Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation

Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation

G Model ARTICLE IN PRESS APSUSC-27742; No. of Pages 6 Applied Surface Science xxx (2014) xxx–xxx Contents lists available at ScienceDirect Applie...

5MB Sizes 0 Downloads 3 Views

Recommend Documents

No documents
G Model

ARTICLE IN PRESS

APSUSC-27742; No. of Pages 6

Applied Surface Science xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation B.B. Fernandes a,b,∗ , S. Mändl b , R.M. Oliveira a , M. Ueda a a b

Instituto Nacional de Pesquisas Espaciais, LAP/INPE, Av. dos Astronautas, 1758, Caixa Postal 515, São José dos Campos, SP, Brazil Leibniz Institut für Oberflächenmodifizierung, Permoerstr. 15, 04318 Leipzig, Germany

a r t i c l e

i n f o

Article history: Received 28 November 2013 Received in revised form 17 April 2014 Accepted 18 April 2014 Available online xxx Keywords: Stainless steel Plasma immersion ion implantation Tribological properties Hardness

a b s t r a c t The formation of hard and wear resistant surface regions for austenitic stainless steel through different nitriding and nitrogen implantation processes at intermediate temperatures is an established technology. As the inserted nitrogen remains in solid solution, an expanded austenite phase is formed, accounting for these surface improvements. However, experiments on long-term behavior and exact wear processes within the expanded austenite layer are still missing. Here, the modified layers were produced using plasma immersion ion implantation with nitrogen gas and had a thickness of up to 4 ␮m, depending on the processing temperature. Thicker layers or those with higher surface nitrogen contents presented better wear resistance, according to detailed microscopic investigation on abrasion, plastic deformation, cracking and redeposition of material inside the wear tracks. At the same time, cyclic fatigue testing employing a nanoindenter equipped with a diamond ball was carried out at different absolute loads and relative unloadings. As the stress distribution between the modified layer and the substrate changes with increasing load, additional simulations were performed for obtaining these complex stress distributions. While high nitrogen concentration and/or thicker layers improve the wear resistance and hardness, these modifications simultaneously reduce the surface fatigue resistance. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Surface improvement of steels through plasma-based techniques has become suitable for applications in industry [1,2]. However, there is still a pursuit for the most efficient parameters to obtain the best modified layers and also for a more complete knowledge of the nature of these layer properties [3–11]. Wear rates, for example, can be decreased if the maximum stress below the surface layer is eliminated [12], which may be obtained with new, additional layers or sub-layers. Fatigue experiments have shown that coatings below 100 ␮m are preferred [13]. However, ductility and creep resistance must be retained for better performance of coatings applied at high temperatures [14]. For these conditions, adhesion may become a problem. Alternatively, implantation and diffusion processes may be employed to modify surfaces by nitriding, e.g. nitrogen insertion. Model of nitrogen depth evolution

∗ Corresponding author at: Laboratório Associado de Plasma (Instituto Nacional de Pesquisas Espaciais), Av. dos Astronautas, 1758, Sao José dos Campos, Brazil. Tel.: +55 12 3208 6698. E-mail addresses: [email protected], [email protected] (B.B. Fernandes).

accounting for different effects are in development nowadays in order to exactly predict the layer thickness and surface concentration [15,16]. While implantation allows the insertion of foreign atoms independent of the surface activity or surface barriers, high temperatures (over 400 ◦ C) are not accessible by using ion bombardment alone, which is of much importance in the treatment of materials where the diffusion coefficient only becomes significant at high temperature [17,18]. PIII treatment of SS304 samples produces a layer over 5 ␮m with nitrogen using temperature of 390 ◦ C, pulses of 10 kV for 10 h [19]. Despite a shallower layer (around 800 nm) obtained through PIII treatment with 8–16 kV and 350 ◦ C, this same work shows that the implantation for only 2 h reduces similarly the friction coefficient and wear rate of SS304 samples [19]. Several mechanical surface treatments (e.g. shot penning or surface mechanical attrition) can be applied before plasma implantation to enhance the nitrogen diffusion on SS304, and layers with more than 30 ␮m are obtained. Nevertheless, such methods are not industrially suitable or have low efficiency [20]. While wear and hardness testing of thin films are well established, lifetime investigations are still sparse. One of the most recent techniques used to study mechanical properties of modified surfaces is nanoindentation. As a further development, cyclic

http://dx.doi.org/10.1016/j.apsusc.2014.04.142 0169-4332/© 2014 Elsevier B.V. All rights reserved.

