- Email: [email protected]
Improved friction and wear performance of micro dimpled ceramic-on-ceramic interface for hip joint arthroplasty
Author's Accepted Manuscript
Improved friction and wear performance of micro dimpled ceramic-on-ceramic interface for hip joint Arthroplasty Taposh Roy, Dipankar Choudhury, Subir Ghosh, Azuddin Bin Mamat, Belinda Pingguan-Murphy
www.elsevier.com/locate/ceramint
PII: DOI: Reference:
S0272-8842(14)01364-9 http://dx.doi.org/10.1016/j.ceramint.2014.08.123 CERI9102
To appear in:
Ceramics International
Received date: Revised date: Accepted date:
27 June 2014 27 August 2014 28 August 2014
Cite this article as: Taposh Roy, Dipankar Choudhury, Subir Ghosh, Azuddin Bin Mamat , Belinda Pingguan-Murphy, Improved friction and wear performance of micro dimpled ceramic-on-ceramic interface for hip joint Arthroplasty, Ceramics International, http: //dx.doi.org/10.1016/j.ceramint.2014.08.123 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Improved friction and wear performance of micro dimpled ceramic-on-ceramic interface for hip joint arthroplasty Taposh Roy1, Dipankar Choudhury2, 3, Subir Ghosh1, Azuddin Bin Mamat4 and Belinda Pingguan-Murphy1* 1
2
3
4
Department of Biomedical Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia; E-Mails: [email protected] (T.R.); [email protected] (S.G.) Faculty of Mechanical Engineering, Brno University of Technology, Technicka 2896/2, 616 69 Brno, Czech Republic; E-Mail: [email protected] CEITEC-Central European Institute of Technology, Brno University of Technology, Technicka 3058/10, 616 00 Brno, Czech Republic Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia; E-Mail: [email protected]
* Author to whom correspondence should [email protected]; Tel.: +603-79674491; Fax: +603-79674579.
be
addressed;
E-Mail:
Abstract: The purpose of this study is to investigate the tribological effect of microdimpled surface textures for application in ceramic-on-ceramic hip prostheses. A rectangular array of circular dimples was selected as the texture pattern. A CNC micro drilling machine was used to fabricate these dimples on a flat ceramic substrate. Tribology tests were performed by using a pin-on-disk method with selected hertzian contact pressures and speeds based on normal gait of a hip joint, and results compared with non-dimpled surface. A dimpled surface with large dimple diameters and a high dimple density (Ø400 µm and density 15%) showed significant tribological performance gains, including a nearly 22% friction and 53% wear reduction. Keywords: micro dimple; ceramic on ceramic hip joints; friction; wear 1. Introduction
Hip joint replacement is one of the most successful orthopaedic surgery procedures, and is a widely practised method for restoring mobility to hip joints. It is estimated that more than 0.6 million hip replacements are performed per year worldwide, and the number will increase 170% by the year 2030 [1]. Despite the excellent success of hip arthroplasty, it is still affected by a need for revision surgery. Therefore, there is scope to improve it, either from an engineering or patient point-of-view. In general, the durability of artificial hip joints (either metal-on-polyethylene or ceramic-onpolyethylene) is 15 years with active lifestyles, which may not be acceptable to patients under 60 years. These younger people make up 44% of overall osteoarthritis patients, and longer lasting hip prostheses would be much more suitable for them [2-5]. A ceramic-on-ceramic (CoC) hip prosthesis is a potential candidate to fulfil the demands of young patients; however, aseptic loosening and fracture rates are still concerns with their use [6-9]. Otherwise, ceramic hip joints have better mechanical properties, such as hardness, and a superb polishing capability, and better biocompatibility. Aseptic loosening is a complex mechanism which is mostly associated with an excessive wear rate and morphology of wear debris. The wear of the articulating surfaces and adverse tissue reactions to wear debris causes loosening and implant failure [10]. CoC hip joints have lower wear rates, compared to metal-on-metal and metal-on-polyethylene hip joints. Removal of micro or nano sized wear debris from the contact interface reduces further abrasive wear of contact interface and possible reaction with surrounding tissue [3, 11]. The application of a surface texture, such as micro-dimples or grooves, is already established as a way to improve the tribological performance of sliding surfaces. For example, dimpled golf balls have a much higher aerodynamic lift force [12]; textured
engine cylinders have lower friction thus are more effective in engine performance [1316]; and sliding bearings have a longer life with a well-defined micro dimple. Yu et al. [17] showed that micro-textured surface geometry is capable of producing a higher hydrodynamic pressure and thus increasing the load carrying capacity. Dimpled surfaces decrease the surface contact area, and thus the coefficient of friction deceases. Further, such dimples can trap wear debris, preventing wear of contacting surfaces from third body hard particles. Previous studies [4, 18-22] mentioned a thicker tribological film due to the micro dimples. The application of a micro dimpled surface texture to artificial hip joints has been investigated recently. A numerical model, established by Gao et al. [23], indicates that dimpled surfaces have potentially beneficial effects on the lubrication performance of metal-on-metal hip replacements, particularly under predominant boundary lubrication conditions. The experimental investigation conducted by Sawano et al. [24] with metalon-polyethylene demonstrated that a 1 µm deep micro dimple was successful at reducing the amount of wear by up to 61% compared to a polished surface. Ito et al. [11] conducted tribological tests to identify the effect of a dimple surface on metal (Co– Cr)-on-polyethylene (UHMWPE) hip joints on the hip simulator that represented the load and motion of hip joints. A significant percentage of friction (30%) and wear rate (68%) was reduced after 10 x 105 cycles with a well-designed dimpled surface (diameter of 0.5 mm, pitch of 1.2 mm, and depth of 0.1 mm). Choudhury et al. [25, 26] conducted both theoretical and experimental investigation of simulated metal-on-metal hip joints, and came to the conclusion that a well-defined honed surface has high potential for use as a hip joint interface, since it was found to increase hydrodynamic pressure, captured wear debris, and lower the friction coefficient.
Our previous study [27] also revealed that the friction coefficient can be significantly reduced by using a micro dimpled ceramic surface in simulated CoC hip joints. Two sets of dimple parameters (diameter and dimple density) were utilized and it focused mainly on the precision of micro tool induced dimples and their effects on the mechanical properties of the surfaces. The aim of the present study is to conduct a detailed tribological investigation (friction, wear and wear debris) on 3 different types of dimple array, and compare their performance to non-dimpled ceramic surface with selected loads and speed based simulated hip joints.
2. Material and Methods 2.1. Sample preparation A pin on disk experiment was conducted using a tribometer to measure the frictional coefficient, as it enables to measure frictional force. Modern hip simulators enable the replication of many operating parameters, including dynamic loading, multidirectional sliding direction, controlled temperature, and both multi-station and long-run programmes. However, none of them provides in-situ friction coefficients. Thus, most research into friction relies on the more conventional pin-on-disk tribometer method to understand the fundamental mechanisms of biotribology, such as friction coefficient and film thickness variation over time. For the present study, we have used a pin-on-disk tribometer to partially replicate a hip joint, in terms of contact pressure, speed and lubrication [27, 28]. The rectangular disk (dimension 15 × 15 × 6 mm3, 99.6% Al2O3) and the cylindrical pin (dimension Ø6.35 mm × L6 mm, 99.6% Al2O3) were prepared based on the specification of the tribometer. All the samples were polished in a grinder with different grades of diamond polycrystalline suspension to achieve a mirror surface finish.
2.2. Dimple fabrication The key parameters for this experiment are the dimple array patterns and the dimple profile, because it was previously reported that these parameters affect the tribology of a system [29-34]. 3 types of dimple array patterns were chosen for this study. The fixed variables in this study are the diameter of the dimple, the dimple depth, and the dimple pitch (centre distance between two dimples). Table 1 summarises the dimple parameters. CATIA V5 design software was used to draw the dimple array patterns. According to these designs, a CNC micro drilling machine (Mikrotools DT110, Singapore) was programmed. For this study, diamond drill bits (UKAM Industrial Superhard Tools, U.S.) were used to produce the micro dimples via a drilling machine. The polished disk samples were placed inside the CNC micro drilling machine. For effective machining, the spindle speed and feed rate were set at 55,000 rpm and 334 mm/min respectively, as per supplier recommendations. To preserve the life of the drill bit, three steps were used (10µm feed for each steps), to produce a total dimple depth of 30 µm.
2.3. Surface roughness The surface roughness of the samples was measured by a 3D optical profiler (Alicona Infinite Focus, Chicago, USA). To measure surface roughness, a random surface area was chosen. An average of 10 readings were collected from each sample. For the dimpled surface, the roughness was measured between dimples.
2.4. Dimple measurement The dimple diameter, pitch and depth were measured by a 3D optical profiler. An average of 10 readings were collected from each sample. Field emission scanning
electron microscopy (FESEM; AURIGA, Zeiss Singapore) was used to look at the dimple morphology.
2.5. Micro hardness, residual stress and wettability A Vickers micro indenter (HVS-1000) was used to measure the micro hardness of the dimpled and non-dimpled surface area. Seven selected areas were indented for each sample. On the dimpled surface, each selected area was further divided in 3 zones, viz. A- very near to the edge of the dimple; B- 50 µm from the edge of the dimple; and C100 µm from the edge of the dimple (Figure 1). Applied load for the indentation was set to 980.7 mN for a duration of 10 sec. The average of the data was used to measure the micro hardness of the samples. Residual stresses are the stresses that remain in a solid material after the original cause of the stress has been removed. Many researchers have reported the feasibility of X-ray diffraction method to measure the residual stress of materials [35-38]. In our study, we used an X-ray diffractometer (XRD: PA Nalytical Empyrean) to measure the residual stress of the samples generated at the time of the dimple fabrication process. The xrd-sin2
technique was employed using CuKα radiation (0.1540598 nm) at 40 kV
and 40 mA. Different tilt angles ( ) were employed ranging from 0º to 40º by a computer controlled Omega-goniometer. The wettability measurement determines the hydrophobic or hydrophilic property of the sample surface. Wettability was measured by water contact angle measurement with a goniometer (OCA15EC, Dataphysics Instrument, Germany). The static sessile drop method was used with a single 2µl drop of water on the surface. High resolution cameras and software were used to capture and analyse the contact angle. For dimpled surfaces, the drop was placed between the dimples and all the experiments were
repeated 10 times, with the average value of the readings showing the wettability of the surface.
