Copyright © 2013 American Scientific Publishers All rights reserved Printed in the United States of America

Journal of Surfaces and Interfaces of Materials Vol. 1, 71–76, 2013

Friction Behavior of High Density Polyethylene (HDPE) Against 304L Steel: An Experimental Investigation of the Effects of Sliding Direction, Sliding History and Sliding Speed S. Dhouibi1 , M. Boujelbene1 2 , M. Kharrat1 3 ∗ , M. Dammak1 4 , and A. Maalej1 RESEARCH ARTICLE

1

Laboratoire des Systèmes Electromécaniques, Ecole Nationale d’Ingénieurs de Sfax, Route de Soukra km 3.5, B.P. 1173, 3038 Sfax, Tunisia 2 Institut Supérieur de Gestion Industrielle de Sfax, Route Mharza Km 1.5, B.P. 954, 3018 Sfax, Tunisia 3 Faculté des Sciences de Gafsa, Campus Universitaire Sidi Ahmed Zarrouk, 2112 Gafsa, Tunisia 4 Institut Préparatoire aux Etudes d’Ingénieurs de Sfax, Rte Menzel Chaker Km 0.5, B.P. 1172, 3018 Sfax, Tunisia Polymers integrated in mechanical systems are often the seat of friction and wear phenomena when they are in contact with metal counterparts. Experimental studies developed recently have shown that these phenomena depend on the experimental conditions such as the normal load, the contact temperature and the sliding speed. It was also found that these phenomena depend on the physical properties of the polymer surface such as roughness and molecular chains orientation. Our work consists in an experimental investigation the friction behavior of User High Density Polyethylene Delivered by PublishingofTechnology to: Guest (HDPE) cube following alternated friction against 304L stainless steel cylinder. IP: 41.229.113.130 On: Mon, 24 Feb 2014 10:36:04 The effects of the sliding direction (i.e., molecular orientation anisotropy), the Publishers sliding history and the test frequency Copyright: American Scientific have been considered in our experimental analysis. Friction tests between 304L steel cylinder and HDPE cube have been carried out in various directions using an experimental device developed in our laboratory. For fresh HDPE surfaces, we have found that the friction coefficient evolution with the sliding direction for a given number of cycles presents an elliptical feature. This friction anisotropy of the HDPE surface has been associated to the anisotropy of the polymer surface microstructure. Whatever the sliding direction is, the friction coefficient undergoes a progressive increase when the number of cycles increases until the end of the test. Friction of fresh HDPE surfaces in one direction affects the friction behavior in the other directions. From our experimental analysis, we can also outline that increasing the test frequency causes an appreciable increase of the friction coefficient for the whole test duration.

Keywords: High Density Polyethylene, HDPE, Stainless Steel, 304L, Friction, Sliding Direction, Sliding History, Sliding Speed.

1. INTRODUCTION The use of plastic materials as substitutes for metallic materials has been common in industry over the twenty last years. The advantages encouraging the use of plastic materials are their low cost, the easiness to be shaped, the low density and the high recyclable character if compared to metal materials. However, they cannot endure high compulsion. The polymers integrated in the mechanical systems are often the seat of friction and wear phenomena. Several works were developed for the analysis of wear and friction processes between polymers and metal ∗

Author to whom correspondence should be addressed.

J. Surf. Interfac. Mater. 2013, Vol. 1, No. 1

counterparts. It was shown that these processes depend on several parameters related to the experimental conditions such as the test temperature, the environment relative humidity, the sliding speed, the normal load    1–6 Experimental studies developed recently established correlations between the friction and wear characteristic of polymer materials and their surface physical properties such as roughness and molecular chains orientation.7–12 The effects of molecular orientation on the friction and wear properties of polymers have been the subject of some previous researches.9–12 The tribological behavior of a steel slider on PTFE polymer was experimentally examined by Tabor and Wynne Williams.9 The authors have found

