Surface & Coatings Technology 200 (2006) 5807 – 5811 www.elsevier.com/locate/surfcoat
Effect of surface roughness on the properties of the layer formed on AISI 304 stainless steel after plasma nitriding Gajendra Prasad Singh a , J. Alphonsa b,⁎, P.K. Barhai a , P.A. Rayjada b , P.M. Raole b , S. Mukherjee b b
a Birla Institute of Technology, Department of Applied Physics, Mesra, Ranchi—835215, India Facilitation Center for Industrial Plasma Technologies, Institute for Plasma Research, B-15-17/P, Sector 25, GIDC Electronic Estate, Gandhinagar—382044, Gujarat, India
Received 3 April 2005; accepted in revised form 24 August 2005 Available online 25 October 2005
Abstract Experiments were performed with an aim of studying an effect of initial surface roughness with different gas compositions in plasma nitriding, using pulse D.C. glow discharge plasma in presence of nitrogen and hydrogen gas mixtures. Samples were prepared with different mechanical treatments: polishing, rough polishing, machining and grinding. Plasma nitriding was carried out on AISI 304 stainless steel at 560 °C under 4mbar pressures for 24 h in presence of N2 : H2 in 20 : 80 and 80 : 20 ratios. After plasma nitriding, surface roughness, micro hardness, case depth and phase formation were evaluated by using stylus profilometer, Vickers micro hardness tester, optical microscope and X-ray diffraction techniques, respectively. After plasma nitriding, hardness and case depth variation are observed with variation in surface roughness as well as gas compositions. Maximum hardness i.e. 1325 HV and case depth i.e. 110 μm are achieved on mirror polished samples at 80N2 : 20H2. The diffraction patterns show the most dominant phase formation of CrN, Fe4N and Fe3N which is responsible for this increase. © 2005 Elsevier B.V. All rights reserved. Keywords: Plasma nitriding; Surface roughness; Stainless steel
1. Introduction Stainless steel in particular austenitic grade are often used in chemical and food industries because of its excellent corrosion resistance. However, its hardness and wear resistance are relatively poor. Therefore, many attempts have been made to increase the surface hardness of this material by using techniques like plasma nitriding, plasma source ion implantation and plasma immersion ion implantation [1–3]. Plasma nitriding (PN) is one of the widely used surface engineering technology to improve the mechanical and tribological properties of steels [4– 8]. It is well known that not only do the mechanical strength, toughness and geometry affect the mechanical behaviour of the component, but also the final surface finish [9–12]. As a result, there would be a significant effect of the original surface finish of the component on the layer formation during nitriding and also on the properties after nitriding. It has been reported that smaller ⁎ Corresponding author. Tel.: +91 79 23235018; fax: +91 79 23235024. E-mail address: [email protected]
(J. Alphonsa). 0257-8972/$ - see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2005.08.149
nitrided layer depths were formed in ground and sandblasted samples when compared to the polished samples. This was primarily associated with high compressive residual stresses at and below the surface, together with some amount of plastic deformation. The residual stresses are present during the nitriding and do not relax due to time and temperature . In the present investigation, an attempt has been made to evaluate the effect of different surface finish prior to plasma nitriding on the properties of AISI 304 stainless steel. After plasma nitriding the surface roughness, surface hardness, nitrogen diffusion layer thickness and the phases formed have been analyzed using stylus profilometer, Vickers micro hardness tester, optical microscope and X-ray diffraction (XRD), respectively. 2. Experimental procedure 2.1. Materials and preparation AISI 304 austenitic stainless steel was selected for the present study and its chemical composition is shown in Table 1.
G.P. Singh et al. / Surface & Coatings Technology 200 (2006) 5807–5811 1400
Table 1 Chemical composition (wt.%) of AISI 304 stainless steel Mn
The samples were disc shaped with a diameter of 25 mm and a thickness of 5 mm. All the samples were initially obtained in the machined condition after sectioning from a 25 mm diameter rod. Samples were prepared with four different surface finish: (a) Polished samples (Ra = 0.05); (b) rough polished samples (Ra = 0.075); (c) machined samples (Ra = 0.47); (d) samples after grinding (Ra = 1.02) using different mechanical surface treatments. They are described as follows: (a) Polished samples were prepared by metallographic standard procedure with a last polishing step using 1-μ diamond paste. (b) The rough polished samples were ground with 240-grade emery paper. (c) Machining was done using the lathe machine. (d) Grinding was carried out on a grinding machine using SiC grinding wheel rotating at 540 rpm. Before plasma nitriding, the prepared samples were rinsed with acetone and dried. All the samples were plasma nitrided in two batches using two different gas compositions.
