- Email: [email protected]
Subsurface zone studies in the AK 12 alloy and metal matrix composite AK 12 with Al2O3 using positron annihilation spectroscopy
Wear 261 (2006) 549–555
Subsurface zone studies in the AK 12 alloy and metal matrix composite AK 12 with Al2O3 using positron annihilation spectroscopy Jerzy Dryzek a,b,∗ , Ewa Dryzek a a
b
Institute of Nuclear Physics PAN, ul. Radzikowskiego 152, 31-342 Krakow, Poland University of Zielona Gora, Institute of Physics, ul. Prof. Szafrana 4a, 65-516 Zielona Gora, Poland Received 29 March 2005; received in revised form 25 November 2005; accepted 12 January 2006 Available online 20 February 2006
Abstract The paper addresses studies of the subsurface region originated in dry sliding of aluminium casting alloy AK 12 and alloy composite; AK 12 with Al2 O3 particles. We found the inverse correlation between the wear rate in these materials and the total depth of the subsurface region detected by the positron annihilation technique. The microhardness profile and scanning electron microscopy used in our studies, generally underestimated the total depth of this region in comparison to that obtained using the positron annihilation technique. The total depth of this region for AK 12 was about 300 m and for the composite it was 60 m. In the latter, the immersed reinforcement particles effectively prevented deeper expansion of the damaged region. © 2006 Elsevier B.V. All rights reserved. Keywords: Aluminium alloy composite; Wear; Positron annihilation; Subsurface zone
1. Introduction Tribological processes, responsible for a friction force and wear create also structural changes below the worn surface: work hardening, mechanical and chemical mixing, wear and debris formation. At the depth of tens or hundreds of micrometers below the surface the defects induced by plastic and elastic deformation are observed. Thus, the subsurface region has a complex structure. In our former papers, the entire region below the worn surface, which exhibits different physical properties than interior was called the subsurface zone. In the literature many other names exist: “mechanically mixed layer”, “transfer layer” or “tribological layers”, but they are rather related to the near-surface region directly responsible for the wear process and debris formation. Indeed, all the changes leading to wear are caused by the forces acting on the real contact points, which determine the subsurface stress field and strain distribution, which result in accumulation of damage leading to the removal of particles. Therefore, the changes in the worn surface are essential for the wear. This corresponds to the accepted mechanism of wear debris formation in ductile materials which
∗
Corresponding author. Tel.: +48 12 66 28 370; fax: +48 12 66 28 458. E-mail address: [email protected] (J. Dryzek).
0043-1648/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.wear.2006.01.003
explains progressive extraction of material from beneath the contact into thin slivers, which subsequently break off to produce debris [1,2]. Another mechanism proposed by Suh assumes that the fracture of the surfaces layer leads to flake-like debris production by delamination (hence it is called “delamination wear”), i.e. by propagation of subsurface cracks and spalling of the surface [3]. The cracks are caused by dislocation pile-ups and nucleation of voids especially at inclusions or second phase particles in the plastically deformed subsurface zone. However, dislocations can move deeper producing certain defect distribution and finally the work hardened zone bellow the worn surface. This delamination theory also seems to be not appropriate for description and explanation of the wear itself and does not give any quantitative description of the subsurface zone expansion [4]. Nevertheless, this theory points at the zone below the worn surface as the zone with defects, which finally can be responsible for the wear process and from that, it would be interesting for our consideration. In the literature, one can find experimental correlation between the wear volume and the thickness of the subsurface zone. It was found in Cu and its alloys [5] and Al–Cu alloy [6]. Another theoretical model of wear proposed by Kapoor and Franklin [4] takes into account pressure distribution below the worn surface. Optical or electron microscopy techniques are commonly used for studies of the subsurface zone. They are useful for
550
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
observation of grains and dislocation structures and chemical changes. Using the displacement of selected microstructural features towards the direction of sliding one can deduce the subsurface shear strain [7,8]. Combining this with the measurements of the microhardness profile the stress–strain curves in the subsurface zone can be deduced. Those measurements are interesting from engineering point of view. For several years, the positron annihilation methods have been also applied to the studies of subsurface zone [9]. They are a very sensitive tool for detection of open volume defects, the great amount of which is created during plastic deformation in the crystalline lattice. From that, the positron annihilation methods can be more suitable for detection of basic processes, which are responsible for constitution of the subsurface zone. In this paper, we intend to demonstrate that they are more accurate in the detection of total depth of the subsurface zone than the microhardness depth profile and scanning electron microscopy and can shed light on the processes, which occurs below the worn surface. As an object of our studies we have chosen the aluminium casting alloy rich in silicon. Microstructure of casting aluminium–silicon alloys with near-eutectic composition consists in presence of hard silicon particles distributed in the Al matrix and differs from the microstructure of the wrought alloys with lower contents of alloying elements studied previously [1]. In this case, we expect the significant effect of hard precipitations on the expansion of defect depth profile generated in sliding process. This effect should be even more marked for metal matrix composite (MMC) consisted of aluminium alloy reinforced with ceramic particles. It is worth noticing that the near-eutectic aluminium–silicon alloys and aluminium matrix composite materials fabricated by the addition of the reinforcement phase such as continuous boron or graphite fibres or hard particles, e.g. SiC and Al2 O3 are important for engineering applications. The paper reports the results of the positron lifetime measurements of casting aluminium alloy AK 12 and aluminium alloy composite AK 12–Al2 O3 after dry sliding against steel. The objective is to investigate the influence of the ceramic reinforcement on the defect distribution in the subsurface region. The subsurface have been also examined using scanning electron microscopy and microhardness measurements. 2. The experimental procedure 2.1. Sample preparation The samples were machined from the commercial aluminium casting alloy AK 12 (with the following composition: Si 12.0–13.5 wt.%, Cu 0.5–1.5 wt.%, Mg 1.0–1.5 wt.%, Ni 0.5–1.5 wt.%, Mn 0.2 wt.%, Zn 0.2 wt.%, Fe 0.6 wt.%, Al balance) and composite AK 12 with (15 ± 4) vol.% of Al2 O3 in the form of cylinders 15 mm high and 10 mm in diameter. The specimens were heat treated (515 ◦ C for 6 h, warm water quenching 60 ◦ C, ageing at 175 ◦ C for 16 h). In the case of composite the volume fraction of reinforcement was quantified from the density of material. The dry sliding tests were carried out in the pin-on disk apparatus in air at room temperature. The
