Nuclear Instruments and Methods in Physics Research B21 (1987) 591-594 North-Holland, Amsterdam
W E A R B E H A V I O U R OF N I T R O G E N I M P L A N T E D S T A I N L E S S STEEL Sadhna SHRIVASTAVA*, K.L. CHOPRA +
A m i t a b h J A I N * , A. S E T H U R A M I A H * ,
V.D. V A N K A R * *
Indian Institute of Technology, Hauz Khas, New Delhi 110016, India
Samples of 304 stainless steel were implanted with N_,+ ions. Wear tests were carried out on a sample implanted with a dose of 5 )< 1016 N~ cm -2 at 130 keV and stored at room temperature for 10 months. Several other samples were tested within a few days after implantation. The former showed remarkable wear resistance whereas the latter failed to exhibit wear resistance. X-ray diffraction revealed the formation of )"-Fe4N in the aged sample whereas other samples showed little or no evidence of it. Artificial aging of one of the discs was also tried by heating it at various temperatures in the range 85-135°C but it failed to exhibit wear resistance. The oxide on each sample is found to contain Fe304 and/or Cr302. The existence of oxide is confirmed by Auger electron spectroscopy.
1. Introduction Ion implantation has considerable effect on the near-surface mechanical properties of metals. In particular the role of ion implantation in producing wear resistance is the subject of extensive study. Hirvonen  found that, under lubricated conditions, nitrogen implanted type 304 stainless steel showed a considerably lower wear rate than unimplanted reference material. It has been found that process parameters such as dose rate, temperature attained during implantation and target chamber pressure are critical in their influence on the wear behaviour of a specimen implanted with a given species at a specified dose and energy. The works of G o o d e and Baumvol  and of Ecer et al. , for example, demonstrate this fact. In the latter work, changing from one implanter to another with a different dose rate and target chamber pressure resulted in failure to produce wear resistance. A complete understanding of the dependence on process parameters is not yet available. Since the relationship between specimen temperature and dose rate varies from implanter to implanter, a study is required in each case to establish the use of a given implanter for the production of a wear resistant surfacc. Depending on implantation temperature and temperatures attained during the wear test, postimplantation treatment such as aging could also be critical [1,4].
*Industrial Tribology Machine Dynamics and Maintenance Engineering Centre. **Centre for Materials Science and Technology. " Physics Department. 0168-583X/87/$03.50 © Elsevier Science Publishers B.V. ( N o r t h - H o l l a n d Physics Publishing Division)
We describe here results obtained on stainless steel samples implanted in a Varian DF3000 implanter with a standard platen.
2. Experimental methods Various discs of 76 mm diameter were cut from 304 stainless steel sheets. Pins with a diameter of 5 mm at the nominally flat tip were made from a 410 stainless steel rod. The discs were cleaned ultrasonically in acetone, benzene and then in isopropanol. They were then implanted with N ~ ions. The target chamber pressure was of the order of 10-5 Torr. Wear tests were carried out using a pin on disc configuration on a friction and wear testing machine (Swansea Tribology Centre, UK). The pins were positioned 120 ° apart on a circle of diameter equal to 40 mm. The lubricant used was clean kerosene. The total load was 13.5 N and the disc was rotated at 200 rpm. The pins were polished before the tests on emery papers of successive grit sizes in the range 120-600 in the pin and disc machine just described. At several time intervals during a test, the pins were weighed on an electronic balance. The pins were cleaned ultrasonically in acetone and benzene before each weighing. The loss in the weight of the pins was used to determine the wear parameter K. The wear parameter is defined as K = V H / ( L W ) , where V is the volumetric wear loss, L is the sliding distance, H is the hardness value of the pin material and W is the load. The disc wear was measured by profilometry on a Talysurf (Taylor-H0bson) at various stages of a test. The wear was estimated by the X. METALS/CERAMICS/SEMICONDUCTORS
S. Shrioastava et al. / Wear behaviour of N implanted stainless steel
mean depth of the wear track and was determined by averaging over four locations 90 ° apart. Samples were analysed using an X-ray diffractometer (.Rigaku D / m a x yB-RU200H) using Cu K~ (k = 1.54 A) radiation at a glancing angle of ct = 3 ° and scanning angle, 20 (where 0 is the Bragg angle), from 10 ° to 90 ° at a scan speed of 5°/rain. Samples were also analysed using an Auger electron spectroscope PHI model 590A at a sputtering rate of approximately 100 ,~k/min.
