lubricants in an X-ray microfocus instrument, Wear, 211 (1997), p 35-43.

19) Information obtained from the website: http://www.pspglobal.com/prop-chemical-compatibility.html

20) Patil, A. O., Coolbaugh, T. S., A LITERATURE REVIEW WITH EMPHASIS ON OIL RESISTANCE, Rubber Chemistry and Technology, (Jul/Aug 2005), 78- 3, p 516

21) Hofmann, W., Rubber Technology Handbook, Hanser, Munich, 2001 22) Van der Waal, G., The Relationship Between the Chemical Structure of Ester Base

Fluids and their Influence on Elastomer Seals, and Wear characteristics, J. Synthetic Lubrication, 1-4, (1984), p 35-47.

23) Torbacke, M., Johansson, A., Seal Material and Base Fluid compatibility: An overview, J. Synthetic Lubrication, 22-2, (2005), p 123-142.

24) Patir, N., Cheng, H. S., An average Flow Model for Determining Effects of Tree- Dimensional Roughness on Partial Hydrodynamics Lubrication, J. Lubrication Technology, 100,(1978), p 12-17.

http://www.pspglobal.com/prop-chemical-compatibility.html



Paper B

1

Two body abrasive wear and frictional characteristics of NBR elastomer under lubricated sliding conditions

Mohammadreza Mofidi, Braham Prakash*

Division of Machine Elements, Luleå University of Technology, Luleå SE-971 87 Sweden

* Tel: +46 920 493055, Fax: +46 920 491047, [email protected]

_________________________________________________________________________

AbstractUnderstanding the mechanisms of abrasion of an elastomer in the presence of lubricants is of importance in sealing applications. In this research a block on ring configuration was used to study the influence of lubrication on the abrasion of acrylonitrile butadiene rubber (NBR), the most commonly used seal material. The friction force and the worn mass of the samples were recorded and the worn surfaces were investigated using an optical microscope. Both scratches (parallel to the direction of sliding) and ridges (perpendicular to the direction of sliding) were observed on the worn surfaces. The scratches appeared on the contact surfaces at the beginning of sliding and the ridges were formed after a certain sliding duration. The worn surfaces of NBR samples sliding against finer abrasives were characterized by more defined ridges. Increasing the contact pressure accelerated the formation of ridges but the presence of a lubricant in the contact decelerated the ridge formation. The wear of the NBR elastomer in the lubricated condition is slightly lower than that in the dry condition. The friction coefficient decreased significantly in presence of the lubricant, especially during sliding against finer abrasives and at higher contact pressure.

Keywords: Elastomer, Abrasive wear, Friction

1 Introduction Most seals operate in the presence of lubricants during their service life. Friction and wear are two important factors in seal performance and the overall efficiency of the machine. Wear reduces the sealing ability. Therefore, the influence of lubricants on wear mechanisms of elastomers has an important role in seal performance.

1.1 Friction The coefficient of friction of a rubber surface against a hard surface can be expressed in terms of the contribution of adhesion, deformation (hysteresis), viscous and cohesion (tearing) components [1, 2]. However; most texts consider only two terms for friction

mailto:[email protected]

2

components. They suggest that the tearing and viscous components can be represented by deformation and adhesion respectively [1]. Adhesion is generally recognized as the making and breaking of junctions at a molecular level [3]. Hysteretic friction is a consequence of energy loss associated with internal damping within the viscoelastic body [4]. The cohesion component of friction is the contribution of wear to the bulk losses and the viscous component is the viscous drag under wet condition [1]. Presence of fluid between rubber and hard substrate reduces not only the adhesion but also the hysteretic component of friction. The lubrication decreases the real contact area between the rubber and hard counterface resulting in a decrease in friction coefficient. This effect is more pronounced at higher velocities due to hydrodynamic effects. On a lubricated surface, the valleys turn into fluid pools which are sealed off and thus make the surface smooth (Figure 1). This smoothening reduces the viscoelastic deformation caused by the surface asperities, and reduces rubber friction [5, 6].