Please cite this article in press as: B.B. Fernandes, et al., Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.04.142

G Model APSUSC-27742; No. of Pages 6

ARTICLE IN PRESS

2

B.B. Fernandes et al. / Applied Surface Science xxx (2014) xxx–xxx

2. Experimental procedures SS304 disks of 15-mm diameter and 3-mm thickness were cleaned in ultrasound acetone bath. The samples were grounded with SiC papers and polished with diamond pastes, and the arithmetic mean surface roughness (Ra ) measured by AFM and Optical Profilometry were Ra = 3.2 ± 0.3 nm and Ra = 18.4 ± 5.7 nm, respectively. The experimental set-up for the nitrogen PIII without auxiliary heating is shown elsewhere [26] and the parameters are shown in Table 1. For PIII performed in heated substrates, the experimental apparatus used is mentioned elsewhere [27,28]. For that case, SS samples were tied up on a tungsten wire that plays the role of the discharge anode, being positively polarized by DC voltages in the 700 V range in relation to the grounded chamber wall. A thermionic oxide cathode generates primary electrons to help the breakdown of the glow discharge, which happens in a pressure range of 10−3 Torr, suitable for PIII. Electrons produced by the oxide cathode are drawn to the samples leading to substrate heating. When additional high negative voltage pulses are applied to the samples, ion implantation takes place. PIII was carried out with the process parameters listed in Table 1. The nitrogen depth profiles were measured using time-of-flight secondary ion mass spectrometry (ToF-SIMS). Conversion of sputtering time to depth was done by measuring depth of the craters and assuming a linear sputter rate within the craters. Tribological evaluations of the sample surfaces were conducted with a CSM tribometer. Measurements of dry friction coefficient were accomplished in an oscillating ball-on-disk tribometer. Parameters used were: load of 1 N with a 4.7-mm diameter alumina ball as a counterpart material, average speed of 10 cm/s and track length of 2 mm. This applied load results in a Hertzian contact pressure of 0.5–0.7 GPa, depending on the thickness of the modified layer. The surfaces after the wear test were examined by SEM in the secondary electron detector mode to find out about the wear mechanisms in PIII treated and untreated samples. Roughness and wear profiles were measured by an optical profilometer. The hardness of the SS samples was measured using the quasicontinuous stiffness measurement (QCSM) module of the ASMEC nanoindenter. This device stays in a vibration-free isolated cabinet and a Berkovich three-sided pyramidal diamond indenter was used

70 1,4

o

SS I (300 C-180 minutes) o SS II (270 C-600 minutes) 60 o SS III (480 C-120 minutes) o SS IV (480 C-60 minutes) 50 o SS V (650 C-60 minutes)

N/Fe Ratio

1,2 1,0

40

0,8

30

0,6 0,4

20

0,2

10

0,0 0

500

1000

1500

2000

2500

3000

3500

Nitrogen Concentration (%)

indentation fatigue is a possibility where a cyclic load is applied on the sample via an indenter, as an alternative approach for measuring the fatigue properties of materials [21]. Investigation of indentation fatigue behavior is crucial to build the relationship between this nanoscale test and conventional fatigue testing of bulk material [14,22,23]. Compared to large-scale testing, nanofatigue is relatively non-destructive, the maximum load is situated within the surface layer, which allows the determination of mechanical property changes locally in materials, for example in coatings, weld seams and other localized areas of interest. Hence, traditional and costly operations can be replaced with small material samples or performed for the first time [14]. Cyclic loading determines the resistance of the entire coating to cracking and delamination [24]. However, such experiments must be performed with spherical indenters, because sharper indenters produce virtually “saturated” deformation and are not suited for examining damage evolution [25]. The present work utilizes nanoindentation techniques to characterize different nitrogen-rich layers produced through PIII treatments on SS304 samples. Mechanical properties of these surface layers at nanoscale levels were measured and evaluated besides physico-chemical characterizations, to demonstrate the potential of the plasma apparatus and new techniques used here.