20% Nitrogen /
80% Nitrogen /
Fig. 2. Surface hardness values of all the four mechanically prepared plasma nitrided AISI 304 samples treated with different gas compositions.
cleaning process was performed for 1 h at 250 °C to remove the native oxide layer and contamination so as to expose a fresh surface of the samples for plasma nitriding. The temperature of the samples was monitored with a Fe-Constantan-Fe thermocouple (J-type), which is attached to the sample holder through the tubular support. There is an error of ± 10 °C in the temperature readings. After completing the sputter cleaning process, the mixture of nitrogen and hydrogen gas was introduced in the reactor for
2.2. Plasma nitriding treatment Plasma nitriding was carried out in a 500 mm diameter and 500 mm height bell shaped stainless steel vacuum chamber. Mechanically prepared four different surface finish samples were grouped and mounted on the sample holder of the reactor. Initially, the vacuum chamber was evacuated to a base pressure of 0.09 mbar by a rotary pump. The samples were first sputter cleaned using Ar–H2 gas mixture in 80 : 20 ratios. The gas flow rate was controlled by the mass flow controller to maintain a pressure of 1 mbar. Plasma was generated using a D.C. pulsed power supply having a repetition rate of 10 kHz. Sputter 1.6 1.4
Ra in micro ns
20% Nitrogen / 80% Hydrogen 80% Nitrogen / 20% Hydrogen
0.8 0.6 0.4 0.2 0
Rough Machined polished
Fig. 1. Surface roughness of all the four mechanically prepared untreated and plasma nitrided AISI 304 samples with different gas compositions.
Fig. 3. Optical microscope observation of plasma nitrided samples. (a) Mirror polished sample plasma nitrided at 80% N2 / 20% H2; (b) mirror polished sample plasma nitrided at 20% N2 / 80% H2; (c) ground sample plasma nitrided at 80% N2 / 20% H2; (d) ground sample plasma nitrided at 20% N2 / 80% H2.
G.P. Singh et al. / Surface & Coatings Technology 200 (2006) 5807–5811 120
Case depths in microns
20% Nitrogen / 80% Hydrogen
60 80%Nitrogen / 20% Hydrogen
G ro un d
in ed M ac h
Ro ug h
M irr or
Po lis h
Fig. 4. Case depth values of all the four mechanically prepared plasma nitrided AISI 304 samples treated with different gas compositions.
Microhardness measurements were performed on nitrided surfaces with a Leitz Vickers Hardness tester using a load of 100 g and a dwell time of 20 s with the major axis of the indenter parallel to the free surface. The microhardness depth profile was taken by taking the hardness measurements from the edge towards the core of the cross-sectioned samples. Case depth of the modified layer was examined by Leitz make optical microscope at a magnification of 100×. The depth of modified layer was measured from the optical microscope by etching the samples with a 2% nital solution. X-ray diffraction (XRD) studies have been carried out using Bragg–Brentano geometry (Seifert XRD-3000 PTS Diffractometer). Cu anode X-ray tube was operated at 40 kV and 30 mA to get Cu Kα radiation (λ = 1.5418 Å). The diffraction patterns were obtained in the 2θ ranges of 35–90° with the step size of 0.1° and counting time of 3 s per step. 4. Results
plasma nitriding. Plasma nitriding was carried out using two different gas mixtures (20% and 80% Nitrogen with balance Hydrogen) under a pressure of 4 mbar at 560 °C for 24 h. The discharge voltage for 20% and 80% nitrogen is 620 and 540 V, respectively. After 24 h, these nitrided samples were cooled in vacuum chamber under the flow of the gas mixture till the temperature decreased to 180 °C. This is mostly done to eliminate formation of oxide layer on the surface of the samples. 3. Characterization techniques Initial (untreated samples) and final (after plasma nitriding) surface roughness of the samples has been measured by Hommelwork LV-100 made stylus profilometer tester that has a tracking length of 4.8 mm and least count of 0.01 μm. It measures average roughness (Ra) over the entire sample.