Figure 1: Smoothing the counterface in presence of lubricant [6]

1.2 Wear Any estimation of abrasion of rubber needs to take into account the mechanism of wear. Three different mechanisms of wear can be identified when an elastomer slides against a hard counterface [1]. During sliding against a hard countersurface with a sharp texture, abrasive wear takes place as a result of tearing of the sliding surface of the elastomer.Fatigue wear is another mechanism of wear which occurs on the surface of an elastomer sliding against blunt projections on the hard counterface. When a highly elastic elastomer slides against a smooth surface, roll formation occurs. In this type of wear, the high frictional force shears a projection on the rubber surface, tears and then rolls the tongue along the direction of sliding [1]. A critical value of shear stress can be defined for each rubber above which roll formation occurs, and below which wear is mainly due to fatigue. Thus the friction coefficient is one of the most important properties of rubber governing the type of wear [7]. In practice, a combination of three forms of wear occurs and it is difficult to separate the contribution of each mechanism to the overall wear [1].

Another classification of wear of elastomers introduces two mechanism of wear of elastomers sliding against hard counterface. Mechanochemical decomposition of the molecular network to a low molecular weight leading to a tar-like wear product (smearing) and cohesive rupture (tearing) [8]. The oily decomposition product which forms during smearing protects the underlying rubber from tearing and thus decreases the rate of wear [9]. Experiments show that the rate of wear during smearing decreases by introducing antioxidants [9, 10].

Rubber

Hard counterface

Fluid

Hard counterface

3

The worn surface of an elastomer may exhibit different appearance. When rubber is abraded without a change of direction, sets of parallel ridges, perpendicular to the direction of sliding, are often found on the surface [11]. The surfaces of elastomers worn by fatigue wear exhibit pitting marks and the surfaces of harder elastomers, sliding against sharp asperities, exhibit scratches parallel to the direction of sliding [12]. The mechanism of abrasion leading to the ridge formation has been studied extensively [11, 13-17]. Mostof previous experiments on the mechanism of ridge formation have been carried out using a line contact configuration. The formation of ridges starts by initiation the cracks at the rear of the contact region, due to the high shearing stress, and continues by growing the cracks under repetitive loading [14]. The scratches, parallel to the sliding direction, occur on the surface of elastomers sliding in point contacts with sharp asperities. The elastomer surface is pulled in the direction of sliding and fails in tension behind the contact perpendicular to the tensile stress field (Figure 2) [18].

Figure 2: Mechanism of scratch and ridge formation on the sliding surface of elastomer [14, 18]

Muhr et al. have studied the influence of lubrication on the abrasion of rubber by a blade in line contact. They observed that when a lubricant is applied, a finer pattern develops and the rate of abrasion is reduced, but the horizontal force on the blade does not decrease as dramatically [19, 20]. Chandrasekaran and Batchelor have studied the friction and wear of butyl rubber sliding on abrasive paper as a function of temperature and load. They conducted dry and lubricated unidirectional sliding tests and reported that the presence of lubricant reduced the coefficient of friction but accelerated wear due to chemical degradation of rubber [21].

1.3 Elastomer - oil compatibility When an elastomer and a base fluid are brought in contact with each other, the elastomer material may absorb the base fluid or the base fluid may extract soluble constituents of the elastomer. The base fluid may also react with the elastomer [22]. Presence of the polar side-groups in the backbone chain increases the oil resistance of the polymer [23]. Crosslinking also limits the degree of polymer swelling by providing tie points (constrains) that limit the amount of solvent that can be absorbed into the polymer [23]. Nitrile rubber (NBR) is a copolymer of acrylonitrile and butadiene and provides a low-cost elastomer with good mechanical properties in sealing application. The concentration of acrylonitrile in the copolymer has a considerable influence on the polarity and swell resistance of the vulcanizate in non-polar solvents. The greater the acrylonitrile content, the lower the amount of the swell in motor fuels, oils, fats, etc [24].

4

Abrasion of sealing elastomer is a type of seal failure and reduces the sealing ability. This study aims at investigating the effect of lubrication on the mechanisms of abrasion and their contributions to the overall wearing of elastomer surfaces.

2 Experimental The experiments were carried out using Micro-Tribometer UMT-2. The rubber specimen glued to a metal backing plate was pressed against an abrasive tape glued on to the circumferential surface of a rotating steel ring. The normal and frictional forces were recorded by piezoelectric sensors. The schematic of the test configuration is shown in Figure 3. The rubber specimens’ dimensions were 16 mm×4 mm×2 mm (the width of contact area was 4 mm). The steel rings were of ø35 and 8 mm thick. The rubber specimens were washed in industrial petroleum for 3 minutes using an ultrasonic cleaner and dried in an oven for 10 minutes at 40 ºC and then weighed. The same procedure was repeated after running the test for each specimen to quantify wear. All the tests were performed at room temperature (22 ± 2 ºC). The worn rubber surfaces were examined using an optical microscope.