0 4000

Depth (nm) Fig. 1. SIMS nitrogen profiles of PIII treated SS304.

for these measurements. Loading is stopped for a period of time of 3 s and the voltage for the piezoelectric element is overlaid with a sinusoidal oscillation. An average of 10 single measurements was used to determine the hardness and the elastic modulus. The nanofatigue tests were conducted in the same ASMEC equipment with a spherical diamond indenter (tip radius of 10 ␮m). This equipment provides several kinds of information, with load versus depth data being analyzed in the present work. The goal of this analysis is to verify abrupt changes (or clear steps in the load–depth curves) in the depths that are generally related to the initial stages of crack formation. Samples were subjected to cyclic contact tests by repeatedly indenting the same area at loads of 20–750 mN with minimum loads of 0.2–7.5 mN, respectively. Because of software limitations, only a maximum of 300 cycles are allowed without the indenter leaving the surface. Nevertheless, a larger number of cycles was performed at the same position. Simulations of Von Mises stresses of the spherical indentations were performed in the software Elastica for the different layers studied here. 3. Results and discussion As a summary of XRD analyses, the treated samples showed expanded austenite phase at the surface, with the exception of the SS V, that showed some CrN peaks. SIMS experiments show that the implantation time is not the main parameter to produce thicker layers by PIII. Indeed, temperature is the determining parameter when the ions are implanted with energies above 5 keV. It was expected to find significant CrN formation in the SS V due to the higher temperature (see Table 1), however, a surface layer of 60 nm with many elements (Ti, Mg, Si, O, C) was detected over an oxide layer of approximately 200 nm. These layers have been formed because of inappropriate cleaning of the PIII chamber plus the higher surface activity at high temperatures. These unexpected coatings do not prevent the nitrogen implantation that allows the formation of a thick layer below them (see Fig. 1 and Table 2). One explanation could be a late deposition during the process itself. The more wear resistant layers are those with higher nitrogen content or layer thickness as can be seen in Fig. 2. Normally, longer treatment times will lead to higher surface concentrations while lower temperatures correlated with a lower diffusion rate will also increase the nitrogen surface concentration. SS II and IV present similar behavior to untreated samples, i.e. after less than 1000 cycles, the harder surface layers (nitrogen- or oxygen-rich) were removed. Meanwhile, the layer created in the SS I is removed only after 4000 cycles and, moreover, such layer allows that the reduction on wear be kept even after its removal. Despite a smaller

Please cite this article in press as: B.B. Fernandes, et al., Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.04.142

G Model

ARTICLE IN PRESS

APSUSC-27742; No. of Pages 6

B.B. Fernandes et al. / Applied Surface Science xxx (2014) xxx–xxx

3

Table 1 PIII parameters. Conditions

Sample SS I

DC source (V × A)

SS II 200 2–3 – –

Voltage Current Voltage Current

AC filament (V × A) HV pulses (kV;␮s;Hz) Time (min) Temperature (◦ C) Substrate heating Nitrogen pressure (10−4 Torr)

12;50;300 180/270 300 No

9;50;300 600 270 No

SS III

SS IV

SS V

10–-60 0.6–1.1 5 15–20 8;30;500 120 480 Yes 7.8

50 0.6 2.5–3 7–8 5;40;400 60 480 Yes

100 0.9–1.2 4 10–11 5;40;400 60 650 Yes

Table 2 Roughness, surface nitrogen content and layer thickness of SS304 samples. Property

Sample Untreated

SS I

SS II

SS III

SS IV

SS V

Roughness – Ra (nm) Surface N content (%) Layer thickness (nm)

18 0 0

66 47 500

23 20 300

37 22 1100

28 11 350

139 2 >4000

untreated surface SS I SS II SS III SS IV SS V

2

Wear Rate (µm /cyc)

1

0,1

0,01

1E-3 0

5000

10000

15000

20000

25000

30000

Cycles Fig. 2. Wear rates versus cycles for untreated and PIII treated SS304 samples.

maximum nitrogen concentration in the SS III and V, their wear resistances were slightly better than that of the layer produced by the treatment I. Profilometry showed that the wear tracks were formed mainly by ploughing, which results in pile-up at the track edges. A

discontinuous oxide formation is observed at shorter wear experiments, which must be related to a specific nitrogen and/or oxygen concentration at the surface. One possible explanation is a native oxide layer formed during removal of the samples from the treatment chamber. After 1000 cycles in the wear reciprocating tests, the untreated surface together with the SS II and SS IV samples present significant abrasive and adhesive wear. Whilst, the SS I presents slight abrasive wear with some fatigue cracks and the SS III only slight abrasive wear (Fig. 3). SEM analyses of the tracks after 2000 cycles show that wear occurs through abrasive and adhesive mechanisms in the untreated sample, as well as in the samples with thinner layers. SS I surfaces also presented these two mechanisms, but with less adhesive wear and more fatigue cracks (Fig. 4), whilst in SS III and V surfaces only abrasive wear was detected. After 10,000 cycles, fatigue cracks can be visualized in the untreated and SS III surfaces, and the latter presents now adhesive wear too. QCSM hardness measurements show the effect of the shallow layers with nitrogen at the SS surface (Fig. 5). The practical tip of the diamond indenter is of a finite radius of a sharp point, which is the reason for the initial increase in hardness observed in all measurements [29]. SS I presented the hardest surface due to the higher nitrogen concentration obtained by PIII. SS V samples showed interesting hardness values which are related to a 200-nm oxide layer