cylinder was a pin whose flat end surface was sliding against a disk of diameter 50 mm made from the martensitic steel (SW18 hardness about 670 HV0.1). The sliding velocity 50 mm/s was maintained during the test. The tests were carried out with the load of 106 N for different sliding distances. 2.2. Positron annihilation measurements The positron lifetime spectra were measured using the conventional fast-fast spectrometer with BaF2 scintillators. The time resolution of the system was 240 ps (FWHM). All obtained spectra containing about 2 × 106 counts were deconvoluted using the LT code subtracting the background and the source component [10]. Positron lifetime spectrum consists of one or more components in the form of exponential decay lines, described by their decay rates λi and intensities Ii , convoluted with the instrumental resolution function. Deconvolution is necessary to extract from the spectra the positron lifetimes τ i = 1/λi whose values reflect the positron annihilation states and hence also the presence and type of defects. The positrons emitted from the 22 Na source have sufficient energy (Emax = 544 keV) to penetrate certain depth of the specimen. The inverse value of the linear absorption coefficient in aluminium is equal to 94.2 m [11]. It means that 63% of positrons are stopped to this depth and annihilate contributing mainly to the obtained value of the positron lifetime. Thus, in our measurements we ignore the near-surface inhomogeneities, mechanically mixed layer and oxide effects. At this stage, it is difficult to deconvolute obtained experimental dependencies. The samples were measured as follows. An envelope made of 7 m thick kapton foil containing the positron source 22 Na (activity 38 Ci) was sandwiched between the basal surfaces of two cylinders treated by the same procedure in a pin-on-disc tester and the positron lifetime spectrum was measured. After that, a 30 m thick layer was removed from the worn surfaces of the samples and the next measurement was performed. The reduction of the specimen thickness was checked with ±5 m accuracy using a micrometer screw. The sequenced procedure allows us to monitor the depth distribution of the positron lifetime, which tags the defect distribution in the subsurface zone. The layers were removed by grinding using SiC pastes on thick iron plate. The removed layer thickness is referred to as the depth below the surface exposed to sliding. The grinding process may induce some additional defects in the layer very close to the surface of thickness much smaller than the positron implantation range in the alloy. Assuming that the contribution to the positron lifetime spectrum coming from these defects is small and its magnitude is similar for each measurement, changes in the positron lifetime spectra can be attributed to sliding. 2.3. Microhardness test For the measurements of the microhardness profile of the subsurface zone the universal microhardness test in the range of microloads was performed. This test is based on the measurement of the indentation depth under dynamic load for a controlled load–unload cycle [12]. It provides understanding not
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
551
micrograph of the MMC subsurface shows fragmented both: precipitates rich in Si and particle of Al2 O3 . The range of its occurrence is comparable to that in AK 12 alloy. 3.3. Strain distribution below worn surface
Fig. 1. Weight loss of the pins as a function of sliding distance against the SW18 martensitic steel disk at the speed 5 cm/s and load 106 N.
only of the total hardness of the tested sample but also of the plastic component, Young’s modulus and percent elastic recovery. This test is suitable for study of thin films and due to the low load is more sensitive to the change of the material properties at the atomic scale, and hence it is also suitable for studies of the subsurface zone. In our experiment the commercially available Mikro– Combi–Tester produced by CSEM was used. The Vickers indenter with the maximum load of 20 mN was applied. The velocity of load increase was equal to 40 mN/min. 3. Results 3.1. Wear rates The specific wear rate, defined as worn volume per unit sliding distance per unit load was calculated from the weight loss measurement for sliding distances presented for two samples in Fig. 1. Extracted from the slop of the dependences, the specific wear rate for AK 12 alloy and MMC is equal to 0.66 × 10−13 and 1.32 × 10−13 m3 /Nm, respectively. For AK 12 the obtained value is comparable with the value measured by Reddy et al. [12] for Al–Si binary alloys in the mild wear region. It is clearly visible that the MMC exhibits almost double value of specific wear of AK 12 alloy. The measured values of the friction coefficient for both materials were similar, for AK 12 it was equal to 0.46 ± 0.02 and for MCC 0.42 ± 0.02. 3.2. The cross-sectional micrographs The cross-sectional scanning electron micrographs for AK 12 and MMC samples whose surfaces were sliding at two distances of 252 and 504 m with the same load are given in Fig. 2a and b, respectively. The view of AK 12 subsurface reveals fragmented particles of precipitates rich in Si, which are rotated and shifted toward the direction of sliding. Within the subsurface band of about 20 m fragments of the silicon particles have the size from 1 m to a few micrometers. This band exhibits features of a mixed layer below of which the plastically deformed region is extended to the depth of about 50 m. Even application of higher load in the case of binary Al–13% Si alloy as reported in Ref. [13], leading to severe wear or seizure, gave the changes in microstructure to a depth of 75 m. The
The closer inspection of the displacement of the microstructural markers from Fig. 2 resulting directly from the shear deformation allows us to estimate the equivalent plastic strain ε deduced from the shear angel between grain boundaries and the normal to the worn surface θ [7,8], i.e. √ 3 ε= tan(θ). (1) 3 The equivalent strain as a function of the depth below the worn surface is given in Fig. 3 for AK 12 and MMC samples for the two sliding distances. Near to the worn surface the flow lines parallel to the surface implied a strain of infinity (θ ∼ = 90◦ ), which therefore could not be measured accurately. Then we ignored the shear angle measurements in this region. Maximum shear strain of 3.5 was found for AK 12 alloy sample after sliding at the distance of 252 m at the depth of 24 m below the worn surface. The two-fold increase of the sliding distance reduced the strain to the value of 0.4. Similar behavior was observed for the MMC samples. Nevertheless, the strains observed for these samples were lower than for AK 12 alloy. The total depth of the subsurface zone deduced from the strain depth dependency is ranged from 50 to 70 m below the worn surface for both types of samples. 3.4. Microhardness profile Microhardness profiles were measured at minimum five different locations along the worn surface of each sample. During the measurement, regions of precipitations and Al2 O3 particles were avoided to focus on the microhardness of the host. The averaged values as the function of depth are given in Fig. 4. It is well visible that the microhardness decreases and at certain depth reaches the bulk value characteristic for the interior region. There are no significant differences between the microhardness profiles measured for AK 12 samples whose surface were exposed to sliding at different distances, Fig. 4a. The total depth of the subsurface zone deduced form this dependency for AK 12 is about 50 m. A slightly lower total depth, of 40 m was observed for MMC samples, Fig. 4b. In this case, the microhardness at the depth of 10 m increases when the sliding distance increases. Together with the microhardness the Young’s modulus depth profile was also measured. The results are given in Fig. 5. It is characteristic that only for the MMC samples this modulus increases in the region near to the worn surface at the depth less than 20 m. For AK 12 no depth dependency was observed. 3.5. Positron annihilation depth profile The positron annihilation technique is based on the fact, that when an energetic positron enters a solid it rapidly loses its
552
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
Fig. 2. Scanning electron micrographs of the subsurface of AK 12 alloy (left) and the composite AK 12–Al2 O3 (right). The samples were exposed to sliding with the load of 106 N for the distances of: 252 m (a) and 506 m (b) (see text). Arrows indicate the sliding direction. In micrographs one can see particles of precipitates rich in Si (bright) and particles of Al2 O3 (dark grey).
energy and becomes thermalized in a time of a few picoseconds. Having thermal energy, it continues its random walk through the lattice, and finally annihilates with an electron of the sample, emitting mainly two photons of energy about 511 keV in opposite directions. The average lifetime ranges from 100 to 500 ps, depending on the material. If the sample contains defects such as vacancies, microvoids or dislocations that have a negative effective charge, the positron may be trapped and form a bound state. Due to the lower than in the host electron density inside these defects, the positron lifetime is longer than in the bulk. Therefore, measuring the positron lifetime constitutes an appropriate technique for the detection of such defects, because each type of defect is tagged by a lifetime component. However, defects repelling positrons, like interstitial atoms, do not appreciably modify its lifetime. The method is sensitive to defect concentrations as low as 10−7 atomic sites.
