Electrochemical studies of stainless steel implanted with nitrogen and oxygen by plasma immersion ion implantation

Electrochemical studies of stainless steel implanted with nitrogen and oxygen by plasma immersion ion implantation

Surface & Coatings Technology 201 (2007) 7873 – 7879 www.elsevier.com/locate/surfcoat Electrochemical studies of stainless steel implanted with nitro...

641KB Sizes 6 Downloads 231 Views

Surface & Coatings Technology 201 (2007) 7873 – 7879 www.elsevier.com/locate/surfcoat

Electrochemical studies of stainless steel implanted with nitrogen and oxygen by plasma immersion ion implantation C. Anandan ⁎, V.K. William Grips, V. Ezhil Selvi, K.S. Rajam Surface Engineering Division, National Aerospace Laboratories, Post Bag no.1779, Bangalore, 560 017, India Received 9 January 2007; accepted in revised form 15 March 2007 Available online 30 March 2007

Abstract Plasma Immersion Ion Implantation (PIII) of stainless steel with nitrogen at temperatures lower than 400 °C has been reported to increase the hardness of the material by several times. However, expectations that the corrosion resistance will remain unaffected after implantation were not found to be so. In the present study the influence of post-oxygen implantation on the corrosion resistance of nitrogen implanted stainless steel is presented. Stainless steel samples were subjected to oxygen, nitrogen and post-oxygen ion implantation at different temperatures. GIXRD and microRaman studies of the implanted samples showed that oxygen implantation leads to the formation of an oxide layer consisting of corundum and spinel structures. The corrosion properties of the implanted samples were studied by potentiodynamic polarization and electrochemical impedance techniques in 3.5% NaCl solution. After nitrogen implantation the corrosion current increased and the corrosion potential shifted to the less noble side to −0.486 V as compared to −0.284 V for the substrate. Oxygen implantation at 400 °C shifted the corrosion potential to the nobler side to − 0.2 V with decrease of corrosion current. For post-oxygen ion implantation at temperatures lower than 400 °C, the corrosion current was higher than the substrate and the corrosion potential was also on the less noble side. However, post-oxygen ion implantation at 400 °C after nitrogen ion implantation resulted in improved corrosion resistance as the corrosion potential shifted to nobler side and the corrosion current was lower than that of substrate. © 2007 Elsevier B.V. All rights reserved. Keywords: Plasma immersion ion implantation; Post-oxygen implantation; Corrosion study

1. Introduction The corrosion resistance of stainless steel is attributed to the presence of alloying elements, especially chromium. It has been generally found that chromium content in excess of 12 wt.% provides good corrosion resistance. This corrosion resistance of chromium rich steels is due to the formation of a coherent chromium oxide layer on the surface. The role of this naturally formed oxide layer in corrosion resistance has been studied in a number of investigations in actual working environments and also in laboratory environments using electrochemical techniques such as potentiodynamic polarization studies and electrochemical impedance spectroscopic methods [1]. It has also been suggested that a coherent spinel type chromium oxide can give good corrosion resistance even at higher temperatures [2]. ⁎ Corresponding author. Tel.: +91 80 2508 6247; fax: +91 80 2521 0113. E-mail address: [email protected] (C. Anandan). 0257-8972/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2007.03.034

Implantation of nitrogen into austenitic stainless steel, either by conventional implantation or by plasma immersion ion implantation has been shown to produce a hard, wear resistant surface on the softer substrate. At implantation temperatures close to or below 400 °C, nitrogen enters into solid solution form in the austenitic structure without formation of nitrides of chromium and iron. The resulting structure, known as expanded austenite, possesses the high hardness and is expected to have good corrosion resistance also since chromium is not lost as nitrides [3]. However, no general agreement has been found on this issue [4–9]. While some of the reports claim improved or no change in corrosion resistance, others present evidence for decreased corrosion resistance. Recently, Mändl et al. have found no clear evidence for the influence of dose or depth of implantation on the corrosion resistance after studying corrosion behaviour of “expanded austenite” prepared under wide implantation conditions [10]. Although post oxidation treatment of plasma nitrided steels is known to facilitate in regaining some

