Some effects of the plasma nitriding process on layer properties

Some effects of the plasma nitriding process on layer properties

17th~ SolM Films. 217(1992) 38 47 38 Some effects of the plasma nitriding process on layer properties M. B. K a r a m l ~ University of Erciyes, Eng...

1MB Sizes 0 Downloads 9 Views

17th~ SolM Films. 217(1992) 38 47


Some effects of the plasma nitriding process on layer properties M. B. K a r a m l ~ University of Erciyes, Engineering Facult.v. Department of Mechanical Engineering, 38090 Kayseri (Turk
Abstract The layer properties of AISI 722M24, AMS 6472 and AISI H13 materials plasma nitrided under various conditions were defined. The materials were plasma nitrided at temperatures between 510 and 590 ~'C for times ranging between 4 and 100 h in a dissociated ammonia atmosphere. The effects of duration, temperature, chemical composition and the shape of sample on the layer properties were determined. It was determined that the total case increased with treatment time and temperature increases the white layer thickness and the effective case depth. The increase in the white layer thickness is very sharp up to a total case depth of 0.3 mm, but it is very slow between 0.3 and 0.75 mm total case. The chemical composition of the material also affects the depth and hardness of the case. High alloy steels have higher hardness but thinner case than low alloy steels. Although the y'-Fe4 N single-phase compound layer was formed on 722M24 steel, a thick and mixed e-7' (Fe2 3N + Fe4N) phase was produced on H13 steel. The hardness distributions from curved and flat surfaces to the core of the test material exhibit different profiles.

1. Introduction The plasma nitriding process has been extensively used to improve the surface properties of a wide variety of materials on a large industrial scale in recent years. A very large proportion of the screws and cylinders of plastic-processing machines, long extrusion cylinders [1-3], crankshafts, gears and pinion of m o t o r cars [2 5], cold and hot working dies [3, 6], tool steels for metal cutting [7], titanium alloys [8] etc. are plasma nitrided to improve wear, fatigue and corrosion resistance with m a n y advantages. It is estimated that over 1000 plasma nitriding units are now in industrial operation worldwide [9], In the case of plasma thermochemical treatment, the energy is introduced directly into the components by b o m b a r d m e n t with ions and fast neutrals in an abnormal d.c. glow discharge. The component to be nitrided is made the cathode and the chamber wall is the anode of the glow discharge. After evacuation of the chamber, a d.c. voltage is applied between anode and cathode. Generally a m m o n i a or an atmosphere of N2 and H2 at a pressure of 1 10 m b a r is introduced. This process provides a very clean and hard layer on the surface without any risk of distortion. When the plasma nitriding process is carried out for up to 100h, a layer thickness of about 1 m m can be obtained with a thin c o m p o u n d layer [ 10]. Although the process increases the wear resistance of the C r - M o steels [11], the white layer phase and thickness affect the wear behaviour of the layer, leading to


spalling of the compound layer [6]. The hardness distribution, white layer thickness and case depth of nitrided surfaces are affected by the variations in process temperature and duration. The lower temperatures, i.e. about 500 "C, produce a harder case while the longer treatment times produce a deeper case. However, the w o r k p i e c e - a n o d e distance plays a significant role in the temperature uniformity [12]. Therefore, somewhat different hardness values can be observed on the same component surface depending on surface shape. In this respect, the thickness variations of the diffusion layer also depend on the shape of the component [13]. The aim of this work is to illustrate some layer properties of a gearing steel (AISI 722M24), AMS 6472 and a hot working die steel (H13).

2. Experimental work The samples to be treated were machined to a diameter of 25 m m with a 50 m m length from AISI 722M24, AMS 6472 and AISI H I 3 materials, the chemical compositions of which are given in Table 1. Holes were drilled with a minimum of 30 m m depth along the longitudinal axis of all samples to provide a temperature measuring point. This depth of hole ensures temperature deviations of less than 1.0% [14]. Treatment was carried out with a 20 kW plasma nitriding unit in a cracked ammonia atmosphere with a pressure of 2.5 mbar. The current density during the nitriding was about 2.5 mA cm -2 and the d.c. voltage of the discharge

