On the materials properties of thin film plasma-nitrided austenitic stainless steel

On the materials properties of thin film plasma-nitrided austenitic stainless steel

Surface & Coatings Technology 200 (2006) 4195 – 4200 www.elsevier.com/locate/surfcoat On the materials properties of thin film plasma-nitrided austen...

165KB Sizes 2 Downloads 46 Views

Surface & Coatings Technology 200 (2006) 4195 – 4200 www.elsevier.com/locate/surfcoat

On the materials properties of thin film plasma-nitrided austenitic stainless steel L. Trabzon*, M.C. ˙Ig˘dil Istanbul Technical University, Mechanical Engineering Department, Materials Science Division, I˙stanbul, Turkey Received 25 October 2004; accepted in revised form 22 December 2004 Available online 26 January 2005

Abstract A series of experiments were performed to study the composition and mechanical properties of the surface layers formed on the austenitic stainless steel after plasma nitriding in the temperature range of 400–500 8C with different N2–H2 gas mixtures. The thin layer at the nitrided surface was examined by the glancing-angle XRD and the differential load penetration from the microhardness measurements. The formation of the expanded austenite phase was detected at temperatures below 450 8C and/or 10% N2 in the treatment gas. The distortion of equivalent lattice constant after plasma nitriding was as high as 10%. The modulus of elasticity in the nitrided surface was increased by 33% after the plasma nitriding. The coating-only hardness was measured and it was equivalent to the hardness of coating containing CrN. Thus, it is possible to obtain thin coatings with superior resistance to corrosion and high hardness on the austenitic stainless steel. D 2004 Elsevier B.V. All rights reserved. Keywords: Austenitic stainless steel; Plasma nitriding; Expanded austenite; Nano-indentation

1. Introduction One of the main goals to improve the surface properties of the austenitic stainless steel without deterioration of its excellent corrosion properties is increasing the surface hardness to withstand the conditions inducing excessive wear [1–5]. One method among others is the plasma nitriding, which is a common surface processing method used to introduce nitrogen into steel to improve the surface hardness at temperatures higher than 400 8C [2,6–9]. Since there is a deterioration of corrosion properties when CrN precipitation occurs at certain processing conditions, there is a great interest of introducing nitrogen into the austenite matrix without forming CrN phase [10,11]. Thus, the nitrided steels having high surface hardness, good wear and corrosion properties can be obtained through formation of the expanded austenite phase which is seen in the plasma nitriding at temperatures lower than 500 8C. The mechanical * Corresponding author. Tel.: +90 212 29313 00/2437; fax: +90 212 2450795. E-mail address: [email protected] (L. Trabzon). 0257-8972/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2004.12.012

and materials properties of the expanded austenite have been studied extensively [11,12]. However, it is also important to understand the materials characteristics of the expanded austenite which is grown in the short duration of plasma nitriding. In this study, the materials properties of thin nitrided layer by plasma nitriding at low temperatures were investigated. The true nature of the expanded austenite was studied extensively by the glancing-angle XRD and the microhardness measurements to eliminate any perturbations from the bulk of austenitic stainless steel.

2. Experiments The specimens used in the study were standard austenitic stainless steel that contains only 0.01 C% by weight (316 L). The size of the samples was 10330 mm and they were cleaned by trichloroethylene to free any kind of oil and dust. The plasma nitriding was carried in a dc-discharge plasma system which is a 200-mm-diameter quartz glass vacuum tube. Prior to locating the specimens on the

