L I

10 20 90 40

10 20 90 40

Figure 2. Typical Elemental Depth Profiles Taken a) Before and b) After Friction and Wear Testing of a TiNi Modified SiJi. Disk (Diesel Envfrdn-ment, 800°C).

The results of the tests on Co modified disks show behavior in some ways similar to the Ti-Ni implanted disks. The Co modified zirconia disks have a low coefficient of friction when run against the TiC-Ni-Mo pin, and a much higher value when run against the TiC pin. However, unlike the Ti-Ni case, the SioN4 disks showed no improvement in the coefficient of friction when ion beam mixed with Co.

Auger analysis of the Co modified zirconia disks and the corresponding pins indicates that cobalt transfers from the disk surface to the pins. Elemental depth profiles indicate that both the pin and the disk Co layers are oxidized, implying the presence of a lubricating cobalt oxide film analogous to the Ti-Ni case. Disturbing, however, are the physical characteristics of the wear tracks, especially for the low coefficient of friction disk. Both wear tracks showed del ami nation of the Co layer, leaving exposed substrate. The wear track of the low coefficient of friction disk varies in width and amount of wear; while it is thus possible that the zirconia/ NiMo-bonded TiC results were influenced by the alignment of the pins and disk, the low coefficient of friction obtained indicates the transfer and possible lubricating effect of a cobalt oxide layer.

AES analysis of the surface of the NiMo-bonded TiC pin run against the Co modified disk also show the strong presence of nickel. This indicates that nickel segregates to the pin surface, and may play a role in the friction and wear behavior of the ceramic couples. This also implies that nickel may have segregated from the pins in the case of the TiNi modification tests, but would not be observable due to the Ni component of the modified layer.

The results of tests involving the Ti-Ni and the Co modified disks suggest that Ni may play a particularly critical role in the friction and wear behavior of the ion beam modified ceramics, either as a part of the modified layer, or due to its presence in the pins. Therefore, two Si^N^ disks were ion beam modified with a layer of Ni and tested in diesel environment at 800°C and at room temperature.

The results of the friction and wear tests indicate that Ni improved the coefficient of friction of SijN^, but not to the same degree as did the Ti-Ni modification, or the Co in zirconia modification, see Table 1. Analysis of the wear surfaces of the disks and the pins showed that the Ni layer was completely worn off the disk, and that no nickel transferred to the pins. The wear track shows cracking, especially near the edge of the crack, typical of the wear tracks of unmodified ceramics. The pin surfaces show clear signs of wear, as opposed to the adherence of a transfer layer seen in the case of Ti-Ni and Co. AES analysis of the unworn areas after testing shows, however, that the Ni layer was much better mixed into the substrate than in the two previous cases.

Discussion

The results of this investigation indicate that solid lubrication of ceramic materials at high temperatures is possible through the use of the technique of surface modification by ion beam mixing of oxide-forming metal ions. In order to optimize the conditions for the application of this technique to real systems, several factors must be considered. These include the characteristics and adhesive properties of the modified surface layer as governed by the operating environment, the ion beam mixing conditions, and the choice of material substrates; the mechanism of lubrication by the modified layer, including its chemical state, mechanical and structural properties, and the possibility of its transfer from one material to the other; and the choice of the material to be run against the modified material.

The adhesion at the ion beam modified layer-ceramic boundary is clearly important in the del ami nation-type wear observed, since this will govern the useful lifetime of the component. While this investigation was conducted on a screening basis, emphasizing short-term tests to find promising ceramic-ceramic couples for immediate use in an experimental engine [21], the results of the analyses of the wear surfaces suggest that much longer lifetimes may be attainable through an optimization of the ion beam mixing conditions and/or the choice and condition of the substrate materials and metal ions to be mixed. The previous study [21] showed that the Ti-Ni modification appeared to mix the best in the Si3N4 substrate. While the amount of wear on the Co modified disks was not satisfactory, the surface morphology of the wear surface of the Co modified zirconia indicates that the degree of initial surface roughness, or initial microstructure in terms of grain size also may play an important role in the adhesion of the modified layer.

The results of the tests conducted on disks modified only with Ni indicate that while Ni alone does not improve the friction and wear behavior of ceramics as well as other species, it may help provide better bonding of those species and/or their lubricating oxides. In the case of the Ti-Ni modification, titanium oxide appears to be responsible for the improved friction and wear behavior, but the Ni layer appears to be the bonding species which mixes into the substrate, especially in the case of Si3N4. In the case of the Co modification, segregation of Ni from the NiMo-bonded TiC pin may be responsible for the transferred Co adhering to the pin surface at least temporarily, thus delaying the degradation of the friction and wear properties caused by removal of the Co layer from the disk surface. Further work is required to determine the influence of Ni on the bonding and/or friction and wear behavior of ion ceramics. Also of interest would be to determine those species which can be coupled with Ni in a modified layer to provide those beneficial results. For example, the ion beam mixing of Cr alone apparently had no beneficial influence on the friction and wear behavior of either Si3N^ or zirconia, even when run against NiMo-bonded TiC pins. Cr ion-mixed with Ni may, however, provide an improvement similar to TiNi modification.

The transfer of material, or the formation of a surface oxide has been shown to be beneficial to the wear characteristics of the TiNi-modified ceramics, as well as to many other material systems in general [21-22, 27-28]. From the results of the present investigation, Ni appears to have been transferred from the disks to the pins, or vice versa, as indicated by the results obtained from the pins run against TiNi or Co modified disks. Although in the case of TiNi modification, it is highly likely that Ti also transferred to the pins, it was not possible to determine definitely whether or not this was the case; a radioactive tracer technique would be useful for this determination for either Ni or Ti transfer. There appears to be no doubt that the modified and transferred layers had been oxidized, and that this oxide contributed to the improvement in friction and wear properties of the modified ceramics.