Fig. 3. Worn tracks after 1000 cycles for (a) untreated SS and (b) SS I.

Please cite this article in press as: B.B. Fernandes, et al., Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.04.142

G Model APSUSC-27742; No. of Pages 6

ARTICLE IN PRESS

4

B.B. Fernandes et al. / Applied Surface Science xxx (2014) xxx–xxx

Fig. 4. Worn tracks after 10,000 cycles for (a) untreated SS and (b) SS I.

o

16

12 10 8 6

Hardness (GPa)

untreated surface o SS I (300 C-180 minutes) o SS II(270 C-600 minutes) o SS III(480 C-120 minutes) o SS V(650 C-60 minutes)

14

Hardness (GPa)

SS I (300 C-180 minutes) o SS III (480 C-120 minutes) o SS IV (480 C-60 minutes)

14 12 10 8 6 4

4

2

2 0

0 0,0

0,2

0,4

0,6

0,0

0,1

Depth (µm)

0,2

0,3

0,4

Depth (µm)

Fig. 5. QCSM hardness profiles of untreated and PIII treated SS304 samples.

formed at the surface before the nitrogen-rich region. The lower hardness for the SS III and SS V samples is due to their higher surface roughness (average over 40 nm). This explanation must be also applied to the SS I sample that registered unexpected moderate hardness. For low indentation loads and high roughness values, a systematic underreporting of the true surface hardness is obtained [30]. At higher loads, the effect of surface roughness on hardness is not so significant. Nanofatigue experiments are often reported using a Berkovichlike tip-shaped indenter [24,31–34]. However, as the stress distribution diverges to infinity for an ideal tip with a zero radius, any quantitative interpretation of the results is rather difficult. On the other hand, while a direct calculation of the stress distribution is possible for spherical indenters [35], the direct observation of the onset of fatigue would require detecting a volume or surface area near zero. Hence, experimental limitation in SEM (secondary electrons detector for imaging the surface) will always lead to an artificial overreporting of the fatigue limit in nanofatigue experiments. Similar to another study [24], clear steps in the depth data were not observed with the spherical indenter in nanofatigue experiments, which would have been an indication of a sudden crack initiation (Fig. 6). Brittle materials are characterized by sinking-in behavior, which is confirmed in these moderate load experiments. However, radial cracks can be visualized through SEM in the SS I. The higher nitrogen concentration at the surface of this sample is the reason for such happening, i.e. this layer is more brittle. Axial cracks are identified inside the contact area in higher load (500 mN),

but they were presumably formed because of defects, like pores, grain boundaries or scratches [36]. The first loop of all indentations higher than 1 mN is open, indicating residual permanent plastic deformation. However, the subsequent indentations at the same location result in fully reversible hysteresis loops. Higher loads produce only elastic deformation on the PIII treated samples (SS I – 6 mN; SS II – 3 mN and SS III – 2 mN). Such elastic behavior must be attributed to the

Fig. 6. Fatigue experiments of SS I (SEM of 600 cycles applying 50 mN). White lines indicate the approximate limits of the indentation.

Please cite this article in press as: B.B. Fernandes, et al., Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.04.142

G Model APSUSC-27742; No. of Pages 6

ARTICLE IN PRESS B.B. Fernandes et al. / Applied Surface Science xxx (2014) xxx–xxx

5

Fig. 7. SEM of (a) untreated SS (12,000 cycles with 750 mN) and (b) SS III (300 cycles with 500 mN).