Figure 3: Test configuration

Two sets of SiC abrasive tapes with different grit sizes (#120 and #500) were used in these experiments and the effect of lubricant on the abrasion of the rubber was studied. The speed of the ring was 1 r.p.m. and the tests were run for 1, 2 and 3 minutes. For longer tests, the abrasive tape was changed after 1 revolution of the ring, so that, the rubber always slid against the fresh abrasive tape during the entire test duration. The tests were run at two different loads. The test parameters are shown in Table 1.

Table 1: Test parameters

Test parameters Load, N Abrasive grit size, # Number of revolutions

Level 1 1.5 500 1 Level 2 15 120 2 Level 3 - - 3

5

The elastomeric material studied is acrylonitrile butadiene rubber (NBR), the most commonly used seal material. The rubber samples were cut out from the elastomeric sheets of 2 mm thickness. The nominal hardness, tensile strength, elongation at break and density of NBR elastomer are given in Table 2.

Table 2: The propertiesof NBR

Hardness, Shore A

Tensile strength, MPa

Elongation at break, %

Density, g/cm3

76.1 25.4 466 1.31

The lubricant used in these experiments was a mineral oil (paraffinic oil) with a density of 870 kg/m3 and a viscosity of 34.1 cSt at 40 °C.

All the experiments were performed twice to check for repeatability.

3 Results and discussion To understand the effect of a lubricant on the wear mechanism, the lost mass and friction coefficient were measured and the worn surfaces were examined by using an optical microscope.

3.1 Worn surfaces The experiments were carried out at two levels of contact pressure. It can be observed that a combination of both ridges (perpendicular to the direction of sliding) and scratches (along the direction of sliding) were formed on the contact surfaces of elastomer. Figure 4 shows the typical worn surfaces and the position of ridges. The ridges were concentrated close to the front part of contact region which is at higher contact pressure. As the sliding distance increases, the worn surface is characterized by a larger area of ridges (Figure 5).

Figure 4: The worn surfaces of rubber samples sliding against (a) coarse abrasives (left) and (b) fine abrasives (right) in dry condition, Load = 15 N, number of revolutions = 2 rev.

6

Figure 5: The worn surfaces of rubber samples sliding against coarse abrasives in dry condition, (a) after 1 revolution, (b) after 3 revolutions, Load = 1.5 N

The formation of the ridges on the surfaces sliding against fine abrasives was more recognizable than that on the surfaces sliding against coarse abrasives (Figure 6).

Figure 6: The worn surfaces of rubber samples after 2 revolutions, Load = 15 N

The ridges formed more rapidly in the dry condition than in the lubricated condition. The presence of lubricant results in a decrease of the real contact area between the asperities of abrasives and the rubber surface (Figure 1). Therefore, the shearing stress is concentrated on the areas in contact with the top of asperities and consequently the surface of the rubber

7

is pulled in the direction of sliding more locally. The local shearing stress results in the formation of scratches parallel to the direction of sliding. In dry condition, the shear stress is distributed more uniformly on the apparent contact area resulting in more uniform tensile stress at the rear of the apparent contact area and consequently the formation of continuous ridges (Figure 2).

3.2 Wear The wear measurements do not show a strong repeatability, possibly owing to the short test duration. Figure 6 shows the lost mass of the worn elastomer as a function of number of revolutions at low load. The results show a negative lost mass of the rubber sample sliding against a coarse abrasive paper at one revolution of sliding distance. This may be due to the absorption of the lubricant into the rubber. For the short sliding distance on the coarse abrasives, tearing occured on the surface of rubber and the net mass loss of rubber was minimal. The lubricant was easily absorbed into the torn surface of the rubber and was not completely removed during washing process. As shown in the Figure 7, the presence of the lubricant reduced the mass loss in most test conditions. However, this does not mean that the net mass loss of rubber in lubricated condition was necessarily less than that in dry condition. The results show that the worn mass of rubber sliding against fine abrasives over a short sliding distance at low load in lubricated condition is slightly higher than that in dry condition.

Figure 7: Lost mass of the elastomer as a function of number of revolutions

8

3.3 Friction As shown in Figure 8, the friction coefficient in the dry condition was higher than that in lubricated condition. The reduction in friction coefficient can be explained from the adhesive and hysteretic friction components. The separating lubricant layer in the contact decreases the adhesive friction and the sealing effect decreases the hysteretic friction. The difference between dry and lubricated friction coefficients at high load is more pronounced than that at low load, especially on fine abrasives. The apparent contact area at high load is larger than that at low load and the lubricant which is trapped in the middle of the contact region has less opportunity to escape from the contact. Thus, sealing the oil in the void spaces between asperities was more effective at decreasing friction coefficient.

