Southwest Research Institute

An experimental program has been conducted to investigate the friction and wear behavior of ceramic materials being considered for use in high-temperature adia-batic engines. Pin on disk type tests were conducted in a simulated diesel environment at 800°C using ceramic-ceramic couples made up of SiC, TiC, and TiC-Ni-Mo pins tested against Si3N4 and partially stabilized zirconia disk surfaces modified by ion beam mixing with Co, Cr, Ni, or Ti-Ni. The coefficients of friction for each couple were determined from these tests, and the amount of wear determined by profilometry of the wear surfaces. Scanning electron microscopy, Auger electron spectroscopy, and energy dispersive spectroscopy were used to characterize the wear surfaces.

The results of these tests show that most non-ion/mixed ceramic-ceramic couples have relatively high wear rates, and coefficients of friction above about 0.25. Ion beam mixing of disk material surfaces with Ti-Ni produced coefficients of friction as low as 0.06, values close to that for liquid lubricated metal couples run at much lower temperature. Surface modification with Co or Ni produced some promising results, while Cr did not improve the wear characteristics of the disk materials. Auger analysis suggests that the formation of lubricious oxide layers is responsible for the superior friction and wear behavior of the Ti-Ni ion-implanted disks.

Introduction

Improvements in low heat rejection engines and the development of a new generation of lighter, more energy efficient engines and power systems has seen the increased application of ceramic materials to their construction, materials capable of withstanding the high operating temperatures and stresses in such engines. The high temperature mechanical and chemical stability of many types of ceramics, most notably silicon nitride (SioN4) and zirconium oxide (ZK^J, has made them candidates for use not only in stationary structural and insulating parts, but, with the recent consideration of all-ceramic engines, in moving parts as well [1,2].

One of the major difficulties in the application of ceramics to the moving parts of high temperature engines has been the poor friction,and wear performance of these materials. This problem is especially critical to applications such as cylinder liners and piston rings for high temperature adiabatic engines, or high temperature bearings in gas turbine engines, where engine design requirements include high temperatures, uncooled operation, and friction and wear properties of key parts comparable to those of conventional, lubricated engines. In the last few years, numerous investigators have explored the friction and wear of ceramic-ceramic couples [3-5]. Although couples exhibiting minimum wear could be identified, wear was never negligible, and the unlubricated sliding friction coefficient was usually dis-couragingly high, i.e., uF > 0.2 [3-11]. In fact, it has been generally concluded by some investigators that ceramic components will not be used unlubricated in sliding contact engine applications at either low [3] or elevated [4] temperatures.

While this is a discouraging conclusion, especially in light of the breakdown of conventional lubricants at elevated temperature, certain observations made during these tests and other related investigations provide some insight as to how the wear properties of these materials might be improved to acceptable levels. In particular, the test environment has been shown to play a major role in the friction and wear behavior of ceramics. Coefficients of friction and wear rates have been found to be reduced when tests are conducted in oxidizing (air, water vapor) environments as compared with tests conducted in inert environments (vacuum, inert gas) [12-14]. Surface analysis of the wear surfaces produced in these environments has shown that the formation of a thin oxide layer during testing in the oxidizing environments apparently provides a form of solid lubrication between the ceramic surfaces, and is responsible for the improvement in friction and wear behavior of the ceramics tested [12-16]. An especially good example of this behavior is demonstrated by the titanium based ceramics TiC and Ni-Mo bonded TiC cermet. When one of these materials is used as a member of a ceramic/ceramic couple, it has been shown that the coefficients of friction of these pairs often are reduced as compared to other ceramic/ceramic combinations. This reduction is apparently the result of the formation and transfer of a titanium oxide layer from the titanium based ceramic member to the other member of the couple [5,17-22].

Oxygen does not necessarily have to be the lubrication causing species for improved ceramic friction and wear behavior, however. Myristic acid, for instance, enhances surface plasticity and surface cracking during sliding of steel on lithium fluoride [23], while the increased surface segregation of carbon as a graphite film on silicon carbide has been shown to lead to a dramatic decrease in the coefficient of friction at temperatures above 800°C [24].

From the results of these studies, it appears that one promising direction toward the improvement of the friction and wear behavior of ceramics is to try to enhance these apparent self-lubricating properties. This could be achieved by a modification of the material surfaces which would, for example, enhance the growth of a lubricating oxide film. One such surface modification technique which has shown initial success is the ion beam mixing of certain metals into ceramic surfaces, in particular, metals which are known to form stable, continuous metal oxides and might induce this type of lubricating behavior [21-22]. It has been shown that the surface ion mixing of a double layer of titanium and nickel in silicon nitride (SI3N4) or partially stabilized zirconia (ZrC^) has improved the friction and wear . properties of these materials in simulated diesel environments (at 800°C) to levels approaching that of conventional lubricated engines run at lower temperatures [21]. Auger electron spectroscopy indicates that the modified surface layer is apparently oxidized, and provides a lubricating layer between the ceramic surfaces [22]. The objective of this report is to discuss further results on this preliminary investigation, as well as to report on results of the ion beam mixing of other metals including Co, Cr, and Ni.

Experimental Procedure

Details of the friction and wear experiments and subsequent surface analysis are given in detail elsewhere [21-22] and are summarized briefly here. Silicon nitride and partially stabilized zirconia disks were surface modified by the ion beam mixing of Co, Cr, Ni, or a Ti-Ni double layer. The choices of metal ions to be ion beam mixed were based on the results of unimplanted ceramic-ceramic tests (Ti and Ni) as discussed previously, and on work reported in the literature [25] which seemed to indicate possible lubricating properties for oxides of Co and Cr. The metals were vapor deposited on the ceramic disk substrates, and then mixed into the surface by bombardment with argon ions. The argon ions were accelerated using a beam voltage of 140 keV, with a fluence of 10 ions/cm2, and a flux of ~ 1012 ions/cm -S. The thicknesses of the modified layers were estimated to be on the order of <400 nm, based on subsequent Auger analysis. The modified disks, as well as unmodified disks, were tested in sliding contact in a three pin-on-disk arrangement where the pin materials were titanium carbide or a Ni-Mo bonded TiC cement. Tests were run in a simulated diesel environment or an inert argon atmosphere at 800?C. The results of these tests are reported primarily in terms of the coefficients of friction of the various couples.

Surface analysis of the wear surfaces of the pins and disks were conducted using secondary electron microscopy (SEM) and

Auger electron spectroscopy (AES). The morphology of the wear surfaces was analyzed using SEM, while the chemistry of various features of the wear surfaces'was analyzed using AES. The results of these analyses were then correlated with the results of the mechanical friction and wear testing to provide a preliminary mechanism for the lubricating properties of these modified surfaces.

Results Friction and Wear Testing

Results of the friction and wear testing of the unmodified and modified ceramic-ceramic couples run in the simulated diesel environment are surranarized in Fig. 1 in terms of the steady-state coefficient of friction, pipit is clear from these results that while the coefficient of friction for pins run against unmodified (bare) disks ranges above 0.2, surface modification of the disks by the ion beam mixing of metals reduced the coefficients of friction to values below 0.1 for certain specific cases.

0 0

Post a comment