Materials Letters 12 ( 1991) 16-20 North-Holland
Studies of point defects and tribological properties of nitrogen-implanted stainless steel J. Dtyzek a, J. Wiezorek b, S. Wollschliger b, J. Lekki ‘, A. Gottdang b and B. Cleff b ’ Institute~1NuclearPhysics, ul. Radzikawsk~ego1.52,PL-31-342 Krakow,Pa~a~d b InstituteofNuclear Physics,UniversityofM&zster, Wilhelm-Klemm-Strasse 9,0-4400 Miinster,FRG Received I8 January 199 1; in final form 1 May 199 1
An austenitic stainless steel (X lOCrNiTil89) was implanted with 3x lOi cm-* and 8 X IO” cm-’ 100 keV N+. Wear, friction and microhardness were measured and correlated with the positron trap density determined by temperature-dependent positron annihilation spectroscopy. Whereas wear and hardness were improved by a combination of implantation and tempering, no effect on the friction coefftcient was found. These results are considered to be due to the temperature-dependent formation of pointdefect agglomerates and nitride precipitates.
1. In~u~on In this paper we present investigations of nitrogenimplanted austenitic stainless steel ( SS) (X 1OCrNiTi189) performed by positron annihilation spectroscopy (PAS), microhardness testing, as well as by friction and wear measurements. PAS techniques are useful in determining the behavior of vacancy-type defects. This is due to the selective sensitivity of positrons to local changes in the electron density of the investigated lattice. One can observe these changes by the positron lifetime or by the Doppler broadening of the annihilation quanta. The presence of vacancies, clusters of vacancies and voids causes a pronounced narrowing of the angularcorrelation curve of annihilation quanta. Selecting looking at small angles of the curve by measuring the peak height one has the opportunity to observe the behavior of these defects with changing physical conditions, e.g. temperature. There were promising prospects for successful application of the PAS technique because of the homogeneity of the austenitic structure of a SS widely used as an engineering material. Positive effects on the tribological characteristics of SS by ion implantation were reported by different authors [ 1,2]. Inherent in the process of ion beam modification of materials are microstructural, com16
positional and topog~phical changes of the implanted layers which affect the macroscopic mechanical properties [ 3 ]. It was the aim of this work to correlate information obtained by the different methods used here, in order to get more insight into processes introduced by thermal treatment of nitrogen-implanted stainless steel.
2. Experiment In our investigations we used disc-shaped SS samples (20 mm in diameter) polished mechanically to a mean surface roughness < 0.02 pm and cleaned by supersonic bath techniques before use. Homogeneous surface implantation was done by computercontrolled scanning of the ion beam across the sample at the 400 kV heavy-ion accelerator at the Institute of Nuclear Physics, University of Mtinster. A hollow cathode ion source (HVEE-SO-55) provided ion current densities between 100 and 200 l.tAcmw2 for nitrogen ions. The implantation energy was fixed at 200 keV for Nt+ ions. Tribological properties were measured under high-vacuum conditions (total pressure D< 1 x 10e5 mbar) using a ball-on-disc tribotester. The time characteristics of friction and wear, measured by the horizontal and vertical displacements of the load arm during the tribotests with
0167-577x/911$03.50 0 1991 Eisevier Science Publishers B.V. AU rights reserved.
Volume 12, number 1,2
September 199 1
a force transducer and a displacement transducer, respectively, were simultaneously recorded by a computer-controlled data acquisition system. The normal load of 0.5 N was applied between the steel balls ( 1OOCr6) and the treated SS discs. The average relative sliding speed was 11 cm/s. No lubricants were used in our tribological measurements. To obtain the positron annihilation characteristics, an apparatus for measurement of the peak height of the two-quanta angular correlation of positron annihilation as a function of temperature was used [ 4,5] at the INP Krakow. In this experimental arrangement a “Na source ( 1 mCi) emitted positrons into the sample and the annihilation radiation was detected by two detectors placed directly opposite each other around the sample. The geometric angular resolution was 3.3 mrad in the y direction and 378.8 mrad in the x direction. The sample holder could be heated to temperatures of 320°C. Source, sample and heater were placed in a vacuum chamber evacuated to pressures p < 10 -’ mbar. Additionally, microhardness measurements were made with a standard Vickers hardness tester using a load of 5 g (HV 0.005).