Figure 8: Friction coefficient as a function of number of revolutions

4 Conclusion Abrasive wear and friction of NBR under lubricated sliding condition were studied. Both scratches (parallel to the direction of sliding) and ridges (perpendicular to the direction of sliding) were observed on the worn surfaces. The ridges on the worn surfaces were formed close to the zone of maximum contact pressure. Increasing the contact pressure and/or using finer abrasives accelerated the formation of ridges. Presence of a lubricant in the contact decelerated the ridge formation. The wear of the NBR sliding under lubricated condition is slightly lower than that in dry sliding. The lubricant decreases the friction coefficient, especially in sliding against finer abrasives at higher contact pressures.

9

AcknowledgementsThe elastomeric material used in this work was supplied by Mr. Stellario Barbera (SKF Sealing Solutions, Italy) and Mr. Joop Vree (SKF Engineering Research Centre, The Netherlands) and we thankfully acknowledge their support. The authors express their gratitude to Dr, Richard Schaake (SKF ERC) for his usfull suggestions concerning this manuscript. Finally, the authors also thank the SKF ERC management for their permission to publish this work.

References1. Moore, D. F., “The Friction and Lubrication of Elastomers”, 1st ed, Pergamon

Press, New York, 1972, p. 14, 198, 264, 265, 266 2. Ludema, K. C., PHYSICAL FACTORS IN TYRE TRACTION, Physics in

Technology 6 (1975) 11-17 3. Moore, D. F., A review of adhesion theories for elastomers,Wear 22 (1972) 113-

1414. Moore, D. F., A review of hysteresis theories for elastomers, Wear 30(1974)1-34 5. Persson, B. N. J., Albohr, O., Tartaglino, U., Volokitin, A.I., Tosatti, On the nature

of surface roughness with application to contact mechanics, sealing, rubber friction and adhesion, E., J. Phys.-Condes. Matter 17 (2005) R1-R62

6. Persson, B. N. J., Tartaglino, U., Albohr, O., Tosatti, E., Sealing is at the origin of rubber slipping on wet roads, Nature Materials 3 (2004) 882-885

7. James, D. I., Jolley, M., E., Abrasion of rubber, MacLaren & Sons Ltd., London, 1964, p. 48

8. Gent, A. N., Pulford, C. T. R., Mechanisms of rubber abrasion, J. Appl. Polym. Sci. 28 (1983) 943-960

9. Pulford, C.T.R., Antioxidant Effect during Blade Abrasion of Natural Rubber, J. Appl. Polym. Sci. 28 (1983) 709-713

10. Schallamach, A., Abrasion, Fatigue, and Smearing of Rubber, J. Appl. Polym. Sci. 12 (1968) 281-293

11. Schallamach, A., Friction and abrasion of rubber, Wear 1 (1958) 384-417 12. Thavamani, P., Khastgir, D., Bhowmick, A. K., Microscopic studies on the

mechanisms of wear of NR, SBR and HNBR vulcanizates under different conditions, J. Mater. Sci. 28 (1993) 6318-6322

13. Uchiyama, Y., Ishino, Y., Pattern abrasion mechanisms of rubber, Wear 158, 141 (1992)

14. Southern, E., Thomas, A. G., Studies of rubber abrasion,, Rubber Chem. Technol. 52 (1979) 1008-1018

15. Zhang, S. W., Yang, Z., Energy theory of rubber abrasion by a line contact, Tribol. Int. 30 (1997) 839-843

16. Iwai, T., Uchiyama, Y., Shimosaka, K., Takase, K., Study on the formation of periodic ridges on the rubber surface by friction and wear monitoring, Wear 259 (2005) 669-675

10

17. Fukahori, Y., Yamazaki, H., Mechanism of rubber abrasion. I: Abrasion pattern formation in natural rubber vulcanizate, Wear 171 (1994) 195-202

18. Schallamach, A., Abrasion of rubber by a needle, J. Polymer Sci. 9 (1952) 385-404 19. Muhr, A. H., Pond, T. J., Thomas, A. G., Abrasion of rubber and the effect of

lubricants, J. Chim. Phys. 84 (1987) 331-334 20. Muhr, A. H., Roberts, A. D., Rubber abrasion and wear, Wear 158 (1992) 213-228 21. Chandrasekaran, M., Batchelor, A. W., In situ observation of sliding wear tests of

butyl rubber in the presence of lubricants in an X-ray microfocus instrument, Wear 211 (1997) 35-43