2164-7542/2013/1/071/006

doi:10.1166/jsim.2013.1002

71

RESEARCH ARTICLE

Friction Behavior of High Density Polyethylene (HDPE) Against 304L Steel

that the friction is consistently higher when sliding occurs across the molecular chains than when sliding along them. Other authors have observed a directional dependence on the wear rate of ultra-high molecular weight polyethylene (UHMWPE). Laurent et al.10 have shown that cumulative wear during walking gait cycles in the medial compartment of knee implants, where the directionality of sliding is greater, is over 100% higher than in the lateral compartment despite similar loading on both components. Hip simulators are used to the understanding of the wear process of UHMWPE. Bragdon et al.11 have shown that physiological motion pathways produce very different wear rates and morphology of the wear surface than unidirectional reciprocating pathways. Wang12 developed a theoretical wear model based on the effective frictional work concept for UHMWPE in lubricated multi-directional sliding. This model provides a theoretical explanation for the effects of crosslinking and multi-directional motion on the wear of UHMWPE in total hip and total knee replacements. The objective of our paper is to investigate experimentally the effects of sliding direction (i.e., molecular orientation anisotropy), sliding history and test frequency on the tribological behavior of thermoplastic polymers following alternated friction against rigid antagonist.

Dhouibi et al.

Normal load Fn

Steel cylinder

HDPE Cube

Sinusoidal translation Fig. 1. Friction test configuration.

2.2. Materials In this study we have chosen for the thermoplastic poly2. MATERIALS AND METHODS Delivered by Publishing Technology to: Guest User High Density Polyethylene (HDPE). Cubic speciIP: 41.229.113.130 On: Mon,mer 24 aFeb 2014 10:36:04 Copyright: American Scientific Publishers mens with 45 × 45 × 15 mm3 were cut from an extruded 15 mm thick sheet. A cylindrical specimen, 70 mm in We have carried out friction tests between a steel cylindiameter and 50 mm in length, was machined from 304L der and a polymer cube in various directions using a wear stainless steel. The Young modulus of 304L is E = and friction device developed in our laboratory.6 The con210 GPa, the Poisson’s coefficient is  = 03 and the meatact between the steel cylinder and the flat polymer, which sured arithmetic roughness of the cylindrical surface is was located beneath the steel cylinder, was done under a Ra = 0125 m. constant applied normal load Fn (Fig. 1). The friction and wear device was regulated to generate a sinusoidal translation of the polymer cube with 10 mm amplitude while 2.3. Methodology the steel cylinder was kept fixed. The translation direction The considered sliding directions of the HDPE surface was perpendicular to the steel cylinder axis. Friction tests are the 0 direction which corresponds to the extrusion were performed under 500 N normal load and for maxidirection, the 45 direction compared to that of extrumum test duration of 10000 cycles. Two values of the test sion and the 90 direction (perpendicular to the extrusion frequency were considered in this study i.e., 0.1 Hz and direction). The friction experiments were conducted in two 1 Hz. Before testing, samples were cleaned with ethanol. All the experiments have been performed at ambient congroups. The first group was developed to compare the fricditions of temperature and humidity. They were carried out tion behavior for the three considered sliding directions under dry conditions. in the case of fresh (unrubbed) HDPE samples. For this Optical Profilometer “FOGALE Nanotech” was used to first group and for each of the considered sliding direction, characterize the surface topography of the polymer sample the friction coefficient is denoted N where N represents before and after friction tests. the number of sliding cycles. The second group of experiIn this study, an experimental machine LLOYD was ments aimed to specify the effect of sliding along another used to characterize the HDPE polymer behavior under direction (i.e., sliding history) on the friction behavior. The compression loading. The displacement speed of the crosscases are the following: bar was fixed to 5 mm/min. Cylindrical polymer specimens • Samples tested along the 0 direction have been already with standard dimensions, i.e., 10 mm in diameter and tested for 10000 cycles along the 90 direction. These test 15 mm in height, were machined from an extruded sheet. conditions are referred to the “90 + 0 ” direction. 2.1. Experimental Devices

72

J. Surf. Interfac. Mater. 1, 71–76, 2013

Dhouibi et al.

Friction Behavior of High Density Polyethylene (HDPE) Against 304L Steel

• Samples tested along the 45 direction have been already tested for 10000 cycles along the 0 direction. These test conditions are referred to the “0 + 45 ” direction. • Samples tested along the 90 direction have been already tested for 10000 cycles along the 0 direction. These test conditions are referred to the “0 + 90 ” direction.

Table I.