7874

C. Anandan et al. / Surface & Coatings Technology 201 (2007) 7873–7879

of the lost corrosion resistance due to nitride formation, only a few reports are available on this issue in the case of ion implanted samples [10]. Previous reports have mainly dealt with structure and composition of the oxygen implanted layers [11,12]. In the present work, the effects of post-oxygen implantation on the corrosion behavior of nitrogen implanted stainless steel as studied in potentiodynamic polarization and electrochemical impedance set up are presented. 2. Experimental details Stainless steel S20500 (AISI 205 grade) sheets of 1.2 mm thickness were used in the form of 25 mm × 12 mm coupons. The composition of the material was estimated by SEM-EDAX in a Leo model 440i scanning electron microscope with EDAX (Oxford Instruments) facility and is as follows (wt.%), Cr— 16.24, Ni—1.14, Mn—8.91, Cu—1.53, Si—0.46 and balance Fe. The sheets were polished by 0.3 μm alumina after a series of grinding operations that ended with 1200 grit silicon carbide paper. They were cleaned ultrasonically in absolute alcohol and dried before loading in to the vacuum chamber. The chamber was pumped to b3 × 10− 4 Pa pressure by a diffusion pump. Process gas, namely nitrogen or oxygen as the case may be, flow was controlled by mass flow controllers (MKS) and the pressure was monitored by Baratron pressure transducer (MKS). Implantation conditions for nitrogen and oxygen are given in Table 1. Nitrogen implantation was carried out at 0.13 Pa pressure and oxygen at 0.17 Pa. Plasma was generated using inductive coupling of RF(13.56 MHz) power at 120 W. The sample surface was sputter cleaned prior to actual implantation. The temperature of the sample was measured using a thermocouple attached to substrate holder and the implantation time was counted once the sample temperature reached the desired value. In the case of post-oxygen implantation, the gas composition was changed to 100% oxygen within a few minutes time while the sample was cooled to the desired temperature. After implantation the samples were cooled to room temperature in vacuum. The implanted samples were characterized by GIXRD in a Rigaku D/max 2200 Powder Diffractometer using CuKα radiation at a glancing angle of 2° and by microRaman spectrometer (LABRAM Model) using 623.4 nm laser. After corrosion study, the sample surface was examined by an optical microscope at different magnifications. Electrochemical measurements were performed using a conventional three-electrode cell, in which the test sample was placed in a Teflon sample holder. A platinum strip of 1-cm2 area served as the counter Table 1 Implantation conditions Condition

A B C D

electrode and Ag/AgCl, 3 M KCl electrode as the reference electrode. The corrosion measurements were carried out in 3.5% NaCl aqueous solution at room temperature, non-stirred and free-air conditions. Potentiodynamic polarization and EIS measurements were performed using Autolab PGSTAT30 galvanostat/potentiostat system. In order to establish the opencircuit potential, prior to the measurements, the sample was immersed in the solution for about 60 min. Impedance measurements were conducted using a frequency response analyzer (FRA). The spectrum was recorded in the frequency range of 10 mHz–100 kHz with data density of 5 points per decade. The applied alternating potential had a rootmean-square amplitude of 10 mV on the EOCP. After each experiment the impedance data was displayed as Nyquist and Bode plots. The Nyquist plot is a plot of real (Z′) vs. imaginary impedance (Z″). At higher frequencies, interception with the real axis is ascribed to the electrolyte bulk resistance (Rs) and at low frequencies an interphase appears whose interception with the real axis is ascribed to the charge-transfer resistance (Rct). The Bode plot is a plot of |Z|vs. log f, and log f vs. phase angle (θ), where |Z| is the absolute impedance and f is the frequency. The advantage of Bode plot is that the data of all the measured frequencies are displayed along with the impedance data. The experimental impedance data can be interpreted on the basis of equivalent electrical circuits using EQUIVCRT program [13]. After getting the stable open-circuit potential, the lower and upper potential limits of liner sweep voltammetry were set at ± 20 mV with respect to EOCP, respectively. The sweep rate was 1 mV/s. For the potentiodynamic polarization studies the upper and the lower potential limits of linear sweep voltammetry were set at ± 200 mV with respect to EOCP, respectively. The sweep rate was 1 mV/s. The Tafel plot was obtained after the electrochemical measurement. The corrosion potential (Ecorr.) and the corrosion current (Icorr.) were deduced from the Tafel plot (that is, log i vs. E plot). The corrosion current is obtained using the Stern–Geary equation [14]   ð1Þ Icorr ¼ ½ba bc =f2:303ðba þ bc Þg 1=Rp ; where ba and bc are Tafel slopes or Tafel constants, expressed in V/decade (or V/dec) and Rp is polarization resistance expressed in Ω/cm2. The polarization resistance is calculated using the following equation: Rp ¼ ð DE=DiÞjDEY0;