22. Torbacke, M., Johansson, A., Seal material and base fluid compatibility: An overview, J. Synthetic Lubrication 22 (2005) 123-142

23. Patil, A. O., Coolbaugh, T. S., A LITERATURE REVIEW WITH EMPHASIS ON OIL RESISTANCE, Rubber Chem. Technol. 78 (2005) 516-535

24. W. Hofmann, Hanser, “Rubber Technology Handbook “, Munich, 2001

Paper C

1

Influence of counterface topography on sliding friction and wear of some elastomers under dry sliding conditions

Mohammadreza Mofidi, Braham Prakash*

Division of Machine Elements, Luleå University of Technology, Luleå SE-971 87 Sweden

* Tel: +46 (0)920 493055, Fax: +46 (0)920 491047

_____________________________________________________________________

ABSTRACTIn this work, the friction and wear behaviour of acrylonitrile butadiene rubber (NBR), hydrogenated acrylonitrile butadiene rubber (HNBR), acrylate rubber (ACM) and fluoroelastomer (FKM) against steel surfaces under unidirectional dry sliding conditions have been studied. The influence of surface roughness of the steel counterface on friction and wear was studied by using a block on ring test configuration. At low load, the friction coefficient decreased after a running-in period and the wear was insignificant, especially for the acrylate rubber and fluoroelastomer. The running-in time in terms of achieving a stable dry friction for the different elastomers, from longest to shortest is: HNBR, NBR, FKM and ACM. An exception is FKM sliding against a smooth steel counterface. At higher contact pressure, powdery worn particles on the ACM and a decrease in friction coefficient were observed but for FKM and HNBR, worn particles with roll shapes were produced. The worn particles of FKM were significantly larger than those of the other tested materials and more severe wear was measured.

Keywords: Elastomer, Friction, Wear

1 INTRODUCTION Elastomers are characterised by low Young’s modulus, large elongation- to- break and high Poisson’s ratio. These properties make them suitable for various sealing applications. Friction and wear are two important factors in seal performance and the overall efficiency of the machine. Most seals operate in the presence of oils during their service life but at times, such as the starting up of the machine, they operate in dry conditions. Although such periods are short, the seal may exhibit high friction coefficient and wear. Wear may affect the sealing ability. Therefore, the friction and wear behaviour of seal material in dry conditions may play an important role in seal performance.

1.1 Friction The coefficient of friction of rubber during sliding against a hard counterface may be attributed to the contribution of adhesion, deformation (hysteresis), viscous and tearing components [1, 2]. However; some researchers considered only two terms of friction components. They considered that the tearing and viscous components can be represented by adhesive and deformation components respectively [2-4]. Adhesion is generally

2

recognized to consist of the making and breaking of junctions at a molecular level. Several theories have been proposed to describe the adhesion component of friction. These studies have shown that the adhesive contribution to friction of an elastomer against a hard surface decreases with decreasing Young’s modulus and it is a function of the viscoelastic properties of the elastomer which in turn depend on temperature and sliding velocity [2, 5].

The deformation component of friction is caused by the flowing action of the elastomer over the rigid asperities of the mating counterface and has been termed as hysteresis losses. The occurrence of hysteretic friction is a consequence of damping within the viscoelastic body [6]. Like the adhesion component of friction, the hysteretic component of friction is a function of viscoelastic properties of the elastomer. Unlike the adhesive friction, the hysteretic friction increases with the decrease in Young’s modulus of the elastomer [2]. If the applied pressure is high then the rubber is squeezed into complete contact with the substrate. The magnitude of the hysteretic component of friction depends on h/ (the ratio of asperity height amplitude to the wavelength of the roughness) [3]. Thus, if the ratio between the amplitude and wavelength is constant, the surface roughness of different length scales contributes equally to the friction force [3]. The minimum length scale which contributes to the friction force is limited by the microstructure of rubber and the contamination of contact area [4].

The contribution of adhesion and hysteresis to friction depends on the geometry and cleanliness of the mating surfaces. The adhesion component is dominant on very clean and smooth surfaces [7, 8]. It can also be significant at low loads, even in lubricated conditions [9] because of the significance of the attractive Van der Waals’ forces between the surfaces in comparison to the normal load [10].