Fig. 8. SEM of SS III sample with cracks attributed to defects.

deformation of surface irregularities (roughness) rather than only to the bulk. When comparing this information with the simulations, which show that plastic deformation should start with loads even below 0.1 mN, several caveats have to be considered. First, the deformed volume will be infinitely small when the yield strength is reached, which is hard or impossible to measure directly. Second, the simulation assumes perfectly smooth surfaces while the

experiment will consist of rough surfaces where a much higher load will be present, leading to a plastic deformation or smoothening of the surface during the initial loading. Third, work hardening effects existing in austenitic steel are not included in the simulation thus exaggerating the calculated deformed volume. The maximum depths registered for the indentations have the same behavior on the other samples (without clear steps), also

Fig. 9. SEM of untreated sample with spallation (a) 57,000 cycles and (b) 700,000 cycles.

Please cite this article in press as: B.B. Fernandes, et al., Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.04.142

G Model APSUSC-27742; No. of Pages 6

ARTICLE IN PRESS

6

B.B. Fernandes et al. / Applied Surface Science xxx (2014) xxx–xxx

without clear cracks which were verified through SEM. Higher loads and cycles were necessary to induce such cracks in these samples, as shown in Fig. 7. More than 100,000 cycles are necessary to induce visible cracks in the samples with less nitrogen. Particularly, cracks in the SS III seem to be initiated in pores or scratches (Fig. 8). The indentation sizes of all samples have not increased when the number of cycles was increased. The untreated sample presented slight spallation after 50,000 cycles, which evolves to a more pronounced one if more cycles are applied (Fig. 9). With such data, subsequent studies can be performed to compare very thin layers of expanded austenite with respect to the fatigue resistance of SS304. Nevertheless, a qualitative increase in the fatigue limit seems to occur after PIII nitrogen implantation of austenitic stainless steel for lower nitrogen contents. 4. Conclusions Optimized process parameters used to perform N-PIII into SS304 samples resulted in improved tribological properties. The wear rate and the friction coefficient of the treated samples were reduced, principally due to the production of a surface layer composed of oxygen and, mainly nitrogen. In fact, nitrided layers over 500 nm of thickness improved significantly the wear resistance of SS304 samples. The results showed here may aid further improvements in food, ethanol and/or metal forming industries [37–39], for example. New insights are also presented here concerning nanofatigue experiments using spherical indenter. One is that cracks cannot be identified simply by depth analysis. Such analysis must be performed with SEM or even TEM. This development is important because conventional fatigue experiments may not reflect certain situations where this property is governed by cyclic contacts and failure results from localized damage accumulation. Despite the fact that high nitrogen concentration improves wear resistance and hardness, it reduces the fatigue resistance of the treated material for rather high nitrogen contents. Acknowledgments We appreciate the financial support from the CNPq, Grant No. 559739/2010-9, National Council of Scientific and Technologic Development, MCTI, Ministry of Science, Technology and Innovation, FAPESP, (Grant Nos. 2011/00872-2 and 2012/21009-3, Sao Paulo Research Foundation). References [1] L. Escalada, J. Lutz, S. Mändl, D. Manova, H. Neumann, S. Simison, Surf. Coat. Technol. 211 (2012) 76–79.