In all positron lifetime spectra measured for the AK 12 alloy a single component was sufficient to obtain a satisfactory fit. Such single component spectra were already reported for the Al–Cu–Mg alloys [15]. This was interpreted as an outcome of competitive trapping in different families of defects giving lifetimes too close to be isolated. In our case the variety of positron trapping sites arises from the microstructure, i.e. presence of precipitates in the alloy and from the plastic deformation of the subsurface during sliding. Thus, the evaluated from the spectra value can be treated as an average one, and it is presented in Fig. 6a as a function of depth for different sliding distances. For the lowest sliding distance of 6.3 m, changes of the positron lifetime with depth are negligible and its value does not depart much from the value equal to 222 ± 1 ps assumed as the bulk value. (Such a value was obtained also for the samples before sliding.) For the highest sliding distances, the positron
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
553
Fig. 3. Equivalent strain as function of depth for the AK 12 alloy (a) and the composite AK 12–Al2 O3 (b) for different sliding distances.
Fig. 5. Young’s modulus as a function of depth for the AK 12 alloy (a) and the composite AK 12–Al2 O3 (b) for different sliding distances.
lifetime measured directly on the worn surface increases but it tends to the bulk value with the depth increase. For the sliding distance of 252 m the value of the positron lifetime measured at the surface is equal to 238 ps and exhibits approximately exponential decay with the depth, d, described by the formula: −d τ = τ0 + A exp . (2) d0
Such a function has been found in our former studies and seems to be a common feature for the subsurface zones created during sliding and detected by positron annihilation spectroscopy [9,14]. The dashed line in Fig. 6a presents the best fit of Eq. (1) to the experimental points. The values of the obtained parameters, τ 0 , A and d0 are gathered in Table 1. The deviation of the experimental points from the exponential decay for the depth 270 and 300 m may suggest the presence of a layer with higher defect concentration. For some cases, the characterization of the defect concentration decay in the subsurface zone by the d0 parameter can be useful. From our former studies, we can conclude that its value depends significantly on the sliding conditions. Nevertheless, still it is not possible to predict its value within other theoretical predictions as it is for the total depth. We argue that the total depth of the subsurface zone is equal to the depth for which the measured positron lifetime reaches the bulk value. In our former paper we have reported that this depth correlates with the von Mises criterion for yield [16]. If we accept this then the onset of Table 1 The values of the parameters from the formula (1) fitted to the experimental dependencies of the positron lifetime on depth depicted in Fig. 6
Fig. 4. Microhardness (Vickers with load of 20 mN) as a function of depth for the AK 12 alloy (a) and the composite AK 12–Al2 O3 (b) for different sliding distances.
Sliding distance (m)
τ 0 (ps)
AK 12 63 126 225 504
222.1 224.8 222.5 223.7
MMC 252 504
224.4 ± 0.5 224.8 ± 0.5
± ± ± ±
A (ps) 0.7 0.7 0.8 0.6
11.0 12.1 14.4 12.2
± ± ± ±
d0 (m) 0.9 1.3 0.7 1.0
18.6 ± 1.2 10.8 ± 1.2
49 59 133 119
± ± ± ±
10 15 20 24
25 ± 4 28.9 ± 8
554
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
MMC and the alloy or the contribution of the positron annihilation in the interface between reinforcement particles and composite matrix. Influence of the lifetime component separation procedure cannot be excluded in this case but the changes of τ 2 are essential. The value of the second lifetime component at the surface is equal to 236 ps, which is slightly lower than the positron lifetime in AK 12 alloy, and decreases sharply with the depth. It implies much smaller total range of the positron lifetime changes, which for the MMC is c.a. 90 m. It is more than three times lower than for the AK 12 alloy sample treated in the same way. As above, the exponential decay function can be also fitted to the experimental points. However, the obtained value of the d0 parameter given in Table 1 is four times smaller than that for AK 12. Such a positron lifetime dependence on depth evidences a steep gradient of defect concentration near the surface. 4. Discussion
Fig. 6. The positron lifetime as a function of depth measured for the specimens of the AK 12 alloy (a) and composite AK 12–Al2 O3 (b) which were treated in the sliding experiment for the different distances at the normal load of 106 N. The dashed lines are the best fits of the relation (1) to the experimental points for the distance of 252 m. The shaded areas mark the bulk values of the positron lifetime.