ð2Þ

where ΔE is the polarization potential and Δi is the polarization current. If the polarization resistance increases the corrosion current decreases. 3. Results

N2

O2

N2 Dose (×1017) ions/cm2

O2 Dose (× 1017) ions/cm2

Bias(KV)

400 °C 400 °C – 380 °C

– 300 °C 400 °C 400 °C

14.4 14.4 – 4.6

– 1.02 11.2 11.2

20 20 20 20

Fig. 1 shows the GIXRD from a sample implanted with nitrogen at 380 °C followed by oxygen at 400 °C in the 2θ range of 25°–65° at an incident angle of 2°. In this figure two broad peaks at 2θ values of 33.3° and 35° can be seen. The peak at 35° is from the spinel oxide consisting of Mn, Fe and Cr and the other at 33.3° is due to the presence of α-Fe2O3, respectively

C. Anandan et al. / Surface & Coatings Technology 201 (2007) 7873–7879

7875

Table 2 Results of linear polarization studies Condition

OCP (V)

Rp (kΩ/cm2)

SS A B C D

−0.261 −0.496 −0.385 −0.216 −0.184

29.0 3.7 7.3 38.4 47.4

[15]. Other peaks due to the substrate, denoted as SS, are also seen in the figure shifted towards lower angles. The substrate peaks are shifted to lower values due to the formation of expanded austenite [3]. Raman spectra from the implanted sample are given in Fig. 2 before and after corrosion experiments. As can be seen in the figure, in spectrum (a), for the nitrogen implanted sample no Raman peaks are found. For the nitrogen implanted sample that had undergone corrosion experiment, spectrum (b), a dominant peak at 650 cm− 1 is observable with a broad feature in 520 cm− 1 range. These are due to the formation Fe3O4 and hydroxides, mainly α-FeOOH during the corrosion process [16]. Raman spectra from the nitrogen followed by post-oxygen implanted and oxygen implanted samples before and after electrochemical studies are also shown in Fig. 2. For these samples a prominent peak at 661 cm− 1 and one at 294 cm− 1 can be seen in the figure in addition to broad features in-between, centered at 405 cm− 1 and 500 cm− 1. After corrosion experiments, the Raman spectra remain essentially the same with the 660 cm− 1 peak dominating the spectra. However, in the 350–590 cm− 1 spectral range there is a noticeable change. Although, no new features can be found, the features now appear well and slightly shifted.

Table 2 lists the OCP and the polarization resistance (Rp) of the substrate (SS) without and after N2 and O2 implantations. The OCP value gives an idea about the reactive nature of the surface. As can be seen in the table, the OCP values for the nitrogen (400 °C) and N2 (400 °C) + O2 (300 °C) implanted samples are more negative compared to the substrate and the Rp values also lower. This shows deterioration of the corrosion resistance after the above mentioned treatments of the samples. However, after oxygen implantation at 400 °C with or without prior nitrogen implantation, the OCP values have shifted towards the nobler side as compared to the substrate. This fact together with the higher Rp values of the 400 °C oxygen implanted samples shows the better corrosion resistance of these samples. Tafel plots and the kinetic parameters of the underlying corrosion processes estimated from the plots are given in Fig. 3 and Table 3, respectively. For the unimplanted substrate material corrosion potential, Ecorr. is − 0.264 V and the corrosion current, Icorr. is 0.3 μA/cm2. For SS samples implanted either with nitrogen at 400 °C or with nitrogen followed by oxygen at 300 °C, the Ecorr. value are more negative as compared to the unimplanted sample i.e. they are at the less noble side. However, when oxygen was implanted at a higher temperature i.e. at 400 °C, without or after nitrogen implantation, the Ecorr. values shift towards the positive side with respect to the unimplanted sample and so they are on the nobler side. The resistance against corrosion is generally proportional to the measured current density, which reflects the dissolution rate through the passive layer. Therefore, an increase in Icorr. means

Fig. 2. Raman spectra from implanted samples: (A) before and (B) after corrosion experiment for 400 °C nitrogen implanted case; (C) before and (E) after corrosion experiment for 400 °C oxygen implanted case; (D) before and (F) after corrosion experiment for 380 °C nitrogen and 400 °C post-oxygen implanted.