~,c, 1992

Elsevier Sequoia. All rights reserved


M. B. Karamt~ / Effect of plasma nitriding on layer properties T A B L E 1. Chemical composition of the test materials

Composition (%) Steel











AISI 722M24 (24 Cr Mo 13) A M S 6472 (38 Cr A 1 M o 6) H 13 (X 40 Cr Mo V 51)































was about 700 V. The samples were degreased in a chemical solution before loading into the treatment chamber. Treatment times up to 100 h were employed with temperatures in the range 510-590 °C. Some of the samples made from AISI 722M24 steel were double treated (i.e. 50 h + 50 h) with a different temperature employed during the second part of the cycle. The samples were heated to the required temperature in about 1 h. After the treatment, the samples were sectioned, mounted and polished for metallographic examination. Case depths were determined from microhardness measurements employing a 300 gf load. The case depth at a hardness value 10% above the core hardness was taken as an accurate measure of the total case depth. The effective case depth is defined as the depth to a hardness value of 500 HV 0.3. The X-ray diffraction (XRD) method was used by employing Cr K s radiation for the determination of the phase structure of the compound layer and wear debris collected during the wear test, details of which were published elsewhere [6, 10, 11]. Microscopic studies were performed to examine microstructures and worn surfaces. Surface profiles of treated samples and worn surfaces were also determined.

12 •

3. Results and discussion

3.1. Properties of the compound layer The compound layer thickness produced on the 722M24 steel surface is increased by treatment time at both of the treatment temperatures (i.e. 550 and 570 °C). Treatment time also increases the total case depth of the diffusion zone [4]. Figure 1 shows the variation in the white (compound) layer thickness with total case depth obtained by the treatment performed at 550 and 570 °C for treatment durations from 4 to 100 h. It can be seen that the white layer thickness increases rapidly with case depth up to 0.3 mm and then goes on slowly. It is well known that thicker compound layers tend to spall by breaking down during service. Therefore, such a thicker


11 10

| °"i 0.5












Total case depth, men

Fig. 1. The variations in white layer thickness and effective case depth with total case depth on 722M24 steel.

layer produced by conventional nitriding processes (i.e. gas and salt bath) needs be ground off. Although the thickness of the white layer produced by the plasma nitriding process increases with case depth, its maximum value is 10.5 gm for this steel treated for 100 h. It is clear that this compound layer was composed of the ?'-Fe4N phase (Fig. 2(a)). This type of layer is the most ductile among all the types of layers and is porous. Thus, it is able to absorb a quantity of oil in service and allows a reduction to be achieved in the friction coefficient [10]. Apart from the thickness the y' compound layer has no detrimental influence on the fatigue limit [141. Although the compound layer produced on AISI 722M24 material is y'-Fe4N, in a thin single phase, in contrast to this, the compound layer of AISI H13 is a thick ~-7'(Fe2 3 N -}- Fe4N) mixed phase (Fig. 2(b)). In a particular atmosphere, the material with the highest carbon content contains greater amounts of e and is more likely to have Fe3C present in the compound layer. This indicates quite clearly that the carbon content of the base material exerts a significant influence on

M. B. Karamt~" / Effect qf plasma nitriding on layer properties


1208. L885, 918. p,,.



748, 625. 518, 395.


288, 1G5. 58.

. . . . . . . . .



. . . . . . . . .




. . . . . .



. . . . . . . . .

9 2 . O0

I ~'''

1 1 6 . O0

. . . . . . .



Angle 20, degrees


488. I








i i iiii ,-0


16i : '


J t




.e~e.' ~"~-


~8 . ~


1BS. 88


l f i S . ~10

Angle 20, degrees

Fig. 2. Phase structures of compound layers on the material surfaces obtained by XRD(treatment conditions: 550 'C, 16 h): (a) AISI 722M24; (b) AISI Ht3. the type o f layer formed [2]. W h e n c o m p a r i n g the chemical compositions o f these two steels, it appears that the extra c a r b o n and alloying elements present in H13 p r o m o t e 8 formation (Table 1). It has still to be determined whether it is the c a r b o n tied up in the material or the c a r b o n sputtered from the surface into

the atmosphere that has most influence on layer formation [2]. This mixed c o m p o u n d layer is extremely brittle and tends to spall. The phase structure o f the comp o u n d layer produced on H13 steel treated at 5 5 0 ' C also includes the Fe203 phase in addition to the above mixed phase.