plastic deformation for the samples after sliding at the distances of 63 and 504 m is equal to 120 and 350 m, respectively, see Fig. 6a. In the case of the MMC two lifetime components were found in the measured spectra. One could expect that one component is attributed to the positron annihilation in the Al2 O3 particles and the second one in the matrix. Indeed, the value of the first lifetime component was close to 155 ps which coincidences with the value reported in the literature, i.e. 150 ps [17] or 159 ± 8 ps [18]. In our research we were interesting in the value of positron lifetime in the matrix hence during the deconvolution we fixed the value of the first lifetime component equal to 155 ps. This is accepted, because we do not expect any changes in the hard Al2 O3 particles during sliding located far away from the worn surface. Fig. 6b shows the dependences of the second component on depth for the MMC samples after sliding in the same conditions as for AK 12 alloy. As for the AK 12 alloy for the sliding distance of 6.3 m the positron lifetime changes with depth are negligible but the value assumed as the bulk lifetime in matrix alloy equal to 225 ps is slightly higher than in the AK 12 alloy. It may be caused by the differences in microstructure of the matrix in the
The striking feature of the experimental results presented is a discrepancy between the total depth of the subsurface zone detected by positron annihilation method and that by the microhardness profile and the strain depth dependency for AK 12 alloy. For the MMC the total depths of the subsurface zone detected by the used methods coincide. This discrepancy can be explained by the higher sensitivity of positrons to the defects induced, e.g. by the plastic deformation than microhardness measurements and scanning electron microscopy observation. In the case of AK 12 alloy the subsurface zone is extended much deeper than in the MMC because the latter contains distributed ceramic particles that prevent movement of dislocations from the worn surface towards the interior. In the dislocation theory it is well known that precipitations, impurities effectively hinder movement of dislocations and in some cases enhance their density as well. Accumulation of dislocations induces hardening of this region. For MMC the microhardness is about 1.5 times higher than for AK 12 measured at the depth of 10 m below the worn surface, Fig. 4. The role of dislocations in wear process has been discussed by Hirth and Rigney [19]. We argue that this can be responsible also for the differences in the wear rate for both materials. The energy dissipated during sliding is used for creation of the subsurface zone. This energy is responsible for plastic deformation, production, and movement of dislocations. It is well known that dislocations with jogs during their motion create a great amount of point defects, i.e. vacancies and interstitial atoms. If the total depth of the subsurface is small then more energy is deposited close to the surface, and one can expect more defects there and hence higher wear rate. If the subsurface zone is extended deeper in the interior thus also the energy is deposited in larger region. This toy model is also supported by another observation, which we did for copper exposed to sliding [9]. The speed or kinetic energy of the pin affected more the expansion of the subsurface zone than the applied load, sliding distance or temperature. The Al2 O3 particles distributed between debris can act as the abrasive particles grinding the worn surface. However, from our experience and experimental results reported in the literature
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
[22] we can suppose that the thickness of the subsurface zone created by grinding using Al2 O3 or other hard particles is of order of a few micrometers but not of a few dozens or hundreds micrometers. This could be even supported, by Figs. 2 and 3, but results from the positron annihilation method indicate that the ranges of the subsurface zone in the alloy and MMC are different. Effect of the precipitations on the subsurface zone depth expansion has been observed already in aluminium alloys; 2017A and 6101A [20]. For the 2017A alloy the total depth of the subsurface zone of about 120 m was hardly affected by the load and sliding distance. In pure aluminium [21] and 6101A alloy [20] the subsurface zone was extended up to 400 m. We believe that in the AK 12 some influence of the hard silicon particles on the subsurface zone constitution takes place. Nevertheless, these particles are not so effective in preventing of the subsurface zone expansion like it is in the case of precipitation in 2017A aluminium or reinforcement particles in the MMC. The total range of the subsurface zone is stressed in our consideration, because this value can be linked with, e.g. the theory of elasticity via the von Mises criterion for yield. The shape of the detected profiles at this stage is difficult to compare with other theoretical predictions. Nevertheless, in all cases we observed gradually decrease of the measured quantities. The positron lifetime profile can be described by an exponential decay function but the microhardness profile exhibits rather a linear dependency. For the strain profile it is difficult to find any relation, mainly because the microstructural markers often became obscured and hence poor accuracy in the measurements of the angle is obtained. The value of the yield strength for the aluminium alloys with 11% of Si reported in the literature is about 165 MPa. This is much above of the static pressure applied to our samples, which was 1.3 MPa. This implies that during sliding local pressure at asperities is much higher. Similarly to our previous results obtained for copper [16], we may estimate the value of this pressure at ten times of the yield strength acting during dynamic strikes of sliding surfaces. It seems that the short strikes play a main the role in initiation and expansion of the subsurface zone. The lifetime value ranged between 243 and 225 ps for both samples. This indicates the presence of single vacancies associated with dislocation rather than isolated vacancies or their clusters [21]. Such vacancies have been observed also in pure aluminium and other aluminium alloys studied by us. The lifetime value obtained in that case is higher and equal to 255 ps. 5. Conclusions In summary, the total range of the subsurface zone in AK 12 alloy, occurred during sliding and detected by positron lifetime method is extended to the depth more than 300 m. The measurements of the microhardness and strain show much smaller range of about 70 m. For the MMC the detected range is even
555
smaller, of about 60 m as it deduced from all applied experimental techniques. We argue that the reinforcement particles distributed in the alloy host play an important role in the prevention of the subsurface zone expansion. This can affect the wear increase. The wear rate for the MMC was about two times larger than for the AK 12 alloy. We recommend the positron annihilation technique as the most suitable for detection of the subsurface zone total range. Acknowledgements The author (ED) expresses her gratitude to the Committee of Scientifics Research (Poland) which support this work by grant No. 2 P03B 008 24 (2003–2005). The author would like to ´ thank Prof. J. Sleziona (Faculty of Materials Science and Metallurgy, The Silesian University of Technology) for providing the samples of aluminium alloy composite. Measurements of the microhardness were performed in Laboratory for Tribology and Surface Engineering, Faculty of Mechanical Engineering and Robotics, AGH University of Science and Technology, Krak´ow. References [1] A. Kjer, Proceedings of the International Conference on Wear of Materials, ASME, New York, 1987, pp. 191–198. [2] Kato, Plenary Lecture at the World Tribology Congress, London, September 8–11, 1997, in: I.M. Hutchings (Ed.), New Directions in Tribology, MEP, UK, 1997, pp. 39–56. [3] N.P. Suh, Wear 44 (1977) 1–16. [4] A. Kapoor, F.J. Franklin, Wear 245 (2000) 204–215. [5] P.J. Blau, J. Mater. Sci. 19 (1984) 1957. [6] C. Perrin, W.M. Rainforth, Wear 203–204 (1997) 171. [7] G.M. Dautzenberg, J.H. Zaat, Wear 23 (1973) 9. [8] M.A. Moore, R.M. Douthwaite, Metall. Trans. A 7 (1976) 1833. [9] J. Dryzek, E. Dryzek, T. Stegemann, B. Cleff, Tribol. Lett. 3 (1997) 269. [10] J. Kansy, Nucl. Instr. Meth. A 374 (1996) 235. [11] J. Dryzek, Appl. Phys. A 81(5) (2005) (on line published: doi:10.1007/s00339-004-2966-6). [12] W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564. [13] A.S. Reddy, B.N. Pramila Bai, K.S.S. Murthy, S.K. Biswas, Wear 171 (1994) 115. [14] E. Dryzek, J. Mater. Sci. 38 (2003) 3755. [15] A. Somoza, A. Dupasquier, I.J. Polmear, P. Folegati, R. Ferragut, Phys. Rev. 61 (2000) 14454. [16] J. Dryzek, E. Dryzek, Tribol. Lett. 15 (2003) 309. [17] M. Forster, W. Claudy, H. Hermes, M. Koch, K. Maier, J. Major, H. Stoll, H.-E. Schaefer, Mater. Sci. Forum 105–110 (1992) 1005. [18] M. Noguchi, T. Mitsuhashi, T. Chiba, T. Tanaka, N. Tsuda, J. Phys. Soc. Jpn. 32 (1972) 1242. [19] J.P. Hirth, D.A. Rigney, in: F.R.N. Nabarro (Ed.), Dislocations in Solids, North-Holland Publishing Company, Amsterdam, New York, Oxford, 1983, p. 3 (Chapter 25). [20] E. Dryzek, J. Dryzek, Tribol. Int., in press. [21] J. Dryzek, E. Dryzek, Tribol. Lett. 17 (2004) 147. [22] F. B¨orner, S. Eichler, A. Polity, R. Krause-Rehberg, R. Hammer, M. Jurisch, J. Appl. Phys. 18 (1998) 2255.