2.6. Contact pressure estimation Hertzian contact theory was used to calculate the maximum contact pressure [39, 40] for cylindrical and planar contact as follows:
Pmax =
2F πbL
(1)
Where, Pmax is the maximum contact pressure, F is the normal load applied to the surface, L is the length of contact, and b is the half width of the rectangular contact area which is calculated as follows:
b=
⎡1 − ϑ12 1 − ϑ22 ⎤ + 4F ⎢ E 2 ⎥⎦ ⎣ E1 ⎛ 1 1 ⎞ πL ⎜ + ⎟ ⎝ R1 R 2 ⎠
(2)
ϑ 1, 2, E1, 2 and R1, 2 are the Poisson’s ratio, elastic modulus and radius of curvature (for planer surface R= α) of the two contacting materials respectively. For calculating maximum contact pressure, the mean elastic modulus of Al2O3 was determined as 375 GPa (for disk and pin), poisson’s ratio for the ceramic material 0.22 (for disk and pin), and radius of the pin 3.175 mm. The radius of the disk specimen was infinite (as it was a plane surface). The calculated result is listed in Table 2.
2.7. Friction and wear measurement
The tribology test was conducted using a tribometer (TR 283 Series, DUCOM, Bangalore, India) with a modified flat-on-line contact configuration. Reciprocating housing driven with an electrical motor provided the reciprocating sliding motion. The
form of sliding speed was rectangular. A load sensor attached to the specimen holder was used to measure the frictional force. For data acquisition, WinDucom software was installed on the computer. The test sequence on the machine, based on temperature and test duration, frequency and amplitude was controlled by the WinDucom software. The pin sample (Al2O3 rod) was installed into a metallic holder, and restricted to zero degrees of freedom. The centre of the pin holder was loaded in a direction normal to the Al2O3 disk located beneath the pins. The pin was moving along in the reciprocating direction in which the disk was fixed. Figure 2 shows the experimental setup. The experimental parameters were selected based on a simulated hip joint, defined in terms of contact pressure, speed, and lubrication. Table 2 lists the experimental conditions for each type of test. Applied loads were calculated with the simulated contact pressure of the hip joint [41, 42]. The stroke length was fixed and the frequency, of 5, 10, 15 or 20 Hz, gave the sliding speed at 20, 40, 60, 80 mm/sec, respectively. Each test was carried out for 180 minutes and an average friction coefficient was calculated for every 900 cycles. Immediately before testing, each sample (pin and disk) was put into an ultrasonic cleaner in iso-propanol for 10 min to remove any debris. Prior to and after testing, the cleaned disks were dried and allowed to stand in an environment with controlled humidity and temperature for 48 h, after which they were weighed to an accuracy of ±10µg. Each disk was weighed five times and the average value taken, allowing the weight loss over a period of testing to be calculated. This weight loss was taken as wear of the samples. Finally an FESEM analysis was carried out to check for any cracking on the sample surfaces due to sliding and wear debris characterization. For these experiments, 25% bovine serum (HCL#SV30160.03; HyClone Fetal Bovine
Serum) was used as the lubricant. All the tests were carried out while the samples were flooded with lubricant and each test was repeated three times.
3. Results and Discussions 3.1. Surface roughness
The measured average surface roughness (Ra) of the samples was 0.12 ± 0.02 µm, indicating the prepared samples had a smoother surface than other engineering surfaces, where dimpled textures were utilized [27]. There was no shoulder or bulge on the edge of dimple as we conducted the polishing after the fabrication of dimple. However, the overall roughness is high compare to a standard ceramic hip prosthesis (20-30 nm). Polishing took place using a 30 μm diamond grinding disc, followed by 9 μm, 6 μm, 1 μm and 0.05 μm diamond polycrystalline suspensions in an individual polishing cloth, attaining 120 nm, being limited by the great hardness of the material. This roughness is in line with that used by Gispert et al.[43], who used UHMWPE (Ra=2.121 µm), AISI 316L (Ra=0.240 µm), CoCrMo (Ra=0.317 µm), and Al2O3 (Ra=0.111 µm) in their research. Since the contact is ‘hard-on-hard’ interface, a portion of roughness will be reduced due to the elastic deformation of the material under loading conditions. Moreover, Vrbka et al.[44] noted that 25% bovine serum can yield an average of a 200 nm thick film in a CoC hip joint interface.
3.2. Dimple profile
Figure 3 displays a 2-D FESEM image and a typical 3-D image of a dimple. Using a 3D profilometer, dimple diameter, depth and pitch were measured. Table 3 summarises the measured and desired values for each. In general, the variation between the
measured and desired values is not more than 5%. This confirms that the CNC micro drilling machine can be used to produce micro dimples with accuracy.
3.3 Micro hardness, residual stress and wettability
The average micro hardness for the non-dimpled area was measured at 18.93 GPa and the average micro hardness for the dimple area was found to be slightly lower, at 18.06 GPa for dimples with 300 µm diameter, and 18.10 GPa for dimples with 400 µm diameter. The micro hardness very near to the dimple (Figure 1 position A) was a little lower compared to 100 µm apart from the dimple edge (Figure 1 position C). For the dimple surface with 300 µm diameter dimples, the micro hardness calculated at positions A, B and C was 17.5, 17.8 and 18.8 GPa respectively. This could be explained as a change in the grain micro structure due to the fabrication process, and the geometric effect of the dimple. The geometric effect is that a 300 µm deep hole is created on the surface so the area around to the hole might be weaker than the other surface. However the average micro hardness was nearly the same as the non-dimpled surface. The residual stress on the dimple surfaces was produced by the micro drilling fabrication process. The residual stress was found to be compressive stress. For the dimple surface with 300 µm and 400 µm dimple diameters, it was 276 MPa (compressive) and 153 MPa (compressive) respectively. Compressive residual stress on the surfaces is beneficial for wear resistance of the samples [45]. The wettability, in terms of water contact angle, for non-dimpled surfaces, was 59º±3 and for the dimpled surface was 61º±3 (Ø300 µm) and 62º±3 (Ø400 µm). The contact angle difference between dimpled and non-dimple was not significant. This shows that the samples are hydrophilic. The dimple fabrication did not alter the chemical properties.
3.4 Tribology testing
3.3.1 Effect of load with friction coefficient The friction coefficient is the one of the most fundamental tribological outputs, and a lower friction coefficient is desirable for artificial hip joints. The present study revealed significant differences in friction coefficient profiles among the experimental specimens at different loads and speeds. Figure 4 shows the friction coefficient profiles of the experimental surfaces at contact pressures of 181.45, 222.23, and 256.61 MPa respectively, applying normal loads of 10, 15 and 20 N. Velocity was maintained at medium walking speed (20 mm/s)[41, 42]. Overall, all the dimple samples exhibited lower friction coefficient profiles compared to non-dimpled samples (Figure 4). By the end of 54,000 cycles, under normal load of 10 N, it was clearly demonstrated that the friction coefficient is affected by the dimple density. Both D300-P15 and D400-P15 samples show the lowest friction coefficient, which is 0.11(D300-P15) and 0.1(D400-P15) respectively (Figure 4a). Both samples have a similar dimple density. These results suggest an effect of dimple area ratio on the friction coefficient behaviour. However, at the higher loads of 15N and 20N the friction coefficient was greatly reduced by the sample having a bigger dimple diameter and density. D400-P15, which has the biggest dimple diameter and highest dimple density produced the lowest friction coefficient (Figure 4 b & c). It is notable that all samples exhibited similar trends in response to load at both 15N and 20N (Figure 4 (b &c)), compared to non-dimpled surfaces. Overall, the non-dimpled samples produced the highest friction, and the value increased with the applied load [29, 46]. With the applied load of 10N, the initial average friction coefficient was 0.108, and at the end of the cycle it was 0.132 for non-
dimpled samples. The increment of friction at the end of the cycle was 22%. For the same experimental condition, sample D400-P15 showed an increment of only 4.8%. Similarly, other dimple samples showed small increments. This shows that samples with dimples have a better adaptability under changing load. It is worth noting that the contact pressure in hip joints is dynamic, and depends on bodyweight and gait. Therefore, it can be speculated that samples with dimpled surfaces are capable of providing lower and steadier friction coefficients in dynamic loading conditions in hip joints. Our previous study [27] also revealed the importance of dimpled surfaces. However, its effectiveness was limited to lower loads (10 N) because of the lower area ratio of dimpling (density 10%). As each of the dimples has the capability to generate hydrodynamic pressure, a higher area ratio dimple surface certainly has a higher hydrodynamic pressure under sliding conditions, which contributes to lowering the friction coefficient. 3.3.2 Effect of speed on frictional behaviors The effect of sliding velocity on frictional behaviour was investigated. 15N was chosen as the fixed load and the tribometer was set at different sliding velocities, such as 20, 40, 60 and 80 mm/s. Figure 5 shows dependence between sliding speed and the mean value of the coefficient of friction, such that dimpled samples reduced friction coefficients. The confidence intervals were visible (p value was <0.05). In a typical Stribeck diagram the coefficient of friction decreases as a function of velocity, reaches a minimum and increases thereafter. The region to the right of the minimum in the friction force curve corresponds to flood lubrication. However in our investigation it was found the maximum friction was at the higher speed and minimum friction was at the lower speed, which is similar to the investigation by Liu et al. [47]. In all cases the
non-dimpled surface showed the maximum friction coefficient and the dimpled surface with a higher area density (area density of 15%) showed the minimum friction coefficient. In figure 5 (a) & (b) a gradual increment of friction coefficient at the sliding speed 20 to 60mm/sec was demonstrated, followed by a sharp increment at 80mm/sec. This happened because the bovine serum was squeezed out from the contact area of the non-dimpled interface due to high speed. Generally, the friction coefficient does not differ with a change in speed under hydrocarbon lubrication [48].
However, the
lubrication mechanism of bovine serum is completely different from hydrocarbon lubrication. For example, Myant et al.[49] revealed that the film thickness of bovine serum decreased with speed because the strength of affinity of protein compositions of bovine serum become weak under higher speed. We consider a similar mechanism happened in our experiment, and so the friction coefficient increased with the sliding speed. However, in case of the dimpled surface, supplementary bovine serum was delivered from the dimple, thus friction coefficient was relatively stable. Adhesive force also can be the reason behind increase of friction coefficient. However results thus verified again that the dimple surface has great potential for reducing friction. These results reconfirmed the importance of dimple parameters, such as dimple diameter and dimple density. Clearly, a bigger dimple diameter and higher dimple density reduces friction coefficient. Samples with a larger dimple diameter have a smaller surface contact area, whereas samples with a higher dimple density have more dimples, meaning more lubricant can be trapped in the dimples. This indirectly helps in lubricating the samples. The benefit of having micro-dimples on the sample surface is obvious, and this is dependent on dimple diameter and dimple density.