M. B. Karaml~ / Effect of plasma nitriding on layer properties

The compound layer thickness produced on H13 steel treated at 510 °C for 25 h is 6.3 lam. The treatment at 530 °C produces compound layer thicknesses of respectively 5.9 pm and 14.1 ixm on H13 steel if the treatment duration is increased from 4 h to 100 h. For the same durations, however, these thicknesses obtained at 550 °C are 5.5 lam and 16.8 ~tm respectively. Thus, it is understood that the effects of treatment temperature on compound layer thickness become conspicuous after longer durations for the H13 material. In contrast to this, the treatments carried out on the 722M24 material show that the higher treatment temperature (i.e. 570 °C) produces a thinner compound layer than the 550 °C treatment does for the same treatment duration (i.e. 10.5 lam at 550 °C, 9.93 lam at 570 °C, for 100 h). Therefore, it can be said that there is no systematic relation between treatment temperature and compound layer thickness. The thickness of the y' compound layer obtained on AMS 6472 steel is 6.7 Ixm under treatment conditions at 530 °C for 36 h. Under the same treatment conditions (550 °C, 4, 16 and 100 h), although 722M24 steel has compound layer thicknesses of 6.7, 7.8 and 10.5 Ixm, the H13 material has 5.5, 11.5 and 16.8 Ixm. Thus it is evident that the e + 7' mixed phase is usually thicker than the y' singlephase compound layer. The thickness of the compound layer affects the wear behaviour of the treated material. Thicker compound layers increase the wear rate owing to breaking off of material. After wear tests carried out on two materials, it was found by X R D that the wear debris of 722M24 and H13 materials consist of iron oxide (Fe203, FeO and Fe304) and ~-Fe2 3N, ~-Fe203 respectively. It is clear that thicker Y'+ ~ phase compound layer will be broken under service conditions. Figure 3 shows the surface profiles of the test material after wear tests. It can be seen that the surfaces of the H13 material are very rough after wear tests. There are some cavities generated by spalling of the white layer. The surface topography of the worn surface of the H13 steel can also be seen in Fig. 4. As illustrated in the figure, the compound layer has fallen off in some areas, and there is also some adhesive shearing of wear plates on the surface. Figure 5 shows the wear scars obtained from the worn surface of the 722M24 material. The pictures were taken from etched worn surfaces. Therefore, the worn compound layer appears white in colour. There is no spalling because of the relatively ductile 7' compound layer produced on 722M24 steel by plasma nitriding. Some oxidation areas can also be seen in the friction zone.

3.2. Properties of the hardened case A typical microstructure of plasma-nitrided 722M24 material treated at 550 °C for 100 h is shown in Fig.


-20 rl0'



(b) Fig. 3. The surface profiles recorded from worn surfaces: (a) 722M24 treated at 570°C for 64h; (b) H13 treated at 550°C for 100h.

6(a). A very uniform compound layer can be seen. The diffusion zone contains a fine dispersion of nitride precipitates within the tempered matrix. The transition zone from nitrided case to core is illustrated in Figs. 6(b) and 6(c). Carbon which derives from the destabilization of the (FeCr)TC 3 phase during nitriding and


M. B. Karamt~" / EfJect of plasma nitriding on layer properties

Fig. 4. Scanning electron micrograph of a worn surface of H13 steel treated at 550 "C for 4 h.

diffuses a h e a d o f the n i t r i d i n g front is likely to exert an influence [4]. H a r d n e s s d i s t r i b u t i o n s o f the test m a t e r i a l s are s h o w n in Fig. 7. A I S I 722M24 steels t r e a t e d at 550 a n d 570 °C for 16 h have a p p r o x i m a t e l y the same surface hardnesses, 850 H V 0.3. H o w e v e r , the t o t a l case o f the s a m p l e t r e a t e d at 570 °C i s d e e p e r t h a n that o f the s a m p l e t r e a t e d at 550 c~C (Fig. 8(a)). A l t h o u g h the H13 steel was n i t r i d e d u n d e r the same t r e a t m e n t c o n d i t i o n s (i.e. 550 °C, 16 h), it exhibits a higher surface h a r d n e s s a n d t h i n n e r case t h a n the 722M24 material. It can be seen f r o m Fig. 7(a) t h a t A M S 6472 steel has a lower h a r d n e s s t h a n H13 t r e a t e d at the s a m e t e m p e r a t u r e (i.e. 530 °C). Because o f the l o n g e r t r e a t m e n t time, it has a d e e p e r case t h a n H13. O n c o m p a r i s o n o f Figs. 7(a) a n d 7(b), the effect o f t r e a t m e n t time on the surface h a r d ness o f the test m a t e r i a l s can be seen clearly. H13 steels t r e a t e d at the s a m e t e m p e r a t u r e , 530 °C, have different surface h a r d n e s s e s f r o m each other. The steel subjected to a long t r e a t m e n t has a lower surface h a r d n e s s (1200 H V 0.3) t h a n the o t h e r which was t r e a t e d for a s h o r t d u r a t i o n (1300 H V 0.3). T h e h a r d n e s s differences between 722M24 steels t r e a t e d at the s a m e t e m p e r a t u r e b u t v a r i o u s d u r a t i o n s are like those o f H13. It is interesting t h a t A M S 6472 steel t r e a t e d at 530 °C for 36 h has a d e e p e r case with a t h i n n e r c o m p o u n d layer t h a n H I 3 steel t r e a t e d at the same t e m p e r a t u r e b u t for 100 h ( F i g . 8(b)). T h e case d e p t h o f the A M S 6472 m a t e r i a l is also deeper t h a n t h a t o f H I 3 steel n i t r i d e d at 550 °C for 100 h. T h e effect o f t e m p e r a t u r e on case d e p t h can be seen f r o m Fig. 8. L o w e r t r e a t m e n t t e m p e r a t u r e s p r o d u c e higher h a r d n e s s a n d t h i n n e r case d e p t h on the test steels. H i g h e r nitriding t e m p e r a t u r e s p r o d u c e c o a r s e r nitride precipitates, leading to lower h a r d n e s s within the case [2, 4]. This