1.2 Wear Wear of elastomers occurs as a result of two processes; local mechanical rupture (tearing) and decomposition of the molecular network to a low molecular weight (smearing) [11]. The mechanical rupture of rubber against smooth hard substrates can be due to fatigue or frictional wear. The strength of rubber has a considerable effect on the wear resistance. A critical value of shear stress can be defined for each rubber above which, roll formation occurs and below which wear is mainly due to fatigue. Thus the friction coefficient is one of the most important parameters governing the type of wear [12]. Fatigue wear as a result of repeated deformation cycles takes place when rubber slides against hard and blunt projections on the hard surface at low frictional force [13]. The surface of rubber which is worn by frictional wear or roll formation is characterized by ridges perpendicular to the direction of sliding but in fatigue wear the worn surface does not bear any visible ridges except pitting marks [14].

1.3 Running-in The determination of friction of rubber requires specification of the history of the two rubbing surfaces. Wear of rubber, through tearing or smearing, can result in changes in the geometry and properties of the surface contact area and consequently the friction force. When an unlubricated rubber slides against the same counterface repeatedly, a decrease in friction may occur until a complete layer of material is deposited from the rubber onto the opposing surface. However, when the rubber slides continuously against a fresh counterface, the friction force increases [15].

3

The metal may also wear during running in period. When a rubber slides against a metallic counterface, molecular segments of the freshly ruptured rubber may adhere to the metallic surface under the action of Van der Waals’ forces forming a lubricating layer of rubbery material. The free radicals of segments in the rubbery layer react with the metallic oxide surface and produce a metal oxide-polymer complex which is weaker than the metal oxide itself and detaches more easily from the surface [16]. This process results in wear of the metallic counter-surface and formation of a layer which has properties quite different from both the rubber and metal affecting the friction force.

In particular situations, such as starting periods, seals may operate in dry conditions which can affect their performance significantly. Understanding the tribological behaviour of seal materials during the initial start-up and run-in periods is relevant for predicting the performance of the seal. Regarding the application, the contact pressure between an elastomeric seal and sealing surface may vary from a few tens of KPa to a few tens of MPa [17]. The aim of this study is to investigate the friction and wear behaviour of four sealing elastomers in dry sliding conditions. The influences of surface topography of countersurface and contact pressure on the friction and wear characteristic during the run- in period have also been studied.

2 EXPERIMENTAL WORK The experiments were carried out using Micro-Tribometer UMT-2. A rubber specimen glued to a metal backing plate was pressed against a rotating steel ring counterface. The normal and frictional forces were recorded by piezoelectric sensors. The schematic of the test configuration is shown in Figure 1. Three sets of bearing steel rings with different ranges of surface roughness were used to study the effect of surface roughness on friction and wear. Figure 2 shows the typical surfaces of the three sets of the steel rings. Each test was run for 12 hours duration. The rubber specimen dimensions were 16 mm×4 mm×2 mm (the width of contact area was 4 mm). The rubber samples were cut from sheets of 2 mm thickness. The counterface bearing steel rings were of ø35mm (outer diameter) and 8 mm thick. The rubber specimens were washed in industrial petroleum for 3 minutes using an ultrasonic cleaner dried in an oven for 10 minutes at 40 ºC and then weighed. The same procedure was repeated after running the test on each specimen in order to quantify wear. Each ring was washed in industrial petroleum for 3 minutes by using the ultrasonic cleaner and dried before the test. A new ring was used for each test. All the tests were performed at room temperature (22 ± 2 ºC).

Figure 1: Test configuration

4

The elastomers studied during this work are commonly used seal materials, acrylonitrile butadiene rubber (NBR), hydrogenated acrylonitrile butadiene rubber (HNBR), acrylate rubber (ACM) and fluoroelastomer (FKM). All the elastomers have a module of elasticity of about 10 MPa at very low speed and room temperature. The nominal hardness, tensile strength, elongation at break and material densities of these elastomers are given in Table 1. The surfaces of the bearing steel rings were characterised by a Wyco 3D optical surface profilometer. Figure 2 shows the typical surface topographies of the three sets of rings and the range of average surface roughness (Ra).