Subsurface zone studies in the AK 12 alloy and metal matrix composite AK 12 with Al2O3 using positron annihilation spectroscopy Jerzy Dryzek a,b,∗ , Ewa Dryzek a a
b
Institute of Nuclear Physics PAN, ul. Radzikowskiego 152, 31-342 Krakow, Poland University of Zielona Gora, Institute of Physics, ul. Prof. Szafrana 4a, 65-516 Zielona Gora, Poland Received 29 March 2005; received in revised form 25 November 2005; accepted 12 January 2006 Available online 20 February 2006
Abstract The paper addresses studies of the subsurface region originated in dry sliding of aluminium casting alloy AK 12 and alloy composite; AK 12 with Al2 O3 particles. We found the inverse correlation between the wear rate in these materials and the total depth of the subsurface region detected by the positron annihilation technique. The microhardness profile and scanning electron microscopy used in our studies, generally underestimated the total depth of this region in comparison to that obtained using the positron annihilation technique. The total depth of this region for AK 12 was about 300 m and for the composite it was 60 m. In the latter, the immersed reinforcement particles effectively prevented deeper expansion of the damaged region. © 2006 Elsevier B.V. All rights reserved. Keywords: Aluminium alloy composite; Wear; Positron annihilation; Subsurface zone
1. Introduction Tribological processes, responsible for a friction force and wear create also structural changes below the worn surface: work hardening, mechanical and chemical mixing, wear and debris formation. At the depth of tens or hundreds of micrometers below the surface the defects induced by plastic and elastic deformation are observed. Thus, the subsurface region has a complex structure. In our former papers, the entire region below the worn surface, which exhibits different physical properties than interior was called the subsurface zone. In the literature many other names exist: “mechanically mixed layer”, “transfer layer” or “tribological layers”, but they are rather related to the near-surface region directly responsible for the wear process and debris formation. Indeed, all the changes leading to wear are caused by the forces acting on the real contact points, which determine the subsurface stress field and strain distribution, which result in accumulation of damage leading to the removal of particles. Therefore, the changes in the worn surface are essential for the wear. This corresponds to the accepted mechanism of wear debris formation in ductile materials which
∗
Corresponding author. Tel.: +48 12 66 28 370; fax: +48 12 66 28 458. E-mail address: [email protected] (J. Dryzek).
0043-1648/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.wear.2006.01.003
explains progressive extraction of material from beneath the contact into thin slivers, which subsequently break off to produce debris [1,2]. Another mechanism proposed by Suh assumes that the fracture of the surfaces layer leads to flake-like debris production by delamination (hence it is called “delamination wear”), i.e. by propagation of subsurface cracks and spalling of the surface [3]. The cracks are caused by dislocation pile-ups and nucleation of voids especially at inclusions or second phase particles in the plastically deformed subsurface zone. However, dislocations can move deeper producing certain defect distribution and finally the work hardened zone bellow the worn surface. This delamination theory also seems to be not appropriate for description and explanation of the wear itself and does not give any quantitative description of the subsurface zone expansion [4]. Nevertheless, this theory points at the zone below the worn surface as the zone with defects, which finally can be responsible for the wear process and from that, it would be interesting for our consideration. In the literature, one can find experimental correlation between the wear volume and the thickness of the subsurface zone. It was found in Cu and its alloys [5] and Al–Cu alloy [6]. Another theoretical model of wear proposed by Kapoor and Franklin [4] takes into account pressure distribution below the worn surface. Optical or electron microscopy techniques are commonly used for studies of the subsurface zone. They are useful for
550
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
observation of grains and dislocation structures and chemical changes. Using the displacement of selected microstructural features towards the direction of sliding one can deduce the subsurface shear strain [7,8]. Combining this with the measurements of the microhardness profile the stress–strain curves in the subsurface zone can be deduced. Those measurements are interesting from engineering point of view. For several years, the positron annihilation methods have been also applied to the studies of subsurface zone [9]. They are a very sensitive tool for detection of open volume defects, the great amount of which is created during plastic deformation in the crystalline lattice. From that, the positron annihilation methods can be more suitable for detection of basic processes, which are responsible for constitution of the subsurface zone. In this paper, we intend to demonstrate that they are more accurate in the detection of total depth of the subsurface zone than the microhardness depth profile and scanning electron microscopy and can shed light on the processes, which occurs below the worn surface. As an object of our studies we have chosen the aluminium casting alloy rich in silicon. Microstructure of casting aluminium–silicon alloys with near-eutectic composition consists in presence of hard silicon particles distributed in the Al matrix and differs from the microstructure of the wrought alloys with lower contents of alloying elements studied previously [1]. In this case, we expect the significant effect of hard precipitations on the expansion of defect depth profile generated in sliding process. This effect should be even more marked for metal matrix composite (MMC) consisted of aluminium alloy reinforced with ceramic particles. It is worth noticing that the near-eutectic aluminium–silicon alloys and aluminium matrix composite materials fabricated by the addition of the reinforcement phase such as continuous boron or graphite fibres or hard particles, e.g. SiC and Al2 O3 are important for engineering applications. The paper reports the results of the positron lifetime measurements of casting aluminium alloy AK 12 and aluminium alloy composite AK 12–Al2 O3 after dry sliding against steel. The objective is to investigate the influence of the ceramic reinforcement on the defect distribution in the subsurface region. The subsurface have been also examined using scanning electron microscopy and microhardness measurements. 2. The experimental procedure 2.1. Sample preparation The samples were machined from the commercial aluminium casting alloy AK 12 (with the following composition: Si 12.0–13.5 wt.%, Cu 0.5–1.5 wt.%, Mg 1.0–1.5 wt.%, Ni 0.5–1.5 wt.%, Mn 0.2 wt.%, Zn 0.2 wt.%, Fe 0.6 wt.%, Al balance) and composite AK 12 with (15 ± 4) vol.% of Al2 O3 in the form of cylinders 15 mm high and 10 mm in diameter. The specimens were heat treated (515 ◦ C for 6 h, warm water quenching 60 ◦ C, ageing at 175 ◦ C for 16 h). In the case of composite the volume fraction of reinforcement was quantified from the density of material. The dry sliding tests were carried out in the pin-on disk apparatus in air at room temperature. The