Further investigations are underway to determine the characteristics of these oxides under the extreme conditions expected at the contact surfaces in order to determine the mechanism of their lubricating effect. Characteristics such as oxide chemistry, mechanical or flow properties, and amorphous versus crystallographic structure would be expected to play important roles in the lubricating phenomenon. Some clues may be derived from phase diagrams [29]. The oxides of nickel show a strong dependence of melting points on stoichiometry. Whereas NiO melts at close to 1600°C, there is a sharp continuous decrease in melting temperature with increase in the fraction of oxygen; Nio03 melts at less than 400°C. It is quite plausible, therefore, that the tribological system of the sliding ceramics was lubricated by a film of fluid Nix0y, where y/x is greater than 1. The relationships in the Co-0 system are less defined. It has been shown, however, that the stoichiometric CoO is the only form stable above 1000°C. The higher oxygen compounds show significant instabilities at high temperatures where C03O4, for example, dissociates at 900°C. In contrast to the above, the oxides of Cr or Ti display high stabilities across the entire range of oxygen potentials. All titanium oxides are stable to at least 1600°C. Similar observations hold for chromium oxides [30].

Related to the material transfer question is the choice of material to be run against the modified ceramic. This choice would be made based not only on the actual chemical processes discussed here, but also on the specific applications. The current investigation was conducted to screen ceramic-ceramic couples for use as cylinder liners and piston rings in high temperature, adiabatic engines. As.such, the material simulating the piston ring (the pin) would see constant sliding contact and thus would be susceptible to more wear than any particular point on the cylinder liner. Thus, for these tests, it was believed that the use of the harder modified ceramic which had some oxidizing capability (Ti) as the pin material would be more advantageous than modifying the pin surface. Other types of sliding applications could, however, benefit from the use of modified ceramics for both members of the couple.

From the previous investigation, the choice of pin material played a confusing role in the combinations 1-4 involving the Ti-Ni modification. Ni was clearly transferred to all four pins [22], but the two pin materials had opposite effects with respect to disk material. The role of Mo comes into question, but because of its volatility at high temperature, was not detected on any wear surfaces using AES. On the other hand, Ni segregated from the NiMo-bonded TiC pins may have played a role in the case of the Co-modified disks.

Conclusions

Pin-on-disk type friction and wear tests have been conducted on a series of ceramic/ceramic couples in the unmodified state, and surface modified by ion beam mixing, in a simulated environment typical of high temperature adiabatic diesel engines. Studies of the morphology and chemistry for the resulting wear surfaces have been conducted and correlated with the friction and wear behavior of these materials.

It has been found that the surface modification of ceramics by the ion beam mixing of Ti-Ni or Co reduces the coefficient of friction (for certain specific pin-disk combinations) to levels falling in the upper end of the range of values considered acceptable for conventional, liquid lubricated engines. On the other hand, the ion beam mixing of Cr or Ni does not appear to have a beneficial effect on the friction and wear behavior of the materials tested. The improvement in friction and wear behavior of the ceramic couples modified with Ti-Ni and Co is apparently the result of the formation of a lubricating oxide of those species. Significant differences among all of the material combinations examined, however, indicate that several factors must be investigated before this surface modification technique can be applied to real systems. These include:

1. the characteristics and adhesion properties of the modified surface layer under operating conditions,

2. the effect of ion beam mixing conditions on the surface modified layer and its ability to provide lubricating properties,

3. the mechanism of lubrication by the oxide layer, and

4. the choice of materials and metals to be ion mixed for optimal lubricating behavior.

Acknowledgment

This work was conducted under a program sponsored by the U. S. Department of Energy and technically monitored by the Lewis Research Center of the National Aeronautics and Space Administration under Contract No. DEN3-352.

References

1. J. J. Burke, et al, editors, Ceramics for High Performance Applications-II Proc. 5th Army Materials Tech. Conf., Newport, RI, March 21-25, 1977 (Brook Hill Pub. Co.; Chestnut Hill, MA) (1978).

2. R. Kamo and W. Bryzik, "Cummins/TACOM Advanced Adiabatic Engine" in Adiabatic Engines; Worldwide Review SP-571, (SAE, Inc.; Warrendale, Wj (1984) pp. 21-34.

3. J. Breznak, E. Breval, and N. H. MacMillan, Journal Mat. Sci. 20 (1985) p. 4657.

4. T. Shimachi, T. Murakami, T. Nakagaki, Y. Tsuya, and K. Umeda, Adiabatic Engines: Worldwide Review SP-571, Society of Automotive Engineers, Warrendale, PA (1984) pp. 35-45.

5. D. J. Boes, Proceedings 22nd CCM, Society of Automotive Engineers, Warrendale, PA (1984) p. 323.

6. E. Rabinowicz, Friction and Wear of Materials, Wiley, NY (1965).

7. K. Miyoshi and D. H. Buckley, in Tri bology in the '80s 1 NASA CP-2300, NASA, Washington, DC, (1984) p. 291.

8. 0. 0. Adewoye and T. F. Page, Wear 70 (1981) p. 37.

9. D. W. Richerson, L. J. Lindberg, W. D. Carruthers and J. Dahn, Ceram. Eng. Sci. Proc. 2 (1981) p. 578.

13. P. Sutor, "Tribology of

Silicon Nitride and Silicon Nitride-Steel Sliding Pairs," in Proc. 9th Annual Conf. on Composites and Advanced Ceramic Materials, F. D. Gac, Chairman, Am. Ceramic Soc., January 20-23, 1985, Cocoa Beach, FL (1985) pp. 460-469.

0 0

Post a comment