L. Trabzon, M.C. I˙g˘dil / Surface & Coatings Technology 200 (2006) 4195–4200

4196

cathode, the samples were mechanically polished by using alumina slurry to obtain mirror-finished surfaces. The samples were exposed to hydrogen contained (H2) glowdischarge to clean the surface before plasma nitriding. The processing temperature was monitored by a thermocouple embedded in the system so that the plasma nitriding was performed if the desired processing temperature was achieved on the specimen surface. The duration of the process was completed by the end of nitrogen contained (N2) glow-discharge. Then the specimens were let cool down under the hydrogen gas ambient. The plasma nitriding was conducted at 1300 Pa with different temperatures of 400, 450 and 500 8C with a maximum 2 kW and cathode current densities ranging from 0.75 to 2.5 mA/cm2. In the experiments, the gas mixtures of 5%, 10% and 25% nitrogen by volume and hydrogen in the rest were used at all three temperatures. The experiments were also run for 15, 30, 60, 120 and 240 min with four samples in the each case. The glancing-angle XRD measurements at a 28 incoming angle to the surface was performed with Cu-Ka (k=1.542 nm) to determine phases in the nitrided thin film on the surface. The microhardness measurements were done by Fischer HP100 ultra-microhardness tool which is characterized by a load range of 0.4–1000 mN, a load resolution of 0.2 mN, an indenter shift resolution of 2 nm and a stepwise increment of load. The Vickers indenter, a four-sided diamond pyramid with a square base and a face angle of 1368, was used for the measurements. The indentation was completed by 60 steps with 1 s waiting period in between. To eliminate artifacts in the experiments due to surface roughness, the specimens were barely polished before the measurements. All experiments were done at least six times and then the evaluation of the results was done by the average of measurements for each set. The derivative of applied load ( F) to penetration depth (s), dF/ds (differential load feed, DLF), was utilized to obtain detailed analysis of microhardness properties of the nitrided thin film on the

γ

γ

(111) n (111)

35

40

γ

(200) n

a.u.

CrN

45

surface of austenite stainless steel. The theory of DLF is based on an energy-related consideration of deformation principles in metals under the pressure of the penetrating indenter [13,14]. The data was also analyzed by Oliver and Pharr method to obtain the reduced modulus of elasticity (E/ (1 m 2), where E is the modulus of elasticity and m is the Poisson’s ratio) of the nitrided surface [15].

3. Results and discussion Fig. 1 shows glancing angle at 28 XRD spectrum of the plasma-nitrided austenitic stainless steel specimens. The glancing angle of XRD spectrum of the pristine sample is also included in Fig. 1. The samples were exposed to plasma at 500 8C for 1 h with the gas compositions of 5%, 10% and 25% N2 content by volume and with 5% N2 gas content by volume for 1 h at 400, 450 and 500 8C on the left and right columns of Fig. 1, respectively. It is seen that the formation of CrN is enhanced by temperature and it is suppressed by a treatment at 400 8C with 5% N2 gas content. CrN formation is initiated at temperatures above 450 8C and/or 25% N2 gas content. Note that there is no any carbide formation in the pristine sample due to low C content. There is a shift in the 2h values and the expansion of peaks of (111) gn planes are pronounced by the amount of nitrogen diffused into the austenite matrix and the higher treatment temperatures. There is a suppression of the expanded austenite phase with 500 8C and 5% N2 gas, but the expanded austenite phase is seen by 10% N2 gas at 500 8C for 1 h treatment. It clearly indicates that the formation of expanded austenite or Sphase (gn) is accelerated by the amount of nitrogen in the treatment gas. The evaluation of expanded austenite formation is also studied as a function of time at 450 8C with 10% N2 gas in the treatment gas mixture as seen in Fig. 2. The glancing-angle XRD spectra in Fig. 2 present the formation and growth of thin film composed expanded

γ

CrN

(200)

γ

γ

(111) n (111)

γ

(200) n

(200)

γ

500 ºC, 5% N2

500 ºC, 25% N2

450 ºC, 5% N2

500 ºC, 10% N2

400 ºC, 5% N2

500 ºC, 5% N2

Before nitriding

Before nitriding

50

55



35

40

45

50

Fig. 1. Glancing angle at 28 XRD spectrum of nitrided samples under different conditions.

55

L. Trabzon, M.C. I˙g˘dil / Surface & Coatings Technology 200 (2006) 4195–4200

CrN

γ

(111) n

γ

(111)

γ

4197

γ

(200) n

(200)

240 minutes

a.u.