Table 1: Tested elastomers and their properties

Elastomeric materials Hardness (Shore A)

Tensile strength(MPa)

Elongation at break (%)

Density (g/cm3)

Nitrile rubber (NBR 3143) 76.1 25.4 466 1.31

Hydrogenated nitrile rubber (HNBR 7611) 71.3 17.5 303 1.24

Acrylic rubber (ACM ) 73.4 7.8 171 1.49

Fluoro rubber (FKM 7327) 72.8 - - 2.03

Figure 2: Surface topographies of steel rings

The experiments were carried out at two contact pressures. The wear and frictional behaviour of the material (running-in and steady state friction) were studied. At low contact pressure the influence of surface roughness was also investigated. Using the Hertz contact theory, the contact pressure at low load (1.5 N) is estimated to be about 240 KPa which is of the order of the contact pressure on a new elastomeric lip seal. At higher load (10 N) a contact pressure of 750 KPa is expected which is in the range of the contact pressure between an elastomeric O-ring and an actuator rod. All the tests were performed twice.

3 RESULTS AND DISCUSSION

3.1 Friction Friction coefficients of the elstomers were affected by changes in the contact surface. These results are presented and discussed below.

Fine surface (0.15<Ra<0.3 μm) Fine surface (0.35<Ra<0.55 μm) Fine surface (0.5<Ra<0.7 μm)

5

3.1.1 Frictional behaviour at high contact pressure Figure 3 shows the worn surfaces of different elastomers tested at high contact pressure of 240 KPa. As shown in the figure, roll formation occurred on the surface of FKM and HNBR. Furthermore, a white powdery layer has been formed on the surface of ACM. The surface of NBR was torn locally but no roll formation occurred (see also Figure 4).

Figure 3: Worn surfaces of the tested rubbers (Normal load: 10 N, speed: 10 r.p.m, duration: 12 hours, surface roughness of rings: 0.35-0.55 )

Figure 4: Worn particles of FKM and HNBR

Figure 5 shows the friction coefficient of the tested rubbers against the bearing steel rings of medium roughness. The highest friction coefficient has been observed in tribological pairs involving FKM followed by HNBR which may, in part, be due to the energy dissipated in the tearing and roll formation. The friction coefficient of ACM increases gradually during the test which may be due to changes in the properties and/or the dimensions of the particles in the powdery layer between the surfaces.

6

Figure 5: Friction coefficient vs. time, (Normal load: 10 N, Speed: 10 r.p.m, Surface roughness, Ra: 0.35 - 0.55 )

3.1.2 Frictional behaviour at low contact pressure Figures 6-9 show the friction coefficients of NBR, HNBR, FKM and ACM respectively. The friction coefficients drop during running-in periods to steady state values with the exception for FKM sliding against fine counter-surface. The longest running-in periods have been observed during sliding against fine counter-surfaces. The shortest running-in time were observed for FKM. The decrease in friction coefficient during running-in periods may be due to the deposition of a layer of filler and low molecular components from the rubber onto the surface of the ring. Smearing as a result of the decomposition of the molecular network to low molecular components may be another reason for the decrease in friction coefficient.

Figure 6: Friction coefficient of NBR vs. time, (Normal load: 1.50 N, Speed: 10 r.p.m.)

7

Figure 7: Friction coefficient of HNBR vs. time, (Normal load: 1.50 N, Speed: 10 r.p.m.)

Figure 8: Friction coefficient of FKM vs. time, (Normal load: 1.50 N, Speed: 10 r.p.m.)

Figure 9: Friction coefficient of ACM vs. time, (Normal load: 1.50 N, Speed: 10 r.p.m.)

The average value of the friction coefficient was calculated from the steady state friction coefficient values during the last 5000 seconds of the test (Figure 10). The error bars in the figure indicate the difference between the results of repeated tests. The steady state values

8

of friction coefficients increase as the surface roughness decreases with exception for FKM. The surface roughness has the minimum and maximum effects on the steady state friction coefficient of ACM and HNBR respectively.

Figure 10: Steady state friction coefficient (average value of friction coefficient)

A low friction layer, including fine particles, has been formed on the sliding surface of ACM and the deformation of the surface of the rubber is more affected by the dimensions of the particles than the surface roughness of the ring. Thus the hysteresis component (which is the most dominant component of steady state friction) is not much affected by the roughness of the ring.

The steady state friction coefficient of FKM against the rough surface is higher than that on the surface with medium surface roughness. Further investigations, taking into account the various material properties and the surface roughness of the ring at nanoscale are required in order to clearly explain this effect.

3.2 Wear At low contact pressure, no severe wear occurred on the surface of elastomers and the lost mass of the tested elastomers were very small (less than 0.5 mg). Figure 11 shows the worn mass of the tested elastomers at high contact pressure. Roll formation occurred on the surface of FKM and HNBR resulting in severe wear; see Figure 4. Although, the roll formation occurred on the surface of HNBR, the amount and dimensions of worn particles are much smaller than those on FKM. Consequently the worn mass of HNBR was much lower than that of FKM. The surface of ACM was also worn and fine worn particles were produced on the sliding surface but the surface of NBR torn locally and the worn mass was low.