cylinder was a pin whose flat end surface was sliding against a disk of diameter 50 mm made from the martensitic steel (SW18 hardness about 670 HV0.1). The sliding velocity 50 mm/s was maintained during the test. The tests were carried out with the load of 106 N for different sliding distances. 2.2. Positron annihilation measurements The positron lifetime spectra were measured using the conventional fast-fast spectrometer with BaF2 scintillators. The time resolution of the system was 240 ps (FWHM). All obtained spectra containing about 2 × 106 counts were deconvoluted using the LT code subtracting the background and the source component [10]. Positron lifetime spectrum consists of one or more components in the form of exponential decay lines, described by their decay rates λi and intensities Ii , convoluted with the instrumental resolution function. Deconvolution is necessary to extract from the spectra the positron lifetimes τ i = 1/λi whose values reflect the positron annihilation states and hence also the presence and type of defects. The positrons emitted from the 22 Na source have sufficient energy (Emax = 544 keV) to penetrate certain depth of the specimen. The inverse value of the linear absorption coefficient in aluminium is equal to 94.2 m [11]. It means that 63% of positrons are stopped to this depth and annihilate contributing mainly to the obtained value of the positron lifetime. Thus, in our measurements we ignore the near-surface inhomogeneities, mechanically mixed layer and oxide effects. At this stage, it is difficult to deconvolute obtained experimental dependencies. The samples were measured as follows. An envelope made of 7 m thick kapton foil containing the positron source 22 Na (activity 38 Ci) was sandwiched between the basal surfaces of two cylinders treated by the same procedure in a pin-on-disc tester and the positron lifetime spectrum was measured. After that, a 30 m thick layer was removed from the worn surfaces of the samples and the next measurement was performed. The reduction of the specimen thickness was checked with ±5 m accuracy using a micrometer screw. The sequenced procedure allows us to monitor the depth distribution of the positron lifetime, which tags the defect distribution in the subsurface zone. The layers were removed by grinding using SiC pastes on thick iron plate. The removed layer thickness is referred to as the depth below the surface exposed to sliding. The grinding process may induce some additional defects in the layer very close to the surface of thickness much smaller than the positron implantation range in the alloy. Assuming that the contribution to the positron lifetime spectrum coming from these defects is small and its magnitude is similar for each measurement, changes in the positron lifetime spectra can be attributed to sliding. 2.3. Microhardness test For the measurements of the microhardness profile of the subsurface zone the universal microhardness test in the range of microloads was performed. This test is based on the measurement of the indentation depth under dynamic load for a controlled load–unload cycle [12]. It provides understanding not
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
551
micrograph of the MMC subsurface shows fragmented both: precipitates rich in Si and particle of Al2 O3 . The range of its occurrence is comparable to that in AK 12 alloy. 3.3. Strain distribution below worn surface
Fig. 1. Weight loss of the pins as a function of sliding distance against the SW18 martensitic steel disk at the speed 5 cm/s and load 106 N.
only of the total hardness of the tested sample but also of the plastic component, Young’s modulus and percent elastic recovery. This test is suitable for study of thin films and due to the low load is more sensitive to the change of the material properties at the atomic scale, and hence it is also suitable for studies of the subsurface zone. In our experiment the commercially available Mikro– Combi–Tester produced by CSEM was used. The Vickers indenter with the maximum load of 20 mN was applied. The velocity of load increase was equal to 40 mN/min. 3. Results 3.1. Wear rates The specific wear rate, defined as worn volume per unit sliding distance per unit load was calculated from the weight loss measurement for sliding distances presented for two samples in Fig. 1. Extracted from the slop of the dependences, the specific wear rate for AK 12 alloy and MMC is equal to 0.66 × 10−13 and 1.32 × 10−13 m3 /Nm, respectively. For AK 12 the obtained value is comparable with the value measured by Reddy et al. [12] for Al–Si binary alloys in the mild wear region. It is clearly visible that the MMC exhibits almost double value of specific wear of AK 12 alloy. The measured values of the friction coefficient for both materials were similar, for AK 12 it was equal to 0.46 ± 0.02 and for MCC 0.42 ± 0.02. 3.2. The cross-sectional micrographs The cross-sectional scanning electron micrographs for AK 12 and MMC samples whose surfaces were sliding at two distances of 252 and 504 m with the same load are given in Fig. 2a and b, respectively. The view of AK 12 subsurface reveals fragmented particles of precipitates rich in Si, which are rotated and shifted toward the direction of sliding. Within the subsurface band of about 20 m fragments of the silicon particles have the size from 1 m to a few micrometers. This band exhibits features of a mixed layer below of which the plastically deformed region is extended to the depth of about 50 m. Even application of higher load in the case of binary Al–13% Si alloy as reported in Ref. [13], leading to severe wear or seizure, gave the changes in microstructure to a depth of 75 m. The
The closer inspection of the displacement of the microstructural markers from Fig. 2 resulting directly from the shear deformation allows us to estimate the equivalent plastic strain ε deduced from the shear angel between grain boundaries and the normal to the worn surface θ [7,8], i.e. √ 3 ε= tan(θ). (1) 3 The equivalent strain as a function of the depth below the worn surface is given in Fig. 3 for AK 12 and MMC samples for the two sliding distances. Near to the worn surface the flow lines parallel to the surface implied a strain of infinity (θ ∼ = 90◦ ), which therefore could not be measured accurately. Then we ignored the shear angle measurements in this region. Maximum shear strain of 3.5 was found for AK 12 alloy sample after sliding at the distance of 252 m at the depth of 24 m below the worn surface. The two-fold increase of the sliding distance reduced the strain to the value of 0.4. Similar behavior was observed for the MMC samples. Nevertheless, the strains observed for these samples were lower than for AK 12 alloy. The total depth of the subsurface zone deduced from the strain depth dependency is ranged from 50 to 70 m below the worn surface for both types of samples. 3.4. Microhardness profile Microhardness profiles were measured at minimum five different locations along the worn surface of each sample. During the measurement, regions of precipitations and Al2 O3 particles were avoided to focus on the microhardness of the host. The averaged values as the function of depth are given in Fig. 4. It is well visible that the microhardness decreases and at certain depth reaches the bulk value characteristic for the interior region. There are no significant differences between the microhardness profiles measured for AK 12 samples whose surface were exposed to sliding at different distances, Fig. 4a. The total depth of the subsurface zone deduced form this dependency for AK 12 is about 50 m. A slightly lower total depth, of 40 m was observed for MMC samples, Fig. 4b. In this case, the microhardness at the depth of 10 m increases when the sliding distance increases. Together with the microhardness the Young’s modulus depth profile was also measured. The results are given in Fig. 5. It is characteristic that only for the MMC samples this modulus increases in the region near to the worn surface at the depth less than 20 m. For AK 12 no depth dependency was observed. 3.5. Positron annihilation depth profile The positron annihilation technique is based on the fact, that when an energetic positron enters a solid it rapidly loses its
552
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
Fig. 2. Scanning electron micrographs of the subsurface of AK 12 alloy (left) and the composite AK 12–Al2 O3 (right). The samples were exposed to sliding with the load of 106 N for the distances of: 252 m (a) and 506 m (b) (see text). Arrows indicate the sliding direction. In micrographs one can see particles of precipitates rich in Si (bright) and particles of Al2 O3 (dark grey).
energy and becomes thermalized in a time of a few picoseconds. Having thermal energy, it continues its random walk through the lattice, and finally annihilates with an electron of the sample, emitting mainly two photons of energy about 511 keV in opposite directions. The average lifetime ranges from 100 to 500 ps, depending on the material. If the sample contains defects such as vacancies, microvoids or dislocations that have a negative effective charge, the positron may be trapped and form a bound state. Due to the lower than in the host electron density inside these defects, the positron lifetime is longer than in the bulk. Therefore, measuring the positron lifetime constitutes an appropriate technique for the detection of such defects, because each type of defect is tagged by a lifetime component. However, defects repelling positrons, like interstitial atoms, do not appreciably modify its lifetime. The method is sensitive to defect concentrations as low as 10−7 atomic sites.