Mechanical properties of HDPE in the 0 and 90 directions. Elastic tangent modulus (MPa) Plastic flow stress (MPa) ± standard deviation ± standard deviation

0 direction 90 direction

808 ± 52 632 ± 36

23.1 ± 0.5 21.9 ± 0.3

3. RESULTS AND DISCUSSION

3.3. Friction Behavior of Fresh HDPE Surface

3.1. Mechanical Characterization of HDPE Polymer Using Compression Test

True stress (MPa)

Friction tests were conducted on fresh HDPE surfaces for 10 mm amplitude, 1Hz frequency, 500 N normal load and 10000 cycles. Evolutions of the friction coefficient with the In order to investigate the variation of mechanical propernumber of cycles for the 0 direction and the 90 direction ties of the HDPE surface with the considered directions, are reported in Figure 4. It can be seen for a given numuniaxial compression tests were performed on cylindrical ber of cycles that the measured friction coefficient for the   samples machined toward the 0 direction and the 90 90 direction is much higher than that measured for the 0 direction. Typical true stress/true strain curves are given in direction. The mean initial values of the friction coefficient Figure 2. The obtained values of elastic tangent modulus  for 304L/HDPE contact Delivered by plastic Publishing Userare 0.065 for the 0 direction and (slope at the origin of stress/strain curve) and flow Technology to: Guest  0.088 for the 90 direction. For both directions, the fric41.229.113.130 On: stress, which corresponds to theIP:point of intersection of Mon, 24 Feb 2014 10:36:04 tion coefficient undergoes a progressive increase when the Copyright: American Scientific Publishers the initial and final tangents of the stress/strain curve, are number of cycles increases until the end of the test. For a presented in Table I. It can be seen that the elastic modulus given number of cycles, experimental values of the friction is more important in the 0 direction (extrusion direction) coefficient measured for the three considered directions while the plastic flow stress is slightly the same for the (i.e., 0 , 45 and 90  can be extrapolated for the other two directions. directions. Results are given in Figure 5 which represents the friction coefficient evolution with the sliding direction for different number of cycles. The elliptical feature which 3.2. Characterization of the Fresh HDPE characterizes this evolution demonstrates the anisotropic Surface Topography character of the friction behavior of the HDPE surface. Typical roughness profiles with both 0 direction cut and As there is no significant difference in the topographical 90 direction cut are shown in Figure 3 for the fresh HDPE parameters between the 0 and the 90 directions, the friction anisotropy of the HDPE surface should be associated to the anisotropy of the polymer surface microstructure. 60 In fact, Briscoe and Stolarski13 observed that the friction 90º 0º and wear of PTFE and HDPE polymers were sensitive to the orientation of their molecular chains with sliding direc40 tion. In their opinion, the investigated polymers wear out by a creation of transferred films highly oriented in the direction of sliding and the sliding occurs between oriented 20 fibrils.

0 0.0

0.1

0.2 True strain

0.3

Fig. 2. Typical true stress/true strain curves for the uniaxial compression test in the 0 and 90 directions.

J. Surf. Interfac. Mater. 1, 71–76, 2013

3.4. Friction Behavior of HDPE Surfaces Having Already Been Tested Along Another Direction Friction tests were conducted on rubbed HDPE surfaces for 10 mm amplitude, 1 Hz frequency, 500 N normal load and 10000 cycles. The considered sliding directions were 73

RESEARCH ARTICLE

For the second group of friction experiments and for each of the considered sliding direction, the friction coefficient is denoted NF where N represents the number of sliding cycles and F indicates that the HDPE surface has been already rubbed for 10000 cycles along another direction.

surface. The analysis of these profiles provides the variation of surface texture between the considered two directions. The selected roughness parameters are arithmetic roughness Ra , square roughness Rq and total roughness Rt . The measured values of the three considered topographical parameters for the 0 direction and the 90 direction are summarized in Table II. We can conclude that the 0 direction is slightly rougher than the 90 direction.

Dhouibi et al.

RESEARCH ARTICLE

Friction Behavior of High Density Polyethylene (HDPE) Against 304L Steel

Fig. 3.

Delivered by Publishing Technology to: Guest User IP: 41.229.113.130 On: Mon, 24 Feb 2014 10:36:04  Typical profiles of roughness; (a): 0 direction, (b): 90 direction. Scientific Publishers Copyright: American

“90 + 0 ”, “0 + 45 ” and “0 + 90 ”. For a given number of cycles, experimental values of the friction coefficient measured for the three considered directions were extrapolated for the other directions. Results are given in Figure 6 which represents the friction coefficient evolution with the sliding direction for different number of cycles. We can conclude from these results that the friction of the fresh HDPE surface in one direction affects the friction behavior

in the others directions. Whatever the friction direction of the rubbed HDPE surface is, the friction coefficient undergoes a progressive increase with the number of cycles until we reach approximately the same value “0.176 ± 0.002” µ90º

0.2

µ45º

0.17

0.1

µ1 µ10 µ10000

0°

Friction coeffecient

90°

0.13

0 0.2

0.09

0

0.1

0.2

µ0º

0.1

0.2

0.05 0

2500

5000 Number of cycles

7500

10000

Fig. 4. Friction coefficient versus number of cycles for the 0 direction and the 90 direction; fresh HDPE surface, Fn = 500 N, 1 Hz.