3.3.3 Wear and surface conditions Wear is the main concern for artificial hip joints, and a reduction of wear is the most desirable tribological outcome of a hip prosthesis. A dimpled surface with a higher area ratio reduces friction coefficient significantly, however, it would not be wise to utilize it as a frictional material unless wear resistance is assured. Wear is a complex mechanism and it is partially related to friction coefficient. Moreover, a higher wear rate refers to a damaged surface, which actually increases friction coefficient to some extent. In this experiment, we considered weight loss of Al2O3 disk only for better consistency. It is to be noted that the weight loss was measured after completing the altogether 162,000 cycles under the 3 types of applied load at 20mm/sec speed ― 54,000 cycles for each of 10, 15, and 20 N loads; total= 3x 54,000 cycles. The experimental test of the weight loss of the Al2O3 disk is presented in the Figure 6. D400-P15 has the lowest wear and is followed by D300-P15, D300-P10, D400-P10, D300-P5, D400-P5 and the non-dimpled surfaces. The difference between the highest and lowest wear is 53% — which demonstrates improved wear by the micro dimpled surface technique. This result was similar to the trend of the experimental work of Sawano et al. [3] and Ito et al. [11]. Both of them investigated on ‘micro dimpled metal’-on-polyethylene whereas the present study focused on ‘micro dimpled CoC’ interface. A different result was revealed by Zhou et al. [50], which demonstrated an increased wear rate by using a dimpled surface; the main limitations of that study were the large diameter of the dimple and the low area ratio. The present study also exhibited a low effectiveness of a 5% density dimpled interface. FESEM image analysis was conducted after the friction tests to investigate the worn surface and size and shape of wear debris due to sliding of the test specimens. Figure 7
demonstrate wear signs in formation of transfer film (material transfer) and grain pullout of the dimpled surface. Among them, the material transfers were identified as much higher incidents compared to grain pull-out. The formation of a transfer film during sliding contact of ceramic materials has been reported by other literature [51-53]. Remarkably, the dimples were found without any major damage because of the high hardness of Al2O3, unlike other published results [30]. Thus, it can be predicted that a dimpled profile on Al2O3 will last longer and perform their tribological endeavours. Along with wear, the size and the shape of wear debris are important factors in the durability of the prosthesis. Cell proliferation, differentiation, and prostanoid are affected by the size, shape, and chemical composition of the particles. Figure 8 shows the morphology of wear debris filtered from used lubricant from the tribology test―the wear debris generated from the non-dimpled specimens is typically bigger in size than that from dimpled specimens. The cause of the smaller size wear debris with the dimpled surface can be described by surface continuity [54]. For the non-dimpled surface, the contact area throughout the sliding time was found to be continuous. For the dimpled surface, due to fabricating dimples, the contact area was not continuous. This may be the cause of bigger wear debris for non-dimple surfaces compared to those of the dimpled surface. However, this is a preliminary and initial assessment of wear debris, and so the results are not yet conclusive in their detail. Nine et al. [55] concluded that nano-sized wear debris is more reactive to the surrounding tissue. Since the average debris size is greater than 500 µm, it should be less harmful to the surrounding tissues; however the toxicity of nano sized debris is not only size depended but also dose dependant. For example, the higher concentrated (dose, mg/mm3) debris yields higher biological response to the surrounding tissue [55]. To this extent, the dimpled interface
is also more reliable for artificial hip joints because it yields a lower number of debris due to low wear rate (Figure 6), and a major portion of debris becomes entrapped inside the dimple [3, 56] .
4 Conclusions:
The tribological effects of various geometric parameters with different dimple densities and diameters were investigated using a reciprocating tribometer under bovine serum lubrication. The comparison between dimpled and non-dimpled samples revealed that dimpled samples reduced both the frictional coefficient and the wear rate in a simulated hip joint condition. The friction test revealed that the measured coefficient of friction for a dimpled surface with dimple diameter of 400 µm, density of 15% and depth of 30 µm lowered the maximum coefficient of friction by 22%, and wear by 53%. The frictional coefficient varied with respect to velocity and load. At 15 N load, the dimpled surfaces lowered the maximum coefficient of friction with compare to nondimpled surface. The wear debris were found to be in smaller in size for dimple surface with compared to the non-dimpled surface. Therefore, after considering the coefficient of friction, wear, and wear debris, we conclude that the micro dimpled surface (profile: Ø400µm, depth 30µm and density 15%) has an improved tribological performance when compared to the non-dimpled polished surface.
Acknowledgements
The research was supported by Faculty of Engineering, University of Malaya, from research grant (UMRG; RG147-12AET), Ministry of Higher Education High Impact research grant (UM.C/HIR/MOHE/ENG/44), Malaysia, and excellent young researcher project (CZ.1.07/2.3.00/30.0039), Brno University of Technology.
Conflicts of Interest
The authors declare no conflict of interest. References [1] S. Kurtz, K. Ong, E. Lau, F. Mowat, M. Halpern, Projections of Primary and Revision Hip and Knee Arthroplasty in the United States from 2005 to 2030, The Journal of Bone & Joint Surgery 89 (4) (2007) 780‐785. [2] B. AS, Hip new world, ASME Mech Eng 128 (2006) 28‐33. [3] H. Sawano, S. Warisawa, S. Ishihara, Study on long life of artificial joints by investigating optimal sliding surface geometry for improvement in wear resistance, Precis Eng 33 (4) (2009) 492‐498. [4] L. Mattei, F. Di Puccio, B. Piccigallo, E. Ciulli, Lubrication and wear modelling of artificial hip joints: A review, Tribology International 44 (5) (2011) 532‐549. [5] S. Ghosh, D. Choudhury, N.S. Das, B. Pingguan‐Murphy, Tribological role of synovial fluid compositions on artificial joints — a systematic review of the last 10 years, Lubrication Science (2014) n/a‐n/a. [6] A.S. Shanbhag, J.J. Jacobs, T.T. Glant, J.L. Gilbert, J. Black, J.O. Galante, Composition and morphology of wear debris in failed uncemented total hip replacement, Journal of Bone & Joint Surgery, British Volume 76 (1) (1994) 60‐67. [7] A. De Aza, J. Chevalier, G. Fantozzi, M. Schehl, R. Torrecillas, Crack growth resistance of alumina, zirconia and zirconia toughened alumina ceramics for joint prostheses, Biomaterials 23 (3) (2002) 937‐945. [8] J. Fisher, A. Galvin, J. Tipper, T. Stewart, M. Stone, E. Ingham, Comparison of the Functional Biological Activity and Osteolytic Potential of Ceramic on Ceramic and Cross Linked Polyethylene Bearings in the Hip, Bioceramics and Alternative Bearings in Joint Arthroplasty (2005) 21‐24. [9] H.S.A. Rahman, D. Choudhury, N.A.A. Osman, H.N. Shasmin, W. Abas, In vivo and in vitro outcomes of alumina, zirconia and their composited ceramic‐on‐ceramic hip joints, Journal of the Ceramic Society of Japan 121 (4) (2013) 382‐387. [10] D. Dowson, C. McNie, A. Goldsmith, Direct experimental evidence of lubrication in a metal‐on‐metal total hip replacement tested in a joint simulator, Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science 214 (1) (2000) 75‐ 86. [11] H. Ito, K. Kaneda, T. Yuhta, I. Nishimura, K. Yasuda, T. Matsuno, Reduction of polyethylene wear by concave dimples on the frictional surface in artificial hip joints, The Journal of Arthroplasty 15 (3) (2000) 332‐338. [12] P. Bearman, J. Harvey, Golf ball aerodynamics, (1975). [13] G. Ryk, I. Etsion, Testing piston rings with partial laser surface texturing for friction reduction, Wear 261 (7–8) (2006) 792‐796. [14] H. Ogihara, T. Kido, H. Yamada, M. Murata, S. Kobayashi, Technology for reducing engine rubbing resistance by means of surface improvement, Honda R&D Technical Review 12 (2) (2000) 93‐98. [15] I. Etsion, Y. Kligerman, A. Shinkarenko, Improving tribological performance of piston rings by partial surface texturing, Transaction of the ASME, Journal of Tribology 123 (2005) 632‐638. [16] J. Keller, V. Fridrici, P. Kapsa, S. Vidaller, J.F. Huard, Influence of Material Nature and Surface Texturing on Wear of Heavy‐Duty Diesel Engine Cylinder Liners, Tribology Transactions 52 (1) (2008) 121‐126.
[17] H. Yu, X. Wang, F. Zhou, Geometric Shape Effects of Surface Texture on the Generation of Hydrodynamic Pressure Between Conformal Contacting Surfaces, Tribology Letters 37 (2) (2010) 123‐130. [18] V. Marian, Lubrication of textured surfaces, in: International Conference" UNIVERSITARIA ROPET 2002", Mechanical Engineering, Petrosani, 17‐19 Octomber 2002, Vol. 2, 2006. [19] I. Křupka, M. Hartl, The effect of surface texturing on thin EHD lubrication films, Tribology international 40 (7) (2007) 1100‐1110. [20] I. Křupka, M. Hartl, Effect of Surface Texturing on Very Thin Film EHD Lubricated Contacts, Tribology Transactions 52 (1) (2008) 21‐28. [21] I. Krupka, P. Svoboda, M. Hartl, Effect of surface topography on mixed lubrication film formation during start up under rolling/sliding conditions, Tribology International 43 (5) (2010) 1035‐1042. [22] D. Choudhury, H.A. Ching, A.B. Mamat, J. Cizek, N.A.A. Osman, M. Vrbka, M. Hartl, I. Krupka, Fabrication and characterization of DLC coated micro dimples on hip prosthesis heads, Journal of Biomedical Materials Research: Part B - Applied Biomaterials (In press) (2014). [23] L. Gao, P. Yang, I. Dymond, J. Fisher, Z. Jin, Effect of surface texturing on the elastohydrodynamic lubrication analysis of metal‐on‐metal hip implants, Tribology International 43 (10) (2010) 1851‐1860. [24] E.A. Swanson, T.P. Schmalzried, F.J. Dorey, Activity Recommendations After Total Hip and Knee Arthroplasty: A Survey of the American Association for Hip and Knee Surgeons, The Journal of Arthroplasty 24 (6, Supplement) (2009) 120‐126. [25] D. Choudhury, R. Walker, T. Roy, S. Paul, R. Mootanah, Performance of honed surface profiles to artificial hip joints: An experimental investigation, International Journal of Precision Engineering and Manufacturing 14 (10) (2013) 1847‐1853. [26] D. Choudhury, R. Walker, A. Shirvani, R. Mootanah, U. Chelmsford, The Influence of Honed Surfaces on Metal‐on‐Metal Hip Joints, Tribology Online 8 (3) (2013) 195‐202. [27] T. Roy, D. Choudhury, A. Bin Mamat, B. Pingguan‐Murphy, Fabrication and characterization of micro‐dimple array on Al2O3 surfaces by using a micro‐tooling, Ceramics International 40 (1) (2014) 2381‐2388. [28] D. Choudhury, T. Roy, I. Krupka, M. Hartl, R. Mootanah, Tribological investigation of ultra‐ high molecular weight polyethylene against advanced ceramic surfaces in total hip joint replacement, Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology (2014). [29] A. Amanov, I. Cho, Y. Pyoun, C. Lee, I. Park, Micro‐dimpled surface by ultrasonic nanocrystal surface modification and its tribological effects, Wear 286 (2012) 136‐144. [30] A. Amanov, S. Sasaki, A Study On The Tribological Characteristics Of Duplex–Treated Ti– 6al–4V Alloy Under Oil–Lubricated Sliding Conditions, Tribology International (2013). [31] P. Andersson, J. Koskinen, S.e. Varjus, Y. Gerbig, H. Haefke, S. Georgiou, B. Zhmud, W. Buss, Microlubrication effect by laser‐textured steel surfaces, Wear 262 (3) (2007) 369‐379. [32] V. Brizmer, Y. Kligerman, I. Etsion, A laser surface textured parallel thrust bearing, Tribology Transactions 46 (3) (2003) 397‐403. [33] J.W. Byun, H.S. Shin, M.H. Kwon, B.H. Kim, C.N. Chu, Surface texturing by micro ECM for friction reduction, International Journal of Precision Engineering and Manufacturing 11 (5) (2010) 747‐753. [34] M. Cho, S. Park, Micro CNC surface texturing on polyoxymethylene (POM) and its tribological performance in lubricated sliding, Tribology International 44 (7) (2011) 859‐867.