4.1. Surface roughness The roughness measurements indicated an increase in roughness after plasma nitriding irrespective of initial roughness as shown in Fig. 1. Plasma nitriding with 20% nitrogen and 80% hydrogen has a higher surface roughness value compared to that treated with 80% nitrogen and 20% hydrogen. There is a large increase in roughness of the plasma nitrided polished and rough polished samples compared to the untreated samples. 4.2. Surface hardness The variation of the micro hardness values of plasma nitrided AISI 304 stainless steel treated with 80% N2 / 20% H2
Fig. 5. XRD patterns (0—Untreated, 1—plasma nitrided at 20% N2 / 80% H2, 2—plasma nitrided at 80% N2 / 20% H2).
G.P. Singh et al. / Surface & Coatings Technology 200 (2006) 5807–5811
and 20% N2 / 80% H2 is shown in Fig. 2. Fig. 2 shows that the surface hardness increases marginally with the increase in nitrogen concentration irrespective of the initial roughness of the samples. The hardness increase in mirror polished sample is higher when nitrided with 80% nitrogen as compared to other samples. 4.3. Cross-section analysis by optical microscope The optical micrographs of cross-sectioned plasma nitrided AISI 304 stainless steel of mirror polished and ground samples treated with both the gas compositions is shown in Fig. 3. A case depth of 110 μm and 50 μm was observed in mirror polished and ground sample respectively when nitrided with 80% N2 / 20% H2. A case depth of 90 μm and 40 μm was observed in mirror polished and ground sample, respectively when treated with 20% N2 / 80% H2. The case depths of the modified layers of all the different mechanically prepared plasma nitrided samples with different gas compositions are shown in Fig. 4. Figs. 3 and 4 show that the case depth decreases with increase in initial surface roughness as well as increase in hydrogen concentration in plasma nitriding. 4.4. Phase analysis by XRD diffractometer XRD patterns of the plasma nitrided samples were plotted for all the four mechanically prepared samples treated with different gas compositions as shown in Fig. 5. The untreated austenitic stainless steel sample after subjecting to different mechanical surface treatments revealed the presence of γ-Fe (austenite) and ∝-Fe (ferrite) phase except for the ground samples which showed only γ phase. After plasma nitriding, XRD patterns show ɛ-Fe3N, γ′-Fe4N, CrN, γ and ∝ phases for all the four samples. Mirror polished samples show high intensity peaks of Fe3N and Fe4N when treated with 80% nitrogen as compared to the other samples. High intensity ∝ phase peaks are observed in ground samples after plasma nitriding. 5. Discussion The initial surface roughness had a great effect on the surface roughness, hardness, case depth and structural phase compositions of AISI 304 stainless steel after plasma nitriding. The increased surface roughness in higher hydrogen concentration plasma can be attributed to the high chemical sputtering rate of hydrogen plasma. This is also known to be a consequence of the discharge voltage increment, which occurs when the nitrogen percentage in the processing atmosphere is reduced . The increase in surface roughness also depends on the processing gas compositions in addition to the treatment time and temperature [3,15] where the latter two factors are practically constant in this experiment. The XRD patterns of all the untreated samples suggest that α (ferrite) phase is formed on the top surface probably due to stress induced γ (austenite) to α transition during machining and polishing . However, this stress induced modified layer
seems to be taken off by the grinding process, which results in pure γ phase on the ground untreated sample. Strikingly after plasma nitriding α phase peak is the most intense in the ground sample irrespective of the gas composition. On the other hand the relative intensity of Fe3N, Fe4N, CrN peaks with respect to γ phase peak is the least in ground sample indicating that nitride formation is very less, and is maximum on mirror polished sample. These observations along with known AISI 304 properties lead to a possible understanding that pure γ phase resists nitride formation whereas presence of α phase on the top surface facilitates nitride formation. Moreover, at the process temperature of 560 °C austenite phase may prefer to decompose slowly into CrN and ferrite . These XRD results are consistent with the hardness