In all positron lifetime spectra measured for the AK 12 alloy a single component was sufficient to obtain a satisfactory fit. Such single component spectra were already reported for the Al–Cu–Mg alloys [15]. This was interpreted as an outcome of competitive trapping in different families of defects giving lifetimes too close to be isolated. In our case the variety of positron trapping sites arises from the microstructure, i.e. presence of precipitates in the alloy and from the plastic deformation of the subsurface during sliding. Thus, the evaluated from the spectra value can be treated as an average one, and it is presented in Fig. 6a as a function of depth for different sliding distances. For the lowest sliding distance of 6.3 m, changes of the positron lifetime with depth are negligible and its value does not depart much from the value equal to 222 ± 1 ps assumed as the bulk value. (Such a value was obtained also for the samples before sliding.) For the highest sliding distances, the positron
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
553
Fig. 3. Equivalent strain as function of depth for the AK 12 alloy (a) and the composite AK 12–Al2 O3 (b) for different sliding distances.
Fig. 5. Young’s modulus as a function of depth for the AK 12 alloy (a) and the composite AK 12–Al2 O3 (b) for different sliding distances.
lifetime measured directly on the worn surface increases but it tends to the bulk value with the depth increase. For the sliding distance of 252 m the value of the positron lifetime measured at the surface is equal to 238 ps and exhibits approximately exponential decay with the depth, d, described by the formula: −d τ = τ0 + A exp . (2) d0
Such a function has been found in our former studies and seems to be a common feature for the subsurface zones created during sliding and detected by positron annihilation spectroscopy [9,14]. The dashed line in Fig. 6a presents the best fit of Eq. (1) to the experimental points. The values of the obtained parameters, τ 0 , A and d0 are gathered in Table 1. The deviation of the experimental points from the exponential decay for the depth 270 and 300 m may suggest the presence of a layer with higher defect concentration. For some cases, the characterization of the defect concentration decay in the subsurface zone by the d0 parameter can be useful. From our former studies, we can conclude that its value depends significantly on the sliding conditions. Nevertheless, still it is not possible to predict its value within other theoretical predictions as it is for the total depth. We argue that the total depth of the subsurface zone is equal to the depth for which the measured positron lifetime reaches the bulk value. In our former paper we have reported that this depth correlates with the von Mises criterion for yield [16]. If we accept this then the onset of Table 1 The values of the parameters from the formula (1) fitted to the experimental dependencies of the positron lifetime on depth depicted in Fig. 6
Fig. 4. Microhardness (Vickers with load of 20 mN) as a function of depth for the AK 12 alloy (a) and the composite AK 12–Al2 O3 (b) for different sliding distances.
Sliding distance (m)
τ 0 (ps)
AK 12 63 126 225 504
222.1 224.8 222.5 223.7
MMC 252 504
224.4 ± 0.5 224.8 ± 0.5
± ± ± ±
A (ps) 0.7 0.7 0.8 0.6
11.0 12.1 14.4 12.2
± ± ± ±
d0 (m) 0.9 1.3 0.7 1.0
18.6 ± 1.2 10.8 ± 1.2
49 59 133 119
± ± ± ±
10 15 20 24
25 ± 4 28.9 ± 8
554
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
MMC and the alloy or the contribution of the positron annihilation in the interface between reinforcement particles and composite matrix. Influence of the lifetime component separation procedure cannot be excluded in this case but the changes of τ 2 are essential. The value of the second lifetime component at the surface is equal to 236 ps, which is slightly lower than the positron lifetime in AK 12 alloy, and decreases sharply with the depth. It implies much smaller total range of the positron lifetime changes, which for the MMC is c.a. 90 m. It is more than three times lower than for the AK 12 alloy sample treated in the same way. As above, the exponential decay function can be also fitted to the experimental points. However, the obtained value of the d0 parameter given in Table 1 is four times smaller than that for AK 12. Such a positron lifetime dependence on depth evidences a steep gradient of defect concentration near the surface. 4. Discussion
Fig. 6. The positron lifetime as a function of depth measured for the specimens of the AK 12 alloy (a) and composite AK 12–Al2 O3 (b) which were treated in the sliding experiment for the different distances at the normal load of 106 N. The dashed lines are the best fits of the relation (1) to the experimental points for the distance of 252 m. The shaded areas mark the bulk values of the positron lifetime.