(c) Fig. 5. Optical micrographs of worn surfaces of 722M24 steel treated at 550 C , 36 h after etching: (a) a view of the friction zone with a white layer near the edge; (b) high magnification picture of the worn white layer from the friction zone; (c) wear scars on the white layer after wear tests carried out on the sample treated at 570 C for 9 h.

M. B. Karamt¢ / Effect of plasma nitriding on layer properties


for 16h.


M. B. Karamt~" / Effect o/plasma nitrMing on layer propertk's


,4® -{ 1300 t

I~ P~..~

[] A

530oC,16h(H13) 550oC,,6h(H13)

,2®] 1,00t



5,00C,25h(H, 3)

~'~\ 1"~


e: ,000 ~







'~ ~






500 4o0 300


o'.i" "o5"

o%" "o'.5' o%


Distance from surface, mm




[ A

11001 ~

722M24 [ •

,000 -1 N :

I 9001



~m,~"~ •

g 7°°t -

H13 [" I.


5500C, 100h 570oC, 100h

(sT0+ss0)oc,(50+5o)13 5300C, 100 13 550oC,




300- ~ 200

. . . . . i . . . . . | . . . . . l . . . . . ! . . . . . ! . . . . . n. . . . . ! . . . . . u. . . . . I . . . . . i . . . . . 0.1 0,2 0,3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1,l Distance


surface, mm

(b) Fig. 7. Typical hardness distributions of the test materials: (a) samples treated for a short time; (b) samples treated for a long time.

reduction in surface hardness level can be offset by employing a double nitriding cycle. The first cycle is carried out at a certain temperature (i.e. 570 °C) followed by the remainder at a lower temperature, of say 550°C (compare the treatments 570°C, 100h, and 570 °C, 50 h + 550 °C, 50 h) (Fig. 7(b)). The effective case depths defined as the depth at which a hardness value of 500 HV 0.3 is found are increased by the depth of the diffusion zone. As illustrated in Fig. 8(a), the effective case grows with total case depth, which grows with increasing treatment duration. On the contrary, for treatment at higher temper-