[35] M.M. Arafat, A.S.M.A. Haseeb, B. Dinan, S.A. Akbar, Stress enhanced TiO2 nanowire growth on Ti–6Al–4V particles by thermal oxidation, Ceramics International 39 (6) (2013) 6517‐ 6526. [36] Z.A. Khan, M. Hadfield, Manufacturing induced residual stress influence on the rolling contact fatigue life performance of lubricated silicon nitride bearing materials, Materials & Design 28 (10) (2007) 2688‐2693. [37] Z.A. Khan, M. Hadfield, S. Tobe, Y. Wang, Ceramic rolling elements with ring crack defects—A residual stress approach, Materials Science and Engineering: A 404 (1–2) (2005) 221‐226. [38] Z.A. Khan, M. Hadfield, S. Tobe, Y. Wang, Residual stress variations during rolling contact fatigue of refrigerant lubricated silicon nitride bearing elements, Ceramics International 32 (7) (2006) 751‐754. [39] H. Hertz, On the contact of elastic solids, J. Reine. angew. Math (92) (1982) 156–171. [40] N. Tay, M. Minn, S. Sinha, A Tribological Study of SU‐8 Micro‐Dot Patterns Printed on Si Surface in a Flat‐on‐Flat Reciprocating Sliding Test, Tribology Letters 44 (2) (2011) 167‐176. [41] M.R. Widmer, M. Heuberger, J. Vörös, N.D. Spencer, Influence of polymer surface chemistry on frictional properties under protein‐lubrication conditions: implications for hip‐ implant design, Tribology Letters 10 (1‐2) (2001) 111‐116. [42] S. Mischler, A.I. Muñoz, Wear of CoCrMo alloys used in metal‐on‐metal hip joints: A tribocorrosion appraisal, Wear (2012). [43] M.P. Gispert, A.P. Serro, R. Colaço, B. Saramago, Friction and wear mechanisms in hip prosthesis: Comparison of joint materials behaviour in several lubricants, Wear 260 (1–2) (2006) 149‐158. [44] M. Vrbka, I. Krˇupka, M. Hartl, T. Návrat, In situ measurements of thin films in bovine serum lubricated contacts using optical interferometry, Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine (2014) 0954411913517498. [45] N.M. Alanazi, A. El‐Sherik, S.H. Alamar, S. Shen, Influence of Residual Stresses on Corrosion and Wear Behavior of Electrodeposited Nanocrystalline Cobalt‐Phosphorus Coatings, International Journal of Electrochemical Science 8 (8) (2013). [46] M. Wakuda, Y. Yamauchi, S. Kanzaki, Y. Yasuda, Effect of surface texturing on friction reduction between ceramic and steel materials under lubricated sliding contact, Wear 254 (3) (2003) 356‐363. [47] H. Liu, R. Niu, Y. Meng, The Effect of Laser Texturing of Steel Surfaces on Film Lubriction Based on Stribeck Curves, in: Advanced Tribology, Springer, 2010, pp. 685‐687. [48] E. Rabinowicz, Friction and wear of materials, (1965). [49] C. Myant, R. Underwood, J. Fan, P. Cann, Lubrication of metal‐on‐metal hip joints: the effect of protein content and load on film formation and wear, Journal of the mechanical behavior of biomedical materials 6 (2012) 30‐40. [50] X. Zhou, A.L. Galvin, Z. Jin, X. Yan, J. Fisher, The influence of concave dimples on the metallic counterface on the wear of ultra‐high molecular weight polyethylene, Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 226 (6) (2012) 455‐462. [51] D.A. Rigney, L.H. Chen, M.G.S. Naylor, A.R. Rosenfield, Wear processes in sliding systems, Wear 100 (1–3) (1984) 195‐219. [52] M. Kerridge, J. Lancaster, The stages in a process of severe metallic wear, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 236 (1205) (1956) 250‐264. [53] E. Rabinowicz, D. Tabor, Metallic transfer between sliding metals: an autoradiographic study, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 208 (1095) (1951) 455‐475.
[54] B. Bhushan, Modern Tribology Handbook, Two Volume Set, CRC Press, 2010. [55] M.J. Nine, D. Choudhury, A.C. Hee, R. Mootanah, N.A.A. Osman, Wear Debris Characterization and Corresponding Biological Response: Artificial Hip and Knee Joints, Materials 7 (2) (2014) 980‐1016. [56] M. Tarabolsi, T. Klassen, F. Mantwill, F. Gärtner, F. Siegel, A.P. Schulz, Patterned CoCrMo and Al2O3 surfaces for reduced free wear debris in artificial joint arthroplasty, Journal of Biomedical Materials Research Part A 101 (12) (2013) 3447‐3456.
List of Figures:
Figure 1: a) FESEM micrograph of dimpled surface with defined 3 zones for measuring hardness tests; b) 3-D optical image of a dimple on dimpled surface; c) dimple topography of selected area on the dimpled surface. Figure 2: a) Image of reciprocating friction and wear machine; b) schematic diagram of loading and motion configuration of experimental setup. Figure 3: a) 3D image of dimpled surface produced by 3D optical profilometer b) 2D image of dimpled surface produced by FESEM Figure 4: Frictional coefficient profile at sliding speed 20 mm/s. a) With 10 N load, b) with 15 N load, and c) with 20 N load Figure 5: Friction coefficient produced by a) samples with dimple diameter of 300 µm and b) samples with dimple diameter of 400 µm. Error bar shows the standard deviation of friction coefficient for different speed Figure 6: Comparison of mass loss for different samples. Error bar shows the standard deviation of the weight loss Figure 7: FESEM images of the dimpled surface after tribology testing. a) Image of wear track on the dimpled sample, b) image of wear track near to the dimple after tribology test. Figure 8: FESEM images of the collected wear debris after tribology testing. a) Wear debris non-dimpled sample, b) wear debris from D300-P15, c) wear debris of D400-P15
List of Tables:
Table 1: Dimple parameters Table 2: Tribology testing conditions Table 3: Comparison of dimple parameters of setting and measured values (all dimensions are in µm, ± shows the standard deviation)
Tables: Table 1: Dimple parameters
Samples
Diamet er, Ø (µm)
Nondimple 1: D300P5 2 :D300P10 3: D300P15 4: D400P5 5: D400P10 6: D400P15
0 300
De Aspect pth, h ratio, λ=D/h (µm) 0 30
Pit ch ,(µm) 0
0.1
100
Dimple density, A* (%) -
Total no. of dimples 0
5
121
10
225
15
324
5
49
10
100
15
144
0 300
30
0.1 550
300
30
0.1
400
30
0.075
400
30
0.075
400
30
0.075
400 180 0 120 0 900
Table 2: Tribology testing conditions No. 1 2
Load (N) 10 15
3
20
Speed Max. contact (mm/s) (MPa) 20 181.45 20, 40, 60, 222.23 80 20 256.61
pressure
Lubricant Bovine serum
Table 3: Comparison of dimple parameters of setting and measured values (all dimensions are in µm, ± shows the standard deviation) Sample
D300-P5 D300-P10 D300-P15 D400-P5 D400-P10
Diameter (setting value) 300 300 300 400 400
Diameter (measured value) 311.61±6. 939 311.12±5. 172 310.11±3. 825 408.70±2. 870 409.87±2.
Depth (setting value) 30 30 30 30 30
Depth (measured value) 32.97±1 .34 33.80±1 .60 32.80±0 .93 34.52±2 .11 34.18±1
Pitch (setting value) 1000 550 400 1800 1200
Pitch (measured value) 1021.63±25. 26 536.95±20.2 6 413.05±19.6 6 1816.83±24. 50 1211.91 ±
D400-P15
Figures:
400
836 409.20±3. 842
30
.27 33.52±1 .70
900
4.87 915.87±14.3 5
Figure 1: a) FESEM micrograph of dimpled surface with defined 3 zones for measuring hardness tests; b) 3-D optical image of a dimple on dimpled surface; c) dimple topography of selected area on the dimpled surface.
Figure 2: a) Image of reciprocating friction and wear machine; b) schematic diagram of loading and motion configuration of experimental setup.
Fig gure 3: a) 3D D image of dimpled d surfa face producedd by 3D optiical profilomeeter b) 2D imagee of dimpled surface s produuced by FESE EM
Fiigure 4: Fricctional coeffficient profile at slidingg speed 20 mm/s. a) With 10 N load, b) with 15 N load, and c) c with 20 N load
Fig gure 5: Friction coefficiennt produced byy a) samples with w dimple ddiameter of 300 µm and b) sam mples with dimple d diameeter of 400 µm. Error baar shows thee standard deeviation of frictio on coefficientt for different speed.
Fiigure 6: Com mparison of mass loss for f different samples. Errror bar show ws the standd deviaation of weigght losses.
a)
b)
Figure 7: FESEM images of the dimpled surface after tribology testing. a) Image of wear track on the dimpled sample, b) image of wear track near to the dimple after tribology test.