Fig. 3. Tafel plots for: (A) 400 °C nitrogen implanted sample (B) 400 °C nitrogen and 300 °C post-oxygen implanted sample (C) 400 °C oxygen implanted sample (D) 380 °C oxygen and 400 °C post-oxygen implanted sample and (SS) substrate.

Fig. 1. Glancing angle XRD of a sample implanted with nitrogen at 380 °C and post-oxygen implanted at 400 °C. (C—Corundum, S—Spinel, SS—Substrate.)

7876

C. Anandan et al. / Surface & Coatings Technology 201 (2007) 7873–7879

Table 3 Results of potentiodynamic polarization study Condition

Ecorr. (V)

Icorr. (μA/cm2)

bc (V/dec)

ba (V/dec)

Rp (kΩ/cm2)

SS A B C D

−0.264 −0.486 −0.395 −0.220 −0.202

0.3 3.6 2.8 0.5 0.2

0.083 0.193 0.161 0.087 0.093

0.084 0.06 0.07 0.111 0.095

68.2 5.68 11.9 38.57 92.08

a degradation of the protective properties of the passive film. However, in the case of post-oxygen implantation at 400 °C, Icorr. has decreased to 0.2 μA/cm2 and the corrosion potential has shifted towards the nobler side. This shows the improvement in the corrosion resistance after this treatment. Other parameters to be taken into account are the anodic (ba) and cathodic (bc) slopes that inform about the anodic and cathodic reactions of the corrosion processes. As can be seen in Table 3, compared to unimplanted case the nitrogen implanted samples show a reduction in the ba value indicating a higher inclination towards anodic processes of dissolution. Oxygen implanted

Fig. 4. a: Nyquist plots for: (A) 400 °C nitrogen implanted sample (B) 400 °C nitrogen and 300 °C post-oxygen implanted sample (C) 400 °C oxygen implanted sample (D) 380 °C nitrogen and 400 °C post-oxygen implanted sample and (SS) substrate. b: Bode plots for (A) 400 °C nitrogen implanted sample (B) 400 °C nitrogen and 300 °C post-oxygen implanted sample, (C) 400 °C oxygen implanted sample, (D) 380 °C oxygen and 400 °C post-oxygen implanted sample and (SS) substrate.

samples have higher ba values indicating resistance to anodic dissolution of the samples. On the other hand, from the table it appears that the cathodic slope bc not to have much influence on the corrosion kinetics. For example, though the 400 °C nitrogen implanted and 300 °C post-oxygen implanted samples have higher bc values their corrosion resistance was poor. Therefore, it can be concluded that oxygen implantation and also the temperature of implantation play a major role in improving the corrosion protection by increasing the impediment to the anodic reaction. Whereas potentiodynamic polarization studies can give information about the susceptibility of the surface to corrosion, a comparison of this result with EIS experiment can give more information on the corrosion mechanism. Nyquist and Bode plots obtained from EIS measurements are given in Fig. 4a and b, respectively. Inset in Fig. 4a shows the data in the high frequency region. It shows that the substrate, 400 °C oxygen implanted and 400 °C post-oxygen implanted samples behave electrochemically in a similar way in this frequency region. The difference lies in the shifting of the curves for the implanted samples along the real axis that is due to the series resistance comprising eletrolyte and contact resistances. In the low frequency side, the Nyquist plots for all samples end up in a semicircular form, with much different radii. The inset also shows that the nitrogen implanted sample and the 300 °C post-oxygen implanted sample exhibit a semicircle in the high frequency region in addition to that in the low frequency region. This indicates the presence of two different electrochemical processes operating in two different time scales. In the high frequency region the process takes place at the electrolyte/sample surface and at lower frequencies the species are able to

Fig. 5. Equivalent circuits for: (A) substrate, 400 °C oxygen implanted sample, 380 °C nitrogen and 400 °C post-oxygen implanted sample and (B) 400 °C nitrogen implanted sample and 400 °C nitrogen and 300 °C post-oxygen implanted sample.