atures (i.e. 570 °C) for a long time, the effective case depth decreases after 50 h of treatment. T.he reason for this apparent reduction in the effective case depth is associated with the production of chromium-rich carbides ahead of the nitriding front [2, 4]. The hardness and depth of the diffusion layer are also related to the amount of alloying element in the metal. It can be stated as a general rule that, under the same nitriding conditions, the hardness of the diffusion layer will increase as the alloy content rises but its thickness will fall. Since H13 steel includes about 5 % C r , 1.94'7o Mo and 0.87%V, which are strong nitride-forming elements, it possesses a mubh longer capability to pick up nitrogen than low alloyed metals. That is why nitrogen builds up on or near the surface as nitride and it does not diffuse very deeply under the surface. Thus the white layer thickness and surface hardness increase but the case depth decreases. If a deeper case is desired, it is necessary to supply more nitrogen during nitriding [ 15]. Although Edenhofer [16] claimed that it was hardly possible to obtain a diffusion layer on high alloy steels with a depth above 0.3ram, a total case depth of 0.5 m m is obtained on H I 3 steel by means of plasma nitriding at 550 ~:C for 100 h. The aluminium-containing material, AMS 6472, has better nitridability, and develops higher hardness and a deeper case than 722M24 steel. Although its hardness is lower than that of the H I 3 material, the depth of the diffusion zone of the steel treated for 36 h is nearly equal to those of H13 and 722M24 steels treated at 530 C for 100 h and 550 C for 100 h respectively. This is evidence for the better nitridability of aluminiumcontaining steels, which results because aluminium attracts the nitrogen strongly. However, the case produced is generally considered too brittle for gearing applications. It is surprising that, as illustrated in Fig. 9, the hardness distributions obtained from axial (A) and radial (B) directions in sections parallel to the axis of the test steels treated at 550 '~C for 4 and 100 h exhibit a deviation according to the measurement direction. It is interesting that the hardness value in the A direction of the sample treated for 4 h is nearly the same as that of the B direction in the nearest areas to the treated surface. Beyond a depth of 0.1 ram, the profile at B follows below the profile at A to a depth of about 0.25 mm and then rises to the same value as thc other. In contrast to this, for the treatment tbr 100h, the hardness values in the B direction are lower than those in the A direction up to a 0.1 m m depth. The hardnesses obtained in the two directions have about the same values at depths from 0.1 to 0.25 mm. After this point, the B direction has a higher hardness and deeper case than the A direction according to the core hardness. The same comments can also be made lbr the

M. B. Kararnt¢ / Effect of plasma nitriding on layer properties


50 + 50 50 + 50 50 + 50 100 ,c¢,~

100 64

, ,r|,~ff


, o,w

49 36 36





16 16 •

9 •


White layer thickness


Effective case

• + [~ + •




4 0.1












100 100 49 49 .2


36 25

16 •

White layer thickness

• + • I

0.0 (b)






Total case depth










Depth b e l o w surface, m m

Fig. 8. (a) White layer thickness and effective and total case depths of 722M24 material treated under different conditions. (b) White layer thicknesses and total case depths of AISI H13 and A M S 6472 materials treated under different conditions.

hardness deviation of the 722M24 material (Fig. 9). The reason for these observations may be a shape effect of the sample, because nitrogen may diffuse much more in the B direction than in the A direction during long treatment durations. Park et al. [ 13] had investigated the effect of geometry on the growth of the nitride layer in ion nitriding

and determined that the thickness of the diffusion layer on grooves (more distant from the anode) decreased and the growth of the layer on the land (closer to the anode) increased with corrugation depth. This was because the current density in a given area depends on several factors, one of which is the radius of curvature of components. The distribution of the components in

M. B. Karaml~" / Effect of plasma nitriding on layer propertie,s


1400 1300 1200 ' 1100 -"


Flat surface, H13 ( A direction )

Round surface,H13

Round surface,722M24

Flat sufface,722M24

section exhibit a higher gradient than the gradient for the round surface after a depth of 0.1 mm (Fig. 9).

( B direction ) ( B direction ) ( A direction )

4. Conclusions

10002 90o -" 80o 700" 600" 500" 400" 300' 200

. . . .

~ 0.1


. . . .

u 0.2

. . . .

n 0.3

. . . . 0.4

Distance from surface, m m

(a) 1300 •

Flat surface,HI3

( A direction)


Fiat surface,722M24

1200 1100 1000

( A direction)

9o0 .



5oo 400 3O0 200

. . . .



. . . .


. . . .




. . . .



. . . .

[ . . . .


i . . . .




. . . .



. . . .





Distance from surface, r a m


Fig. 9. Hardness depth profiles in the section parallel to the axis of the test materials treated at 550 'C for (a) 4 h and (b) 100 h (flat surface, A direction, hardness distribution in axial direction; round surface, B direction, hardness distribution in radial direction).

the treatment chamber can also affect the temperature uniformity. Also when the c a t h o d e - a n o d e distance increases, the surface temperature decreases at high gas pressures ( 5 - 1 0 T o r r ) [12]. It is clear that the flat surface of the treated sample has lower temperature owing to the distance to the anode. This may cause a higher hardness near the surface and a shallower case than that of the other face of the sample for a long treatment time. In the short treatment cycle, 4 h, the temperature deviation from the flat surface to the core may be higher and thus the hardnesses in the cross-

