Sliding wear of quasicrystalline coatings - Springer Link

0 downloads 0 Views 61KB Size Report
Sliding wear of quasicrystalline coatings. R.P. Matthewsa, C.I. Langa and D. Shechtmanb a Department of Materials Engineering, University of Cape Town, ...
Tribology Letters 7 (1999) 179–181

179

Sliding wear of quasicrystalline coatings R.P. Matthews a , C.I. Lang a and D. Shechtman b a Department b

of Materials Engineering, University of Cape Town, Private Bag, Rondebosch, 7701 South Africa Department of Materials Engineering, Technion – Israel Institute of Technology, Haifa 32000, Israel

Received 12 June 1999; accepted 14 October 1999

The mass loss during sliding wear of several quasicrystalline (QC) coatings has been measured. QC coatings of the Al–Cu–Fe, Al–Cu–Fe–Cr and Al–Pd–Mn systems have been investigated and compared with a hardened tool steel and a WC–6% Co hardmetal. The wear rates of the coatings are in general comparable to conventional metallic materials. There is some variation in sliding wear behaviour of different QC coatings. Keywords: quasicrystals, sliding wear

1. Introduction

2. Experimental

Quasicrystals (QCs) have unique quasiperiodic atomic arrangements, including icosahedral (ψ) and decagonal (D) structures, and also exhibit properties different to conventional metallic materials. They generally exhibit high hardness and stiffness but low fracture toughness, e.g., [1], and low electrical and thermal conductivities, e.g., [2,3]. Their low fracture toughness compromises the usefulness of QCs in engineering applications, but this can be overcome by utilisation of QCs in the form of surface coatings on metallic substrates. The friction coefficients of QCs are also reported to be very low, which, in combination with their high hardness, has led to suggestions that QC materials are potentially useful as wear resistant materials, e.g., [4]. Although friction and scratch test results have been widely reported, quantitative data on wear rates determined from mass loss measurements are relatively uncommon in the scientific literature. Sordelet et al. [5] conducted abrasive wear testing of Al–Cu–Fe quasicrystalline coatings containing varying amounts of FeAl, and reported on volume loss as a function of composition. Inoue and Kimura [6] have reported wear rates for the sliding wear of bulk quasicrystals as a function of sliding velocity. In the work presented here, multiple-pass sliding wear tests have been performed on several QC coatings and steady-state wear rates have been determined, in order to evaluate the potential of QC materials for sliding wear applications. The coatings evaluated are representative of a range of quasiperiodic structures and include Al65 Cu23 Fe12 (ψ phase) [7], Al70.9 Cu9 Fe10 Cr10 B0.1 (ψ phase + D phase) [8] and Al70 Pd20 Mn10 (ψ phase + D phase) [9] alloys. These QC alloys have high hardness and low friction coefficient, e.g., [2,4,10–12], and are hence good candidates for further investigation as wear resistant materials.

Quasicrystalline coatings of the five compositions listed in table 1 were prepared by a plasma spray technique, and were deposited on metallic substrates including aluminium, steel, nickel and titanium alloys, to a thickness of 0.5 mm. Substrate composition did not appear to influence the structure and properties of the coatings. The phase structure of the QC coatings, determined by X-ray diffraction, is also shown in table 1. All the coatings contained quasicrystalline phases together with periodic phases. In the Al65 Cu23 Fe12 and Al65 Cu23 Fe12 + 1 wt% FeAl QC coatings the majority phase was the periodic β phase, together with the icosahedral phase. The other QC coatings were mostly icosahedral with lesser amounts of decagonal and periodic phases: the Al70.9 Cu9 Fe10 Cr10 B0.1 QC coating contained icosahedral and periodic phases with a small amount of decagonal phase, whereas the Al70 Pd20 Mn10 and Al70 Pd20 Mn10 + 1 wt% FeAl contained icosahedral and decagonal phases with a small amount of periodic phase. Table 1 also shows measured hardness and the density values used to calculate volume loss. For comparison purposes, AISI 01 hardened tool steel and WC–6% Co specimens were also prepared. Specimens were cut to a size of 4.7 mm × 4.7 mm, giving a specimen surface area of 22 mm2 . Surfaces were polished to a surface roughness between Ra = 0.1 and 0.2. Three specimens of each composition were subjected to sliding wear tests. Sliding wear tests were performed at ambient temperatures in air, using a standard pin-on-disk apparatus. Sliding was carried out against an AISI D3 cold worked tool steel disk counterface, heat treated to produce a hardness in excess of 620 HV and ground to an average roughness between Ra = 0.1 and 0.2. The specimens were loaded against the counterface disk with a force of 4.2 N. The sliding speed was 0.4 m/s. Tests were interrupted at predetermined sliding distance intervals (ranging from 500 to 2000 m, depending on the coating composition) followed by