[2] D. Manova, I.-M. Eichentopf, S. Heinrich, S. Mändl, E. Richter, H. Neumann, B. Rauschenbach, Nucl. Instr. Meth. Phys. Res. B 257 (2007) 442–446. [3] C. Blawert, B.L. Mordike, Y. Jirásková, O. Schneeweiss, Surf. Coat. Technol. 116–119 (1999) 189–198. [4] D. Manova, F. Scholze, S. Mändl, H. Neumann, Surf. Coat. Technol. 205 (2011) S286–S289. [5] D. Manova, T. Höche, S. Mändl, H. Neumann, Nucl. Instr. Meth. Phys. Res. B 267 (2009) 1536–1539. [6] D. Manova, J.W. Gerlach, F. Scholze, S. Mändl, H. Neumann, Surf. Coat. Technol. 204 (2010) 2919–2922. [7] D. Manova, S. Mändl, H. Neumann, B. Rauschenbach, Surf. Coat. Technol. 201 (2007) 6686–6689. [8] S. Mändl, Plasma Process. Polym. 4 (2007) 239–245. [9] D.L. Williamson, J.A. Davis, P.J. Wilbur, Surf. Coat. Technol. 103–104 (1998) 178–184. [10] J. Lutz, D. Manova, J.W. Gerlach, M. Störmer, S. Mändl, IEEE Trans. Plasma Sci. 39 (11) (2011) 3056–3060. [11] K. Köster, P. Kaestner, G. Bräuer, H. Hoche, T. Troßmann, M. Oechsner, Surf. Coat. Technol. 228 (1) (2013) S615–S618. [12] D. Manova, D. Hirsch, E. Richter, S. Mändl, H. Neumann, B. Rauschenbach, Surf. Coat. Technol. 201 (2007) 8329–8333. [13] A.K. Lynn, D.L. DuQuesnay, Biomaterials 23 (2002) 1937–1946. [14] M. Eskner, Doctoral Thesis, Royal Institute of Technology, Sweden, 2004. [15] T. Moskalioviene, A. Galdikas, J.P. Rivière, L. Pichon, Surf. Coat. Technol. 205 (2011) 3301–3306. [16] D. Manova, C. Günther, A. Bergmann, S. Mändl, H. Neumann, B. Rauschenbach, Nucl. Instr. Meth. Phys. Res. B 307 (2013) 310–314. [17] L. Marot, M. Drouet, F. Berneau, A. Straboni, Surf. Coat. Technol. 156 (2002) 155–158. [18] C.B. Mello, M. Ueda, M.M. Silva, H. Reuther, L. Pichon, C.M. Lepienski, Wear 267 (2009) 867–873. [19] C.B. Mello, M. Ueda, C.M. Lepienski, H. Reuther, Appl. Surf. Sci. 256 (2009) 1461–1465. [20] L. Shen, L. Wang, Y. Wang, C. Wang, Surf. Coat. Technol. 204 (2010) 3222–3227. [21] N. Faisal, R. Ahmed, Y. Fu, Proceedings of the Advanced Technologies Conference, October 17–19, KASCT, Riyadh, Saudi Arabia, 2011, pp. 36–44. [22] B.X. Xu, Z.F. Yue, J. Wang, Mech. Mater. 39 (2007) 1066–1080. [23] J.M. Cairney, R. Tsukano, M.J. Hoffman, M. Yang, Acta Mater. 52 (2004) 3229–3237. [24] S. Saber-Samandari, K.A. Gross, Acta Biomater. 9 (2013) 5788–5794. [25] F. Guiberteau, N.P. Padture, H. Cai, B.R. Lawn, Philos. Mag. A 68 (5) (1993) 1003–1016. [26] M. Ueda, L.A. Berni, J.O. Rossi, J.J. Barroso, G.F. Gomes, A.F. Beloto, E. Abramof, Surf. Coat. Technol. 136 (2001) 28–31. [27] R.M. Oliveira, J.A.N. Gonc¸alves, M. Ueda, J.O. Rossi, P.N. Rizzo, Surf. Coat. Technol. 204 (2010) 3009–3012. [28] R.M. Oliveira, C.B. Mello, G. Silva, J.A.N. Gonc¸alves, M. Ueda, L. Pichon, Surf. Coat. Technol. 205 (2011) S111–S114. [29] X. Li, B. Bhushan, Scr. Mater. 42 (2000) 929–935. [30] M. Asgari, A. Barnoush, R. Johnsen, R. Hoel, Mater. Sci. Eng. A 529 (2011) 425–434. [31] S. Saber-Samandari, K.A. Gross, Acta Biomater. 7 (2011) 4235–4241. [32] N.H. Faisal, R. Ahmed, R.Y. Fu, Coatings 2 (2012) 195–209. [33] A. Duszová, P. Hvizdoˇs, F. Lofaj, Ł. Major, J. Dusza, J. Morgiel, Int. J. Refrac. Metals Hard Mater. 41 (2013) 229–235. [34] L. Llanes, E. Tarrés, G. Ramírez, C.A. Botero, E. Jiménez-Piqué, Procedia Eng. 2 (2010) 299–308. [35] N. Schwarzer, F. Richter, G. Hecht, Surf. Coat. Technol. 114 (1999) 292–304. [36] T. Karsch, H. Bomas, H.-W. Zoch, S. Mändl, Int. J. Fatigue 60 (2014) 74–89. [37] M. Mauermann, U. Eschenhagen, T. Bley, J.-P. Majschak, Trends Food Sci. Technol. 20 (2009) S8–S15. [38] W.S. Labiapari, C.M. Alcântara, H.L. Costa, J.D.B. de Mello, Wear 302 (2013) 1536–1545. [39] B. Podgornik, J. Jerina, Surf. Coat. Technol. 206 (2012) 2792–2800.

Please cite this article in press as: B.B. Fernandes, et al., Mechanical properties of nitrogen-rich surface layers on SS304 treated by plasma immersion ion implantation, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.04.142