taken before and after each PAS cycle. Work hardening by the friction treatment was studied by measuring the microhardness outside and inside of the wear trace. Figs. 1 and 2 show the results of the PAS measurements, figs. 3 and 4 show the wear and friction behavior after 4 h of testing ( > 20000 cycles). The Vickers microhardness results for the differently treated samples are given in table 1. In fig. 1, the PAS
3. Results and discussion Samples of austenitic SS (X10CrNiTi189) were implanted with 100 keV N+ of doses 3x 10” ions cm-2 and 8x 10” ions cm-2. The influence of the mechanical treatment of the samples before implantation, i.e. the manufacturing and polishing procedures, was investigated by the use of samples without and with stress annealing where the SS was heated up to 550°C for 90 min followed by cooling down over 16 h. Positron annihilation spectroscopy (PAS) using the peak height method was done for these various SS samples at different temperatures in a heating cycle from room temperature to 320°C and then cooling down again. Each data point was obtained for 2 h to obtain good statistics. The friction and wear measurements were done for nitrogen-implanted SS as well as for non-implanted material. In order to correlate these measurements with the PAS data, samples also were tempered under high-vacuum conditions at 100, 2 10 and 3 10°C. The Vickers microhardness HV 0.005 was
Fig. 1. Peak height in temperature-dependent PAS of austenitic SS (XI OCrNiTi 189) 0 without, and 0 with stress annealing.
.c c 3
tttttt, -;r;++; + +t t
)144 + 4 1
4 0 2
m ._E Y : a
tf I 444
+tt t+ t4 4441 :,,4444 44
444& 5 tt
b ‘tfi t4’
Fig. 2. Peak height in temperature-dependent PAS of austenitic SS implanted with 3x IO” crne2 (0) and 8x IO” crnm2 (0) 100 keV N+.
Volume 12, number 1,2
September 199 1
Fig. 3. Dependence of displacement of the tribotester load arm on tempering temperature for austenitic SS.
3.2 3.1 3.0 2.9 2.8 2.7 2.6
Fig. 4. Dependence of the coefficient of friction on tempering temperature for austenitic SS.
peak height measurement of the non-implanted, not stress annealed material does not show a pronounced dependence on the temperature on heating (about 1.08 arbitrary units). When cooling down, the peak counts diminish from z 1.07 at 320’ C to x 1.00 at 40°C. After being stress annealed the material shows peak counts of x 1.02 over the whole PAS cycle. Different behavior was found when im-
planting the SS by nitrogen. Fig. 2 shows that the PAS peak height of the implanted, but not stress annealed, SS is x 1.04 for 3 x 10” cm-* N+ and slightly higher(~1.06)for8x10”cm-*N+.AlongthePAS cycle the peak height rises to a maximum of = 1.08 for the lower doses and z 1.12 for the higher one. Both maximum values are reached at about 200°C. At about 300°C a value of x 1.06 is reached which
Volume 12, number 1,2
September 199 I
Table I Sample treatment and Vickers microhardness HV 0.005 for austenitic SS (Xl OCrNiTi 189) Sample treatment implant a)
Microhardness HV 0.005 dose (particles cme2)
_ N N _ N
3x IO” 8x 10”
550°C 90 min b, _
before/after PAS cycle
outside/inside wear trace
4401352 3751350 340/550 480/740 _
a) 100 keV. ‘) Cooling down I6 h.
is the same value as found in the case of the non-implanted, not stress annealed SS (cf. fig. 1). When cooling, the implanted samples show the same PAS peak height as the non-implanted ones. The wear (fig. 3) during friction under high-vacuum conditions depends in a different way on sample treatment. Whereas the non-implanted SS shows the same wear as the SS implanted with 3 X 10” cme2 N+ before stress annealing, the wear increases for the non-implanted SS but diminishes for the implanted sample after stress annealing. By tempering the samples, the non-implanted steel exhibits only a small reduction of wear compared to the implanted steel reaching a minimum of wear when tempered at about 200°C. The coefficient of friction measured for the different austenitic SS samples (cf. fig. 4) was found to be nearly independent of stress annealing, tempering and nitrogen implantation. To interpret these results, the production of defect agglomerates and nitrogen precipitates has to be considered together with the values of Vickers microhardness (cf. table 1). ( 1) After manufacturing, the non-implanted SS samples have a great amount of defects that leads to the temperature-independent positron trap density, i.e. peak height, shown in fig. 1. Above 200°C this defect structure becomes somewhat smaller by annealing of point defects. During cooling down in the PAS cycle, as well as by the stress annealing procedure, the manufacturing defect structure disappears and the positron trap densities are reduced to the values inherent in the austenitic SS lattice properties, e.g. thermal expansion. This can also be seen