 J.C. Baltzer AG, Science Publishers

180

R.P. Matthews et al. / Quasicrystalline coatings Table 1 Test materials. Test specimens QC coating composition QC phase structure Al65 Cu23 Fe12 Al65 Cu23 Fe12 + 1 wt% FeAl Al70.9 Cu9 Fe10 Cr10 B0.1 Al70 Pd20 Mn10 Al70 Pd20 Mn10 + 1 wt% FeAl

ψ + periodic phase ψ + periodic phase ψ + D + periodic phase ψ + D + periodic phase ψ + D + periodic phase

AISI 01 tool steel WC–6% Co

Hardness HV30 kg 350 366 477 501 406 774 1600+

Density (kg/m3 ) 4000 4000 4000 4000 4000

(est.) (est.) (est.) (est.) (est.)

7580 1495

Figure 1. Wear rates and hardness of QC coatings and conventional materials.

ultrasonic cleaning and weighing to an accuracy of 0.1 mg. Testing was continued until steady-state wear was well established. Mass loss was converted to volume loss and wear rates were calculated from the linear portion of the volume loss versus distance curves.

3. Results Figure 1 shows the steady-state wear rates and hardness of the five experimental QC coatings, together with the results from the tool steel and WC–Co specimens. For comparison purposes, volume wear rates for all the QC coatings were calculated using an estimated density of 4 g/cm3 . Differences in density between different QC compositions are not likely to be of sufficient magnitude to alter the relative performance observed in the QC coatings. Wear rates shown for each composition represent an average of the steady-state volume wear rate of three specimens. Wear rate error bars show the maximum and minimum wear rate for each composition; hardness error bars show plus and minus standard deviation. The Al65 Cu23 Fe12 and Al65 Cu23 Fe12 + 1 wt% FeAl QC coatings exhibit similar wear rates, which are considerably higher than those of the Al70.9 Cu9 Fe10 Cr10 B0.1 , Al70 Pd20 Mn10 and Al70 Pd20 Mn10 + 1 wt% FeAl QC coatings. The latter three QC coatings all perform similarly, and have steady-state wear rates comparable with the tool

steel specimen. In view of the fact that there is considerable interest in the potential of QC materials for wear resistant applications, it is of interest to note that the QC coatings investigated here do not approach the exceptionally low wear rates of materials such as WC–Co hardmetals. Nevertheless, QC materials may offer benefits over metallic materials, with which their wear rates are comparable, owing to their other unique properties.

4. Discussion Under the test conditions described, the QC coatings exhibit wear rates which are similar to, or higher than, the metallic materials tested. This is not consistent with previous (friction and scratch test) comparisons between QC materials and metallic materials which have suggested that QC materials have improved wear resistance compared to metals, e.g., [11,13]. These studies have, however, compared QC materials with metallic materials of relatively low hardness. The QC coatings investigated in the present study differ in having a lower hardness than the metallic specimens with which they are compared. It is of interest to note that the Al70.9 Cu9 Fe10 Cr10 B0.1 QC coating shows a wear rate comparable with hardened tool steel, even though the hardness of the Al70.9 Cu9 Fe10 Cr10 B0.1 QC coating is significantly lower than that of the counterface whereas that of the tool steel specimen is higher. Inoue and Kimura [6] also