120 minutes 60 minutes 15 minutes Before nitriding 35

40

45

50

55

2θ θ Fig. 2. Glancing angle at 28 XRD spectrum of nitrided samples by time at 450 8C and 10% N2 by volume in the treatment gas.

4,1 a(200) 400, ºC

1 hour treatment 4,05 4

a(200) 450, ºC

a(200) 500, ºC

a(111) 400, ºC

a(111) 450, ºC

a(111) 500, ºC

3,95 3,9 3,85 3,8 3,75

%5

%10

%25

Amount of N2 gas by volume

Fig. 3. Equivalent cubic lattice constant versus percentage of nitrogen gas in the treatment gas mixture at different temperatures for 1 h.

time at 450 8C and 10% N2 gas by volume as shown in Fig. 4. The filled and empty symbols are for the equivalent lattice constant from (111) and (200) planes, respectively. The equivalent lattice constant from (111) plane stays fairly constant; however, that from (200) plane increases with the treatment time until 120 min plasma nitriding. It may be explained by anisotropic dependence of the stress on the strain [16]. The evaluation of microhardness by the conventional Vickers microhardness measurements introduces some rough understanding of microhardness of the nitrided surface since there are always some problems with very low applied loads on the surface. The observed microhardness is an underestimation of the surface hardness due to substrate effect in the conventional Vickers microhardness measurements. The mechanical properties of coating-only and prediction of the thickness of plasma nitriding can be obtained by the depth-sensing indentation DSI analysis [13]. The derivative of applied force by the indentation depth (dF/ds) in the DSI measurements provides combination of linear curves as shown in Figs. 5 and 6 [14]. A change of the slope is an indication of the substrate effect and the slope

Equivalent cubic lattice constant, Å

Equivalent cubic lattice constant, Å

austenite without any CrN formation at the given conditions. The characteristics of the expanded austenite phase (gn) are further studied by calculating equivalent cubic lattice constant from corresponding d spacing of (111) and (200) planes in the glancing-angle XRD spectra as shown in Fig. 3. The filled and empty symbols stand for the equivalent lattice constants obtained from (111) and (200) planes, respectively. Since the distortion on cubic lattice is small enough, the equivalent cubic lattice constant from d spacing of (111) and (200) planes is done by cubic lattice approximation. The equivalent cubic lattice constant is higher in the (200) case for all gas compositions and temperatures. It is interpreted as the tetragonal distortion of fcc lattice [4–7]. The distortion is becoming higher at elevated temperatures as depicted by 10% N2 gas in the treatment atmosphere. The equivalent lattice constant calculated from (200) plane is not changed by the amount of N2 gas. On the other hand, there is a clear trend in the equivalent lattice constants by (111) plane with the amount of nitrogen in the treatment plasma ambient. In other words, there is an increase in equivalent lattice constants from (111) plane by the higher N2 by volume. The change of distortion of fcc lattice is further studied as a function of

3,98

450 ºC, 10% N2

3,96

120

240

3,94 60

3,92

γn(111) γn(200)

3,9 3,88 3,86

30

3,84 3,82

240

60

120

30

3,8 3,78 0

50

100

150

200

250

Time (min) Fig. 4. Equivalent cubic lattice constant versus treatment duration at 450 8C and 10% N2 by volume in the treatment gas.

L. Trabzon, M.C. I˙g˘dil / Surface & Coatings Technology 200 (2006) 4195–4200

4198 600

%10 N2, 450 Cº

240 min 120 min

500

Unnitrided

60 min 15 min

dF/ds

400

300

200

100

0 0

2

1

3

4

s (μm) Fig. 5. Differential load feed plots for nitrided samples with different processing durations at 450 8C and 10% N2 by volume in the treatment gas.