values on these samples as higher content of nitrides on the mirror polished sample treated with higher nitrogen concentration gives highest hardness value among all the samples. However, the variation in the case depth on the four samples seems to be very interesting, as the case depth on the mirror polished sample is almost double than that on ground sample at the given temperature, treatment time and gas composition. This clearly indicates that the surface roughness and/or its consequent intrinsic effects on material properties heavily alter the diffusion mechanism. Since the modification of material properties by the means of generating the various surface roughnesses may not be taking place beyond a few microns depth, the diffusion coefficient of the bulk material should not change. This suggests that only top few micron layer should have a vast difference in its diffusion coefficient and nitriding properties to cause such a big difference in case depths of mirror polished and ground samples. The apparent difference in the top layers of mirror polished and ground sample, from XRD data, is that α phase is not present on untreated ground sample. Thus, it seems that pure austenite not only resists iron nitride formation but also the nitrogen diffusion. This may be due to higher packing fraction of it as compared to ferrite. There is a slight increase in the case depth due to the increase in the nitrogen concentration at the given initial surface roughness, temperature and process time. This may be due to the increase in more nitrogen incorporation and its subsequent effect on nitrogen diffusion. 6. Conclusion The surface roughness before plasma nitriding significantly affects the properties of the nitriding layer. Mirror polished samples exhibited high surface roughness increment, hardness and case depth after plasma nitriding compared to the rough polished, machined and ground samples. This is due to the presence of ∝ phase on the surface of the untreated samples. Its presence facilitated diffusion of nitrogen into the sample during plasma nitriding thereby increasing the case depth. Hence, the results indicate that transformation of austenite phase to ferrite and CrN phase may significantly affect the diffusion and nitriding mechanism. Higher percentage of nitrogen in the processing gas resulted in higher surface hardness and case depths. Since the surface roughness increases with increase in hydrogen gas, it is advisable to use
G.P. Singh et al. / Surface & Coatings Technology 200 (2006) 5807–5811
a gas having higher concentration of nitrogen. These results indicate that stainless steel samples having surface roughness of 0.05 μm (Ra), i.e. mirror polished, when plasma nitrided at 560 °C with 80% N2 and 20% H2 gives the highest hardness and the highest case depth. However, the corrosion resistance of the layer will be poor at this temperature because of CrN precipitation. Further work at lower temperature range and their influence with surface roughness for corrosion resistance will be taken. References  Z.L. Zhang, T. Bell, Surf. Eng. 1 (2) (1985) 131.  D.L. Williamson, O. Ozturk, R. Wei, P.J. Wilbur, Surf. Coat. Technol. 65 (1994) 15.  G.A. Collins, R. Hutchings, K.T. Short, J. Tendys, X. Li, M. Samandi, Surf. Coat. Technol. 74/75 (1995) 417.  E. Menthe, K.T. Rie, Surf. Coat. Technol. 116–119 (1999) 199.  Liang Wang, Xiaolei Xu, Zhiwei Yu, Zukun Hei, Surf. Coat. Technol. 124 (2000) 93.
 H. Michel, T. Czerwiec, M. Gontois, D. Alblitzer, A. Ricar, Surf. Coat. Technol. 72 (1995) 103.  Clodomiro Alves Jr., Josede Anchieta Rodrigues, Antonio Eduardo Martindli, Mat. Sci. Eng. A279 (2000) 10.  N. Renevier, P. Collignon, H. Michel, T. Czerwiec, Surf. Coat. Technol. 111 (1999) 128.  S.R. Bradbury, D.B. Lewis, M. Sarwar, Surf. Coat. Technol. 85 (1996) 215.  D.B. Lewis, S.R. Bradbury, M. Sarwar, Surf. Coat. Technol. 82 (1996) 187.  Y.U. Geller, Tool Steels, MIR Publishers, Moscow, 1978, p. 556.  G. Hoyle, High Speed Steels, Butterworth and Co. Ltd., London, 1988, p. 107.  A.daS. Rocha, T. Strohaecker, T. Hirsch, Surf. Coat. Technol. 165 (2003) 176.  C. Ruset, S. Ciuca, E. Grigore, Surf. Coat. Technol. 174–175 (2003) 1201.  B. Larisch, U. Brusky, H.J. Spies, Surf. Coat. Technol. 116–119 (1999) 205.  C. Blawert, A. Weisheit, B.L. Mordike, F.M. Knoop, Surf. Coat. Technol. 85 (1996) 15.  X.L. Xu, L. Wang, Z.W. Yu, Z.K. Hei, Surf. Coat. Technol. 132 (2000) 270.