R.P. Matthews et al. / Quasicrystalline coatings

measured the wear rate of a quasicrystalline alloy (icosahedral Al–Fe–Cr–Ti) against a hard steel, and obtained a wear rate similar to that of the Al70.9 Cu9 Fe10 Cr10 B0.1 QC coating. The differences in performance between the different QC compositions are of interest. There is a clear difference in wear rate between the two Al65 Cu23 Fe12 QC coatings (Al65 Cu23 Fe12 and Al65 Cu23 Fe12 + 1 wt% FeAl) and the three other coatings: the Al70.9 Cu9 Fe10 Cr10 B0.1 coating and the Al70 Pd20 Mn10 QC coatings (Al70 Pd20 Mn10 and Al70 Pd20 Mn10 +1 wt% FeAl). Consideration of minor variations in composition shows that the addition of 1 wt% FeAl (to the Al65 Cu23 Fe12 and the Al70 Pd20 Mn10 QCs) has little effect on the measured properties. In contrast, Sordelet et al. [5] measured the influence of FeAl on the abrasive wear of Al65 Cu23 Fe12 and found that the addition of 1 wt% FeAl resulted in a decrease in mass loss. The performance of these QC coatings in sliding wear is thus less sensitive to the addition of FeAl than the abrasive wear performance. Although the Al65 Cu23 Fe12 and Al65 Cu23 Fe12 +1 wt% FeAl QC coatings show a slightly lower hardness than the other QC coatings, it is unlikely that this small hardness difference alone accounts for the significant differences in wear resistance. Similarly, although differences in oxidation rate could possibly contribute to differences in wear resistance, they are unlikely to account for the results observed. Previous work has shown that as-sprayed Al65 Cu23 Fe12 QC coatings exhibit lower friction coefficients than as-sprayed Al70.9 Cu9 Fe10 Cr10 B0.1 and Al70 Pd20 Mn10 QC coatings [14]. Since the Al65 Cu23 Fe12 QC coatings have the highest wear rate, clearly a low coefficient of friction does not necessarily confer a low sliding wear rate in QC coatings. Consideration of the phase structure of the coatings shows that the QC coatings which exhibit the lowest wear rates exhibit a similar structure (predominantly icosahedral, with decagonal and periodic phases). This suggests that a predominantly quasicrystalline structure offers advantages in sliding wear. In conclusion, the unusual properties of QCs do not necessarily result in exceptional sliding wear resistance, nor do

181

QC materials exhibit uniformly low wear rates. Indeed, QC coatings with different phase structures behave differently under the same wear testing conditions. Further work on the relationship between QC structure and wear resistance will be necessary in order to determine the potential of QCs as wear resistant materials.

Acknowledgement The coatings were prepared with assistance from F.S. Biancaniello of the Metallurgy Division, National Institute of Standards and Technology, MD, USA and D. Sordelet of Ames Lab., Ames, IA, USA.

References [1] U. Koster, W. Liu, H. Liebertz and M. Michel, J. Non-Cryst. Solids 153/154 (1993) 446. [2] J.M. Dubois, S.S. Kang and A. Perrot, Mater. Sci. Eng. A 179/180 (1994) 122. [3] Y. Yokayama, A. Inoue, A.-P. Tsai, A. Inoue and T. Masumoto, Mater. Trans. JIM 33 (1992) 97. [4] J.M. Dubois, S.S. Kang and J. von Stebut, J. Mater. Sci. Lett. 10 (1991) 537. [5] D.J. Sordelet, M.R. Besser and J.L. Logsdon, Mater. Sci. Eng. A 255 (1998) 54. [6] A. Inoue and H. Kimura, Nanostruct. Mater. 11 (1999) 221. [7] A.-P. Tsai, A. Inoue and T. Masumoto, Jpn. J. Appl. Phys. 29 (1987) L1505. [8] C. Dong and J.M. Dubois, J. Mater. Sci. 26 (1991) 1647. [9] C. Beeli, H.-U. Nissen and J. Robadey, Philos. Mag. Lett. 63 (1991) 87. [10] J.M. Dubois, S.S. Kang and Y. Massiani, J. Non-Cryst. Solids 153/154 (1993) 443. [11] S.S. Kang, J.M. Dubois and J. von Stebut, J. Mater. Res. 8 (1993) 2471. [12] A. Inoue, Y. Yokayama and T. Masumoto, Mater. Sci. Eng. A 181/182 (1994) 850. [13] Y. Ding, D.O. Northwood and A.T. Alpas, Surf. Coat. Technol. 96 (1997) 140. [14] C.I. Lang, D. Shechtman and E.G. Gonzalez, Bull. Mater. Sci. 22 (1999) 101.