9

Figure 11: Worn mass of the tested rubbers (Normal load: 10 N, speed: 10 r.p.m, duration: 12 hours, surface roughness Ra: 0.35-0.55 )

It is important to keep in mind that conditions in the above tests and some sealing applications may differ and therefore the rating of the performance in terms of wear and friction could be different, e.g. a shear layer, such as observed in the ACM measurements, may not form in another environment, such as in presence of a lubricant.

4 CONCLUSION The frictional characteristics of four different seal materials (NBR, HNBR, ACM, and FKM) during sliding against steel countefaces of varying roughness values have been investigated. At low contact pressure, the results show that the friction coefficients drop during running-in periods to steady state values. The longest running-in periods were observed during sliding against smooth surfaces. Apart from FKM, the steady state friction coefficient increases as the surface roughness decreases. The surface roughness has the least and most effects on the steady state friction coefficient of ACM and HNBR respectively.

At high contact pressure, roll formation occurred on the surfaces of FKM indicating severe wear. The roll formation also occurred on the surface of HNBR but its worn mass was much lower than that of FKM. The worn particles of the ACM had a powdery form but the worn particles of FKM and HNBR were in roll shapes.

AcknowledgementsAll elastomeric materials and steel rings used in this work were supplied by Mr. Stellario Barbera (SKF Sealing Solutions, Italy) and Mr. Joop Vree (SKF Engineering Research Centre, The Netherlands) and we thankfully acknowledge their support. The authors express their gratitude to Dr. Richard Schaake (SKF ERC) for his useful suggestions concerning this manuscript. Finally, the authors also thank the SKF ERC management for their permission to publish this work.

REFERENCES1. Ludema, K. C., Physical factors in tyre friction, Physics in Technology 6 (1975),

p 11

10

2. Moore, D. F., The Friction and lubrication of elastomers, 1st ed., Pergamon Press, New York, 1972

3. Persson, B. N. J., Theory of rubber friction and contact mechanics, J. Chem. Phys. 115-8 (2001), p 3840-3861

4. Persson, On the nature of surface roughness with application to contact mechanics, sealing, rubber friction and adhesion, J. Phys.: Condens. Matter 17 (2005), R1–R62

5. Moore, D. F., A review of adhesion theories for elastomers, Wear 22 (1972), p 113-141

6. Moore, D. F., A review of hysteresis theories for elastomers, Wear 30 (1974), p 1-34

7. Persson, B. N. J., Volokitin, A. I., Rubber friction on smooth surfaces, Eur. Phys. J. E 21-1 (2006), p 69-80

8. Fuller, K. N. G., Tabor, D., The Effect of Surface Roughness on the Adhesion of Elastic Solids, Proceedings of the Royal Society of London. Series A, Mathematical and Physical Science 345- 1642 (Sep. 1975), p 327-342

9. Greenwood, J. A., Tabor, D., The Friction of Hard Sliders on Lubricated Rubber: The Importance of Deformation Losses, Proc. Phys. Soc. 71 (1958) p 989-1001

10. Johnson, K. L., Kendall, K., Roberts, A. D., Surface energy and the contact of Elastic Solids, Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, 324- 1558 (Sep. 1971), p 301-313

11. Gent, A. N., Pulford, Mechanisms of rubber abrasion, J. Appl. Polym. Sci. 28 (1983), p 943

12. James, D. I.., Jolley, M., E., Abrasion of rubber, MacLaren & Sons Ltd., London, 1964

13. Zhang, S.W., Tribology of Elastomers, 1st ed., Elsevier, Netherlands, 2004 14. Thavamani, P., Khastgir, D., Bhowmick, A. K., Microscopic studies on the

mechanisms of wear of NR, SBR, and HNBR vulcanizates under different conditions, J. Mater. Sci. 28 (1993), p 6318

15. Cooper, R. P., Ellis, B, The effect of Run-in on rubber friction, J. Appl. Polym. Sci. 27 (1982), p 4735

16. Ab-Malek, K., Stevenson, A., On the lubrication and wear of metal by rubber, J. Mater. Sci. 19 (1984), p 585

17. ASM Handbook, Vol. 18, Friction, lubrication, and wear technology, ASM

Documents

Tribology of Elastomers