3. Results and
Weight loss versus time for pins rubbed against an unimplanted disc and a disc implanted to a dose of 5 >( 1016 N~ cm-2 at 130 keV at a dose rate of 0.77 ~ A c m - : and stored at room temperature for about ten months is plotted in fig. 1. There is a remarkable reduction in wear following ion implantation. In the first half hour of the test, for the unimplanted case K is calculated to be 7.6 x 10 -4 and for the implanted case
,2 o G
l' m E
K equals 3.3 X 10-5. In the unimplanted case one observes a running-in period of 120 rain after which the wear rate reduces. In the implanted case, the running-in stage is less pronounced. Wear rates are so low that each pin had made contact over an area of 1 mm2 only at the end of 240 min. The implantation induced wear resistance persists for 2520 min which corresponds to a sUding distance of 63.3 kin. Beyond this sliding distance the wear becomes higher than that in the unimplanted case. The wear depth for the unimplanted disc during the first half hour is 15 /~m and in subsequent tests it increases slowly and reaches a value of 25 /zm. In contrast the groove depth on the implanted disc is insignificant at the end of the first half hour and does not increase appreciably as long as wear resistance persists. The wear tracks on the unimplanted and the implanted disc were examined under an optical microscope. In the unimplanted case, the wear track shows ruts typically 300/~m long with a pile of material at the end of each rut. On the implanted disc, wear tracks were smooth by comparison; they only showed continuous striations. Other implanted discs were tested within one to six days of implantation and did not exhibit wear resistance. The wear behaviour and nature of the wear track observed on the disc which showed low wear are similar to that reported by H_irvonen . The X-ray diffractogram (fig. 2) of this disc suggests the presence of y'-Fe4N through the existence of a peak at 41.1 °. Table 1 shows positions of peaks that are not due to the steel lattice together with reference data that identify these peaks.
2 ~D o
Fig. 1. Weight loss as a function of time for pins rubbing against an unimplanted ((D) and an implanted (~,) (5 x 1016 N~ cm -2, 130 keV) disc.
= 5 x 1016N~ cm 2
Fig. 2. Glancing angle X-ray diffractogram for a 304 stainless steel disc implanted with 5 x 1016 N~ cm -2 at 130 keV.
S. Shrivastava et a L / Wear behat,iour of N implanted stainless steel
Table 1 Experimental and reference 20 peak positions in X-ray diffraction. Values in brackets indicate peak intensities. Reference peaks of various compounds
35.5 41.1 57.1 62.9
In contrast the samples which failed to exhibit wear resistance upon being tested soon after implantation showed little or no evidence of Fe4N in their X-ray diffractograms. This suggests that nitride formed as a result of aging plays an essential role in achieving wear resistance. Another sample was implanted at a dose rate of 0.44 /~Acm -2 and annealed isochronally for 8 h at temperatures ranging from 85°C to 135°C at intervals of 10°C. The sample was tested after each anneal but failed to exhibit wear resistance even though the anneal temperatures towards the higher end of the range would be expected to produce an amount of diffusion comparable to that in the aged sample. Also the X-ray diffractograms after the 135°C anneal did not show nitride formation to the extent of the aged sample. This is not yet understood. Nitrides of Cr and Fe have been reported previously in stainless steel following nitrogen implantation [5,6]. Baron et al.  detected CrN by electron diffraction. In
----..... .......... .....
IRON CHROMIUM NITROGEN NICKEL OXYGEN
3 20 o ~ < 10
J }: ....
the present work the strongest peak of CrN (at 43.7 °) would be masked and other known peaks are weaker and may not show up for that reason. All X-ray diffractograms in the present work show the presence of Fe304 a n d / o r Cr302 following ion implantation. The peaks leading to this identification lie at 35.5 °, 57.1 ° and 62.9 °, the last one appearing as a shoulder on the left of a peak due to the steel lattice (see table 1 for a comparison with reference data). The existence of oxide has been mentioned by Hartley  who pointed out that the mechanisms leading to oxidation are beam induced. As in the previous work, we see that the colour associated with an oxide layer ends abruptly at the edges of implanted regions. Fig. 3 shows the Auger depth profiles for the sample which exhibited wear resistance. The depth profiles confirm the existence of oxide. X-ray diffractograms for samples implanted with 1017 N~ c m - 2 showed twice the amount of oxide compared to samples implanted with 5 x 1016 N~ cm-2 independent of the dose rate over a wide range. This is in general agreement with the observations and the model of Goode and Banmvol . It is seen in the present work that the effects of aging can be of importance in samples ion implanted to produce wear resistance. A similar observation has been made previously in connection with the use of ion implantation for improving fatigue life . The present work has demonstrated the use of glancing angle X-ray diffraction in the identification of phases formed on ion implantation. As indicated above previous studies have relied on other techniques such as electron diffraction which involve more complicated sample preparation. Further work is underway to understand further the influence of process and postimplantation treatment parameters on the wear behaviour of implanted solids.
Fig. 3. AES depth profiles for 304 stainless steel implanted with 5 x 1016 Nf crn -2 at 130 keV.
We are grateful to Prof. A.B. Bhattacharyya for his kind interest and to Dr. Urea Jain for a helpful discussion. X. METALS / CERAMICS / SEMICONDUCTORS
S. Shrivastava et a L / Wear behat,iour of N implanted stainless steel
References  J.K. Hirvonen, J. Vac. Sci. Technol. 15 (1978) 1662.  P.D. Goode and I.J'. Baumvol, Nucl. Instr. and Meth. 189 (1981) 161.  G.M. Ecer, S. Wood, D. Boes and J. Schreurs, Wear 89 (1983) 201.
 G. Dearnaley and P.D. Goode, Nucl. Instr. and Meth. 189 (1981) 117.  M. Baron, A.L. Chang, J. Schreurs and R. Kossowsky, Nucl. Instr. and Meth. 182/183 (1981) 531.  F.G. Yost, S.T. Picraux, D.M. Follstaedt, L.E. Pope and J.A. Knapp, Thin Solid Films 107 (1983) 287.  N.E.W. Hartley, Thin Solid Films 64 (1979) 177.