plastic deformation for the samples after sliding at the distances of 63 and 504 m is equal to 120 and 350 m, respectively, see Fig. 6a. In the case of the MMC two lifetime components were found in the measured spectra. One could expect that one component is attributed to the positron annihilation in the Al2 O3 particles and the second one in the matrix. Indeed, the value of the first lifetime component was close to 155 ps which coincidences with the value reported in the literature, i.e. 150 ps [17] or 159 ± 8 ps [18]. In our research we were interesting in the value of positron lifetime in the matrix hence during the deconvolution we fixed the value of the first lifetime component equal to 155 ps. This is accepted, because we do not expect any changes in the hard Al2 O3 particles during sliding located far away from the worn surface. Fig. 6b shows the dependences of the second component on depth for the MMC samples after sliding in the same conditions as for AK 12 alloy. As for the AK 12 alloy for the sliding distance of 6.3 m the positron lifetime changes with depth are negligible but the value assumed as the bulk lifetime in matrix alloy equal to 225 ps is slightly higher than in the AK 12 alloy. It may be caused by the differences in microstructure of the matrix in the
The striking feature of the experimental results presented is a discrepancy between the total depth of the subsurface zone detected by positron annihilation method and that by the microhardness profile and the strain depth dependency for AK 12 alloy. For the MMC the total depths of the subsurface zone detected by the used methods coincide. This discrepancy can be explained by the higher sensitivity of positrons to the defects induced, e.g. by the plastic deformation than microhardness measurements and scanning electron microscopy observation. In the case of AK 12 alloy the subsurface zone is extended much deeper than in the MMC because the latter contains distributed ceramic particles that prevent movement of dislocations from the worn surface towards the interior. In the dislocation theory it is well known that precipitations, impurities effectively hinder movement of dislocations and in some cases enhance their density as well. Accumulation of dislocations induces hardening of this region. For MMC the microhardness is about 1.5 times higher than for AK 12 measured at the depth of 10 m below the worn surface, Fig. 4. The role of dislocations in wear process has been discussed by Hirth and Rigney [19]. We argue that this can be responsible also for the differences in the wear rate for both materials. The energy dissipated during sliding is used for creation of the subsurface zone. This energy is responsible for plastic deformation, production, and movement of dislocations. It is well known that dislocations with jogs during their motion create a great amount of point defects, i.e. vacancies and interstitial atoms. If the total depth of the subsurface is small then more energy is deposited close to the surface, and one can expect more defects there and hence higher wear rate. If the subsurface zone is extended deeper in the interior thus also the energy is deposited in larger region. This toy model is also supported by another observation, which we did for copper exposed to sliding [9]. The speed or kinetic energy of the pin affected more the expansion of the subsurface zone than the applied load, sliding distance or temperature. The Al2 O3 particles distributed between debris can act as the abrasive particles grinding the worn surface. However, from our experience and experimental results reported in the literature
J. Dryzek, E. Dryzek / Wear 261 (2006) 549–555
[22] we can suppose that the thickness of the subsurface zone created by grinding using Al2 O3 or other hard particles is of order of a few micrometers but not of a few dozens or hundreds micrometers. This could be even supported, by Figs. 2 and 3, but results from the positron annihilation method indicate that the ranges of the subsurface zone in the alloy and MMC are different. Effect of the precipitations on the subsurface zone depth expansion has been observed already in aluminium alloys; 2017A and 6101A [20]. For the 2017A alloy the total depth of the subsurface zone of about 120 m was hardly affected by the load and sliding distance. In pure aluminium [21] and 6101A alloy [20] the subsurface zone was extended up to 400 m. We believe that in the AK 12 some influence of the hard silicon particles on the subsurface zone constitution takes place. Nevertheless, these particles are not so effective in preventing of the subsurface zone expansion like it is in the case of precipitation in 2017A aluminium or reinforcement particles in the MMC. The total range of the subsurface zone is stressed in our consideration, because this value can be linked with, e.g. the theory of elasticity via the von Mises criterion for yield. The shape of the detected profiles at this stage is difficult to compare with other theoretical predictions. Nevertheless, in all cases we observed gradually decrease of the measured quantities. The positron lifetime profile can be described by an exponential decay function but the microhardness profile exhibits rather a linear dependency. For the strain profile it is difficult to find any relation, mainly because the microstructural markers often became obscured and hence poor accuracy in the measurements of the angle is obtained. The value of the yield strength for the aluminium alloys with 11% of Si reported in the literature is about 165 MPa. This is much above of the static pressure applied to our samples, which was 1.3 MPa. This implies that during sliding local pressure at asperities is much higher. Similarly to our previous results obtained for copper [16], we may estimate the value of this pressure at ten times of the yield strength acting during dynamic strikes of sliding surfaces. It seems that the short strikes play a main the role in initiation and expansion of the subsurface zone. The lifetime value ranged between 243 and 225 ps for both samples. This indicates the presence of single vacancies associated with dislocation rather than isolated vacancies or their clusters [21]. Such vacancies have been observed also in pure aluminium and other aluminium alloys studied by us. The lifetime value obtained in that case is higher and equal to 255 ps. 5. Conclusions In summary, the total range of the subsurface zone in AK 12 alloy, occurred during sliding and detected by positron lifetime method is extended to the depth more than 300 m. The measurements of the microhardness and strain show much smaller range of about 70 m. For the MMC the detected range is even
555
smaller, of about 60 m as it deduced from all applied experimental techniques. We argue that the reinforcement particles distributed in the alloy host play an important role in the prevention of the subsurface zone expansion. This can affect the wear increase. The wear rate for the MMC was about two times larger than for the AK 12 alloy. We recommend the positron annihilation technique as the most suitable for detection of the subsurface zone total range. Acknowledgements The author (ED) expresses her gratitude to the Committee of Scientifics Research (Poland) which support this work by grant No. 2 P03B 008 24 (2003–2005). The author would like to ´ thank Prof. J. Sleziona (Faculty of Materials Science and Metallurgy, The Silesian University of Technology) for providing the samples of aluminium alloy composite. Measurements of the microhardness were performed in Laboratory for Tribology and Surface Engineering, Faculty of Mechanical Engineering and Robotics, AGH University of Science and Technology, Krak´ow. References [1] A. Kjer, Proceedings of the International Conference on Wear of Materials, ASME, New York, 1987, pp. 191–198. [2] Kato, Plenary Lecture at the World Tribology Congress, London, September 8–11, 1997, in: I.M. Hutchings (Ed.), New Directions in Tribology, MEP, UK, 1997, pp. 39–56. [3] N.P. Suh, Wear 44 (1977) 1–16. [4] A. Kapoor, F.J. Franklin, Wear 245 (2000) 204–215. [5] P.J. Blau, J. Mater. Sci. 19 (1984) 1957. [6] C. Perrin, W.M. Rainforth, Wear 203–204 (1997) 171. [7] G.M. Dautzenberg, J.H. Zaat, Wear 23 (1973) 9. [8] M.A. Moore, R.M. Douthwaite, Metall. Trans. A 7 (1976) 1833. [9] J. Dryzek, E. Dryzek, T. Stegemann, B. Cleff, Tribol. Lett. 3 (1997) 269. [10] J. Kansy, Nucl. Instr. Meth. A 374 (1996) 235. [11] J. Dryzek, Appl. Phys. A 81(5) (2005) (on line published: doi:10.1007/s00339-004-2966-6). [12] W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564. [13] A.S. Reddy, B.N. Pramila Bai, K.S.S. Murthy, S.K. Biswas, Wear 171 (1994) 115. [14] E. Dryzek, J. Mater. Sci. 38 (2003) 3755. [15] A. Somoza, A. Dupasquier, I.J. Polmear, P. Folegati, R. Ferragut, Phys. Rev. 61 (2000) 14454. [16] J. Dryzek, E. Dryzek, Tribol. Lett. 15 (2003) 309. [17] M. Forster, W. Claudy, H. Hermes, M. Koch, K. Maier, J. Major, H. Stoll, H.-E. Schaefer, Mater. Sci. Forum 105–110 (1992) 1005. [18] M. Noguchi, T. Mitsuhashi, T. Chiba, T. Tanaka, N. Tsuda, J. Phys. Soc. Jpn. 32 (1972) 1242. [19] J.P. Hirth, D.A. Rigney, in: F.R.N. Nabarro (Ed.), Dislocations in Solids, North-Holland Publishing Company, Amsterdam, New York, Oxford, 1983, p. 3 (Chapter 25). [20] E. Dryzek, J. Dryzek, Tribol. Int., in press. [21] J. Dryzek, E. Dryzek, Tribol. Lett. 17 (2004) 147. [22] F. B¨orner, S. Eichler, A. Polity, R. Krause-Rehberg, R. Hammer, M. Jurisch, J. Appl. Phys. 18 (1998) 2255.