Corrosion passivation of plasma nitrided iron

Corrosion passivation of plasma nitrided iron

Surface Science 189/190 (1987) 221-225 North-Holland, Amsterdam 221 CORROSION PASSIVATION OF PLASMA NITRIDED IRON M. JURCIK-RAJMAN and S. VEPREK Ins...

190KB Sizes 3 Downloads 46 Views

Surface Science 189/190 (1987) 221-225 North-Holland, Amsterdam

221

CORROSION PASSIVATION OF PLASMA NITRIDED IRON M. JURCIK-RAJMAN and S. VEPREK Institute of Inorganic Chemistry, University of Ziirich, Winterthurerstrasse 190, CH-8057 Ziirich, Switzerland Received 31 March 1987; accepted for publication 7 May 1987

Corrosion passivation of iron nitrided in a glow discharge has been studied by means of potentiodynamic and galvanostatic methods in NaC1 water solution. The results show that the corrosion resistance of iron nitride is due to a formation of an anodic passivation layer.

1. Introduction Plasma nitriding in glow discharge is industrially used to produce wear and corrosion resistant diffusion coatings on steel. Many studies were done on the formation of the coatings and on their mechanical and tribological properties [1-3], but relatively little is known regarding the chemical nature of the corrosion inhibition. It has been shown by other authors that the corrosion potential increases with increasing nitrogen content in the near-surface layer of steels, but it was not clear if this is due to passivation or to an increase of the normal potential [2,4-7]. In order to answer this question we have undertaken a systematic study of the nitriding process and of the corrosion behaviour using as starting material pure, polycrystalline a-iron. Because of limited space available we shall report in the present paper only the results of the electrochemical studies which reveal the nature of the corrosion inhibition being anodic passivation. These studies were complemented by gravimetric, X-ray diffraction, scanning electron microscopic and atmospheric corrosion investigations, results of which will be reported in a paper to follow.

2. Experimental Samples of 18 x 10 x 0.5 mm 3 in size were cut from a cold rolled iron sheet of a purity of 99.5 wt%. The samples were nitrided in an radio-frequency glow discharge in a mixture of nitrogen and hydrogen ( p ( H 2 ) : p ( N 2 ) = 3:1) and in pure nitrogen at a pressure of about 1.6 Torr and a temperature up to 0039-6028/87/$03.50 9 Elsevier Science Publishers B.V. (North-Holland Physics Publishing Division)

222

M. Jur?ik-Rajman, S. VepPek / Corrosion passivation of plasma nitrided iron

350 ~ C. The extent of nitriding was followed by gravimetry and X-ray diffraction. The surface morphology was investigated by means of optical and scanning electron microscopy. The corrosion behaviour was studied by means of potentiodynamics and by galvanostatic methods following the potential changes associated with the time-dependent corrosion in 2 and 0.05 molar aqueous solution of NaC1 (pH = 7, at RT, air free under Ar flow). All electric potential data are reported with respect to the saturated calomel electrode

(SEE). 3. Results The corrosion potential of pure, untreated iron with respect to SCE in 0.05M water solution amounts to - 850 _ 20 inV. The value of the corrosion potential of nitrided iron depends on the course of the measurement as illustrated by fig. 1. Usually one finds, in such a measurement, a value of the corrosion potential which is several hundred of mV higher than that of pure iron. This is due to formation of an anodic passivation layer. Only a strong cathodic polarization of about - 1 0 0 0 mV removes this layer. Starting the potentiodynamic scan with such a reduced surface one finds the value of corrosion potential equal to that of a pure iron (see Ucl, fig. 1) irrespective of the degree of nitriding as shown in fig. 1, curve a. The passivation occurs at a value of several hundred mV above the corrosion potential of pure iron (see 1 in fig. 1). With further increase of the applied potential the potentiodynamic curve passes through passive (see 2) and transpassive (3) regions. When the potential is scanned down, starting from the transpassive region, curve b is found which yield a corrosion potential Uc2 significantly higher than that of pure iron. It should be emphasized that curve b is reversible as long as the applied potential does not decrease more than about 200 mV below the respective value of the corrosion potential and as long as the nitrided layer remains compact, i.e. before the onset of pitting corrosion (see below). The value of the corrosion potential of anodically passivated nitride layers increases with increasing degree of nitriding. This is illustrated by fig. 2 which shows Evans' diagrams of pure iron (curve a) and of y'-Fe4N and c-Fe2_3 N phases (curves b and c, respectively). This also explains, at least partially, the well known increase of the corrosion resistance of steels with increasing degree of nitriding. Another factor explaining this dependence is found in the mechanism of the corrosion as illustrated by fig. 3. The initial potential U0(FexN) increases with increasing nitrogen content in the sample. The sudden decrease of the applied potential a t to.p. is due to the onset of pitting corrosion. After this onset, the applied potential, which is necessary to keep a constant value of the anodic current of 5 m A / c m 2, decreases only slowly during dissolution of a total amount of iron equivalent to a thickness of several hundred of/~m.