defines hardness equivalent, i.e. the higher the slope, the higher the hardness [13]. It is now generally accepted that the effect of substrate on hardness measurements is eliminated if the indenter depth is only 1/10 to 1/7 of the coating thickness [14]. Thus, it is very easy to estimate the thickness of nitridation at given conditions precisely by distinguishing a change of slopes without any destructive methods. The DSI measurements can also be analyzed using the method developed by Oliver and Pharr to calculate the reduced modulus of elasticity [15]. The differential load feed characteristics of the plasma-nitrided austenitic stainless steel as a function time at 450 8C and with 10% N2 gas in the gas mixture are shown in Fig. 5. It is very clear to see the substrate effect by a deviation of

slope from the first linear part on the graph. Since the effect of substrate is eliminated when the load is penetrated within 1/10 to 1/7 of thickness of the nitrided surface, it is easily calculated that the thickness of nitrided layer for the 15 min plasma nitriding is in the range of 1.7 to 2.4 Am. On other hand, the thickness of nitrided layer by the conventional microhardness is overestimated by more than 3 Am (not shown here). With the longer plasma nitriding at the same conditions, there is an increase in the thickness of the nitrided surface. The striking observation is that there is no change in slope with the plasma nitriding duration so that the microhardness of the coating is equivalent in all cases. Thus, the expanded austenite is obtained in all plasma nitriding conditions on the surface. Contrary to the

800

1 hour nitrided 700

%10, 500C˚ %25, 500C˚

600

%5, 500C˚ dF/ds

500

%10, 450C˚

%10, 400C˚ %5, 400C˚ %5, 450C˚

400

Unnitrided 300 200 100 0 0

0,5

1

1,5

2

2,5

3

3,5

s (μm) Fig. 6. Differential load feed plots for nitrided samples for 1 h in different treatment conditions.

4

L. Trabzon, M.C. I˙g˘dil / Surface & Coatings Technology 200 (2006) 4195–4200

E/(1-υ2) (Gpa)

250 240 min

120 min

200 60 min 30 min 150

4199

500 8C with 10% N2 gas by volume in the plasma ambient for 1 h. The increase in the reduced modulus of elasticity is not dependent on plasma nitriding conditions, not shown here.

15 min

450 C, %10 N, 300 mN 100 0

50

100

150

200

250

4. Conclusions

300

Treatment time (min) Fig. 7. Reduced modulus of elasticity versus time for samples nitrided at 450 8C and 10% N2 by volume in the treatment gas.

conventional microhardness measurement methods [2], DSI measurements precisely provide exact mechanical properties of the surface. Depth-sensing indentation measurements of the coatings grown in different plasma conditions are shown in Fig. 6. There is no considerable penetration of nitrogen into the austenite lattice at 400 8C and with 5% N2 gas in the treatment gas mixture as DLF data are identical to the pristine sample. On the other hand, the slope of dF/ds is not altered by a formation of CrN at the elevated temperatures and with the higher nitrogen content by volume. Therefore, it is possible to grow a thin film by the plasma nitriding at low temperatures with very high microhardness as well as superior resistance to corrosion [1–3]. In the measurement of hardness by conventional Vickers indentation, there is a differentiation on hardness with different microstructures due to substrate effect. The coating-only microhardness properties are only obtained by eliminating substrate effect with the DSI method contrary to the conventional methods. The reduced modulus of elasticity from the same measurements is plotted in Fig. 7 to observe the mechanical properties of the nitrided region in depth. The reduced modulus of elasticity of the sample nitrided with 10% N2 gas and 450 8C for 240 min is increased by 33% from that of the pristine sample. The expansion of fcc lattice with nitrogen atoms changes the nature of the bonding and lattice geometry so that it results in an increase in the modulus of elasticity and strength [15–17]. The modulus of elasticity is found to be increased by more than 50% due to plasma nitriding and it is independent of processing conditions and depth of nitridation [8,18]. Since the penetration depth in the DLF experiments was the same for all cases to obtain reduced the modulus of elasticity (E/(1 m 2), the reading is the combination of both the nitrided layer and the unnitrided substrate. The thickness of nitrided layer is increased by the duration of plasma nitriding, so that there is an increase of the reduced modulus of elasticity due to the elimination of substrate effect with a thicker formation of the nitrided layer. The effect of temperature and amount of N2 gas in the treatment ambient is presenting the same trend, not shown here. Namely, there is an increase in the reduced modulus of elasticity by temperature and nitrogen gas by volume. Note that the formation of CrN increases the reduced modulus of elasticity as high as 227 GPa at