C. Anandan et al. / Surface & Coatings Technology 201 (2007) 7873–7879 Table 4 EIS data obtained by equivalent circuit simulation of different implanted layers studied in the present work Condition RS Qlayer n (Ω/cm2) (μF/cm2) SS A B C D

7.68 7.5 6.0 9.37 10.2

– 13.56 20.9 – –

– 0.75 0.77 – –

Rlayer (Qdl – Y0) (Ω/cm2) (μF/cm2)

ndl

– 18.0 22.1 – –

0.88 45.0 0.78 4.2 0.85 5.4 0.84 9.4 0.84 13.1

6.59 184.3 194.3 19.30 25.4

Rct (kΩ/cm2)

interact with the bulk below the implanted layer as they have longer time to interact since reversal of field takes place over a longer period at low frequencies. This shows that the layer modified by 300 °C post-oxygen implantation is unable to prevent the interactions between electrolyte and the bulk below the nitrogen implanted layer. The equivalent circuits and the fitting parameters derived from these data are shown in Fig. 5A and B and Table 4, respectively. Fig. 5A shows the commonly proposed equivalent circuit model (Randle's circuit) for a simple corrosion system, which is entirely charge-transfer controlled. The double layer capacitance provides information about the polarity and the amount of charge at the substrate/electrolyte interface. The capacitance (Cdl) is replaced with constant phase element (CPE) Q in the EQUIVCRT program for a better quality fit [13]. CPE accounts

7877

for the deviation from ideal dielectric behavior and is related to the surface inhomogeneities. This element is written in admittance form as: Y ðxÞ ¼ Y0 ð jxÞn ;

ð3Þ

where Y0 is an adjustable parameter used in the non-linear leastsquares fitting and n is also an adjustable parameter that lies between 0.5 and 1. The value of n is obtained from the slope of log|Z| vs. log f plot (Fig. 4b). The phase angle (θ) can vary between 90° for a perfect capacitor (n = 1) and 0° for a perfect resistor (n = 0). The circuit description code (CDC) for the equivalent circuit proposed for the simple Randle's circuit (Fig. 5A) is R(QR), where R is the resistive element. The substrate and both the 400 °C oxygen and 400 °C post-oxygen implanted samples show this Randle's circuit behavior. This implies that both native oxide and the oxide formed by oxygen ion implantation at 400 °C behave in a similar way electrochemically. The charge-transfer resistance of the oxygen implanted sample is found to be lower compared to the substrate. This may be due to the differences in the native oxide formed under ambient conditions on the substrate and the one by ion implantation. If the modified layer present on the substrate surface, such as a coating or the implanted layer in the present case, provides a direct conduction path for the corrosive media then the

Fig. 6. Optical micrographs of the corroded area from (A) 400 °C nitrogen implanted, (B) 400 °C nitrogen and 300 °C post-oxygen implanted sample, (C) 400 °C oxygen implanted sample, (D) 380 °C nitrogen and 400 °C post-oxygen implanted sample.

7878

C. Anandan et al. / Surface & Coatings Technology 201 (2007) 7873–7879

electrochemical interface can be divided into two sub-interfaces: electrolyte/modified layer and electrolyte/unimplanted substrate. The proposed equivalent circuit for such a system is shown in Fig. 5B. The parameters in the equivalent circuit Rlayer and Qlayer are related to the resistance and constant phase element (CPE) of the implanted layer present on the outer surface. Rct and Qdl are related to the charge-transfer reaction at the electrolyte/unimplanted substrate interface. The CDC of the proposed equivalent circuit for this case is R(Q[R(QR)]). Fig. 4a and b show the impedance response of 400 °C nitrogen implanted and 400 °C nitrogen with 300 °C post-oxygen implanted layer. The obtained data for these implanted samples show that the implanted layer has affected the corrosion characteristics of the substrate. The Rct values are less showing that the charge transfer is facilitated after the implantation. Table 4 shows the EIS data obtained by equivalent circuit simulation. The Rct values obtained from the study gives the true picture of the corrosion resistance. The native/naturally occurring oxide layer on the SS has the highest Rct value of 45 kΩ/cm2 and any ion implantation has reduced the resistance to charge transfer and lowered the corrosion resistance. Nitrogen implantation has affected the corrosion resistance much more and the EIS results confirm the results obtained from the potentiodynamic polarization studies. Optical micrographs of the corroded areas of different samples are shown in Fig. 6A to D. It can be observed from these micrographs that the size of the corrosion spots on the nitrogen implanted and low temperature (300 °C) post-oxygen implanted samples are much larger than that in either the 400 °C oxygen implanted or 400 °C post-oxygen implanted samples. The nitrogen implanted samples showed extensive surface corrosion in addition to larger number of corrosion pits. Raman spectra taken on these corrosion spots displayed features that are due to Fe3O4 and hydroxide as shown in Fig. 2. Further, the 400 °C oxygen implanted and 400 °C post-oxygen implanted samples presented a surface that appeared smoother and to have undergone uniform corrosion with fewer isolated corrosion pits. 4. Discussion In general, our results on the corrosion resistance of samples implanted with nitrogen under different conditions showed degradation irrespective of whether chromium nitride was present or an expanded austenite was formed. For example, in Table 3, the corrosion current Icorr. has increased to 3.6 μA/cm2 from 0.3 μA/cm2 and the corrosion potential Ecorr. has shifted to less noble side, towards − 0.486 V. The polarization resistance has also decreased from 68.2 kΩ/cm2 to 5.68 kΩ/cm2. Even though there are conflicting reports on the effect of expanded austenite on the corrosion resistance of nitrogen implanted stainless steels, it is generally believed that defects and strain produced due to lattice expansion in the formation of the expanded austenite can cause a lower corrosion resistance [11]. If chromium nitride forms after nitrogen implantation, obviously corrosion resistance will decrease as chromium is not available to form the protective native oxide.