The results obtained from this study may be summarized as follows. (1) The thickness of the white layer and effective case on the 722M24 material increase with increasing nitrided case depth• Although the white layer thickness decreases with treatments at higher temperatures and longer durations, the effective case depth exhibits little increase• The reason for these effects is associated with the production of chromium-rich carbides ahead of the nitriding front. On the contrary, the white layer thickness on high alloy steel (i.e. H13) increases with increasing treatment time and temperature, because the steel possesses a much longer capability to pick up nitrogen than low alloyed steels. Therefore, nitrogen builds up on the surface or near the surface as a nitride and does not diffuse very deeply under the surface. Thus the white layer thickness and surface hardness increase but the case depth decreases. It should be stated that the aluminium content of the steel has a marked effect on the case depth. Consequently, it can be said that the white layer thickness, case hardness and depth also depend strongly on the chemical composition of the steel and steels with much more strongly nitride-forming elements develop high hardnesses. (2) With the increasing treatment temperature, the hardness of the layer decreases and a small increase is also observed in the total case. This is because coarser nitride precipitates which are essentially non-hardening are produced by elevated-temperature nitriding. (3) Die steels nitrided by the plasma process may produce a mixed-phase white layer. This thick white layer is extremely brittle and it increases the wear rate of the material. Extra carbon and alloying elements present in the steel cause ~ phase formation in long treatment durations. (4) The shape of the surface to be treated has a minor effect on the hardness distribution in the layer• The surfaces that have longer distances from the anode have higher hardnesses and shallower cases than other surfaces of the component because they have a lower temperature.


The author wishes to thank Professor T. Bell and Dr. A. M. Staines for their invaluable help with and comments on the work.

M. B. Karamt~ / Effect of plasma nitriding on layer properties

References 1 T. Bell, Plasma heat treatment of tooling for plastics industry, Plastics Rubber Process., 1 (4) (1976) 161 - 166. 2 A. M. Staines, Trend in plasma-assisted surface engineering process, Heat Treat. Met., 4 (1990) 85-92. 3 B. Edenhofer, The ionitriding process--thermochemical treatment of steel and cast iron materials, Metall. Mater. Technol., 8 (8) (1976) 421-426. 4 M. B. Karaml~ and A. M. Staines, An evaluation of the response of 722M24 steel to high temperature plasma nitriding treatments, Heat Treat. Met., 3 (1989) 79--82. 5 K. Keller, Dimensional changes in gear wheels due to the nitriding, Antriebstechnik, 10 (3) (1971) 1-7 (in German). 6 M. B. Karaml~, An investigation of the properties and wear behaviour of plasma-nitrided hot working steel (HI3), Wear, 150 (1991) 331-342. 7 J. I. Onate, J. K. Dennis and S. Hamilton, Nitrogen implantation of tool steels, Heat Treat. Met., 3 (1987) 77-82. 8 A. S. Korhonen, J. M. Molarius and M. S. Sulonen, Plasma processing in nitrogen atmospheres: 10 years development, Surf. Eng., 4 (1) (1988) 44-50.


9 T. Bell, Surface engineering: past, present and future, Surf. Eng., 6 (1) (1990) 31-40. 10 M. B. Karam~, Friction and wear behaviour of plasma-nitrided layers on 3% C r - M o steel, Thin Solid Films, 203 (1991) 49-60. 11 M. B. Karaml~, Tribological behaviour of plasma-nitrided 722M24 material under dry sliding conditions, Wear, 147 (1991) 385-399. 12 C. Ruset, The influence of pressure on temperature uniformity in the plasma nitriding process, Heat Treat. Met., 3 (1991) 81-84. 13 M. J. Park, W. S. Back, S. C. Kwon and M. C. Yoo, Effect of geometry on growth of nitride layer in ion nitriding, Proc. Conf. on Ion Nitriding and Ion Carburizing, Cincinnati, OH, September 1989, ASM International, Metals Park, OH, 1990, pp. 203-209. 14 T. Bell and N. L. Loh, The fatigue characteristics of plasma-nitrided 3% C r - M o steel, J. Heat Treat., 2 (3) (1982) 232-237. 15 B. Edenhofer, Applications and advantages of nitriding treatments outside the normal range of temperatures. Part 2: treatments at high temperatures (above 580 °C), Hiirterei-Tech. Mitt., 30 (4) (1975) 204-208 (in German). 16 B. Edenhofer, Comparison of different nitriding processes with special attention given to the ionitriding process, Fachber. Oberfl~ichentech., 12 (4) (1974) 97-102 (in German).