74

0.1

Fig. 5. Friction coefficient versus sliding direction for different number of cycles; fresh HDPE surface, Fn = 500 N, 1 Hz (0 , 45 and 90 represent respectively the friction coefficient in the 0 , 45 and 90 sliding direction).

J. Surf. Interfac. Mater. 1, 71–76, 2013

Dhouibi et al.

Friction Behavior of High Density Polyethylene (HDPE) Against 304L Steel Table II.

µ“0º+90º”

0.2

Direction

µ“0º+45º” µ1F µ10F µ10000F

0.1

0.1

0

0.1

0 90

Table III.

µ“90º+0º”

0 0.2



0.2

0.1

0.2

after 10000 cycles. For this maximum number of cycles, the evolution of the friction coefficient with the sliding direction reaches a circular feature.

Ra (m)

Rq (m)

Rt (m)

3.2 ± 0.1 2.6 ± 0.1

3.8 ± 0.1 3.2 ± 0.1

16.8 ± 0.2 12.1 ± 0.1

Topographical parameters of rubbed HDPE surfaces.

Direction

Ra (m)

Rq (m)

Rt (m)

0 45 90 “90 + 0 ” “0 + 45 ” “0 + 90 ”

10.3 ± 1.3 12.8 ± 0.8 16.5 ± 1.5 5.4 ± 0.7 10.4 ± 0.8 5.1 ± 0.5

12.8 ± 1.9 15.6 ± 0.3 19.4 ± 1.8 7.2 ± 0.7 12.2 ± 0.9 6.5 ± 0.4

45.3 ± 2.6 60.4 ± 1.1 73.3 ± 2.1 16.8 ± 1.2 49.3 ± 1.9 15.6 ± 1.1

to move toward those measured for the fresh HDPE surface (Table II). 3.6. Effect of the Test Frequency on the Friction Behavior

Friction coefficient

In order to investigate the effect of the test frequency on the friction behavior, friction tests were performed on fresh HDPE surfaces for 0.1Hz frequency, 10 mm amplitude, 3.5. Characterization of the Rubbed HDPE Surface 500N normal load and 3000 cycles. Evolutions of the fricTopography Delivered by Publishing Technology to: Guest tion coefficient with User the number of cycles are reported in IP: 41.229.113.130 On: Mon, 24 Feb 2014 10:36:04  Figure 7 for the 0 sliding direction and for the two conTopographical parameters of the rubbed HDPE surfaces Copyright: American Scientific Publishers sidered test frequencies (i.e., 0.1 Hz and 1 Hz). It appears are listed in Table III for the different considered slidfrom this result that increasing the test frequency for one ing directions. It can be seen that the roughness paramedecade causes a considerable increase of the friction coefters measured for the 0 , 45 and 90 directions are much ficient for the whole test duration. This finding falls in higher than those measured for the fresh HDPE surface line with those of several researchers how reported that (Table II). From the results of Table III we can also conthe increased sliding speed had an apparent impact on the clude that the friction on a HDPE surface having already increase of friction coefficient.14 15 In their experimental been tested along other directions will reduce the surface study carried on polymer bearings made from different roughness. In fact the roughness parameters measured for plastic materials, Feyzullahoglu and Saffak15 found that the “90 + 0 ”, “0 + 45 ” and “0 + 90 ” directions seem increasing the sliding speed led the friction coefficient to be augmented. According to them, the increase of the sliding speed causes the increase of surfaces heat, which leads 1Hz 0.08 to the formation of a transfer film layer on the steel surface 0.1Hz and the escalation of friction coefficient.

4. CONCLUSIONS

0.07

0.06

0.05 0

1000

2000

3000

Number of cycles

Fig. 7. Friction coefficient versus number of cycles for two values of the test frequency; fresh HDPE surface, Fn = 500 N, 0 sliding direction.