The materials properties of the austenitic stainless steel after the plasma nitriding were examined at the temperatures from 400 to 500 8C with N2 gas compositions of 5%, 10% and 25% by volume in the treatment ambient. The formation of the expanded austenite phase is pronounced at the temperatures below 450 8C and/or 10% N2 gas, otherwise, CrN is precipitated. The effect of temperature in forming CrN is more effective than the volume percentage of nitrogen gas in the treatment ambient. By the formation of the expanded austenite phase, there is a lattice expansion about 10% along [100] direction and it is also more pronounced by the higher temperatures. By eliminating the substrate effect in the hardness measurements, the hardness properties of the nitrided thin film on the surface were observed. The hardness is the same as long as the expanded austenite phase is probed and it is seen that the hardness of expanded austenite is equivalent to that of the nitrided thin films containing CrN. The modulus of elasticity is also seen to be increased by 33% from the unnitrided case after the plasma nitriding.

Acknowledgement The authors are grateful for the help received from Turgut Gqlmez on the use of plasma nitriding equipment ¨ rgen for the use of XRD and microhardness and Mustafa U facilities and KqrYat KazmanlV for the discussions on DLF analysis.

References [1] E. Menthe, A. Bulak, J. Olfe, A. Zimmermann, K.-T. Rie, Surf. Coat. Technol. 133 (2000) 259. [2] S. Picard, J.B. Memet, R. Sabot, J.L. Grosseau-Poussard, J.P. Riviere, R. Meilland, Mater. Sci. Eng., A Struct. Mater.: Prop. Microstruct. Process. 303 (2001) 163. [3] L. Marot, E. Le Bourhis, A. Straboni, Mater. Lett. 56 (2002) 76. [4] E. Menthe, K.-T. Rie, J.W. Schultze, S. Simson, Surf. Coat. Technol. 74–75 (1995) 412. [5] A. Saker, Ch. Leroy, H. Michel, C. Frantz, Mater. Sci. Eng., A Struct. Mater.: Prop. Microstruct. Process. 140 (1991) 702. [6] T. Czerwiec, N. Renevier, H. Michel, Surf. Coat. Technol. 131 (2000) 267. [7] B. Larisch, U. Brusky, H.-J. Spies, Surf. Coat. Technol. 116–119 (1999) 205. [8] V. Singh, K. Marchev, C.V. Cooper, E.I. Meletis, Surf. Coat. Technol. 160 (2002) 249. [9] O. Ozturk, D.L. Willamson, J. Appl. Phys. 77 (1995) 3839. [10] E. Menthe, K.-T. Rie, Surf. Coat. Technol. 116–119 (1999) 199.

4200

L. Trabzon, M.C. I˙g˘dil / Surface & Coatings Technology 200 (2006) 4195–4200

[11] X.L. Xu, L. Wang, Z.W. Yu, Z.K. Hei, Surf. Coat. Technol. 132 (2000) 270. [12] M.P. Fewell, D.R.G. Mitchell, J.M. Priest, K.T. Short, G.A. Collins, Surf. Coat. Technol. 131 (2000) 300. [13] B. Rother, D.A. Dietrich, Surf. Coat. Technol. 74–75 (1995) 614. [14] M.K. Kazmanli, B. Rother, M. Urgen, C. Mitterer, Surf. Coat. Technol. 107 (1998) 65.

[15] [16] [17] [18]

G.M. Pharr, W.C. Oliver, MRS Bull. (1992 (July)) 28. S. M7ndl, B. Rauschenbach, J. Appl. Phys. 88 (6) (2000) 3323. H.M. Ledbetter, M.W. Austin, Mater. Sci. Technol. 3 (1987) 101. J.F. Lin, K.W. Chen, C.C. Wei, C.-F. Ai, Surf. Coat. Technol. (2004) (in press).