Figure 8: FESEM images of the collected wear debris after tribology testing. a) Wear debris non-dimpled sample, b) wear debris from D300-P15, c) wear debris of D400-P15
Improved friction and wear performance of micro dimpled ceramic-on-ceramic interface for hip joint Arthroplasty Taposh Roy, Dipankar Choudhury, Subir Ghosh, Azuddin Bin Mamat, Belinda Pingguan-Murphy
www.elsevier.com/locate/ceramint
PII: DOI: Reference:
S0272-8842(14)01364-9 http://dx.doi.org/10.1016/j.ceramint.2014.08.123 CERI9102
To appear in:
Ceramics International
Received date: Revised date: Accepted date:
27 June 2014 27 August 2014 28 August 2014
Cite this article as: Taposh Roy, Dipankar Choudhury, Subir Ghosh, Azuddin Bin Mamat , Belinda Pingguan-Murphy, Improved friction and wear performance of micro dimpled ceramic-on-ceramic interface for hip joint Arthroplasty, Ceramics International, http: //dx.doi.org/10.1016/j.ceramint.2014.08.123 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Improved friction and wear performance of micro dimpled ceramic-on-ceramic interface for hip joint arthroplasty Taposh Roy1, Dipankar Choudhury2, 3, Subir Ghosh1, Azuddin Bin Mamat4 and Belinda Pingguan-Murphy1* 1
2
3
4
Department of Biomedical Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia; E-Mails: [email protected] (T.R.); [email protected] (S.G.) Faculty of Mechanical Engineering, Brno University of Technology, Technicka 2896/2, 616 69 Brno, Czech Republic; E-Mail: [email protected] CEITEC-Central European Institute of Technology, Brno University of Technology, Technicka 3058/10, 616 00 Brno, Czech Republic Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia; E-Mail: [email protected]
* Author to whom correspondence should [email protected]; Tel.: +603-79674491; Fax: +603-79674579.
be
addressed;
E-Mail:
Abstract: The purpose of this study is to investigate the tribological effect of microdimpled surface textures for application in ceramic-on-ceramic hip prostheses. A rectangular array of circular dimples was selected as the texture pattern. A CNC micro drilling machine was used to fabricate these dimples on a flat ceramic substrate. Tribology tests were performed by using a pin-on-disk method with selected hertzian contact pressures and speeds based on normal gait of a hip joint, and results compared with non-dimpled surface. A dimpled surface with large dimple diameters and a high dimple density (Ø400 µm and density 15%) showed significant tribological performance gains, including a nearly 22% friction and 53% wear reduction. Keywords: micro dimple; ceramic on ceramic hip joints; friction; wear 1. Introduction
Hip joint replacement is one of the most successful orthopaedic surgery procedures, and is a widely practised method for restoring mobility to hip joints. It is estimated that more than 0.6 million hip replacements are performed per year worldwide, and the number will increase 170% by the year 2030 [1]. Despite the excellent success of hip arthroplasty, it is still affected by a need for revision surgery. Therefore, there is scope to improve it, either from an engineering or patient point-of-view. In general, the durability of artificial hip joints (either metal-on-polyethylene or ceramic-onpolyethylene) is 15 years with active lifestyles, which may not be acceptable to patients under 60 years. These younger people make up 44% of overall osteoarthritis patients, and longer lasting hip prostheses would be much more suitable for them [2-5]. A ceramic-on-ceramic (CoC) hip prosthesis is a potential candidate to fulfil the demands of young patients; however, aseptic loosening and fracture rates are still concerns with their use [6-9]. Otherwise, ceramic hip joints have better mechanical properties, such as hardness, and a superb polishing capability, and better biocompatibility. Aseptic loosening is a complex mechanism which is mostly associated with an excessive wear rate and morphology of wear debris. The wear of the articulating surfaces and adverse tissue reactions to wear debris causes loosening and implant failure [10]. CoC hip joints have lower wear rates, compared to metal-on-metal and metal-on-polyethylene hip joints. Removal of micro or nano sized wear debris from the contact interface reduces further abrasive wear of contact interface and possible reaction with surrounding tissue [3, 11]. The application of a surface texture, such as micro-dimples or grooves, is already established as a way to improve the tribological performance of sliding surfaces. For example, dimpled golf balls have a much higher aerodynamic lift force [12]; textured
engine cylinders have lower friction thus are more effective in engine performance [1316]; and sliding bearings have a longer life with a well-defined micro dimple. Yu et al. [17] showed that micro-textured surface geometry is capable of producing a higher hydrodynamic pressure and thus increasing the load carrying capacity. Dimpled surfaces decrease the surface contact area, and thus the coefficient of friction deceases. Further, such dimples can trap wear debris, preventing wear of contacting surfaces from third body hard particles. Previous studies [4, 18-22] mentioned a thicker tribological film due to the micro dimples. The application of a micro dimpled surface texture to artificial hip joints has been investigated recently. A numerical model, established by Gao et al. [23], indicates that dimpled surfaces have potentially beneficial effects on the lubrication performance of metal-on-metal hip replacements, particularly under predominant boundary lubrication conditions. The experimental investigation conducted by Sawano et al. [24] with metalon-polyethylene demonstrated that a 1 µm deep micro dimple was successful at reducing the amount of wear by up to 61% compared to a polished surface. Ito et al. [11] conducted tribological tests to identify the effect of a dimple surface on metal (Co– Cr)-on-polyethylene (UHMWPE) hip joints on the hip simulator that represented the load and motion of hip joints. A significant percentage of friction (30%) and wear rate (68%) was reduced after 10 x 105 cycles with a well-designed dimpled surface (diameter of 0.5 mm, pitch of 1.2 mm, and depth of 0.1 mm). Choudhury et al. [25, 26] conducted both theoretical and experimental investigation of simulated metal-on-metal hip joints, and came to the conclusion that a well-defined honed surface has high potential for use as a hip joint interface, since it was found to increase hydrodynamic pressure, captured wear debris, and lower the friction coefficient.
Our previous study [27] also revealed that the friction coefficient can be significantly reduced by using a micro dimpled ceramic surface in simulated CoC hip joints. Two sets of dimple parameters (diameter and dimple density) were utilized and it focused mainly on the precision of micro tool induced dimples and their effects on the mechanical properties of the surfaces. The aim of the present study is to conduct a detailed tribological investigation (friction, wear and wear debris) on 3 different types of dimple array, and compare their performance to non-dimpled ceramic surface with selected loads and speed based simulated hip joints.
2. Material and Methods 2.1. Sample preparation A pin on disk experiment was conducted using a tribometer to measure the frictional coefficient, as it enables to measure frictional force. Modern hip simulators enable the replication of many operating parameters, including dynamic loading, multidirectional sliding direction, controlled temperature, and both multi-station and long-run programmes. However, none of them provides in-situ friction coefficients. Thus, most research into friction relies on the more conventional pin-on-disk tribometer method to understand the fundamental mechanisms of biotribology, such as friction coefficient and film thickness variation over time. For the present study, we have used a pin-on-disk tribometer to partially replicate a hip joint, in terms of contact pressure, speed and lubrication [27, 28]. The rectangular disk (dimension 15 × 15 × 6 mm3, 99.6% Al2O3) and the cylindrical pin (dimension Ø6.35 mm × L6 mm, 99.6% Al2O3) were prepared based on the specification of the tribometer. All the samples were polished in a grinder with different grades of diamond polycrystalline suspension to achieve a mirror surface finish.
2.2. Dimple fabrication The key parameters for this experiment are the dimple array patterns and the dimple profile, because it was previously reported that these parameters affect the tribology of a system [29-34]. 3 types of dimple array patterns were chosen for this study. The fixed variables in this study are the diameter of the dimple, the dimple depth, and the dimple pitch (centre distance between two dimples). Table 1 summarises the dimple parameters. CATIA V5 design software was used to draw the dimple array patterns. According to these designs, a CNC micro drilling machine (Mikrotools DT110, Singapore) was programmed. For this study, diamond drill bits (UKAM Industrial Superhard Tools, U.S.) were used to produce the micro dimples via a drilling machine. The polished disk samples were placed inside the CNC micro drilling machine. For effective machining, the spindle speed and feed rate were set at 55,000 rpm and 334 mm/min respectively, as per supplier recommendations. To preserve the life of the drill bit, three steps were used (10µm feed for each steps), to produce a total dimple depth of 30 µm.
2.3. Surface roughness The surface roughness of the samples was measured by a 3D optical profiler (Alicona Infinite Focus, Chicago, USA). To measure surface roughness, a random surface area was chosen. An average of 10 readings were collected from each sample. For the dimpled surface, the roughness was measured between dimples.
2.4. Dimple measurement The dimple diameter, pitch and depth were measured by a 3D optical profiler. An average of 10 readings were collected from each sample. Field emission scanning
electron microscopy (FESEM; AURIGA, Zeiss Singapore) was used to look at the dimple morphology.
2.5. Micro hardness, residual stress and wettability A Vickers micro indenter (HVS-1000) was used to measure the micro hardness of the dimpled and non-dimpled surface area. Seven selected areas were indented for each sample. On the dimpled surface, each selected area was further divided in 3 zones, viz. A- very near to the edge of the dimple; B- 50 µm from the edge of the dimple; and C100 µm from the edge of the dimple (Figure 1). Applied load for the indentation was set to 980.7 mN for a duration of 10 sec. The average of the data was used to measure the micro hardness of the samples. Residual stresses are the stresses that remain in a solid material after the original cause of the stress has been removed. Many researchers have reported the feasibility of X-ray diffraction method to measure the residual stress of materials [35-38]. In our study, we used an X-ray diffractometer (XRD: PA Nalytical Empyrean) to measure the residual stress of the samples generated at the time of the dimple fabrication process. The xrd-sin2
technique was employed using CuKα radiation (0.1540598 nm) at 40 kV
and 40 mA. Different tilt angles ( ) were employed ranging from 0º to 40º by a computer controlled Omega-goniometer. The wettability measurement determines the hydrophobic or hydrophilic property of the sample surface. Wettability was measured by water contact angle measurement with a goniometer (OCA15EC, Dataphysics Instrument, Germany). The static sessile drop method was used with a single 2µl drop of water on the surface. High resolution cameras and software were used to capture and analyse the contact angle. For dimpled surfaces, the drop was placed between the dimples and all the experiments were
repeated 10 times, with the average value of the readings showing the wettability of the surface.