J. Surf. Interfac. Mater. 1, 71–76, 2013

An experimental investigation of the friction behavior of HDPE cube following alternated friction against 304L steel cylinder has been developed in this paper. The effects of the sliding direction (i.e., molecular orientation anisotropy), the sliding history and the test frequency have been considered in the experimental analysis. For fresh HDPE surfaces, whatever the sliding direction is the friction coefficient undergoes a progressive increase when the number of cycles increases until the end of the test. The friction coefficient evolution with the sliding direction for 75

RESEARCH ARTICLE

Fig. 6. Friction coefficient versus sliding direction for different number of cycles; HDPE surfaces having already been tested along another direction, Fn = 500 N, 1 Hz (“90 +0 ” , “0 +45 ” and “0 +90 ” represent respectively the friction coefficient in the “90 + 0 ”, “0 + 45 ” and “0 + 90 ” sliding direction).

Topographical parameters of fresh HDPE surface.

RESEARCH ARTICLE

Friction Behavior of High Density Polyethylene (HDPE) Against 304L Steel

a given number of cycles presents an elliptical feature. As there is no significant difference in the topographical parameters between the different sliding directions, the friction anisotropy of the HDPE surface is associated to the anisotropy of the polymer surface microstructure. This result is supported by the difference of the measured HDPE Young modulus between the 0 direction (extrusion direction) and the 90 direction. The friction of fresh HDPE surfaces in one direction affects the friction behavior in the other directions. For the HDPE surface having already been tested along another direction, whatever the friction direction is, the friction coefficient undergoes a progressive increase with the number of cycles until we reach approximately the same value “0176 ± 0002” after 10000 cycles. For this maximum number of cycles, the evolution of the friction coefficient with the sliding direction reaches a circular feature. From the experimental analysis, it can also be outlined that increasing the test frequency causes an appreciable increase of the friction coefficient for the whole test duration.

Dhouibi et al.

topography and for the constructive discussions developed around this work.

References and Notes 1. M. Vaziri, R. T. Spurr, and F. H. Stott, Wear 122, 329 (1988). 2. C. H. Da Silva, D. K. Tanaka, and A. Sinatora, Wear 225–229, 339 (1999). 3. L. Tongsheng, C. Peihong, L. Xujun, T. Jiang, and X. Qunji, J. Mater. Sci. 35, 2597 (2000). 4. T. Schwenke, L. L. Borgstede, E. Schneider, T. P. Andriacchi, and M. A. Wimmer, Wear 259, 926 (2005). 5. M. Flannery, T. Mc Gloughlin, E. Jones, and C. Birkinshaw, Wear 265, 999 (2008). 6. S. Dhouibi, M. Boujelbene, M. Kharrat, M. Dammak, and A. Maalej, J. Mater. Sci. and Eng. C 29, 1521 (2009). 7. S. Bahadur and D. Tabor, Polymer Wear and Its Control, edited by L. H. Lee, ACS Symposium Series, Washington (1985), Vol. 287, pp. 253–268. 8. I. Sekiguchi, Y. Yamaguchi, K. Tamura, and A. Deguchi, Japanese J. Tribology 38, 1097 (1993). 9. D. Tabor and D. E. Wynne Williams, Wear 4, 391 (1961). 10. M. P. Laurent, T. S. Johnson, J. Q. Yao, C. R. Blanchard, and R. D. Crowninshield, Wear 255, 1101 (2003). 11. C. R. Bragdon, D. O. O’conner, J. D. Lowenstein, M. Jasty, and W. D. Syniuta, Proc. Instn. Mech. Engrs. 210, 157 (1996). 12. A. Wang, Wear 248, 38 (2001). 13. B. J. Briscoe and T. A. Stolarski, Wear 104, 121 (1985). 14. H. Unal and A. Mimaroglu, Mater. Design. 24, 183 (2003). 15. E. Feyzullahoglu and Z. Saffak, Mater. Design. 29, 205 (2008).

Acknowledgments: The authors would like to thank Dr. Antoine Chateauminois from the “Ecole Supérieure de Physicochimie Industrielle de la Ville de Paris” for his contribution in the characterization of the HDPE surface Delivered by Publishing Technology to: Guest User IP: 41.229.113.130 On: Mon, 24 Feb 2014 10:36:04 Received: 14 June 2011. Accepted: 30 August 2011. Copyright: American Scientific Publishers

76

J. Surf. Interfac. Mater. 1, 71–76, 2013