Results of corrosion experiments on samples post-implanted with oxygen ions at 300 °C and 400 °C show the influence of implantation temperature on the ability of the oxide formed by oxygen ion implantation in improving the corrosion properties. As can be seen in Table 3, after post-oxygen implantation at 300 °C, the corrosion potential has shifted towards the nobler side with respect to that of the nitrogen implanted sample and the corrosion current is comparable to that of the nitrogen implanted sample. At 400 °C, oxygen implantation results in marked improvement in all the corrosion parameters. The corrosion potential has shifted to nobler side with respect to the substrate and the corrosion current is also lower. The anodic and cathodic slopes are also closer to the substrate values. The polarization resistance has also increased substantially. Thus, post-oxygen ion implantation after nitrogen implantation has a beneficial effect on the corrosion resistance if carried out at appropriate temperature. The reason for this is discussed below. From oxidation of Cr containing stainless steels in the temperature range 250 to 900 °C, it was found that at lower oxidation temperatures, the oxide layer at the surface was enriched with iron oxide, Fe2O3 and Cr oxide, Cr2O3 was present below the surface. At temperatures N 700 °C the oxide again became Cr2O3 rich due to diffusion of chromium to the surface [17–19]. However, if heated to 400 °C in vacuum Cr enrichment was found at the surface[18]. Further, it was observed that the resistance of stainless steel against pitting corrosion improved after oxidation in air at an oxidation temperature N350 °C. It was believed that at these temperatures the oxide formed was compact [1]. From Raman studies of oxide scales formed at different temperatures on stainless steels of varying Cr content it was found that at lower temperatures these oxides were mainly spinel oxide and of α-Fe2O3 with the dominant Raman peak in the 660–680 cm − 1 [20–23]. Raman features of α-Cr2O3 appeared at higher temperatures, N800 °C, only if the Cr content was N12%. In the present study, for oxygen implanted samples a prominent Raman peak is observed at ≈ 660 cm− 1 along with Raman features of α-Fe2O3. The GIXRD in Fig. 1 presents evidence for the presence of both forms of oxides in the oxygen implanted layers. Optical micrographs presented in Fig. 6A–D show fewer and smaller corrosion pits in samples that are implanted with oxygen at 400 °C. Based on these evidences, it can be said that in the present case also post-oxygen and oxygen implantation lead to the formation of compact oxide composed of corundum and spinel form of oxides consisting Mn, Fe and Cr. Further, as the oxide in the present study is formed by ion implantation under vacuum conditions this may help in forming a Cr rich compact oxide [10,18]. 5. Conclusions Stainless steel S20500 (AISI grade 205) samples were implanted with oxygen, nitrogen and nitrogen followed by oxygen by plasma immersion ion implantation and were characterized by GIXRD and microRaman spectroscopy. Corrosion properties of the implanted samples were studied using potentiodynamic polarization and electrochemical impedance methods. The impedance spectra of the implanted samples