2.6. Contact pressure estimation Hertzian contact theory was used to calculate the maximum contact pressure [39, 40] for cylindrical and planar contact as follows:
Pmax =
2F πbL
(1)
Where, Pmax is the maximum contact pressure, F is the normal load applied to the surface, L is the length of contact, and b is the half width of the rectangular contact area which is calculated as follows:
b=
⎡1 − ϑ12 1 − ϑ22 ⎤ + 4F ⎢ E 2 ⎥⎦ ⎣ E1 ⎛ 1 1 ⎞ πL ⎜ + ⎟ ⎝ R1 R 2 ⎠
(2)
ϑ 1, 2, E1, 2 and R1, 2 are the Poisson’s ratio, elastic modulus and radius of curvature (for planer surface R= α) of the two contacting materials respectively. For calculating maximum contact pressure, the mean elastic modulus of Al2O3 was determined as 375 GPa (for disk and pin), poisson’s ratio for the ceramic material 0.22 (for disk and pin), and radius of the pin 3.175 mm. The radius of the disk specimen was infinite (as it was a plane surface). The calculated result is listed in Table 2.
2.7. Friction and wear measurement
The tribology test was conducted using a tribometer (TR 283 Series, DUCOM, Bangalore, India) with a modified flat-on-line contact configuration. Reciprocating housing driven with an electrical motor provided the reciprocating sliding motion. The
form of sliding speed was rectangular. A load sensor attached to the specimen holder was used to measure the frictional force. For data acquisition, WinDucom software was installed on the computer. The test sequence on the machine, based on temperature and test duration, frequency and amplitude was controlled by the WinDucom software. The pin sample (Al2O3 rod) was installed into a metallic holder, and restricted to zero degrees of freedom. The centre of the pin holder was loaded in a direction normal to the Al2O3 disk located beneath the pins. The pin was moving along in the reciprocating direction in which the disk was fixed. Figure 2 shows the experimental setup. The experimental parameters were selected based on a simulated hip joint, defined in terms of contact pressure, speed, and lubrication. Table 2 lists the experimental conditions for each type of test. Applied loads were calculated with the simulated contact pressure of the hip joint [41, 42]. The stroke length was fixed and the frequency, of 5, 10, 15 or 20 Hz, gave the sliding speed at 20, 40, 60, 80 mm/sec, respectively. Each test was carried out for 180 minutes and an average friction coefficient was calculated for every 900 cycles. Immediately before testing, each sample (pin and disk) was put into an ultrasonic cleaner in iso-propanol for 10 min to remove any debris. Prior to and after testing, the cleaned disks were dried and allowed to stand in an environment with controlled humidity and temperature for 48 h, after which they were weighed to an accuracy of ±10µg. Each disk was weighed five times and the average value taken, allowing the weight loss over a period of testing to be calculated. This weight loss was taken as wear of the samples. Finally an FESEM analysis was carried out to check for any cracking on the sample surfaces due to sliding and wear debris characterization. For these experiments, 25% bovine serum (HCL#SV30160.03; HyClone Fetal Bovine
Serum) was used as the lubricant. All the tests were carried out while the samples were flooded with lubricant and each test was repeated three times.
3. Results and Discussions 3.1. Surface roughness
The measured average surface roughness (Ra) of the samples was 0.12 ± 0.02 µm, indicating the prepared samples had a smoother surface than other engineering surfaces, where dimpled textures were utilized [27]. There was no shoulder or bulge on the edge of dimple as we conducted the polishing after the fabrication of dimple. However, the overall roughness is high compare to a standard ceramic hip prosthesis (20-30 nm). Polishing took place using a 30 μm diamond grinding disc, followed by 9 μm, 6 μm, 1 μm and 0.05 μm diamond polycrystalline suspensions in an individual polishing cloth, attaining 120 nm, being limited by the great hardness of the material. This roughness is in line with that used by Gispert et al.[43], who used UHMWPE (Ra=2.121 µm), AISI 316L (Ra=0.240 µm), CoCrMo (Ra=0.317 µm), and Al2O3 (Ra=0.111 µm) in their research. Since the contact is ‘hard-on-hard’ interface, a portion of roughness will be reduced due to the elastic deformation of the material under loading conditions. Moreover, Vrbka et al.[44] noted that 25% bovine serum can yield an average of a 200 nm thick film in a CoC hip joint interface.
3.2. Dimple profile
Figure 3 displays a 2-D FESEM image and a typical 3-D image of a dimple. Using a 3D profilometer, dimple diameter, depth and pitch were measured. Table 3 summarises the measured and desired values for each. In general, the variation between the
measured and desired values is not more than 5%. This confirms that the CNC micro drilling machine can be used to produce micro dimples with accuracy.
3.3 Micro hardness, residual stress and wettability
The average micro hardness for the non-dimpled area was measured at 18.93 GPa and the average micro hardness for the dimple area was found to be slightly lower, at 18.06 GPa for dimples with 300 µm diameter, and 18.10 GPa for dimples with 400 µm diameter. The micro hardness very near to the dimple (Figure 1 position A) was a little lower compared to 100 µm apart from the dimple edge (Figure 1 position C). For the dimple surface with 300 µm diameter dimples, the micro hardness calculated at positions A, B and C was 17.5, 17.8 and 18.8 GPa respectively. This could be explained as a change in the grain micro structure due to the fabrication process, and the geometric effect of the dimple. The geometric effect is that a 300 µm deep hole is created on the surface so the area around to the hole might be weaker than the other surface. However the average micro hardness was nearly the same as the non-dimpled surface. The residual stress on the dimple surfaces was produced by the micro drilling fabrication process. The residual stress was found to be compressive stress. For the dimple surface with 300 µm and 400 µm dimple diameters, it was 276 MPa (compressive) and 153 MPa (compressive) respectively. Compressive residual stress on the surfaces is beneficial for wear resistance of the samples [45]. The wettability, in terms of water contact angle, for non-dimpled surfaces, was 59º±3 and for the dimpled surface was 61º±3 (Ø300 µm) and 62º±3 (Ø400 µm). The contact angle difference between dimpled and non-dimple was not significant. This shows that the samples are hydrophilic. The dimple fabrication did not alter the chemical properties.
3.4 Tribology testing
3.3.1 Effect of load with friction coefficient The friction coefficient is the one of the most fundamental tribological outputs, and a lower friction coefficient is desirable for artificial hip joints. The present study revealed significant differences in friction coefficient profiles among the experimental specimens at different loads and speeds. Figure 4 shows the friction coefficient profiles of the experimental surfaces at contact pressures of 181.45, 222.23, and 256.61 MPa respectively, applying normal loads of 10, 15 and 20 N. Velocity was maintained at medium walking speed (20 mm/s)[41, 42]. Overall, all the dimple samples exhibited lower friction coefficient profiles compared to non-dimpled samples (Figure 4). By the end of 54,000 cycles, under normal load of 10 N, it was clearly demonstrated that the friction coefficient is affected by the dimple density. Both D300-P15 and D400-P15 samples show the lowest friction coefficient, which is 0.11(D300-P15) and 0.1(D400-P15) respectively (Figure 4a). Both samples have a similar dimple density. These results suggest an effect of dimple area ratio on the friction coefficient behaviour. However, at the higher loads of 15N and 20N the friction coefficient was greatly reduced by the sample having a bigger dimple diameter and density. D400-P15, which has the biggest dimple diameter and highest dimple density produced the lowest friction coefficient (Figure 4 b & c). It is notable that all samples exhibited similar trends in response to load at both 15N and 20N (Figure 4 (b &c)), compared to non-dimpled surfaces. Overall, the non-dimpled samples produced the highest friction, and the value increased with the applied load [29, 46]. With the applied load of 10N, the initial average friction coefficient was 0.108, and at the end of the cycle it was 0.132 for non-
dimpled samples. The increment of friction at the end of the cycle was 22%. For the same experimental condition, sample D400-P15 showed an increment of only 4.8%. Similarly, other dimple samples showed small increments. This shows that samples with dimples have a better adaptability under changing load. It is worth noting that the contact pressure in hip joints is dynamic, and depends on bodyweight and gait. Therefore, it can be speculated that samples with dimpled surfaces are capable of providing lower and steadier friction coefficients in dynamic loading conditions in hip joints. Our previous study [27] also revealed the importance of dimpled surfaces. However, its effectiveness was limited to lower loads (10 N) because of the lower area ratio of dimpling (density 10%). As each of the dimples has the capability to generate hydrodynamic pressure, a higher area ratio dimple surface certainly has a higher hydrodynamic pressure under sliding conditions, which contributes to lowering the friction coefficient. 3.3.2 Effect of speed on frictional behaviors The effect of sliding velocity on frictional behaviour was investigated. 15N was chosen as the fixed load and the tribometer was set at different sliding velocities, such as 20, 40, 60 and 80 mm/s. Figure 5 shows dependence between sliding speed and the mean value of the coefficient of friction, such that dimpled samples reduced friction coefficients. The confidence intervals were visible (p value was <0.05). In a typical Stribeck diagram the coefficient of friction decreases as a function of velocity, reaches a minimum and increases thereafter. The region to the right of the minimum in the friction force curve corresponds to flood lubrication. However in our investigation it was found the maximum friction was at the higher speed and minimum friction was at the lower speed, which is similar to the investigation by Liu et al. [47]. In all cases the
non-dimpled surface showed the maximum friction coefficient and the dimpled surface with a higher area density (area density of 15%) showed the minimum friction coefficient. In figure 5 (a) & (b) a gradual increment of friction coefficient at the sliding speed 20 to 60mm/sec was demonstrated, followed by a sharp increment at 80mm/sec. This happened because the bovine serum was squeezed out from the contact area of the non-dimpled interface due to high speed. Generally, the friction coefficient does not differ with a change in speed under hydrocarbon lubrication [48].
However, the
lubrication mechanism of bovine serum is completely different from hydrocarbon lubrication. For example, Myant et al.[49] revealed that the film thickness of bovine serum decreased with speed because the strength of affinity of protein compositions of bovine serum become weak under higher speed. We consider a similar mechanism happened in our experiment, and so the friction coefficient increased with the sliding speed. However, in case of the dimpled surface, supplementary bovine serum was delivered from the dimple, thus friction coefficient was relatively stable. Adhesive force also can be the reason behind increase of friction coefficient. However results thus verified again that the dimple surface has great potential for reducing friction. These results reconfirmed the importance of dimple parameters, such as dimple diameter and dimple density. Clearly, a bigger dimple diameter and higher dimple density reduces friction coefficient. Samples with a larger dimple diameter have a smaller surface contact area, whereas samples with a higher dimple density have more dimples, meaning more lubricant can be trapped in the dimples. This indirectly helps in lubricating the samples. The benefit of having micro-dimples on the sample surface is obvious, and this is dependent on dimple diameter and dimple density.