M. Jur~ik-Rajman, S. VepPek / Corrosion passivation of plasma nitrided iron

223

i (mAlcm 2)

nV,SEE)

Fig. 1. A typical potentiodynamic curve of a nitrided iron. The arrows indicate the direction of the potential scan. 1: Passivation; 2: passive region; 3: transpassive region. 2N NaC1 water solution, pH = 7, oxygen free; scan speed 2 m V / s .

With increasing nitrogen content in the iron sample the value of U0 increases and it reaches about 1300 mV for c-Fe2_3 N. The applied potential needed to keep the constant anodic current after the onset of pitting corrosion increases with increasing nitrogen content as well. However there seems to be no change of the onset of pitting (to.p. , s e e fig. 3) with increasing nitrogen content. The value of the time of the onset of pitting together with the value of the anodic current do not give any meaningfull measure of the thickness of the passivation layer: the apparent thickness of the removed layer before onset of the pitting (assuming a uniform dissolution which is not justified because of the pitting nature of the corrosion) amounts to several hundred nanometers whereas the thickness of the oxygen containing passivation layer, estimated from sputter profiling combined with XPS, is only a few nm. The presently available XPS data do not allow to draw more conclusion about the nature of the passivation layer.

224

M. Jur?ik-Rajman, S. Vep~ek / Corrosion passioation of plasma nitrided iron

i(mAlcm2) 10 Fe4N

Fe2_3N

1(3~

1C)3

1()4

"1000-800 "600 -400 -200 0 200 U(mV,-SEE]

Fig. 2. Evans diagrams of pure iron (a), "t'-Fe4N (b), and of (-Fe2_3N (c). 0.05N NaCI water solution, further conditions as in fig. 1.

It is remarkable to notice that, except for the a p p e a r a n c e of some holes through which the pitting corrosion occurs, the surface of the nitrided sample at areas where no holes had developed shows n o changes i n the m o r p h o l o g y even after such an extensive corrosion (SEM study). This illustrates the excellent resistance of iron nitrides against corrosion.

U{mV,SCE) I

UolFe4N) /+00 200! o

-200

2

Fe4N ,

V 1o-"------~_

~o

'

s'o tim~(mi~T

fo.p.

UoCF l Fig. 3. Galvanostatictime dependence of applied potential of pure iron (a), and of y'-Fe4N (b). 2M NaC1, i = 5 mA/cm -2, constant. Conditions as in fig. 1.

M. Jur~ik-Rajman, S. Vep~ek / Corrosion passivation of plasma nitrided iron

225

4. Conclusion W e have shown that the c o r r o s i o n resistance of n i t r i d e d iron is due to f o r m a t i o n of an anodic passivation layer. R e m o v a l of this layer b y cathodic polarization results in a loss of the corrosion protection. U n d e r anodic polarization, the nitride layer remains r e m a r k a b l y resistant a n d the corrosion of the u n d e r l a y i n g iron takes place due to pitting.

Acknowledgements This work has been s u p p o r t e d in p a r t b y the Swiss N a t i o n a l Research F o u n d a t i o n . W e should like to acknowledge valuable discussions with Serge Rambert.

References [1] K. Bergel and B. Leidel, H~terei Tech. Mitt. 35 (1980) 11. [2] R. Chaterjee-Fischer, H~irtereiTech. Mitt. 38 (1983) 35. [3] B. Edenhofer, H~terei Tech. Mitt. 29 (1974) 105; B. Edenhofer, Fachber. Oberfl~ichentech. 12 (1974) 97. [4] E.J. Knokoly, L. FiSldvhryand A. Nag),, Fachber. Oberfl~ichentech. 35 (1980) 75. [5] E.J. Mittemeier and P.F. Colijn, Fachber. Oberfl~ichentech.49 (1985) 77. [6] E. Angelini, B. De Benedetti and F. Zucchi, Met. Ital. 11 (1984) 499. [7] D.J. Coates, B. Mortimer and A. Hendry, Corrosion Sci. 22 (1982) 951,973.