from the microhardness values (cf. table 1) which increases after sample manufacturing (365-440) and then reduces to the standard values of about 350 for austenitic SS by stress annealing or by PAS cycling. On the other hand, although the PAS values after cooling down are identical with and without stress annealing (cf. fig. 1 ), the corresponding wear rates, as measured by the displacement of the tribotester load arm, are quite different (cf. fig. 3 ). This means that the temperatures reached in the PAS cycle are not high enough for total stress annealing which can also be seen from the somewhat smaller peak height at 320’ C for the stress annealed SS compared to the corresponding values of the untreated SS (cf. fig. 1). ( 2 ) By implanting with nitrogen, the point-defect structure of the SS is changed as well as the formation of nitrides. Especially the increase in temperature also gives rise to the formation of metastable nitrides that form coherent precipitates in the SS matrix [ 6,7]. This leads to the creation of additional positron traps at the boundaries between austenite and nitride phases [ 8 1. Fig. 2 shows that after the nitrogen implantation the positron trap density is reduced. This seems to be due mainly to ion beam annealing of the residual manufacturing stress as can be seen from the microhardness of the SS implanted with 3 x 10” crnp2 N+ (HV 0.005 = 340, cf. table 1). On the other hand, the wear rate is not increased as found for the non-implanted but stress annealed SS (cf. fig. 3). When raising the nitrogen dose to 8~ 10” cmw2 N+, the trap density at the start of the PAS cycle is slightly increased; the microhardness has a somewhat higher 19
Volume 12, number 1,2
value (HV 0.005 = 480) than the stress annealed SS. During the PAS cycle (cf. fig. 2), the temperature dependence of the trap density of both the 3x IO” cmv2 and the 8 x IO” cmp2 N+ implanted samples, is nearly the same as that found for the thermal behavior of the SS matrix. At about 2OO”C, the maximum peak height is reached because at higher temperatures defect annealing also occurs. Similar results for defect annealing of deformed and ion-implanted SS at temperatures up to 600°C were reported earlier . Formation of nitride precipitates during PAS cycling cannot be inferred from the trap density behavior (cf. fig. 2). This process is superposed by the formation of vacancy clusters and agglomerates due to the increased point-defect mobility at higher temperatures. Moreover, at x 320°C and during cooling the same peak height is always found. This shows that point defects created during the nitrogen implantation and their mobility govern the trap density in the temperature-dependent PAS. Only the microhardness measured after PAS cycle for nitrogen-implanted SS (cf. table 1) establishes the formation of nitride precipitates as the values are dose dependent and much higher than for non-implanted SS (HV 0.005=740 for 8x10” cm-2 N+). It should be mentioned that work hardening was increasing the microhardness up to HV 0.005 = 649 for non-implanted SS but 883 for implanted SS which may also be due to nitride precipitates formed at the high local temperatures present during adhesive friction. The interpretation of the wear and friction values measured for nitrogen-implanted SS (figs. 3 and 4) is now more or less straightforward. The wear rate clearly correlates with the amount of vacancy clusters and agglomerates as shown by the wear minimum at the tempering temperature of 2 IO’ C where the peak height maximum in the PAS occurs. The coefficient of friction does not show any significant dependence on the different treatments applied to the SS in this work, because in adhesive contacts the friction depends on the ratio of shear strength and microhardness which have similar atomic origins and
September I99 1
therefore change proportionally to each other.
4. Conclusions The combination of nitrogen implantation with annealing of austenitic stainless steel (X 1OCrNiTi 189) resulted in a beneficial effect on microhardness and wear resistance. By tempering the SS implanted with 3 x IO” crnm2N+, 100 keV, at a critical temperature of about 200°C the wear rate could be reduced by a factor of 3 compared to non-implanted SS tempered at the same temperature, and by a factor of 5 compared to SS only implanted but not tempered. The microhardness was improved to HV 0.005 =550 and 740 for 3x IO” cmp2 and 8x IO” cme2 N+, respectively, after heating to 320°C during the temperature-dependent positron annihilation spectroscopy. These results were discussed as being determined by the formation of point-defect agglomerates and, to some extent, by nitride precipitates. In contrast, no reduction of the friction coefficient was found which is in agreement with the model that adhesive forces between the SS sample and the testing ball in dry sliding contact under vacuum conditions, are dominant.
References [ 1] H. Dimigen, K. Kobs, R. Leutenecker, H. Ryssel and P. Eichinger, Mater. Sci. Eng. 69 ( 1985) 18 1.  F.G. Jost, L.E. Pope, D.M. Follstaedt, J.A. Knapp and S.T. Picraux, Thin Solid Films 107 (1983) 287. [ 31 G.S. Was, Progr. Surface Sci. 32 ( 1990) 2 11. [ 41 F. Heinrich and U. Zimmermann, Helv. Phys. Acta 52 ( 1979) 200. [ 51 J. Dryzek, Proceedings ofthe 20th Polish Seminar on Positron Annihilation (1989) p. 73.  S. Fayeulle and D. Treheux, Nucl. Instr. Methods B 19/20 (1987) 216.  R. Leutenecker and J. Weidhaas, paper presented at the 7th International Conference on Ion Beam Modification of Materials, Knoxville, Tennessee, September 1990.  K. Ghazi Wakili, Ph.D. Thesis, ETH, Zurich (1989).  B. Viswanathan, W. Trifhshauser and G. Kogel, Radiation Effects 19 (1983) 231.