3.3.3 Wear and surface conditions Wear is the main concern for artificial hip joints, and a reduction of wear is the most desirable tribological outcome of a hip prosthesis. A dimpled surface with a higher area ratio reduces friction coefficient significantly, however, it would not be wise to utilize it as a frictional material unless wear resistance is assured. Wear is a complex mechanism and it is partially related to friction coefficient. Moreover, a higher wear rate refers to a damaged surface, which actually increases friction coefficient to some extent. In this experiment, we considered weight loss of Al2O3 disk only for better consistency. It is to be noted that the weight loss was measured after completing the altogether 162,000 cycles under the 3 types of applied load at 20mm/sec speed ― 54,000 cycles for each of 10, 15, and 20 N loads; total= 3x 54,000 cycles. The experimental test of the weight loss of the Al2O3 disk is presented in the Figure 6. D400-P15 has the lowest wear and is followed by D300-P15, D300-P10, D400-P10, D300-P5, D400-P5 and the non-dimpled surfaces. The difference between the highest and lowest wear is 53% — which demonstrates improved wear by the micro dimpled surface technique. This result was similar to the trend of the experimental work of Sawano et al. [3] and Ito et al. [11]. Both of them investigated on ‘micro dimpled metal’-on-polyethylene whereas the present study focused on ‘micro dimpled CoC’ interface. A different result was revealed by Zhou et al. [50], which demonstrated an increased wear rate by using a dimpled surface; the main limitations of that study were the large diameter of the dimple and the low area ratio. The present study also exhibited a low effectiveness of a 5% density dimpled interface. FESEM image analysis was conducted after the friction tests to investigate the worn surface and size and shape of wear debris due to sliding of the test specimens. Figure 7
demonstrate wear signs in formation of transfer film (material transfer) and grain pullout of the dimpled surface. Among them, the material transfers were identified as much higher incidents compared to grain pull-out. The formation of a transfer film during sliding contact of ceramic materials has been reported by other literature [51-53]. Remarkably, the dimples were found without any major damage because of the high hardness of Al2O3, unlike other published results [30]. Thus, it can be predicted that a dimpled profile on Al2O3 will last longer and perform their tribological endeavours. Along with wear, the size and the shape of wear debris are important factors in the durability of the prosthesis. Cell proliferation, differentiation, and prostanoid are affected by the size, shape, and chemical composition of the particles. Figure 8 shows the morphology of wear debris filtered from used lubricant from the tribology test―the wear debris generated from the non-dimpled specimens is typically bigger in size than that from dimpled specimens. The cause of the smaller size wear debris with the dimpled surface can be described by surface continuity [54]. For the non-dimpled surface, the contact area throughout the sliding time was found to be continuous. For the dimpled surface, due to fabricating dimples, the contact area was not continuous. This may be the cause of bigger wear debris for non-dimple surfaces compared to those of the dimpled surface. However, this is a preliminary and initial assessment of wear debris, and so the results are not yet conclusive in their detail. Nine et al. [55] concluded that nano-sized wear debris is more reactive to the surrounding tissue. Since the average debris size is greater than 500 µm, it should be less harmful to the surrounding tissues; however the toxicity of nano sized debris is not only size depended but also dose dependant. For example, the higher concentrated (dose, mg/mm3) debris yields higher biological response to the surrounding tissue [55]. To this extent, the dimpled interface
is also more reliable for artificial hip joints because it yields a lower number of debris due to low wear rate (Figure 6), and a major portion of debris becomes entrapped inside the dimple [3, 56] .
4 Conclusions:
The tribological effects of various geometric parameters with different dimple densities and diameters were investigated using a reciprocating tribometer under bovine serum lubrication. The comparison between dimpled and non-dimpled samples revealed that dimpled samples reduced both the frictional coefficient and the wear rate in a simulated hip joint condition. The friction test revealed that the measured coefficient of friction for a dimpled surface with dimple diameter of 400 µm, density of 15% and depth of 30 µm lowered the maximum coefficient of friction by 22%, and wear by 53%. The frictional coefficient varied with respect to velocity and load. At 15 N load, the dimpled surfaces lowered the maximum coefficient of friction with compare to nondimpled surface. The wear debris were found to be in smaller in size for dimple surface with compared to the non-dimpled surface. Therefore, after considering the coefficient of friction, wear, and wear debris, we conclude that the micro dimpled surface (profile: Ø400µm, depth 30µm and density 15%) has an improved tribological performance when compared to the non-dimpled polished surface.
Acknowledgements
The research was supported by Faculty of Engineering, University of Malaya, from research grant (UMRG; RG147-12AET), Ministry of Higher Education High Impact research grant (UM.C/HIR/MOHE/ENG/44), Malaysia, and excellent young researcher project (CZ.1.07/2.3.00/30.0039), Brno University of Technology.
Conflicts of Interest
The authors declare no conflict of interest. References [1] S. Kurtz, K. Ong, E. Lau, F. Mowat, M. Halpern, Projections of Primary and Revision Hip and Knee Arthroplasty in the United States from 2005 to 2030, The Journal of Bone & Joint Surgery 89 (4) (2007) 780‐785. [2] B. AS, Hip new world, ASME Mech Eng 128 (2006) 28‐33. [3] H. Sawano, S. Warisawa, S. Ishihara, Study on long life of artificial joints by investigating optimal sliding surface geometry for improvement in wear resistance, Precis Eng 33 (4) (2009) 492‐498. [4] L. Mattei, F. Di Puccio, B. Piccigallo, E. Ciulli, Lubrication and wear modelling of artificial hip joints: A review, Tribology International 44 (5) (2011) 532‐549. [5] S. Ghosh, D. Choudhury, N.S. Das, B. Pingguan‐Murphy, Tribological role of synovial fluid compositions on artificial joints — a systematic review of the last 10 years, Lubrication Science (2014) n/a‐n/a. [6] A.S. Shanbhag, J.J. Jacobs, T.T. Glant, J.L. Gilbert, J. Black, J.O. Galante, Composition and morphology of wear debris in failed uncemented total hip replacement, Journal of Bone & Joint Surgery, British Volume 76 (1) (1994) 60‐67. [7] A. De Aza, J. Chevalier, G. Fantozzi, M. Schehl, R. Torrecillas, Crack growth resistance of alumina, zirconia and zirconia toughened alumina ceramics for joint prostheses, Biomaterials 23 (3) (2002) 937‐945. [8] J. Fisher, A. Galvin, J. Tipper, T. Stewart, M. Stone, E. Ingham, Comparison of the Functional Biological Activity and Osteolytic Potential of Ceramic on Ceramic and Cross Linked Polyethylene Bearings in the Hip, Bioceramics and Alternative Bearings in Joint Arthroplasty (2005) 21‐24. [9] H.S.A. Rahman, D. Choudhury, N.A.A. Osman, H.N. Shasmin, W. Abas, In vivo and in vitro outcomes of alumina, zirconia and their composited ceramic‐on‐ceramic hip joints, Journal of the Ceramic Society of Japan 121 (4) (2013) 382‐387. [10] D. Dowson, C. McNie, A. Goldsmith, Direct experimental evidence of lubrication in a metal‐on‐metal total hip replacement tested in a joint simulator, Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science 214 (1) (2000) 75‐ 86. [11] H. Ito, K. Kaneda, T. Yuhta, I. Nishimura, K. Yasuda, T. Matsuno, Reduction of polyethylene wear by concave dimples on the frictional surface in artificial hip joints, The Journal of Arthroplasty 15 (3) (2000) 332‐338. [12] P. Bearman, J. Harvey, Golf ball aerodynamics, (1975). [13] G. Ryk, I. Etsion, Testing piston rings with partial laser surface texturing for friction reduction, Wear 261 (7–8) (2006) 792‐796. [14] H. Ogihara, T. Kido, H. Yamada, M. Murata, S. Kobayashi, Technology for reducing engine rubbing resistance by means of surface improvement, Honda R&D Technical Review 12 (2) (2000) 93‐98. [15] I. Etsion, Y. Kligerman, A. Shinkarenko, Improving tribological performance of piston rings by partial surface texturing, Transaction of the ASME, Journal of Tribology 123 (2005) 632‐638. [16] J. Keller, V. Fridrici, P. Kapsa, S. Vidaller, J.F. Huard, Influence of Material Nature and Surface Texturing on Wear of Heavy‐Duty Diesel Engine Cylinder Liners, Tribology Transactions 52 (1) (2008) 121‐126.
[17] H. Yu, X. Wang, F. Zhou, Geometric Shape Effects of Surface Texture on the Generation of Hydrodynamic Pressure Between Conformal Contacting Surfaces, Tribology Letters 37 (2) (2010) 123‐130. [18] V. Marian, Lubrication of textured surfaces, in: International Conference" UNIVERSITARIA ROPET 2002", Mechanical Engineering, Petrosani, 17‐19 Octomber 2002, Vol. 2, 2006. [19] I. Křupka, M. Hartl, The effect of surface texturing on thin EHD lubrication films, Tribology international 40 (7) (2007) 1100‐1110. [20] I. Křupka, M. Hartl, Effect of Surface Texturing on Very Thin Film EHD Lubricated Contacts, Tribology Transactions 52 (1) (2008) 21‐28. [21] I. Krupka, P. Svoboda, M. Hartl, Effect of surface topography on mixed lubrication film formation during start up under rolling/sliding conditions, Tribology International 43 (5) (2010) 1035‐1042. [22] D. Choudhury, H.A. Ching, A.B. Mamat, J. Cizek, N.A.A. Osman, M. Vrbka, M. Hartl, I. Krupka, Fabrication and characterization of DLC coated micro dimples on hip prosthesis heads, Journal of Biomedical Materials Research: Part B - Applied Biomaterials (In press) (2014). [23] L. Gao, P. Yang, I. Dymond, J. Fisher, Z. Jin, Effect of surface texturing on the elastohydrodynamic lubrication analysis of metal‐on‐metal hip implants, Tribology International 43 (10) (2010) 1851‐1860. [24] E.A. Swanson, T.P. Schmalzried, F.J. Dorey, Activity Recommendations After Total Hip and Knee Arthroplasty: A Survey of the American Association for Hip and Knee Surgeons, The Journal of Arthroplasty 24 (6, Supplement) (2009) 120‐126. [25] D. Choudhury, R. Walker, T. Roy, S. Paul, R. Mootanah, Performance of honed surface profiles to artificial hip joints: An experimental investigation, International Journal of Precision Engineering and Manufacturing 14 (10) (2013) 1847‐1853. [26] D. Choudhury, R. Walker, A. Shirvani, R. Mootanah, U. Chelmsford, The Influence of Honed Surfaces on Metal‐on‐Metal Hip Joints, Tribology Online 8 (3) (2013) 195‐202. [27] T. Roy, D. Choudhury, A. Bin Mamat, B. Pingguan‐Murphy, Fabrication and characterization of micro‐dimple array on Al2O3 surfaces by using a micro‐tooling, Ceramics International 40 (1) (2014) 2381‐2388. [28] D. Choudhury, T. Roy, I. Krupka, M. Hartl, R. Mootanah, Tribological investigation of ultra‐ high molecular weight polyethylene against advanced ceramic surfaces in total hip joint replacement, Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology (2014). [29] A. Amanov, I. Cho, Y. Pyoun, C. Lee, I. Park, Micro‐dimpled surface by ultrasonic nanocrystal surface modification and its tribological effects, Wear 286 (2012) 136‐144. [30] A. Amanov, S. Sasaki, A Study On The Tribological Characteristics Of Duplex–Treated Ti– 6al–4V Alloy Under Oil–Lubricated Sliding Conditions, Tribology International (2013). [31] P. Andersson, J. Koskinen, S.e. Varjus, Y. Gerbig, H. Haefke, S. Georgiou, B. Zhmud, W. Buss, Microlubrication effect by laser‐textured steel surfaces, Wear 262 (3) (2007) 369‐379. [32] V. Brizmer, Y. Kligerman, I. Etsion, A laser surface textured parallel thrust bearing, Tribology Transactions 46 (3) (2003) 397‐403. [33] J.W. Byun, H.S. Shin, M.H. Kwon, B.H. Kim, C.N. Chu, Surface texturing by micro ECM for friction reduction, International Journal of Precision Engineering and Manufacturing 11 (5) (2010) 747‐753. [34] M. Cho, S. Park, Micro CNC surface texturing on polyoxymethylene (POM) and its tribological performance in lubricated sliding, Tribology International 44 (7) (2011) 859‐867.