C. Anandan et al. / Surface & Coatings Technology 201 (2007) 7873–7879

were analyzed in terms of various equivalent circuit models. GIXRD and micro Raman studies show that oxygen implantation leads to the formation of corundum and spinel oxides with Mn, Cr and Fe. Corrosion studies show that after nitrogen implantation the corrosion potential shifts to the less nobler side and corrosion current increases as compared to the substrate. Post-oxygen implantation at temperatures lower than 400 °C does not result in improving the corrosion resistance of nitrogen implanted samples as the corrosion potential still remains at less noble side and the corrosion current is higher. Post-oxygen implantation at 400 °C improves the corrosion resistance by shifting the corrosion potential to the nobler side as compared to the substrate and lowering the corrosion current. The impedance data of the substrate, 400 °C oxygen and 400 °C post-oxygen implanted samples show a single time constant model whereas the nitrogen and lower temperature post-oxygen implanted samples follow a double time constant model. Acknowledgements The authors would like to thank the Director, National Aerospace Laboratories, Bangalore for his support and permission to publish the work. The authors would like to thank Dr. Anjana Jain, Materials Science Division, NAL for the XRD measurements. References [1] A. John Sedriks, Corrosion Resistance of Steels, John Wiley and Sons, New York, 1979. [2] M.J. Carmezim, M.O. Figueredo, in: Y. Pauleau, P.B. Barna (Eds.), Protective Coatings and Thin Films, Kluwer Academic Publishers, The Netherlands, 1997, p. 431.

7879

[3] D.L. Williamson, O. Ozturk, R. Wei, P.J. Wilbur, Surf. Coat. Technol. 65 (1994) 15. [4] M.E. Chabica, D.L. Williamson, R. Wei, P.J. Wilbur, Surf. Coat. Technol. 51 (1992) 24. [5] M. Samandi, B.A. Shedden, D.I. Smith, G.A. Collins, R. Hutchings, J. Tendys, Surf. Coat. Technol. 59 (1993) 261. [6] P. Preston, R.A. Smith, J. Buchanan, Roth Reece, S.G. Kamath, J. Vac. Sci. Technol., B 12 (2) (1994) 940. [7] C. Blawert, H. Kalvelage, B.L. Mordike, G.A. Collins, K.T. Short, Y. Jiraskova, O. Schneeweiss, Surf. Coat. Technol. 163 (2001) 181. [8] S. Mukherjee, P.M. Raole, A. Kumar, I. Chattoraj, K.R.M. Rao, I. Manna, Surf. Coat. Technol. 186 (2004) 282. [9] S. Mändl, R. Gunzel, E. Richter, W. Moller, Surf. Coat. Technol. 100/101 (1998) 372. [10] S. Mändl, D. Manova, H. Neumann, M.T. Pham, E. Richter, B. Rauschenbach, Surf. Coat. Technol. 200 (2005) 104. [11] R. Wei, P.J. Wilbur, O. Ozturk, D.L. Williamson, Surf. Coat. Technol. 51 (1992) 133. [12] A. Lacoste, S. Bechu, Y. Pelletier, R. Gouttebaron, Surf. Coat. Technol. 156 (2002) 225. [13] M. Stern, A.L. Geary, J. Electrochem. Soc. 104 (1957) 56. [14] B.A. Boukamp, J. Electrochem. Soc. 142 (1995) 1885. [15] P.A. Tempest, R.K. Wild, Oxid. Met. 17 (1982) 345. [16] D. Thierry, D. Persson, C. Leygraf, D. Delichere, S. Joiret, C. Pallotta, A. Hugot-Le-Goff, J. Electrochem. Soc. 132 (1984) 305. [17] J.C. Hamilton, R.E. Benner, B.E. Mills, J. Vac. Sci. Technol. 20 (1982) 946. [18] G. Betz, G.K. Wehner, L. Toth, A. Joshi, J. Appl. Phys. 45 (1974) 5312. [19] K. Segerdahl, J.E. Svensson, L.G. Johansson, J. Electrochem. Soc. 151 (2004) B394. [20] H. Astemann, J.E. Svensson, L.G. Johansson, J. Electrochem. Soc. 151 (2004) B141. [21] R.L. Farrow, R.E. Benner, A.S. Nagelberg, P.L. Mattern, Thin Solid Films 73 (1980) 353. [22] P. Fabis, R. Heidersbach, C. Brow, T. Rockett, Corrosion 37 (1981) 700. [23] S.C. Tjong, Mater. Res. Bull. 18 (1983) 157.