[35] M.M. Arafat, A.S.M.A. Haseeb, B. Dinan, S.A. Akbar, Stress enhanced TiO2 nanowire growth on Ti–6Al–4V particles by thermal oxidation, Ceramics International 39 (6) (2013) 6517‐ 6526. [36] Z.A. Khan, M. Hadfield, Manufacturing induced residual stress influence on the rolling contact fatigue life performance of lubricated silicon nitride bearing materials, Materials & Design 28 (10) (2007) 2688‐2693. [37] Z.A. Khan, M. Hadfield, S. Tobe, Y. Wang, Ceramic rolling elements with ring crack defects—A residual stress approach, Materials Science and Engineering: A 404 (1–2) (2005) 221‐226. [38] Z.A. Khan, M. Hadfield, S. Tobe, Y. Wang, Residual stress variations during rolling contact fatigue of refrigerant lubricated silicon nitride bearing elements, Ceramics International 32 (7) (2006) 751‐754. [39] H. Hertz, On the contact of elastic solids, J. Reine. angew. Math (92) (1982) 156–171. [40] N. Tay, M. Minn, S. Sinha, A Tribological Study of SU‐8 Micro‐Dot Patterns Printed on Si Surface in a Flat‐on‐Flat Reciprocating Sliding Test, Tribology Letters 44 (2) (2011) 167‐176. [41] M.R. Widmer, M. Heuberger, J. Vörös, N.D. Spencer, Influence of polymer surface chemistry on frictional properties under protein‐lubrication conditions: implications for hip‐ implant design, Tribology Letters 10 (1‐2) (2001) 111‐116. [42] S. Mischler, A.I. Muñoz, Wear of CoCrMo alloys used in metal‐on‐metal hip joints: A tribocorrosion appraisal, Wear (2012). [43] M.P. Gispert, A.P. Serro, R. Colaço, B. Saramago, Friction and wear mechanisms in hip prosthesis: Comparison of joint materials behaviour in several lubricants, Wear 260 (1–2) (2006) 149‐158. [44] M. Vrbka, I. Krˇupka, M. Hartl, T. Návrat, In situ measurements of thin films in bovine serum lubricated contacts using optical interferometry, Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine (2014) 0954411913517498. [45] N.M. Alanazi, A. El‐Sherik, S.H. Alamar, S. Shen, Influence of Residual Stresses on Corrosion and Wear Behavior of Electrodeposited Nanocrystalline Cobalt‐Phosphorus Coatings, International Journal of Electrochemical Science 8 (8) (2013). [46] M. Wakuda, Y. Yamauchi, S. Kanzaki, Y. Yasuda, Effect of surface texturing on friction reduction between ceramic and steel materials under lubricated sliding contact, Wear 254 (3) (2003) 356‐363. [47] H. Liu, R. Niu, Y. Meng, The Effect of Laser Texturing of Steel Surfaces on Film Lubriction Based on Stribeck Curves, in: Advanced Tribology, Springer, 2010, pp. 685‐687. [48] E. Rabinowicz, Friction and wear of materials, (1965). [49] C. Myant, R. Underwood, J. Fan, P. Cann, Lubrication of metal‐on‐metal hip joints: the effect of protein content and load on film formation and wear, Journal of the mechanical behavior of biomedical materials 6 (2012) 30‐40. [50] X. Zhou, A.L. Galvin, Z. Jin, X. Yan, J. Fisher, The influence of concave dimples on the metallic counterface on the wear of ultra‐high molecular weight polyethylene, Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology 226 (6) (2012) 455‐462. [51] D.A. Rigney, L.H. Chen, M.G.S. Naylor, A.R. Rosenfield, Wear processes in sliding systems, Wear 100 (1–3) (1984) 195‐219. [52] M. Kerridge, J. Lancaster, The stages in a process of severe metallic wear, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 236 (1205) (1956) 250‐264. [53] E. Rabinowicz, D. Tabor, Metallic transfer between sliding metals: an autoradiographic study, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 208 (1095) (1951) 455‐475.
[54] B. Bhushan, Modern Tribology Handbook, Two Volume Set, CRC Press, 2010. [55] M.J. Nine, D. Choudhury, A.C. Hee, R. Mootanah, N.A.A. Osman, Wear Debris Characterization and Corresponding Biological Response: Artificial Hip and Knee Joints, Materials 7 (2) (2014) 980‐1016. [56] M. Tarabolsi, T. Klassen, F. Mantwill, F. Gärtner, F. Siegel, A.P. Schulz, Patterned CoCrMo and Al2O3 surfaces for reduced free wear debris in artificial joint arthroplasty, Journal of Biomedical Materials Research Part A 101 (12) (2013) 3447‐3456.
List of Figures:
Figure 1: a) FESEM micrograph of dimpled surface with defined 3 zones for measuring hardness tests; b) 3-D optical image of a dimple on dimpled surface; c) dimple topography of selected area on the dimpled surface. Figure 2: a) Image of reciprocating friction and wear machine; b) schematic diagram of loading and motion configuration of experimental setup. Figure 3: a) 3D image of dimpled surface produced by 3D optical profilometer b) 2D image of dimpled surface produced by FESEM Figure 4: Frictional coefficient profile at sliding speed 20 mm/s. a) With 10 N load, b) with 15 N load, and c) with 20 N load Figure 5: Friction coefficient produced by a) samples with dimple diameter of 300 µm and b) samples with dimple diameter of 400 µm. Error bar shows the standard deviation of friction coefficient for different speed Figure 6: Comparison of mass loss for different samples. Error bar shows the standard deviation of the weight loss Figure 7: FESEM images of the dimpled surface after tribology testing. a) Image of wear track on the dimpled sample, b) image of wear track near to the dimple after tribology test. Figure 8: FESEM images of the collected wear debris after tribology testing. a) Wear debris non-dimpled sample, b) wear debris from D300-P15, c) wear debris of D400-P15
List of Tables:
Table 1: Dimple parameters Table 2: Tribology testing conditions Table 3: Comparison of dimple parameters of setting and measured values (all dimensions are in µm, ± shows the standard deviation)
Tables: Table 1: Dimple parameters
Samples
Diamet er, Ø (µm)
Nondimple 1: D300P5 2 :D300P10 3: D300P15 4: D400P5 5: D400P10 6: D400P15
0 300
De Aspect pth, h ratio, λ=D/h (µm) 0 30
Pit ch ,(µm) 0
0.1
100
Dimple density, A* (%) -
Total no. of dimples 0
5
121
10
225
15
324
5
49
10
100
15
144
0 300
30
0.1 550
300
30
0.1
400
30
0.075
400
30
0.075
400
30
0.075
400 180 0 120 0 900
Table 2: Tribology testing conditions No. 1 2
Load (N) 10 15
3
20
Speed Max. contact (mm/s) (MPa) 20 181.45 20, 40, 60, 222.23 80 20 256.61
pressure
Lubricant Bovine serum
Table 3: Comparison of dimple parameters of setting and measured values (all dimensions are in µm, ± shows the standard deviation) Sample
D300-P5 D300-P10 D300-P15 D400-P5 D400-P10
Diameter (setting value) 300 300 300 400 400
Diameter (measured value) 311.61±6. 939 311.12±5. 172 310.11±3. 825 408.70±2. 870 409.87±2.
Depth (setting value) 30 30 30 30 30
Depth (measured value) 32.97±1 .34 33.80±1 .60 32.80±0 .93 34.52±2 .11 34.18±1
Pitch (setting value) 1000 550 400 1800 1200
Pitch (measured value) 1021.63±25. 26 536.95±20.2 6 413.05±19.6 6 1816.83±24. 50 1211.91 ±
D400-P15
Figures:
400
836 409.20±3. 842
30
.27 33.52±1 .70
900
4.87 915.87±14.3 5
Figure 1: a) FESEM micrograph of dimpled surface with defined 3 zones for measuring hardness tests; b) 3-D optical image of a dimple on dimpled surface; c) dimple topography of selected area on the dimpled surface.
Figure 2: a) Image of reciprocating friction and wear machine; b) schematic diagram of loading and motion configuration of experimental setup.
Fig gure 3: a) 3D D image of dimpled d surfa face producedd by 3D optiical profilomeeter b) 2D imagee of dimpled surface s produuced by FESE EM
Fiigure 4: Fricctional coeffficient profile at slidingg speed 20 mm/s. a) With 10 N load, b) with 15 N load, and c) c with 20 N load
Fig gure 5: Friction coefficiennt produced byy a) samples with w dimple ddiameter of 300 µm and b) sam mples with dimple d diameeter of 400 µm. Error baar shows thee standard deeviation of frictio on coefficientt for different speed.
Fiigure 6: Com mparison of mass loss for f different samples. Errror bar show ws the standd deviaation of weigght losses.
a)
b)
Figure 7: FESEM images of the dimpled surface after tribology testing. a) Image of wear track on the dimpled sample, b) image of wear track near to the dimple after tribology test.
Figure 8: FESEM images of the collected wear debris after tribology testing. a) Wear debris non-dimpled sample, b) wear debris from D300-P15, c) wear debris of D400-P15